Ecology of the subarctic regions
Écologie des régions subarctiques
Proceedings of the Helsinki symposium
Actes du colloque d’Helsinki
,
%
unesco
Ecology a n d conservation / Écologie et conservation
Titles in this series
/ Dans
cette collection
1
I.
Ecology of the subarctic regions. Proceedings of the Helsinki symposium &cologie des régions subarctiques. Actes du colloque d’Helsinki
II.
Methods of study in soil ecology. Proceedings of the Paris symposium Méthodes d’étude de l’écologie des sols. Acies du colloque de Paris
III.
Resources of the Biosphere. Proceeding of the Paris Conference (in preparation)
1
PubIished in 1970 by the United Nations Educational, Scientific and Cultural Organization Place de Fontenoy, 75 P a r i ~ - 7 ~ Printed by Imprimeries Réunies de Chambéry Publié en 1970 par l'organisation des Nations Unies pour l'éducation, la science et la culture place de Fontenoy, 75 Paris-7e Imprimeries Réunies de Chambéry
0 Unesco
1970 Printed in France
SC.SS/XXIV. 1/AF
Preface
The promotion of research in plant and animal ecology has been the objective of m a n y symposia organized or sponsored by Unesco and is one of the main fields of research towards which the Organization directs its attention in relation to its activities in natural resources research. While previous work has been concentrated on problems in the arid zone and in the humid tropics, the present volume is devoted to a completely different environment: the subarctic environment. The area of transition between the temperate zone and the Arctic tundra,comprising tracts of forest,fen and heathland in north-western Canada, Alaska, northern Scandinavia, northern Russia and Siberia, was the subject of a symposium organized by the Government of Finland and Unesco in co-operation with the International Geographical Union. It was held in Otaniemi (near Helsinki) from 25 July to 3 August 1966,and was attended by 70 scientists from 13 countries. The opening speeches were given by the Finnish Minister of Education,Mr.R.H.Oittinen and by Professor Paavo Kallio, of the University of Turku. The papers presented were divided into nine sections : subarctic definition; meteorology of subarctic regions; snow cover as an ecologicalfactor;weathering and geomorphological processes; permafrost as an ecological factor; main features of soil-forming processes; ecology of subarctic vegetation; ecology of important species of the subarctic fauna; conservation of nature and rational use of renewable natural resources of subarctic regions. It is hoped that the publication of these papers will lead to a better understanding of the physical conditions of the regions under study and will open up n e w perspectives for further research and practical applications. M a n y of the papers which were the result of highly specialized research were presented in a w a y
Préface
L a promotion de la recherche concernant l’écologie végétale et animale a fait l’objet d’un grand nombre de colloques organisés OU patronnés par l’Unesco, et c’est un des principaux domaines de la recherche dont s’occupe l’organisation dans le cadre de ses activités relatives aux ressources naturelles. Les travaux antérieurs ont porté essentiellement sur les problèmes de la zone aride et de la zone tropicale humide, mais le présent volume est consacré à un milieu entièrement différent: celui des régions subarctiques. L a région intermédiaire qui est située entre la zone tempérée et la toundra arctique et qui comprend des forêts, des marécages et des landes (nord-ouest du Canada, Alaska, Scandinavie septentrionale,nord de la Russie et Sibérie) a fait l’objet d’un colloque organisé par le gouvernement finlandais et l’Unesco, en coopération avec l’Union géographique internationale. Ce colloque s’est tenu à Otaniemi (près d’Helsinki) du 25 juillet au 3 août 1966 et a réuni soixante-dix savants de treize pays. Les discours d’ouverture ont été prononcés par le ministre finlandais de l’éducation, M. R.H.Oittinen, et par le professeur Paavo Kallio, de l’université de Turku. Les mémoires qui ont été présentés étaient répartis en neuf sections: définition de la zone subarctique; météorologie des régions subarctiques;le tapis neigeux en tant que facteur écologique; phénomènes météorologiques et géomorphologiques; le pergélisol en tant que facteur écologique; principaux aspects des phénomènes de formation du sol; écologie de la végétation subarctique; écologie des espèces importantes de la faune subarctique; conservation de la nature et utilisation rationnelle des ressources naturelles renouvelables des régions subarctiques. On espère que la publication de ces mémoires contribuera à faire mieux comprendre les conditions physiques des régions étudiées et ouvrira de nouvelles perspectives en ce qui concerne la recherche et les
such as to arouse interest a m o n g colleagues working
applications pratiques. Un grand n o m b r e de mémoires,
in entirely different fields.
qui étaient l’aboutissement de recherches hautement
A m o n g the major points arising from the discussions were the need for the establishment of protected control areas ; for international treaties protecting migratory birds, whales, seals, etc., a n d for a better understanding of cyclic fluctuations of subarctic m a m m a l s and birds, the prime problem of boreal terrestrial ecology. T h e meeting urged that research should be carried out o n productivity a n d sustained yield of subarctic fauna a n d flora and that meteorological observations b e extended and improved for microclimatological studies in the subarctic zones. Following the presentation of papers a n d the discussions a visit w a s arranged to the University of Turku, and a four-day field excursion to Lapland to visit the K e v o Research Station. An alternative field excursion w a s arranged for zoologists to the Oulanka Biological Station of the University of Oulu. In presenting this volume, Unesco wishes to offer its thanks to the participants in the s y m p o s i u m whose papers have m a d e it possible, and to Professor Kallio a n d the m e m b e r s of the Finnish organizing committee for their generous a n d efficient organization of the symposium. T h e responsibility for the selection and presentation of facts and for opinions expressed rests with the authors.
spécialisées, ont été présentés de manière à éveiller l’intérêt de collègues travaillant dans des domaines entièrement différents. Les discussions ont fait ressortir n o t a m m e n t la nécessité de créer des zones de contrôle protégées, de conclure des traités internationaux visant à protéger les oiseaux migrateurs, les baleines, les phoques, etc., et de mieux comprendre les fluctuations cycliques des mammifères et des oiseaux subarctiques, problème capital de l’écologie terrestre boréale. Les participants ont estimé qu’il faudrait, d’une part, effectuer des recherches sur la productivité et le rendement perm a n e n t de la faune et de lafloresubarctiqueset’d’autre part, multiplier et améliorer les observations météorologiques en vue de l’étude microclimatologique des zones subarctiques. Après la présentation des mémoires et la conclusion des débats, ont eu lieu une visite à l’université de Turku et u n e excursion de quatre jours en Laponie à destination de la station de recherche de Kevo. Les zoologistes, pour leur part, ont eu la possibilité de participer à une autre excursion jusqu’à la station biologique d’oulanka, qui dépend de l’université
d’Oulu. L’Unesco tient à exprimer sa reconnaissance a u x m e m b r e s d u colloque qui, grâce à leurs mémoires, lui ont permis de publier le présent volume, et à remercier le professeur Kallio et le comité d’organisation finlandais pour la générosité et l’efficacité avec lesquelles ils ont organisé le colloque. Les auteurs sont seuls responsahles du choix et de la présentation des faits, ainsi que des opinions exprimées.
Contents
J. Bliithgen
B. A. Tikhomirov
I. M. Dolgin
S. Huovila
H.Odin and K.Perttu
R. Sarvas
W.O. Pruitt Jr.
Table des matières
Problems of definition and geographical differentiation of the Subarctic with special regard to northern Europe . Problèmes de délfinition et de différenciationgéographique du S u barctique, spécialement en Europe septentrionale [Résumé] . Forest limits as the most important biogeographical boundary in the North . Importance des limites forestières en tant que frontière biogéographique dans le Nord [Résumé] . Subarctic meteorology . Météorologie subarctique [Résumé]
A. Corte
35 38 41
. de
63 65
Radiation measurements near the forest limit in northern S w e d e n . . Mesure du rayonnement aux limites de la forêt dans la Suède septentrionale [Résumé] . Temperature s u m as a restricting factor in the development of forest in the Subarctic . L a température globale en tant que facteur restrictif dans le développement des forêts subarctiques [Résumé] .
Quelques aspects écologiques
30
.4-59
.
S o m e features of the microclimate within hilly regions in Finland Quelques caractéristiques du microclimat dans les régions accidentées Finlande [Résumé] .
S o m e ecological aspects of s n o w
11
.
67
77
79 81 83
de la neige [Résumé]
97
't -
Bioecological aspects of the s n o w plant communities of C a p e Spring, Argentine Antarctica . .
101
Aspects bio-écologiques des communautés de plantes des neiges au cap Spring, Argentine antarctique [Résumé]
A. R a p p
.
S o m e geomorphological processes in cold climates Etude
.
de certains processus géomorphologiques dans les régions à climat
froid [Résumé]
.
.
104
.
105
.
113
J. Biidel
A. J a h n
Denudation and river erosion in the “zone of pronounced valley forma. . tion” o n South-east Spitsbergen Dénudation et érosion fluviale dans la zone de formation accusée de vallées au Spitzberg du Sud-Est [Résumé] . .
117
Soil m o v e m e n t s under the influence of freezing . mouvements du sol sous l’inyuence du gel [Résumé]
. .
119 122
Complexité des notions de faciès morphologique arctique et subarctique (nord-ouest et centre ouest du Groenland). Géographie boréale et anthro. pologie: fondements physiques des notions de lieu de territoire . Complexity of the terms LLArctic” and “Subarctic” as notions of morphological aspect (North-westand Middle-west Greenland) [ S u m m a r y ] .
125
Les
J. Malaurie
R. J. E. B r o w n
.
Permafrost as a n ecological factor in the Subarctic . écologique de la région subarctique [Résumé]
Le pergélisol,facteur
T.L. Péwé
M. Salmi
. .
.
Permafrost and vegetation o n flood-plainsof subarctic rivers (Alaska) : [a summary]. . Le pergélisol et la végétation dans les plaines inondables des cours d’eau . . subarctiques (Alaska) [Résumé]
115
128 129 138 141 141
Investigations o n palsas in Finnish Lapland . . Recherches sur les hydrolaccolithes (palses) de la Laponie Jinlandaise [Résumé] . .
151
143
E. Schenk
Permafrost and frost structures in the subarctic area Formation et structures du pergélisol [Résumé] .
. .
155 158
H.Svensson
Frozen-ground morphology of northeasternmost N o r w a y . . Morphologie des sols gelés dans l’extrême nord-est de la Norvège [RBsumé].
161
P. L. Johnson
R e m o t e sensing as a n ecological tool . L a détection à distance en écologie [Résumé]
J. C. F. T e d r o w
Soils of the subarctic regions Les sols des régions subarctiques [Résumé]
.
.
P7
. .
169 185
. .
189 199
F. di Castri R. Covarrubias E. Hajek
Soil ecosystems in subantarctic regions . Écosystèmes du sol dans les régions subantarctiques [Résumé]
. .
207 221
L. A. Viereck
Soil temperatures in river bottom stands in interior Alaska . . L a température du sol dans les peuplements forestiers du fond des vallées à l’intérieur de l’Alaska [Résumé] . .
223
. .
235 238
I. Hustich
On the study of the Sur l’étude
D. J. Bellamy W.M. Tickle
E. Einarsson
.
ecology of subarctic vegetation
de l’écologie de la végétation subarctique [Résumé]
.
232
A
critical limit of primary production for the survival of arctic alpine . . plants in the northern Pennines of England Un seuil critique de la production primaire pour la survie de plantes alpines arctiques dans les Pennines septentrionales (Angleterre) [Résumé].
241 246
Plant ecology and succession in s o m e nunataks in the Vatnajökull glacier
.
247
Phyto-écologie et évolution du tapis végétal dans certains nunataks du glacier de Vatnajökull, dans le sud-est de l’Islande [Résumé] . .
254
in South-east Iceland
.
.
A. N. F o r m o z o v
Ecologie des plus importantes espèces de la faune subarctique Ecology of the major species of subarctic fauna [ S u m m a r y ] .
A. G. Loughrey
T h e ecology and population dynamics of the barren-ground caribou in Canada Le caribou des toundras du Canada [Résumé] .
J. P. Kelsall
V. A. Peiponen
257 273 275 279
Animal activity patterns under subarctic s u m m e r conditions . Organisation de l’activité des animaux en fonction des caractéristiques de l’été subarctique [Résumé] .
281
Les îles Saint-Pierre et Miquelon, u n e enclave subarctique méridionale The St. Pierre and Miquelon Islands,an enclave of the southern Subarctic [Summary] .
289
. Forests a n d forestry in subarctic regions la sylviculture dans les régions subarctiques [Résumé]
.
295 300
R. Kalliola
. National parks a n d nature reserves in subarctic regions Parcs nationaux et réserves naturelles dans les régions subarctiques [Résumé]
303 307
I. G. S i m m o n s
Problems of the conservation of relict arctic a n d subarctic species in Britain Conservation des survivances botaniques arctiques et subarctiques en . Grande-Bretagne [Résumé]
317
Subarctic peatlands a n d their utilization Les tourbières subarctiques et leur utilisation [Résumé] .
319 325
E. Aubert
de la Riie
P. Mikola
Les forets et
R. Ruuhijärvi E. Ehlers
I
-. . *
-.
T h e rational use of forests and bogs, in view of comparative observations in north-western Canada and northern Finland . Utilisation rationnelle des forets et des marécages d’après des observations comparées effectuéesdans le nord-ouest du Canada et le nord de la Finlande [Résumé] .
286
292
309
327 333
E. Schenk
On the string formation in the aapa
moors a n d raised bogs of Finland . L’origine des tourbières réticulées (aapamoors) [Résumé] . .
335 340
G. Sirén
R e m a r k s o n current silvicultural research in the Subarctic of Finland . Remarques sur les recherches effectuéesen sylviculture dans la région subarctique de Finlande [Résumé] .
343 348
International scientific co-operation in conservation with special reference to peatlands .
349
/ Discussion générale et conclusions
351
T.Pritchard
General discussion and conclusions List of participants
/ Liste des participants
.
363
Problems of definition and geographical differentiation of the Subarctic with special regard to northern Europe Joachim Blüthgen
I should like to express m y appreciation for being invited to read the introductory paper which will set the scene and provide the frame for the w o r k of this s y m p o s i u m on the Subarctic. I a m aware that I can deal with only s o m e of the problems attached to this very complex and variably used term. T h e m o r e I worked o n the subject, the m o r e I doubted whether it is possible at present to give a satisfactory definition of this primarily geographical term which is also used in numerous neighbouring sciences though each gives it a different meaning. Although m y task is to try to define and subdivide the Subarctic I must confess that this-due to contributions which are still contradictory-is not quite objectively possible as yet. It reflects inevitably a great deal of personal opinion. I ask therefore that m y c o m m e n t s be regarded as a still incomplete contribution. It is b o u n d to be of a preliminary character since only m u c h m o r e painstaking investigations into the countless contributions to this subject could lead, in the near future, to a solution of this terminological problem. Besides, m o r e extensive contacts with competent specialists over the world are a primary necessity. T h e entanglement of terms is far m o r e complex than I originally assumed. Therefore I should especially like to acknowledge gratefully the proposal m a d e by Unesco to open the s y m p o s i u m with this topic. It really corresponds to a n urgently felt need to avoid as m u c h as possible all misunderstandings of w h a t is m e a n t by the term “Subarctic”. Thereby help is offered to all those specialists doing research on special problems which are in a n y w a y connected with the Subarctic, so that they might eventually be able to refer the results they have obtained to the right or at least a conventionally agreed u p o n frame. It will be the specific task of the geographer to form building-blocks from all the numerous individual results and interpretations, building-blocks to erect
the terminological structure for “the Subarctic”, to combine t h e m and to integrate t h e m into this structure according to their due weight. T h e geographical term “subarctic” is one of those geographical designations of areas which are often used for the characterization of transitional zones between core areas. Delimiting t h e m , however, is only possible in a complex fashion,i.e., by a combination of different distinctive features. T h e y are defined in different ways, if ever. W i t h subarctic w e can group subpolar, subtropical, suboceanic, subcontinental, subalpine. These “sub” areas form every one of t h e m a transitional zone at the periphery of those core areas which are formed by differentiation of the Earth according to latitude, surface, and height, and of whose n a m e s they are derived (arctic, tropics, etc.). B u t it is not possible to deduce clearly from their n a m e s whether they form parts of these areas or whether they are areas of their o w n outside them. These terms, which derive their n a m e s from geographical situations, are n o w liable to b e connected with the contents of various systems : plant-geographical, climatogeographical, pedological, ecological, etc., or they m a y b e used as terms in landscape or in the geography of population, h u m a n settlement, or econo m y . Their delimitation is not only different in the various branches of learning, but even within one a n d the s a m e field there exist serious discrepancies between different authors, according to which characteristic is stressed. W h e n w e look into geographical literature w e see at once the numerous different uses of these terms. T h e s a m e holds true in those m a n y neighbouring sciences o n which geography is b o u n d to rely. In the following pages I shall try to throw s o m e light o n the term subarctic a n d its use until n o w in various fields of learning a n d according to the geographical differentiation of its area, especially in
11
J. Blüthgen
northern Fennoscandia. Here w e can rely u p o n the w o r k done locally by the T r o m s ö M u s e u m in northern Norway, by the Abisko Research Station of the Royal Swedish A c a d e m y of Sciences a n d by the K e v o Subarctic Research Station of the T u r k u University near Utsjoki in northern Finland. These stations are likely to promote scientific research especially in the fields of natural sciences, ecology and geography. Of course I can select only s o m e instances taken from a rather broad range of this term’s uses. If w e first take a look into a differentiation of c o m plex landscape entities, w e choose the nine different subordinate types, into which Hassinger (1933) differentiated his m a i n type subpolare Landschaftstypen (subpolar types of landscapes) : 1. Lowland tundras (Kolas, Kanin, North Siberian a n d North Laurentian Lowland, North American arctic archip elago). 2. Highland tundras (Lapland, Finnmark, Desert Ural, eastern Siberian mountains, Chukchee Peninsula, Labrador Uplands, northern Kamchatka). 3.Alpine m e a d o w s (above the timber line in the mountains). 4.Cold high mountain steppes (Pamir, South a n d East Tibet, Tien Shan, P u n a of the Andes). 5. Cold high mountain deserts (western a n d northern Tibet). 6. Subpolar m e a d o w Islands (south-western Greenland, Iceland, Faeroe Islands, North-east N e w foundland, Aleutian Islands, Tierra del Fuego, Falkland Islands). 7. Subpolar forested plains (northern Sweden, northern Finland, northern Russian lowland, the taiga of western Siberia, Canadian coniferous belt). 8. Subpolar oceanic mountainous forest regions (northern Scandinavian mountains, North American Sea Alps, Newfoundland mountain area, A n d e s of Southern Chile and Tierra del Fuego, South K a m chatka, Sakhalin, Kuril Islands). 9. Subpolar continental mountainous forest regions (forested Ural, middle and eastern Siberian m o u n tains, Canadian Coast Range). This shows the term subpolar to be used in a very wide sense, as the continental forest regions of northern Eurasia including the western Siberian and Canadian taiga, the alpine m e a d o w s above the timber line in the mountains, the cold high mountain deserts of the Pamir, southern and eastern Tibet, and even the P u n a of the A n d e s are classiûed as subpolar, although being situated in very low latitudes. T o d a y this view is n o longer accepted, neither by geobotanists nor by most climatologists although this equalization of landscapes, which s h o w similarities only in thermal averages, in the related arcto-alpine flora, a n d in convergently similar growth forms (creeping a n d cushion growth), has been used for decades by other authors as well. T h e researches of Troll (1941, 1959) on m a n y tropical and ectropical mountains of the
12
Earth have s h o w n that mountain climates, despite certain similarities of rough averages with those of higher latitudes, are first a n d foremost the mountainous variations of neighbouring lowland climates. T h a t holds trile especially for the m a r c h of radiation a n d its consequence, the m a r c h of temperature. T h e relation between daily a n d yearly m a r c h of temperature is in both regions, i.e., the tropical high mountains and the polar lowlands, basically different. In the European area w e have in addition the historical factor that the arctic floristic element has receded after the post-glacial increase in temperature from the up till then subarctic-periglacial middle European lowland partly into the Alpine height zones, partly into those areas of northern Europe adjacent to the pole. This floristic relationship, forced as it w a s by the immigrational history of the plants, has contributed to the error of equalizing the accompanying climates. Besides, this has been favoured by the fact that, from the Alps to the Barents Sea, the arctoalpine element of the flora is found towards the north in successively lower mountain regions, until it is last m e t with near sea level along the coast of the Barents Sea. W i t h this last point in view w e approach a problem which produces difficulties in delimiting the Subarctic not only in Fennoscandia but everywhere. It holds true for all regions where meridionally arranged mountain ranges of middle to higher latitudes cause the alpine height zone to pass into the subarctic lowland region. That is not only the case in the Scandes but also in the Urals, in s o m e north-east Siberian mountain ranges, on K a m c h a t k a a n d in the North American R o c k y Mountains. In all these regions coldadapted plants m o v e d either to a greater altitude or to the north w h e n the climate b e c a m e milder after the disappearance of ice. T h e time elapsed since then is too short a n d the difference really existing between high alpine a n d subarctic climate is not of such vital importance for these plant associations to cause adaptation by differentiation within the plant life. Besides it is generally k n o w n that there are s o m e circumpolar plant areas which have not taken part in this retreat into greater altitudes, but which have followed the receding ice towards the north. T h e wellk n o w n Atlas of the distribution of vascular plants in N. W. Europe by Hulten (1950) shows instructive examples. Despite the floristic relationship, and s o m e ecological similarities, it is nevertheless important to keep in mind that there are differences between the arctic a n d alpine elements of the flora d u e to the migrational history. In animal geography such differences are perhaps even m o r e striking if w e think of the circumpolar occurrence of the reindeer, the lemming, and other species which d o not exist in the middle European alpine area a n d the transplantation of which has not
Problems of definition and geographical differentiation of the Subarctic
been possible. In the post-glacial time they followed the receding ice to the north exclusively. Let us take a look into the differentiation of the
the “Subalpine Birch W o o d ” as further independent zones. It is clear that he regards the extension of his subarctic proper as very narrow and excludes the
Earth according to landscape belts which Passarge published in 1923. For along time his system played an important part at least within middle European geography. Within the polar areas Passarge distinguishes the Kültewüsten (cold deserts) and the Kültesteppen (cold steppes). The latter only are of interest to us. He divides t h e m into the Tundralünder (tundra areas) and the Subpolare Wiesenlünder (subpolar m e a d o w areas). When dealing with the tundra he mentions its different aspects : Kümmertundra (poor tundra), Torfhügeltundra (peat-hill tundra), Flechtenrundhöcker (lichen covered roches moutonnées), and Waldtundra (forest tundra), but he does not use them for a systematic sub-division. The t e r m subpolar does not appear in this last group, but only with regard to the subpolar m e a d o w areas and outside the polar areas in the sub-division of the so-called “middle belt”. Here are distinguished a m o n g others the “subpolar inland coniferous areas’’, i.e., the taiga, as a separate type within the middle belt, notwithstanding its extension very far south in Canada and Siberia. The “forest tundra” is mentioned as the main living area of the reindeer and is s h o w n on the outline m a p as a narrow strip parallel to the polar border. All these areas, which are first treated as natural landscapes, s h o w a specific form of utilization by m a n also dealt with by Passarge. In plant-geography proper, the term subarctic or subpolar zone is found relatively early. However, it covers not so m u c h ecological zones but mostly areas based o n floristic elements, e.g., Engler’s area of subarctic flora that he places between his tundra zone-which he, by the way, n a m e s “arctic”-and his deciduous forests of the middle latitude (Engler and Drude, 1896-1928). In designating the Boreal Coniferous Belt or the taiga as subpolar, he has been followed by m a n y geobotanists and climatologists as far as they constructed their climate systems according to climatic effects o n vegetation. Hettner (1935) joined this group. Lundegårdh (1957) also identified the southern limit of the polar zone with the northern limit of the forest zone and classiíied the taiga as subpolar coniferous zone without further sub-division. A m o n g plant geographers w h o use the term subarctic, its geographical delimitation is variable. Let us take a few examples only. Walter (1954, p. 151) identifies the area of the subarctic flora with the forest tundra and regards it as a sub-division of the arctic geo-element. Sjörs (1965, p. 50), on the other hand, ascribes a plant-geographically independent character to the subarctic outside the Arctic proper, but his subarctic covers only the northern taiga, to which he adds towards the pole the “Continental W o o d l a n d Tundra”, the “Alpine North Ural”, a n d
Birch W o o d as well as the W o o d l a n d tundra a n d even the real tundra. We can see the huge discrepancy which exists, as regards the conception of the subpolar area, between Engler and Sjörs. In his earlier w o r k (1956) Sjörs criticizes the equalization of “high boreal” and “subarctic”. In accordance with the animal geographer E k m a n he preferred to n a m e the northern taiga regions “high boreal” and equalize the Subarctic only with Grigoriev’s (1956) forest tundra and the “hemiarktis” of Rousseau (1952). But, as w e have seen, Sjörs has altered his definition of subarctic in his new book of 1965. After numerous earlier investigations concerning the question of vegetational zones in Labrador, Rousseau (1964) published an important classifying contribution which differentiated the problem, at least for East Canada (Labrador), to a higher degree. In a biogeographical description, which partly follows up the researches of Hustich (1939) in Labrador, he distinguishes, between his temperate and arctic zones, the Subarctic and, adjacent to it to the north, the hemiarctic zones as transitional zones, both belonging neither to the arctic nor to the temperate,i.e., boreal, zones. The temperate zone is extended to that line where profitable mixed coniferous forests of Picea glauca, Picea mariana, Abies balsamea, Populus tremuloides, Betula papyrifera, partly also of Thuja occidentalis cease to form dense stands. This is the timber line of forest economy. In the subarctic zone coniferous forest of the taiga type is predominant o n dry and wet sites, very often interrupted, however, by bogs a n d generally open and sparse compared to the boreal zone. It is the “parc subarctique”. Real tundra is not found in this area, but w e find scattered permafrost cores already. The tundra is only partly found in the hemiarctic zone where it is associated with sparse forest stands and tufts of Picea mariana and Larix laricina, on h u m i d sites with S p h a g n u m mosses and o n drier sites with Cladonia lichens. This zone is physiognomically and ecologically the equivalent to the Eurasian forest tundra. L o w tree growth is concentrated to the water edge, to s o m e tufts in h u m i d hollows or to sheltered valley stretches. On the exposed treeless areas the flora of the tundra or the north Scandinavian m o u n tain heath, as the case m a y be, is dominant, which are intimately related to the alpine flora. In the arctic zone proper, in conclusion, as far as it can be a habitat for plants at all, w e find humidity- or dryness-adapted tundra, characterized by small Betula glandulosa trees, dwarf bushes ( E m p e t r u m hermaphroditum), Cladonia, Carex, and S p h a g n u m species. For our problem it is of importance that the tundra proper thus is regarded by Rousseau as part of the Arctic, and not as belonging to the Subarctic or hemiarctic.
13
J. Blüthgen
On the other hand it is of principal importance that Rousseau regards the northern transitional belt of the taiga as subarctic. This discriminating description of the plant-geographical transition belts between the arctic vegetation of northern Labrador and the boreal main region is also accepted by Knapp (1965) in his recently published book on the vegetation of northern and middle America. H e already calls subarctic the dissolving open border area of the taiga, which is interrupted by bogs and habitats of arctic elements of flora. This is the “Open Coniferous Forest” of Ritchie (1960)and the “Open BorealWoodland” of Hare (1950), in which w e again find indistinct names and where boreal and subarctic are used as synonyms. Finally w e have, in connexion with the mountain birch forest of northern Europe, the term “subalpine zone” which,since Wahlenberg’swork, Flora lapponica (1812), has been an integral part of plant geography in the north. Hustich (1960,p. 55) rightly protested against the continued use of this term, at least as far as the birch forests of Lapland are concerned, in his contribution to the volume, A Geography of Norden, which was prepared for the International Geographical Congress,1960,at Stockholm. In accordance with Grigoriev, to w h o m I shall return in a moment, his opinion is that the Subarctic ought to be a transitional zone of its o w n between the Arctic and the boreal coniferous zone. The question which Hustich has left out is just which parts of the hitherto Arctic and which parts of the hitherto boreal zone are supposed to form this zone. It looks as if in recent plant-geographical publications the view which has gained more and more ground is to regard the Subarctic as a basically independent belt. This is reflected in the representation of the sub-arctic zone by Schmithüsen (1961), which is widely used in Germany. H e shows the Subarctic to begin with the “northern tree and wood limit and the wooded tundra”. This is followed by further sub-divisionsof the Subarctic, viz., the “tundra belt” and the “belt of subarctic meadows” (Aleutians,Iceland), existing in those parts influenced by the ocean. They are also called “subpolar meadows” though the absence of birch trees in most of these oceanic places is not due to natural causes but to h u m a n influence. As an explanation it has to be added that the “northern wood and tree limit”is thought of not as a line but as a transitionalbelt. Ifw e compare this system with that of Rousseau mentioned above w e see that Schmithüsen,in opposition to Rousseau, regards the tundra proper as part of the Subarctic and not yet of the Arctic. W e shall return to this remarkable point later after having discussed the suggestions of Grigoriev. It is as difficult to define the Subarctic according to climatic criteria as it is according to plant-geographical-ecological criteria. W e shall first examine
14
h o w far a genetic point of view-which differentiates the climates according to the general circulation of air on the earth-is able to supply a basis for the term subarctic. A m o n g the older representatives of this school Hettner (1930) must also be named here. His genetic differentiation of climates gives no quantitative specihations of climatic measurements,i.e., no threshold values nor values of continuancy.Only when limiting the Subarctic (p. 96) he remarks that less than four months show a temperature of more than loo C. According to this statement a great part of the taiga ought to be assigned to this belt. In another of his works, however, (Hettner, 1935, p. 139) he identifies the subarctic zone as the outer belt of the Arctic proper and indicates a July temperature of 60 C to 100 C. A t about the same time Philippson (1933,p. 308) equalized the border between arctic and subarctic with the July isotherm of 50 C. If one examines these different statements for the special case of Fennoscandia, the differentiation made by Hettner is not at all satisfactory, because here he has three meridional belts of climate: the oceanic climate of the North Atlantic to the Scandes and the Varanger Peninsula, the polar climate within the Scandes from southern Norway to the Kola Peninsula, and the cold inland climate from the Bothnian area over the greater part of Finland to northern and eastern Russia. H e attempts no sub-division of his polar climate. Such contradicting statements can be of no help for our problem. A m o n g the younger climatogeographers we shall single out Alissow (1954).His system is based on the distribution of the belts within the general circulation, and on the extension of air masses during the single seasons. H e distinguishes between the arctic and the temperate zones a special sub-arctic zone limited by the position of the arctic frontal zone, i.e., that region where the west-east tracks of the cyclones of the ectropical west drift are nearest to the pole. The arctic frontal zone tends to be situated more to the north in summer than in winter. Where this occurs the result is a transitional zone with arctic air masses in winter and those of the temperate zone in summer. T w o special cases have to p e distinguished: (a) a continental zone with large temperature amplitude between summer and winter (North-east Yakutia), and (b) an oceanic zone with amplitudes of the highest monthly averages reaching 200 C (South Greenland and Alaska). A look into the corresponding m a p in Alissow’s book shows, on the other hand, that the “Subarctic” defined in this w a y is in part substantially different from the “Subarctic” based on plantgeographical and ecological investigations. According to Allisow, Iceland, as well as the whole area from northern Scandinavia to the estuary of the Yenisey is not a part of the Subarctic because in this region, according to Alissow, no seasonal shifting of the arctic frontal zone takes place.
Problems of definition and geographical differentiation of the Subarctic
A “Subarctic” climatically defined in this way, even though genetic, cannot be considered as satisfactory although it makes use of a complex climate constituent, the air masses. For, even if one might be inclined to concede that the subarctic character of northern Lapland is not very marked, this is not at all the case with the tundras of north-easternEurope. The arctic frontal zone or the winter invasion of arctic air alone cannot be deemed satisfactory as a climatic motivation. Besides, northern Eurasia is invaded by arctic air during summer with northern winds and during winter mainly by continental air with south to west winds (boreal zone of Flohn, 1950). A m o n g the numerous effective climate systems, which are based on the effects of climate on soil and vegetation,there is also no unanimity as to the separation of a Subarctic. First w e shall deal with Köppen’s classificationwhich is still widely used. His first classification (Köppen, 1900) was based exclusively on heat and humidity claims of plant life (he used the plant groups of D e Candolle). The term “subarctic” or “subpolar climate” does not appear at all. A m o n g the differentiated areas that of the birch climate of the plant group of microtherms (D)is of importance for northern Europe. During about four months only the temperature is above 100 C, reaching an upper limit of 190 C, while that of the coldest month lies somewhere between 30 C and -520 C, the annual amplitude of the monthly averages must at least be 100 C. Beyond the July isotherm of 100 C begins the realm (E)of the hekistotherms (E)or cold climates with the warmest monthly average between O0 C and 100 C. This is the polar-fox or arctic-tundra climate with a continental,large,annual amplitude of monthly averages between 200 C and 600 C,with a cold, long, and dry winter and a short but stable summer. In the next classification by Köppen (1918) the criteria for the belts under consideration here have been changed as well as their names. T h e coldest month of both areas (Dand E) is n o w below -20 C, the warmest of the snow forest climate (D)is above 100 C, the snow-rich tundra climate (E)is between O0 C and 100 C. Besides there is an explicit differentiation between ET (tundra climate) and EH (climate of altitudes above 3,000 m), which Köppen unfortunately abandoned again later. In his manual (1923) and later, in the wall m a p of the climates of the earth edited by Köppen and Geiger (first published in 1928, latest revision by Geiger in 1961) the D-limitof the snow forest climates against the temperate C-climates has been lowered from -20 C to -30 C of the coldest month; the border against the E-climates, which is of importance for Lapland, has stayed, however, as it was (warmest month 100 C = tree limit).In Lapland we find therefore the birch climate (Dfc) near to the tundra climate (ET).W h a t is important is that the areas in northern Europe appear to be rather arbitrarily limited and show no conformity with biological
border lines in this region. A subarctic climate of its o w n is never mentioned. This Köppen classification&hasbeen modified-several times, and that which Trewartha (1954) undertook became widely known in the last decade.As the coldest subtype of his “humid microthermal climates” (D) Trewnrtha united the types Dcf, Ddf, and D c w to a “subarctic climate”. It forms a large belt which extends into the mountains of Central Asia and which comprises the greater part of the taiga. Stretching from the latitude of Oslo-Stockholm-Helsinkiright to the Barents Sea in northern Europe,it is interrupted only by the “tundra climate” (ET)of the Scandes. As with Köppen this is assigned to the polar climates (E).In this w a y another untenable definition of the Subarctic has been given, for neither is the climate of Helsinki subarctic, nor is the climate of southern Norway arctic, quite apart from the disputable identification of the tundra climate with the arctic climate. At least the Subarctic is not regarded as belonging to the Arctic, but its extension to the border of the boreal forests against the steppes in rather low latitudes need not be discussed. In a similar w a y Blair (1954)defined his subpolar climate, which he regards as a main type of its own, but which he equalizes with the whole taiga as well. In this last respect it really was a progress when von Wissmann (1966) did not extend his “subpolar climate” into the forest zone but equalized it with the climate of the tundra. With this however, the subpolar climate is limited to a rather small fringe along the coasts of the polar seas. As the 100 C isotherm of July is taken by him as the limiting value, the birch forest and the forest tundra, as well as the oceanic “meadows” of Iceland remain outside the subpolar zone of von Wissmann. In the m a p of climates which was published at the same time and in the same book by Creutzburg and Habbe (1966) the term “subpolar climate” is used in a separate way, limited by the isochiones (lines of the same length of snow cover) of 150 and 240 days respectively which means, in terms of temperature, between the July isotherms of 70 C and 110 C in oceanic regions. Taking into consideration the wetter and drier variants the subpolar climate of northern Europe extends towards the equator to the south Norwegian mountains and still includes the Bothnian Gulf and Lake Ladoga. The southernmost point of the subpolar climate on the whole of the Earth is in this presentation the northern point of Hokkaido and the mountain area between Lake Baikal and the sources of the Amur. W h a t was said above about Trewartha’s subarctic climate applies to the subpolar region of Creutzburg and Habbe as well: though it forms a belt of its own, outside the Arctic, it reaches m u c h too far into middle latitudes. T o conclude this climatogeographical comparison w e shall examine the m a p of the seasonal climates of
‘15
J. Bliithgen
the Earth by Troll and Paffen (1965),which is mainly based on the ecological needs of vegetation and which is the most differentiated of all climate classifications. Within Zone I (polar and subpolar zones) it shows the “subarctic tundra climate” and the “subpolar, high oceanic climate”. The first is found, as the n a m e implies, only on the northern hemisphere and there it comprises mainly the tundras north of the polar wood and tree limit.The warmest month of this zone averages 60 C to 100 C,the coldest stays below -80 C. In the Hudson B a y area, as was to be expected, it advances farthest south including the north-western shore of James Bay. In Eurasia it extends,in a margin parallel to the coast, from the neck of Kamchatka to northern Russia and from there to the west slowly narrowing to the north coast of Kola, Varanger Peninsula, and Mageröy, and thereby assigns the greater part of Lapland to the continental boreal climate of the following “cold temperateboreal zone”. T h e oceanic variety, which w e m a y conveniently call “subpolar high oceanic climate” has cool summers (warmest month 50 C to 120 C), only moderately cold winters (coldest month 20 C to -80 C), a relatively small amplitude, and lower temperatures not frequently below 100 C. It is represented in the subantarctic islands, in the southern half of Iceland and in the Aleutians. Vegetational areas are taken as the starting-point for this differentiation of climate for which thermic threshold values, etc., are then found. T h e main problem, i.e., the extension of the Subarctic,is again brought back to the plant-geographical-ecological question of which formation is still “subarctic” and which is already “cold temperate boreal”. T h e area of the Scandinavian mountain birch forest is accordingly, in northern Europe, placed mainly within the oceanic and continental boreal climate, in Iceland within the subpolar high oceanic climate, and in northern Iceland and south-western Greenland within the arctic tundra climate. This statement must not be taken,however, as the proof of a wrong climatic differentiation but is in accordance with the fact, stressed by the present writer in his geographical investigation into the Scandinavian mountain birch forest (Blüthgen, 1960) as a landscape formation, that the birch forest comprises an extensive climatic spectrum and, as a vegetational formation, is therefore not appropriate for a climatological-ecological definition and delimitation of the Subarctic. Only incidentally and for the sake of completeness should it be mentioned that on the ocean a marine subarctic zone can also be differentiated and its delimitation is even m u c h less problematical than the terrestrial one. T h e unanimous opinion of all oceanographers is that those ocean areas are subarctic,where “subpolar convergences”, i.e., mixing zones between cold arctic and warmer water, occur and which show in some cases an oscillating movement in their annual
16
march. At the same time these are the areas where polar drift ice seasonally appears. These are the “polar fronts” of the seas. F r o m year to year these areas show as large and irregular oscillations of situation as their atmosphericalcounterparts. In view of this it is understandable that there is some uncertainty as regards their delimitation in certain areas-for instance in Baffin Bay, as shown by a comparison of the limits of Schott (1942) and Dietrich (1964)-but the criteria as such are fortunately not disputed. T h e subarctic ocean areas are relevant for the problem of delimiting the Subarctic on the continents for yet another reason. In all those cases where the Subarctic on the continents extends very far south of the polar circle, with tundra, forest tundra, birchwood or “subpolar meadows” in its lowland variety, it is neighbouring subpolar ocean areas which are responsiblefor it :Iceland,southern Greenland, Labrador, the northern part of Newfoundland, Kamchatka. T h e sub-arctic ocean areas are therefore a great help to delimit the terrestrial Subarctic as well, where this extends south of the polar circle. By far the most important,ecologically far-reaching and thoroughly substantiated presentation of the Subarctic has been given by the Russian physiogeographer Grigoriev (1956). This well-known author carried out his first field work on this topic in the area of the Bolshezemelskaya Tundra in north-eastern Russia. This dates back to 1904. T h e book containing not only his o w n but also numerous other detailed researches,mainly Russian, and appeared for the first time in 1946. In the second edition, also prepared by the author himself, special attention is given to radiation measurements and calculations published in the meantime by Budyko (1955) and to the ecological field work of Tikhomirov (1960). W e should take a closer look into this fundamental work which is as far as I k n o w the first comprehensive treatment of the term “subarctic” in literature form. I was able to use parts of a translation that has been started by the Verlag H.Haack, at Gotha, whose help I wish to acknowledge here. It is to be hoped that the full translation w ill soon be published, as this work is not widely known. It offers an ideal starting-pointfor a discussion of the ecological findings related to the Subarctic in general. T h e term subarctic as used by Grigoriev originates mainly from plant-geographical-ecologicalfacts and that means from climato-geographical, pedological and hydrological facts too. His book deals with the nature of the Subarctic; however, its importance for h u m a n settlement, economy, and transport is hardly touched upon. The Subarctic covers roughly the areas of the “subarctic and arctic tundras” of the geobotanists. It is of special importance that “subarctic” refers to an independent geographical zone extending between the arctic and the temperate zones, including the boreal zones. The decisive conditions
Problems of definition and geographical differentiationof the Subarctic
for this are: (a) low, but in summer still positive radiation balance, predominantly cold air masses in summer also; (b) cyclonic activity with frontal precipitation; (e) discrepancy between low radiation effect and high humidity surplus in air and soil during summer; (d) low summer temperatures with monthly averages below 100 C to 120 C and very low temperatures of the thawing soil; (e) excessive humidity of the soil in summer. Grigoriev divides his subarctic belt into two zones, the Russian designations of which, priarkticesky and priborealny, might be translated into English as “pararctic” and “paraboreal”. T h e paraboreal zone in its turn is divided into a northern subzone (of moss and lichen tundra) and a southern subzone (of bush and forest tundra). The pararctic zone is equal to the “arctic tundra” of m a n y plant geographers, as the paraboreal is to the “subarctic tundra”. Further on I shall use this division of the Subarctic in a “pararctic” and a “paraboreal” subzone which was also introduced and proposed by Grigoriev, but with some deviation from Grigoriev as regards the paraboreal subzone. For this zonal differentiation based on plant-geographical facts Grigoriev found climatological averages and extremes. The radiation balance which is positive only in the three or four summer months and only near the surface with temperatures above freezing-point has the effect that the frost, which has penetrated the soil more or less deeply during the other eight or nine winter months-there are no transitional months in the sense of spring and autumn in the middle latitudes-is not altogether annihilated. Permafrost exists but regionally differs as the type of soil, drainage, vegetation cover and exposition cause locally highly variable temperature balances. However, it must be said, here that the Subarctic is not identical with the area of the permafrost (Fig.1). In continental regions the permafrost is known to advance far into middle latitudes beyond the Subarctic;but it is altogether absent in some high oceanic regions of the Subarctic. In areas with permafrost the result is that the depth of thawing varies intensely during summer when water concentration is high and evaporation relatively low, because the air is-due to the low temperatures-mostly near saturation point anyway and has therefore practically no power of evaporation. This abundance of cold melting water in the thawing soil, especially in flat areas, is one of the main characteristics of the Subarctic and the decisive phenomenon of the tundra in its various aspects. American scientists, however, have identified the southern boundary of the discontinuous permafrost with the southern limit of the Subarctic itself (see Brown, 1969), but this is a doubtful definition because permafrost depends not only on the arctic radiation balance but also on the degree of continentality.
“Continentality” is a complex term which cannot be derived from arctic conditions alone. In addition to Grigoriev’s list the so-called subpolar meadows should be included in the paraboreal Subarctic. These are mainly free of permafrost and receive their high soil humidity during summer also from abundant precipitations at low air temperatures which forbid high evaporation. They belong potentially to the forest tundra in the form of low birch wood, their present treelessness being the result of man-made deforestation during past centuries. The cold thawing soil prevents the plants from absorbing greater amounts of water or nourishment, with the effect that dry periods or sporadic w a r m days-which even in the Subarctic m a y for a short time cause extremes up to 300 C (cf. Lembke, 1947) with the help of the summer midnight sun and appropriate weather conditions-may cause withering damages. Therefore the exposed plants, especially evergreens,are normally protected against such strain by the xerophytic structure of their organs, by cushion growth, and by being able to adjust the openings of their stomata, though this latter statement has not been verified by other writers. The root net is very often so superficial,that it is restricted to the not very thick layer of raw h u m u s on the mineral soil and is subjected to its extreme variations of humidity. A deeper penetration of the roots into the mineral soil is m u c h too risky for the plant because the fluctuations of temperature around freezing-point which are frequently met with in the Subarctic, even in summer, and the extreme variations of temperature cause breaking damages at the border line between raw humus and mineral soil. Accordingly only a very thin layer of air and soil, on or near the surface, offers enough heat to some subarctic plants, i.e., the “microthermals” of D e Candolle. The characteristic water concentration of the subarctic soil is caused above all by four factors 1. Thawing water in the soil above the permafrost layer. In case of missing permafrost,i.e., in oceanic climates,supersaturation of the soil by high and frequent precipitation compensates lacking melt-water to a certain degree during middle or late summer. 2.Melting water of the winter snow accumulated during eight or nine months outside oceanic regions. 3. L o w evaporation of altogether 10-12per cent only of the annual precipitation, lowest in oceanic and somewhat higher in continental parts of the Subarctic. 4.M a x i m u m precipitation in late summer outside the oceanic regions of the Subarctic, compensating evaporation losses during this period. It is true that global radiation during the summer is high due to the long polar days; in some parts it is even higher than in more southern latitudes. However,the low daily arc of the sun,the high percentage of diffuse light, the effective emission approaching
17 2
FIG.1. Distribution of sea-ice and permafrost (partly after Handbook of Geophysiology, 1960).
Land-ice within arctic and subarctic regions Continuous permafrost Discontinuous permafrost
14C
Sporadic permafrost
I\\y
Subarctic drift-ice High arctic pack-ice
120
100'
80'
O L " - "
500
1000
1500
2000krn L
J. Bliithgen
50 per cent (a consequence of the counter-radiation being reduced by l o w content of absolute water vapour and carbon dioxide), the screening against direct sun radiation during the s u m m e r by fog (mainly in sea a n d coast areas), the consumption of melting heat which continues far into the s u m m e r , a n d the high albedo as long as ice a n d s n o w are covering soil and water surfaces, all contribute to obtain only a small degree of w a r m i n g of the ,air. However, the climatic difference between the layer of air near the ground (or even within vegetation cushions) and the normal height of 2 m within the Stevenson screen (which is normal in climatic measurements) is not smaller than in l o w latitudes. On the contrary, the vertical lapse rate above the ground is often extremely high. In north-eastern Land, that m e a n s even within the high Arctic, D e g e took temperature measurements of 250 C to 280 C within vegetation cushions when the surrounding air at normal height showed only 40 C to 50 It is a specific characteristic of the Subarctic that due to its latitudinal situation the s u m m e r insolation surplus is still large enough to cause absence of s n o w after having subtracted the required melting and evaporation heat. T h e soils surface plays a n important role in the transformation of short-wave radiation energy into heat radiation, which is not so m u c h the case in the Arctic proper or at least not in a significant measure. On the other hand, the heat gain is so small that in this zone the heat s u m which is necessary for tree growth is either not obtained or, if it is, it only allows for a rather scarce a n d slow production of w o o d tissue. This is, as mentioned before, due to several factors. T h e small heating effect is mainly due to water temperatures which remain low in the polar seas and, because of melting ice a n d water turbulence rise only immaterially above freezing-point so that the air over the seas remains cold by conduction and exchange, a n d obstinate s u m m e r fogs are formed. T h e fact that air pressure over boreal continents is in s u m m e r generally s o m e w h a t lower than over the polar seas leads to the frequent inflow of cold, foggy, arctic air into the continental subarctic regions thus seriously and frequently lowering the air temperature. T h e subarctic flora is adapted to these continual temperature setbacks, as s h o w n by Tikhomirov in several studies about plant life in northern U.S.S.R.: it can d o with only low degrees of temperature in the surrounding air, its vegetation period already starts at a little above freezing-point a n d s o m e biological activities within the cells begin even at a time w h e n the air temperature is still s o m e w h a t below freezingpoint. W e k n o w besides, that not only within the cushions but also in cavities beneath the crusted s n o w the temperature m a y reach positive degrees. T h e adaptation of vegetation to a short w a r m period is s h o w n by a certain acceleration of growth due to long
c.
20
daylight and a corresponding reduction-but not abolition-of the daily time of rest and by prolonging biological phases over several summers. Forming of buds, development of leaves, flowering, seed maturation, seeding, a n d germinating are often spread over several years. This leads to the fact that plants of the Subarctic reach a n especially old age. T h e annual growth-rings of hibernating plants are therefore extremely narrow. K i h l m a n (1890) counted 544 growthrings o n a small juniper tree of 83 m m traverse in the Kola forest tundra. T h e formation of seed in the flowering plants is mostly effected by self-pollination helped by the ever-blowing wind as pollinating insects are often lacking. T h e observation of increased vegetative propagation of numerous species of the subarctic flora gives the s a m e result. In the forest tundra this p h e n o m e n o n is also found a m o n g ligneous plants such as birch a n d spruce. M a n y a multi-stemmed isolated spruce tuft can b e traced to the rooting of weighted-down branches of a n o longer existing central mother tree. W i t h the birch it is the power of developing n e w sprouts at the base of the trunk, which leads to polycormia after the trunk h a d either died or been felled (Fig.2). As to the animals’ adaptation to their subarctic environment a n d their w a y of survival w e shall mention the detailed investigations of F o r m o z o v which will soon be published in English (translation by the Boreal Institute, University of Alberta, E d m o n t o n , Canada). In this connexion w e m a y mention only the following facts. Animals that are very characteristic of the terrestrial Subarctic are the reindeer, the polar fox, the tundra wolf, the l e m m i n g a n d other rodents, a great variety of insects (especially gnats), numerous rapacious birds a n d several species of geese, ducks a n d swans, which populate the tundra in s u m m e r after having returned from their winter shelters. Those animals which stay in the Subarctic during winter m u s t be prepared to overcome the long, stormy winter with a densely packed s n o w cover. S o m e species can d o this either by storing fat reserves and reducing their activity during winter or by hibernating under the s n o w cover as the lemmings and other rodents do. S o m e retire to the taiga where weather conditions are m o r e favourable a n d s n o w is not so tightly packed. Along subarctic coasts the birds of prey are often concentrated o n suitable bird cliffs from where they start for their catch flights over the sea, in very m u c h the s a m e w a y as fishermen d o from their crowded villages. W h a t has been said about the plants’ struggle for life in the Subarctic also holds true, although in another manner, for the subarctic fauna. F o r m o z o v (1969)wrote: “ L e m o n d e animal subarctique est très original, pittoresque; ses représentants possèdent d’excellentes capacités d’adaptation a u x dures conditions d’existence”’ The southern limit of the Subarctic is to most authors identical to the forest limit which they think
1
Problems of definition and geographical differentiation of the subarctic
FIG. 2. Subarctic birch wood in northernmost Finland, near Karigasniemi, showing mountain birch (Betula tortuosa) with typical polycormia; in the foreground shrubs of dwarf birch (Betula nana) (Blüthgen, 1960).
of as showing the existence of m o r e or less open stands. T h e considerable disintegration of the taiga forest at its subarctic border reflects the variations of habitats around the absolute limit of existence valid for the whole Subarctic. B u t this delimitation, which already assigns the subregion of the forest tundra as paraboreal to the Subarctic, is at the root genetically unsatisfactory, though physiognomically striking. It is m u c h m o r e important to choose the limit of seed ripening of the forest trees as the southern limit, which is also called rational or reproductive, or generative forest limit. T h e distribution of ripe seeds by the wind over long distances is therefore the deciding additional factor which determines the actual forest limit. In this connexion the high effectiveness of the chionochore transport of last year’s seed, i.e., over a crusted, spring snow-cover, m u s t b e mentioned. A t the extreme periphery of ita ever happening, seed ripening occurs only at intervals of m a n y years. T h e distribution of ripe seeds into the Subarctic also, occurs only at intervals of m a n y years and from isolated favoured mother trees. If the limit of sufficient reproduction or seed ripening-let us say every fifth to tenth year-is taken as the basis for the southern delimitation of the Subarctic, a forest belt of trees still growing without hindrance but which are no longer capable of annual reproduction, belongs to the paraboreal sub-zone of the sub-Arctic. This is the southern belt of this subzone which, besides, is physiognomically interrupted and broken up by bogs a n d areas which are unfavourable to growth. In the northern part, the forest tundra, including the northern parts of the Scandinavian mountain birch forests, is prevalent. This limit of reproduction should coincide with the timber line, that is, the border of economical forest exploitation.
As w e have seen, the existence of permafrost cannot be identified with the Subarctic because this phenom e n o n , in its southern-most occurrence reaches well into the steppes of Transbaikalia. Permafrost is accordingly only one of m a n y factors. H o w e v e r ihere are subarctic areas free of forests which d o not s h o w the development of continuous permafrost. This occurs in the oceanic regions of the Subarctic-as already stated-where winter temperatures are not l o w enough but s n o w cover offers satisfactory protection so that frost cannot penetrate to any depth worth mentioning. In these oceanic regions of the Subarctic the too low s u m m e r temperature is the forestpreventing factor. This applies mainly to Lapland, but perhaps also to a certain degree to Iceland a n d the Aleutians a n d this special problem is still far from being solved. T h e question whether Lapland already belongs to the Subarctic is not easily answered. Here w e encounter the plant-geographical peculiarity of the mountain birch forest, extending as a peripheral belt, at the forest limit from the high mountains and viddas of southern N o r w a y to the Barents Sea. It consists of the birch species Betula tortuosa (see Fig. 2), which in its habitus is different from the other birches, but which is regarded by s o m e botanists only as a subspecies of Betula pubeseens. There are hybrids, in fact, but hardly ever at the periphery of its occurrence. It is easiest to judge the difference of habitus if one happens to c o m e across t w o “pure-bred” specimens near to each other o n the s a m e spot. Its wide-spreading growth and the missing tendency of having branches hanging d o w n is a sure sign of the difference between the mountain birch and the other species. It is able to stand wet, cool s u m m e r s and therefore spreads m o r e easily in this climate than pine or spruce. During the 21
J. Bliithgen
FIG.3. Young pine trees (Pinus silvestris) above the birch region on the top of Kaunispää (540 m), Raututunturit (northern Finland) (Blüthgen, 1960).
climatic amelioration of the last decades, especially
in the 1930s, a striking advance of the coniferous trees has been observed, even overtaking the birch zone. This amelioration shows itself at all seasons and not only in the spring, as s h o w n b y the temperature curves for Karesuando for different time periods (see Blüthgen, 1966, p. 575). Perhaps this long-range thermal variability might be regarded as typical of the S u b arctic. The result was that pine seedlings appeared spontaneously in a wide area, not only within the birch region but even beyond in the regio alpina or pararctic subzone, a fact on which Hustich has also c o m m e n t e d several times. One of the reasons for this p h e n o m e n o n w a s that better germinating conditions were m e t with outside the birch zone than within the almost luxurious, herbaceous, ground vegetation of the birch woods. It is remarkable and material for the critical estimation of these findings that nowhere at that time could older witnesses of such a far advance, neither living nor dead, be found. A visit to such places, where the pine had appeared and taken root in the 1930s, when repeated after fifteen years, showed this clear picture to be muddled up somewhat. It is true that, for instance on Kaunispää in Finnish Lapland (540 m above sea-level) in 1960, numerous pine trees at least 1-2 m high were found (Fig. 3), where in 1939 only young plants had grown under winter s n o w cover.Theselittle trees had obviously endured the process of growing u p from under the protecting winter s n o w cover. The s u m m i t of Kaunispää (in the Raututunturit Mountains) is wind-exposed and does not provide the protection of a thick layer of snow. The especially critical late-winter period with icy arctic north-eastern winds from the polar sea, s n o w being blown over a s n o w crust, low absolute humidity, and already effective sun radiation, has
22
done no d a m a g e to these young pines. However, the young pine and spruce trees, which were found in 1936 on the isolated s u m m i t of Dundret (823 m), near Gällivare in the alpine region above tree limit, were in 1952 partly withered or at least d a m a g e d at the height of the winter s n o w cover. But this isolated summit is n o longer situated in the subarctic ; besides, the withering damages might have been a negative effect of the warming of the last decades (Blüthgen, 1952). The birch (Betula tortuosa), o n a visit in 1952 in the fell areas south of Lake Torneträsk, showed a under large n u m b e r of hand-high seedlings-still s n o w protection during winter-obviously flown there, into open stands nearly man-high (near Yassijaure) and which filled the gaps in the old stands that evidently h a d been thinned by m a n a long time ago. The definition and differentiation of the Subarctic which Grigoriev prepared for North-east Russia cannot entirely be applied to northern European, i.e., Fennoscandian, conditions despite the relative vicinity of these areas. A b o v e all the Scandinavian mountain birch forest in its uniformity against neighbouring formations is a disturbing factor. It is geographically an extremely characteristic landscape p h e n o m e n o n of Scandinavia from the south-westernviddas of southern N o r w a y to roughly the area around and east of M u r m a n s k on the Kola Peninsula. Therefore, as a geographer, one might not be inclined to divide it for the sake of giving a definition of the Subarctic. Physiognomically, it is so clearly a distinct vegetational region of its o w n , in vertical as well as in horizontal extension, that it is a very important geographical characteristic of Fennoscandia which has no counterpart anywhere in the world. E v e n on Kamchatka, which of all the regions of the world might be taken
Problems of definition and geographical differentiation of the Subarctic
into consideration, things have s o m e w h a t different aspects in so far as a broad birch region with Betula ermanii does indeed exist there but not as the uppermost forest belt, for the tree a n d forest limit itself o n K a m c h a t k a is formed by a narrow area of dwarf pines (Pinuspumila) a n d others which are not found in Scandinavia. A closer view of the regio betulina of northern Europe, which w a s studied thoroughly for the first time in 1913 by Fries shows, however, regional differences which allow a differentiation and partition of this area with regard to a delimitation of the S u b arctic. In the above-mentioned contribution to the Stockholm International Geographical Congress in 1960, the author divided the birch forest into several geographical facies. T h e term “facies”, originally derived from geology, has been used in this connexion in a n e w sense for the investigation into geographical landscapes, different also from that peculiar sense m a d e widely k n o w n by the plant sociology of BraunBlanquet. In the above-mentioned contribution t w o large areas, the southern and the northern, were named. T h e northern which begins at the Jämtland mountain pass and the sub-division of which is alone interesting to us, shows seven facies belts, viz., the polar-maritime margin facies, the northern Atlantic fjord facies, the northern subcontinental valley and plateau facies, the northern continental margin facies, the subarctic valley a n d plateau facies, the subarctic margin facies, and the Barents Sea facies. Only the first and the last three mentioned are to be regarded as and combined to the subarctic confacies of the birch forest. A s subarctic characteristics c o m m o n to all these there are s o m e climatically affected traits such as ; the occurrence of cold, late-winter, polar-sea winds over a s n o w cover remaining well into M a y , large daily temperature oscillations immediately above
the s n o w cover due to the rapidly increasing length of day, and increased evaporation power of arctic air poor in water vapour. A s a result of these conditions, only imperfectly reflected in the values recorded by the scattered climatic stations of this area, vegetation, in order to survive, greatly needs the protection of the snow. Therefore shearing damages begin to appear at the height of the early spring s n o w cover (table birches (Fig. 4) a n d analogous forms with other bushes) ; but a superficial journey through northern Lapland (in 1960, along the road from Ivalo via Karigasniemi-Karasjok to the Porsanger Fjord) showed against all expectations surprisingly f e w shearing damages within the regio betulina. T h a t m a y , however, be caused by the road having been built as m u c h as possible across protected areas. A visit in 1966 has s h o w n that such p h e n o m e n a are only scarcely found in the regio betulina of Utsjoki. Nevertheless H ä m e t Ahti (1963)in her detailed plant-sociological analysis of the mountain birch forest of northern Lapland mentioned t h e m as one of the characteristics of her “oceanic subalpine birch forests”, (for instance o n the Varanger Pensinsula and at Utsjoki), while they are not mentioned for the submaritime nor for continental-subalpine subzones. N e a r to the polar sea the exposure to the sun (“sydberg-vegetation”), which at the s a m e time m e a n s protection against the cold polar sea winds, plays a major part in the distribution of higher vegetation, this being a striking feature of the Subarctic. Within the small polar-maritime facies in northern N o r w a y insufficient s u m m e r w a r m t h is the deciding factor, the winters being abnormally mild but rather dark a n d stormy. Besides, the occurrence of palses, i.e., peat hillocks with a n ice core (Figs.5 to 7), must be mentioned here. T h e y are the first precursors of real tundra only stretching farther east, uninterruptedly, over continuous
FIG. 4. Table birches (Betula tortuosa), some with tops, north of Karesuando (Lapland) (Bliithgen, 1936).
23
J. Blüthgen
permafrost and standing in evidence of the S u b arctic. This p h e n o m e n o n also is lacking in the other facies of the regio betulina including the polar-maritime facies. Palses very often have a weathered, dried-up s u m m i t where in a n otherwise boggy country often s o m e mountain birches have taken root. This could be observed by the present writer in 1936 o n the “bush cold steppes” which are mainly formed by Betula nana a n d several Salix species, between Karesuando and Kautokeino beyond the Finnish-Norwegian border, over a wide area so that the tree limit there w a s extremely scattered.
To the natural characteristics of the subarctic part of the regio betulina, m a n - m a d e factors m u s t b e added if w e w a n t to see the Subarctic with the whole scope of geography in mind. This region is pastured by reindeer ; these vagrant animals bite the birches a n d cause trampling damages o n dry locations (damages which often f o r m the starting-points for deflation effects) the subarctic m e a d o w s are used for keeping cows, goats, and sheep in scattered permanent settlements. Although the settlement density at the border of h u m a n settlement as a whole is extremely low (about 2-3lhuman beings o n 10 km2)the influence of h u m a n FIG.5. A group of small palses, i.e., peat hillocks with an ice core, near North Cape on Mageröy (Blüthgen,1960).
FIG.6. A rather big pals near Karlebotn (Varangerfjord, northern Norway) (Blüthgen, 1966).
24
Problems of definition and geographical differentiation of the Subarctic
economy, in the Subarctic, not only in northern Europe, is strongly marked. Despite all the technical improvements in m o d e r n h o m e s the d e m a n d for w o o d to build fisheries, enclosures a n d for heating is heavy because of the very slow gain and very far-between regeneration periods. Today’s picture of the bordering forests in the paraboreal subzone a n d the course of the tree and w o o d limit m u s t therefore be regarded as being influenced by m a n far m o r e than might appear at first sight of these apparently virgin and uninfluenced bordering stands. M a n y of the pine-free birch bush forests are likely to o w e their present appearance to the hewing of scattered pine tufts and of birch mother trees used for heating and making enclosures during the past centuries. T h e power to form sprout from the roots of Betula tortuosa has led, to the formation of multi-trunked trees similar to bushes (polycormia) which over large areas, for instance in inner F i n n m a r k and in Finnish Lapland, determines the appearance of the birch forests. This does not deny that there are natural factors as well which m a y lead to polycormia. T h e increase of settlements during the last decades, though slow, together with over-exploitation of the border forest during and reconstruction w o r k after the heavy damages of the Second World w a r have left w o u n d s which, at best, will be closed after m a n y decades, probably even centuries, by a natural regeneration of the forests. Perhaps continued damages and a final receding of the w o o d and tree limit in the paraboreal subzone must be reckoned with especially along the Barents Sea coast which is m o r e densely populated byfishing settlements. On the other h a n d s o m e afforestation (Fig.8, see also Mikola, 1969) with Scotch pine has taken place in northernmost Finnish Lapland near Utsjoki showing that at least s o m e recovering of devastated mixed birch a n d pine forest is observed at certain places under favourable local climatic conditions. B u t the annual average yield of wood, only in small favourable sites, exceeds 1 cm3/hectare, so that these stands are of n o perceptible economic value in comparison to m o r e southerly stands. In the settlements there is s o m e possibility for cultivating vegetables in greenhouses because of long daylight during the subarctic s u m m e r . B u t this is only of local importance though it helps to improve man’s diet. Finally I should like to set up the following theses which are a s u m m a r y of the results obtained and which m a y be regarded as important for the definition a n d differentiation of the subarctic. 1. T h e Subarctic forms a belt of its o w n between the arctic a n d the middle latitudes. It comprises a marine and a terrestrial part. B o t h m u s t b e defined separately. 2. T h e marine Subarctic is confined to the water areas o n both sides of polar convergences between
-
5-
+
o al
R
-3 0 ~ :m
-
>i
o
U C o) N
2
u-
-8 0 ~ :m
al U .L
-oal
-
U
100 cm
FIG.7. Section through upper part of apal near Karlebotn (Varangerfjord, northern Norway). The uppermost part of the section-some 30 cm-consists of peat, beneath follows frozen post-glacial clay mixed with clear ice near the lowermost part (approx. 1 m) (Bliithgen, 1966). polar a n d temperate sea water with seasonally shifting drift ice. 3. T h e terrestrial Subarctic lies between the arctic discontinuous patch tundra a n d the economic forest limit of the boreal coniferous forest belt (taiga). It thus comprises the plant-geographical areas of the continuous herbaceous tundra at the border of the Arctic, the boggy forest tundra, the northernmost facies of the Scandinavian mountain birch forest d o w n to the rational forest border along the northern periphery of the taiga where not m o r e than every fifth to tenth year is a year of seed ripening a n d where, therefore, there are larger breaks between t w o generations. T h e rational forest border, as it might b e called, is at the s a m e time nearly the limit of economic forest exploitation. 4. In agreement with Grigoriev the terrestrial Subarctic can b e subdivided into a pararctic (pa) and a paraboreal (pb) subzone.
25
J. Blüthgen
FIG. 8. Afforestation of Scotch pine in the mixed birch-pine forest south of Utsjoki near Petsikko, well-drained soil, but northern exposure. The group of young trees have suffered from frost damage in spring during one of the latter years (Blüthgen, 1966).
5. T h e pararctic subzone (pa) comprises the herbaceous tundra in its drier or m o r e h u m i d varieties. It is climatically characterized by average temperatures of the warmest m o n t h between 40 C and 80 C; the average temperature of the coldest m o n t h stays beneath -80 C. 6. T h e paraboreal sub-zone (pb) has a n oceanic and a continental variety. Within the continental area it can be divided into a northern (pbn) and a southern (pbs) belt. 7. T h e oceanic variety of the paraboreal subzone comprises the subpolar m e a d o w s with birch w o o d s in Alaska, south-west Greenland, and Iceland, the Atlantic part of the Scandinavian mountain birch forest north of about 680 N; warmest m o n t h 100 C to 120 C, coldest 20 C to O0 C, low amplitudes. 8. T h e northern belt of the continental variety of the paraboreal sub-zone (pbn) includes in northern Eurasia and North America the forest tundra and the Lapland part of the Scandinavian birch forest. T h e warmest m o n t h here shows 80 C to 100 C, the coldest stays below -80 C, the amplitudes are large, however. 9. T h e southern belt of the continental variety of the paraboreal subzone (pbs) is formed by the northern border forests of the taiga with open, often boggy stands beyond the rational (or reproductive) forest limit where, o n a n average seed ripening is possible only every fifth to tenth year. W a r m e s t m o n t h shows 100 C to 120 C, coldest likewise below -80 C, amplitudes are large here as well. 10. T h e continental Subarctic is characterized climatically by high unperiodic temperature oscillations. N o m o n t h is therefore quite free of frost but in the continental region the m a x i m a of the three s u m m e r months can approach 300 C.
26
11. T h e oceanic Subarctic has in fact lower oscillations but lower s u m m e r averages and therefore total heat is lower in summer. 12. T h e total heat gained by a positive s u m m e r radiation balance of about four months duration is sufficient to produce, after the melting of the s n o w a continuous vegetation cover which forms the habitat of a subarctic fauna adapted to these conditions (reindeer, polar fox, lemming, mosquitoes). 13. Absolute air humidity is very low, the relative humidity mostly high. S u m m e r is therefore foggy, especially near the Arctic Ocean. 14. T h e quantity of precipitation is mostly small a n d is mainly caused by advection activities following cyclonic fronts a n d occlusions rather than by convection, though this m a y occur in s o m e cases attached to labile weather type. Only in the oceanic regions, and there mostly in front of mountain ranges is precipitation relatively high. 15. Protection against wind and s n o w plays a vital part for the vegetation, as the s n o w cpver is mostly very uneven or thin and wind intensity-especially in late winter-is relatively high. 16. Permafrost is developed continuously everywhere in the pararctic subzone, a n d in the paraboreal, only in the continental parts. In the oceanic parts it occurs only in the peripheral form of interrupted peat hillocks with ice cores (palses) or is even lacking altogether where in the coldest m o n t h the temperatures is O0 C or immaterially below freezing-point a n d temperature oscillations are only short-lived and d o not lead to extreme frost periods. 17. T h e open solifluction, polygon soils, stone rings and areal formation of frost debris, typical of the Arctic, occur only sporadically in the Subarctic
Problems of definition and geographical differentiation of the Subarctic
FIG. 9. Solifluidal terraces under close vegetation cover on Mageröy (noirthern Norway) (Blüthgen, 1960).
because of the mostly continuous vegetation cover. Hidden flows of earth, however, under vegetation cover o n slopes (grass pads, fig. 9), and the occasional breaking up of these pads frequently occur. 18. T h e soil is cold in s u m m e f and superhumid; w a r m t h is sufficient only in a very thin surface layer of s o m e centimetres ; plants have therefore superficial roots and are adapted against too high evaporation by the xerophytic structure of their organs or by cushion growth or by both. Large areas in flat regions are rich in bogs of different kinds. 19. In the regional differentiation of the vegetation of the Subarctic according to associations, differences of local exposition and the differentiation of the microclimate near to the ground play the decisive part. 20. T h e shortness of the subarctic s u m m e r together with almost n o transitional seasons require that the sub-arctic vegetation prolongs its reproduction phases over several years, and that vegetative generation is widely found. However, growth is accelerated somew h a t during s u m m e r light conditions. 21. T h e Subarctic is the m a i n exploitation area of reindeer e c o n o m y in so far as the animals find the necessary nourishment, to build up stores of fat, o n the subarctic s u m m e r pastures. In the winter they feed on lichens in the protected border forests towards the taiga only in order to survive until the reopening of the s u m m e r pastures. 22. As the marine subarctic areas are rich in plankton a n d fish, due to the convergence a n d mixing of cold water rich in oxygen with w a r m e r water from the
south, the subarctic coasts along or near to these oceanic regions are characterized by densely inhabited fishing settlements in otherwise almost uninhabited surroundings. 23. Because of the slow and low increase of vegetation¶ man’s influence is extremely noticeable in the Subarctic even though it is sparsely populated. Changes in the subarctic vegetation cover are mainly d u e to the following factors: pines are receding and occur within the birch region; a birch forest is forming d u e to the m o r e bushy appearance of the stands because trees sprout from the roots ; overgrazing by too m a n y herds of reindeer; the coastal settlements’ d e m a n d for w o o d for heating purposes and building enclosures; also, locally, the appearances of natural m e a d o w s is changed as a result of the intensive haymaking in s u m m e r to meet the needs of the increasing c o w staple. . Summing up this tentative definition leads to a n extension of terrestrial a n d marine subarctic areas which m a y be seen in Figure 10.
ACKNOWLEDGEMENTS I
wish to express m y sincere thanks for technical assistance and discussion to my collaborators, R. Lind e r m a n n a n d D. Thannheiser, a n d to mention here m y late friend Hans-Günther Sternberg, fellow of m a n y discussions o n a n d of journeys through the Scandinavian Subarctic.
27
160'
\
FIG.10. M a p of the subarctic and its sub-division (compiled by J. Bliithgen).
Pararctic
Terrestrial Subarctic
Northern
14c
Paraborea I Sou thern , ,
Rational forest limit Marine Subarctic (seasonal shifting sea-ice)
r-.
Mountainous vegetation within the Subarctic
m,
Paratemperate (polar radiation régime in temperate oceanic regions with positive temperature anomaly)
VmJ 4 :
aracontinental (polar radiation . regime in boreal regions)
120'
1000
/
1 O00
e00
60°
LOO
O ,
500 '
,
B
.
,
1000
1500
2000 krn 1
J. Blütbgen
Résumé Problemes de déjnition et de différenciationgéographique du Subarctique, spécialement en Europe septentrionale (J. Bliithgen)
.
L a région subarctique est une zone de paysages située entre l’Arctique et la forêt boréale de conifères; c’est la zone phytogéographique de la toundra herbacée continue qui s’étend de la limite de l’Arctique jusqu’à la région forestière limitrophe du nord de la taïga, où plus d’une année sur d e u x permet la maturation des graines et où il n’y a, par conséquent, pas d’interruption plus grande entre d e u x cycles végétatifs. L a région subarctique comprend diverses sous-zones : continentale et océanique, d’une part, et para-arctique et paraboréale, d’autre part. L a sous-zone subarctique continentale se caractérise climatiquement par de fortes oscillations n o n périodiques de température. A u c u n mois n’y est donc sans gel; mais les m a x i m u m s des trois mois d’été peuvent s’approcher de 300. D a n s la sous-zone subarctique océanique, les oscillations de température sont moins fortes, mais par contre les moyennes estivales sont plus faibles, si bien que les s o m m e s thermiques y sont moindres. L a s o m m e thermique fournie par le bilan radiatif positif d’un été d’environ quatre mois est suffisante pour donner après la fonte de la neige une couverture végétale continue qui constitue l’habitat d’une faune subarctique adaptée à ces conditions (renne, lemming, moustiques). L’humidité absolue de l’atmosphère est très faible; l’humidité relative est le plus souvent élevée. L’été
est donc brumeux. Les précipitations sont généralem e n t faibles et résultent des m o u v e m e n t s d’advection consécutifs a u x fronts cycloniques et a u x occlusions. C e n’est que dans la région océanique, n o t a m m e n t sur le versant des chaînes montagneuses, que les précipitations sont relativement fortes. L e pergélisol est continu dans toute la sous-zone para-arctique et, dans la sous-zone paraboréale, seulement dans les parties continentales. D a n s les parties océaniques, il est discontinu. L a solifluction caractérisée, les sols polygonaux, les cercles de pierres et la formation aérolaire de débris produits par le froid n e s’observent qu’occasionnellement dans la région subarctique. L e sol est froid l’été; les végétaux hygrophiles ont donc des racines superficielles et l’adaptation les protège contre une évaporation trop forte. D a n s la différenciation régionale de la végétation subarctique selon les associations, l’exposition locale et le microclimat (près d u sol) jouent u n rôle décisif. Par suite de la brièveté de l’été subarctique et de l’absence presque complète de saisons de transition, la végétation subarctique doit étendre son cycle de reproduction sur plusieurs années ; en outre, la reproduction végétative est très répandue. P a r m i les principales ressources naturelles renouvelables de la région subarctique, o n peut citer le renne, ainsi que les poissons des zones océaniques riches en plancton. D u fait de la lenteur et de la faiblesse de l’accroissement végétatif dans la région subarctique, l’influence de l’homme y est extrêmement marquée.
Uiscusssion J. MALAURIE. Après le remarquable exposé du professeur Bliithgen, il est souhaité que le congrès aille, sur le plan épistémologique, plus avant, c’est-à-direrecherche la structure logique -indispensable à toute doctrine scientifique -de ce terme “ subarctique”, terme prêtant à des analogies spécieuses et à quelque ambiguïté. Sur le plan géomorphologique, d‘autre part, des définitions quelque peu différentes pourraient être apportées aux limites présentées, surtout phytogéographiques. J’y reviendrai plus loin dans ma communication; mais nombreux sont les faciès subarctiques dans les plateaux du nord-ouest du Groenland, haut arctique. D’autre part, comment, génétiquement, sur le plan géographique, définir une forme subarctique, alors m ê m e que, dans les roches résistantes, les formes sont le plus souvent polygéniques. Les formes spécfiquement, originellement “arctiques” n’ont été observées que dans les roches “meubles” de la terre d’hglefield (nord-ouest du Groenland). Le terme “toundras”, d’après une toute récente publication
30
de Donald A. Dagon (Polar notes, no 6,juin 1966), ayant été redéfini également “structurellement”, il est souhaité que le congrès saisisse en vérité l’occasion de ce rassemblement d‘experts pour constituer rapidement une commission “interdisciplinaire” visant à tenter de mieux définir le t e r m e “subarctique”; dans m o n esprit, voire m a formation de géomorphologue, ce terme a seulement valeur “topologique”. Ce serait avec plaisir que je participerais à ces travaux. A u moment m ê m e où la recherche polaire se développe si rapidement,il serait vraiment regrettable que se développent entre spécialistes des malentendus, fruits d’écoles, d’approches multiples, dans l’utilisation de ce terme. J e rappellerai seulement ce bien fâcheux vocable “périglaciaire”, source de si vives controverses, et toujours, dans son acception “boréale”, bien vivant.
J. B L ~ T H G E N .The problem of defining the subarctic is so complex that the proposal to discuss it between members of an interdisciplinaly commission is an excellent idea.
Problems of definition and geographical differentiation of the Subarctic
E. EINARSSON. O n page 16 Dr. Blüthgen gives a definition of a “subarctic tundra climate” and “subpolar, high oceanic climate”. According to that definition North Iceland has a “subpolar, high oceanic climate” not a “subarctic tundra climate” as stated in the paper. All the Icelandic lowlands have a “subpolar, high oceanic climate”, the mountain areas, however, especially in the central part of the country have a “subarctic tundra climate”.
T h e main report dealing with the variation J. BL~THGEN. of opening of stomata during periods of dryness (or warmth) relied upon is Stocker’s, Darmstadt, w h o has studied the vegetation of Swedish Lapland near Abisko. This is in agreement with xerophylism of m a n y subarctic and arctic plants which in this way are adapted to variations in humidity.
E. HULTEN. In your m a p of the Subarctic you had marked J. BL~THGEN. That is a correction which miist be followed up; in literature the classification of Iceland as subarctic 01 otherwise is rather contradictory. In the first part of m y paper I cited some of these contradictory statements. T. AHTI. I would like to point out that, from the point oi view of comparative circumpolar ecology, tree species m a y be poor indicators of the Arctic and the subarctic in some areas when compared to lesser vegetation, particularly byophytes and fungi, including lichens, the distribution of which has been m u c h less affected and interrupted b y the post-glacial vicissitudes. Perhaps, with these criteria, the Subarctic m a y be distinguished even in Iceland, Greenland and the Aleutian Islands. Of course, it depends on which criteria are given preference. Another more objective criterion for comparison of different areas is, of course, climatic measurements.
J. BL~THGEN. Thank YGU for adding these biological factors to define the Subarctic. A geographical definition of “Subarctic” must of course rely on as m a n y statements from related sciences as possible.
south-western Alaska and the Aleutian Islands with a special dark green colour. I carried out field work in that area last summer. T h e flat ground is an arctic tundra with the same plant communities as on the Arctic Slope of Alaska and with numerous lakes. There is no reason to regard this area as a separate division of the Subarctic, it is an arctic district. Concerning the Aleutian Islands the question is more complicated as they are very mountainous, but they are all completely treeless.
J. BL~THGEN. The dark green colour in south-western Alaska and on the Aleutian Islands represents oceanic “subpolar meadows” after the detailed maps in the Physicogeographical World Atlas (c 1964) from Moscow. S o m e earlier authors also used the term “subpolar meadows” with respect to subarctic regions of high oceanic climate. Thank you for your comment based on pour field observations. Perhaps the term “subpolar meadows” must thus be abondoned in subarctic regions, because even Iceland, hitherto often classified as a land of “subpolar meadows” has been deprived of most of its birch wood in lower districts by man.
R. SARVAS.W h e n considering the definition of the subF. E . ECKARDT. J’ai suivi avec grand intérêt votre exposé, en particulier pour ce qui concerne le comportement physiologique des plantes pendant les périodes de dégel et de sécheresse estivale. Vous avez mentionné à u n certain moment que les plantes ferment les stomates en vue de réduire les pertes hydriques (page 7 du document miméographié). Pourriez-vous m’indiquer les travaux sur lesquels ont été basées ces observations ? Je pose la question parce que j’ai eu l‘occasion personnellement de mesurer la transpiration de trois espèces végétales, à savoir Salix glauca, Vaccinium uliginosum var. microcarpa et Betula nana, au Groenland, pendant l’été. Ces mesures ont été effectuées dans les régions particulièrement sèches de Söndre Strömfiord et de Thulé (précipitations annuelles de l’ordre de 60-70 mm). Au cours de ces études je n’ai jamais p u constater la moindre augmentation dans la résistance diffusive des stomates, la courbe journalière de transpiration ayant pratiquement la m ê m e allure que celle de l’évaporation potentielle mesurée au moyen d’un disque de papier buvard mouillé.
arctic region, there is reason to keep in mind for what purpose the definition is to be used. W h e n it is mainly used to arrange data already available and to make large general surveys, syntheses, there is undoubtedly reason to use a definition of great coverage, even as large as to be able to serve several branches of research, such as, for instance, meteorology, phytogeography, geology, forestry, etc. When, however, n e w information is required, the most important thing is that the definition serves the unravelling of the actual problem in the best way possible. It can often be useful to work out one definition of the subarctic region to be used for the solving of one single problem. Perhaps, however, a feature c o m m o n to all these definitions could be that the sub-arctic region is formed b y an area, where, from the point of view of the problem or object in question, an arctic influence is more or less to be noticed. O n the basis of this, I suppose that there will always be several definitions, some of which serve primarly the synthesis and others the analysis.
31
J. Bliithgen
Bibliography / Bibliographie ALISSOW, B. P. 1954.Die Klimate der Erde. Berlin. 277 p. (Moscow 1950,German translation 1954.) BLAIR, T. A. 1954. Climatology,general and regional. 5th ed. New York. 478 p. BLUTHGEN, J. 1942. Die polare Baumgrenze in Lappland. Berlin. 80 p. (Veröff. Dt.Wiss. Inst. Kopenhagen, Reihe I (Arktis), Nr. 10.) . 1952. Baumgrenze und Klimacharakter in Lappland. Ein Beitrag zur Frage der Veränderlichkeit von Klimawerten und deren geographischer Auswirkung. Bad Kissingen. p. 362-371.)Ber. Dt. Wetterdienst U.S. Zone Nr. 42, Knoch-Heft.) . 1960. Der skandinavische Fjällbirkenwald als Landschaftsformation. Petermanns Geogr. Mitt., vol. 104,p. 119-
-_
__
144.
__. 1966. Allgemeine
Klimageographie. 2nd ed. Berlin.
720 p. (Lehrbuch d. Allg. Geogr., edited by E. Obst, Vol. II.) BROWN, R. J. E. 1969.Permafrost as an ecological factor in the Subarctic. Ecology of the subarctic regions. Proceedings of the Helsinki symposium / Écologie des régions subarctiques. Actes du colloque d’Helsinki. Paris, Unesco. (Ecology and conservation / Ecologie et conservation, I.) BUDYKO, M. I. 1955. Teplovoy balans zemnoy poverknosti. Leningrad Atlas. CREUTZBURG, N. ;HABBE, K.A. 1966.Klimatypen der Erde, M a p 1: 50,000,000.In: J. Blüthgen, Allgemeine Klimageogfaphie,2nd ed. DIETRICH, G. ; KALLEK. Allgemeine Meereskunde. Berlin. 492 p. (See also G.Dietrich, 1964.Ozeanographie. Braunschweig, 94 p.) DOLGIN, I. M. 1969. Sub-arctic meteorology. Ecology of the subarctic regions. Proceedings of the Helsinki symposium Ecologie des régions subarctiques. Actes du colloque d’Helsinki. Paris, Unesco. (Ecology and conservation / Écologie et conservation, I.) EKMAN, S. 1922.Djurvärldens utbredningshisïoriap i Skandinaviska Halvön. Stockholm, 614 p.) . 1944. Djur i de svenska fjällen. Stockholm, 428 p. (STFshandböcker o m det svenska fjället, vol. 3.) ENGLER, A.; DRUDEO. 1896-1928.Die Vegetation der Eide, vol. I-XV.Leipzig. FLOHN, H. 1950. Neue Anschauungen über die Allgemeine Zirkulation der Atmosphäre und ihre klimatische Bedeutung. Erdkunde, vol. 4,p. 141-162. --. 1951.Grundzüge der atmosphärischen Zirkulation und Klimagürtel. Wiss. Abh. Dt. Geogr.-Tag Frankfurt,p. 105-
_-
118. FORMOZOV,N. A. 1969.Écologie des plus importantes espèces de la faune subarctique./ Ecology of the subarctic regions. Proceedings of the Helsinki symposium / Écologie des régions subarctiques. Actes du colloque d’Helsinki. Paris, Unesco. (Ecology and conservation / Écologie et conservation I.) FRIES, Th. C. E. 1913.Botanische Untersuchungen im nöidlichsten Schweden. Ein Beitrag zur Kenntnis der alpinen und subalpinen Vegetation in Torne Lappmark. Uppsala. 361 p. GRIGORIEV,A. A. 1956.Subarktika. 2nd ed. Moscow. 223 p. HÄMETT-AHTI, L. 1963.Zonation of the mountain birch forests
32
in northernmost Fennoscandia. Helsinki. 127 p. (Ann. Bot. Soc. Zool. Bot. Fenn. Vanamo, vol. 34,no. 4.) HARE, F. K. 1950.Climate and zonal divisions of the boreal forest formation in eastern Canada. Geogr. Rev., vol. 40, p. 615-635. HASSINGER,H . 1933.Die Geographie des Menschen (Anthropogeographie). In: F. Klute (ed.), Handb. d. Geogr. Wiss. vol. Allg. Geographie II. Potsdam. p. 167-542. HETTNER, A. 1930. Die Klimate der Erde. Leipzig, Berlin. 115 p. (Geogr. Schriften, vol. 5.) . 1935. Die Pflanzenwelt. Vergleichende Länderkunde, vol. IV, part 5,p. 1-153.Berlin, Leipzig. HULTÉN, E. 1950.Atlas över växternas utbiedning i Norden1 Atlas of the distribution of vascular plants in N W . Europe. Stockholm. 512 p. . 1962. The circumpolar plants. I. Vascular cryptogams, conifers, monocotyledons. Stockholm. 275 p. (Kgl. Av. Vet.-Ak. Handl., Ser. IV vol. 8,no. 5.) HUBTICH, 1. 1939. Notes on the coniferous forest and tree limit in the east of Newfoundland-Labrador. Acta geogr., vol. VII, no. 1, p. 1-77.. . 1950. Notes on the forests on the east coast of Hudson Bay and James Bay. Acta geogr.,vol. XI, no. 1,p. 1-83. . 1958. On the recent expansion of the Scotch pine in northern Europe. Fennia, vol. 82,no. 3,p. 1-25. HUSTICH, I. 1960. Plant geographical regions. In: A. S ö m m e (ed.), A geography of Norden. Oslo. p. 54-62. . 1966.O n the forest-tundra and the northern tree limit. Ann. Uniu. Turku, vol. A, II, 36,41 p. (Reports from the Kevo Sub-arctic Research Station, vol. I.) KALELA, O. 1961.Seasonal change:of habitat in the Norwegian lemming, L e m m u s lemmus (L.).Helsinki, 72 p. (Ann. Acad. Scient. Fennicae, Ser. A, IV, vol. 55.) . 1963.Beiträge zur Biologie des Waldlemmings, Myopus schisticolor (Lillj.) Helsinki, 96 p. (Arch. Soc. Zool. Bot. Fenn. Vanamo, Suppl. 18.) KALLIO, P. 1964. The Kevo Subarctic Research Station of the University of Turku. Ann. Uniu. Turku, vol. A, II, 32,p. 9-40.(Reports from the Kevo Subarctic Research Station, vol. I.) KIHLMAN, A. O. 1890. Pflanzenbiologische Studien aus Russisch Lappland. Helsinki. 256 p. (Acta Soc. Fauna Flora Fenn., vol. VI, 3.) KIMBLE, G. H . T.; DOOD, D. (eds.). 1955. Geography of the Northlands. New York. 534 p. (Amer. Geogr. Soc., Special publ. no. 32.) KNAPP, R. 1965.Die Vegetation von Nord- und Mittelamerika und der Hawaii-Inseln.Stuttgart.373 p. KOPPEN, W.1900.Versuch einer Klassifikation der Klimate vorzugsweise nach ihren Beziehungen zur Pflanzenwelt. Geogr. Zeitschr., vol. 6,p. 593-611; 657-679. . 1918. Klassifikation der Klimate nach Temperatur, vol. 64, Niederschlag und Jahreslauf. Peterm. Geogr. Mitt., p. 193-203;243-248. . 1923,1931.Die Klimate der Erde. Grundriss der Klimakunde. Berlin, Leipzig. 369 p.; 2nd edition under the title: Grundriss der Klimakunde. Berlin, Leipzig, 1931, 388 p. ; GEIGER, R . 1928,1961. Klima der Erde -Climate of the earth. (Wall-map 1: 16,000,000.1st ed., Gotha, 1928; 3rd., Darmstadt, 1961.)
__
__
__ __ _-
-
--
--
__
Problems of definition and geographical differentiation of the Subarctic
L EMBHE, H . 1947. Die mittleren absoluten Maximaltemperaturen in Europa und den Mittelmeerländern. Erkunde, VOL 1, p. 184-189.) LUNDEGARDH, H . 1957. Klima und Boden in ihrer Wirkung auf das Pflanzenleben. Jena, 584 p. MIKOLA, P. 1969. Forests and forestry in subarctic regions. Ecology of the sub-arctic regions. Proceedings of the Helsinki symposium / Écologie des régions subarctiques. Actes du colloque d’Helsinki.Paris, Unesco (Ecology and conservation / Ecoiogie et conservation, I.) PASSARGE, S. 1921. Vergleichende Landschaftskunde. II. Kältewiisten und Kältesteppen. Berlin. 163 p. . 1923. Die Landschaftsgiirtel der Erde. Breslau. 144 p. PHILIPPSON, A. 1931, 1933. Grundziige der Allgemeinen Geographie. Vol. I, 2nd ed., Leipzig, 1933, 379 p.; vol. II, 2nd ed., Leipzig, 1931, 551 p. RITCHIE,J. C. 1960. The vegetation of northern Manitoba. Arctic, vol. 13, p. 211-229. ROUSSEAU, J. 1952. Les zones biologiques de la péninsule Québec-Labrador et l’hémoarctique. Canad. J. Bot., vol. 30, p. 436-474. . 1964. Coupe biogéographique et ethnobiologique de la péninsule Québec-Labrador. Vol. 2, p. 29-94. Paris. Ecole Pratique des Hautes Etudes à la Sorbonne, VIe section, Québec-Labrador et l’hémoarctique. Canad. Bot., vol. 30, p. 436-474. . 1964. Coupe biogéographique et éthnobiologique de la péninsule Québec-Labrador. Vol. 2, p. 29-94. École Pratique des Hautes Etudes à la Sorbonne, VIe section, Bibliothèque arctique et antarctique. SCHMITH~SEN,J. 1961. Allgemeine Vegetationsgeographie. Berlin. 262 p. (Lehrbuch d. Allg. Geogr., ed. by E.Obst, vol. IV.) SCHOTT, G. 1942. Geographie des Atlantischen Ozeans. 3rd ed. Hamburg. 438 p.
--
--
SJÖRS,H.1956. Nordisk väztgeografi.Stockholm. 229 p.
_- . 1965. Forest regions. In: The plant cover of Sweden. p. 48-63.Acta phytogeogr. suecica,vol. 50. TIKHOMIROV, B. A. 1960. Plantgeographical investigations of the tundra vegetation in the Soviet Union. Canad. J. Bût., vol. 38, p. 815-832. . 1967. Environment and the mode of plant adaptation to it in the Far North of U S S R . Oulon University, Finland Aquilo, Söyrinki’s Jubilee volume. TREWARTHA, G. T. 1954. A n introduction to climate. 3rd ed. London. 402 p. TROLL, C. 1941. Studien zur vergleichenden Geographie der Hochgebirge der Erde. Ber. 23. Hauptvers. Ges. Freunde u. Förderer Rhein. Fr.-Wi1h.-Uniu.Bonn, p. 49-96. . 1959. Die tropischen Gebirge. Ihre dreidimensionale klimatische und pflanzengeographische Zonierung. Bonn. 93 p. (Bonner Geogr. Abh. vol. 25.) .1965. Jahreszeitenklimate der Erde. D e r jahreszeitliche Ablauf des Naturgeschehens in den verschiedentn Klimagürteln der Erde/Seasonal climates of the earth. The seasonal course of natural phenomena in the different climatic zones of the earth. With m a p 1: 45 Mill. by C. Troll and K . H.Paffen: Jahreseeitenklimate der Erde / Seasonal climates of the earth. In: Weltkarten zur Klimakunde / World maps of climatology.2nd ed., p. 7-28.Berlin, Göttingen, Heidelberg: the coloured m a p also in J. Blüthgen, 1966, Allgemeine Klimageographie. WAHLENBERG, G. 1812. Flora lapponica. Berlin. 550 p. WALTER, H . 1954. Einführung in die Phytologie. III: Grundlagen der Pflanzenverbreitung. Einführung in die Pflanzengeographie. 2nd. part Arealkunde (Floristisch-historische Geobotanik). Stuttgart. 245 p. W I S S M A N N ,H . von. 1966. Die Klimate der Erde. M a p 1: 50 Mill. In: J. Blüthgen, 1966. Allgemeine Klimageographie.
_-
_-
_-
33 3
Forest limits as the most important biogeographical boundarvJ in the North B. A. Tikhomirov
The transitional,boundary belts or limits are inherent in every geographical phenomenon ; but the biogeographical limits have their o w n specific features. They are characterized, as a rule, by the complexity of the boundary line, the gradual transition1 from one biological complex to the others, the degradation of its typical features and gradual development of others. Here, besides the changes in the indices of the abiotic medium (climate, relief, soil, humidity, etc.) the complicated mechanism of interrelations between organisms and their relations with the medium interferes with the biogeocenological process which lies behind the population inconstancy of plant c o m m u nities. In this intricate complex-the complexity of which is not open to doubt-the important role belongs to the peculiarities of organisms and in the first place their capability for reconstruction and for the adaptive reactions, connected with the changeability and with the relative stability of the medium. The process of the adaptation of organisms to the changing and severe medium at their areas’ limits depends first on the hereditary basis of the organisms, the amplitude of intraspecific variability of the hybridogenous processes and also on the degree of influence of the external factors. Proceeding from these general presumptions let us examine the forest limits as the most important biogeographical limit in the North. Without going into details of the history of the development of this problem, it is worth mentioning that the problems of the forest7slimits,its characteristics, and dynamics, the interrelations between the forest and the tundra and the reasons for the absence of trees from the tundra called the attention of northern Russians and U.S.S.R. explorers more than a century ago. Scientists such as Maidel (1894), Kihlm a n (1890), Schrenk (1855), Middendorf (1867), Tanfil’ev (1911), Pohle (1903, 1917), Grigor’ev (19241,
’
Kaminsky (1924), Gorodkov (1929),Tsinzerling (1932), Tolmachev (1931), Sochava (1940), Leskov (1940), Govorukhin (1947,1956, 1963), Tyulina (1936, 1937), Medvedev (1943, 19521, Leont’ev (1948), Andreyev (1956),Tikhomirov (1953,1956,1962),Tyrtikov (1954), Norin (1961, 1962), Vasskovsky (1958), Kr’uchkov (1963),Yurtsev (1962, 1966), and others have made their contributions to this problem.2 It should be mentioned that several scientists in other countries also touched on these questions (Griezebakh, 1874; Hein, 1932; Nordenskjold, 1882 ; Roder, 1895; Wigge, 1927, and others), but special attention has been drawn during the last two decades to Hare (1950), Hustich (1939, 1950, 1953), Sjors (1963), Rousseau (1952), Sigafoos (1958), Hopkins (1959), Johansen (19631, R o w e (1959), Savile (1963), and others. A comprehensive analysis of the natural conditions influencing the northern forest limits is the main feature of such studies in the U.C.S.R. Lately the necessity of such complex analysis of factors is recognized also by a number of forest-limit investigators in other countries (Hustich,1953; Savile, 1963, and others). The maps of the northern limits of trees have already been made (Hustich, 1953; Tikhomirov, 1962; Vasskovsky, 1958, and others) and there is no need to discuss this problem, for the new data add nothing to the codguration of their areas. For the analysis of botanic-geographical relations in the Subarctic not only is it important to ascertain the northern limits of trees, but also, mainly, the characteristics of the nearly circumpolar transitional 1. This doee not except the sharp limite when Nome biological complexes pa6s into others in connexion with sharp changes in relief, humidity and other external fsctors. 2. Recently, the interest in these problems in the U.S.S.R.w a s especially increased in connexion with the problem of afforestationof the southern parte of the tundra and the preservation of the northern forest limits.
35
B. A. Tikhomirov
belt between the northern taiga and the tundra. This belt c a m e into Russian literature after the second part of the nineteenth century under the n a m e of lesotundru (“forest-tundra” in English “Wald-tundra” in German).l T h e forest-tundra m a y be a good example of a transitional biogeographic boundary, limit or belt of the Earth. I would like to mention s o m e general features of forest-tundra, as a world-wide biogeographical phenomenon. First its circumpolar distribution is noteworthy. T h e forest-tundra represents a zonal natural p h e n o m e n o n caused by the zonal climatic factors. As is generally k n o w n , the m a i n factor responsible for the zonal p h e n o m e n a o n the Earth is the uneven flow of solar heat energy to the different parts of the Earth’s surface. This factor resulting in zonality is inevitable. All other factors influencing the character of zonality merely intensify or distort its display o n the Earth. In particular, the botanic-geographical correlation of the northern Eurasian a n d northern American continents, as well as of the islands bordering o n them, is greatly influenced by the Arctic Ocean ice and the process of its melting. It is enough to mention that the ice area in the Arctic Ocean reaches 8.8 million km2 in winter with only 800,000 km2 less in s u m m e r (Zubov, 1945). T h e heat expended o n thawing of the ice, a n d the cold and humid winds blowing o n to the continent from the north, create unfavourable thermal conditions o n the land. O n e m a y recall the fact that the forest border runs parallel to the Arctic O c e a n shores :on the protruding part of the T a i m y r Peninsula are found the northernmost forest outposts (72040‘N.). On the contrary, the invasion of cold arctic water a n d ice into the H u d s o n Bay, as well as the long time during which the ice actually remains in the bay, bring the northern forest limit as low as 500 N. T h u s the difference between these t w o geographical limits o n the Earth, i.e., the northern forest limit and the southern tundra-limits, amounts to m o r e than 200 of latitude. In other parts of the world the northern forest limit, going generally parallel to the ocean shore-line,oscillates from 720 30’N. (lower reaches of the L e n a river) to 580 55‘ N. (the shore of the Okhotsk Sea). So the ice regions of the Arctic Ocean, the regimes of sea currents governing the water temperature, the processes of surface water evaporation and, finally, the air currents having their origin in the polar basin, form the complex of factors which play a very important role in fixing the limits of the northern forests. 1,ong ago one considered the forest limits to be roughly coincident with the July isoterm of 100 C as well as with the “isoline of K a m i n s k y ” representing the southern limits of areas with a relative humidity in day-time, in s u m m e r , of > 70 per cent. But in connexion with the m o r e detailed study of forest limits, of their configuration a n d also with the careful
36
study of climate regime characteristics in the north, considerable diversity between these lines can be found despite their general coincidence. Still m o r e careful studies a n d confrontations are necessary a n d these are being carried out partly by Hopkins for Alaska. T h e northern forest limit takes m a n y forms a n d it is likely that it follows the biocenological complex not as yet considered which corresponds best to the nature of the different species, forms and populations of trees. Again, the general factors restricting the development of forests in the north being deterioration of the temperature regime a n d shortening of the w a r m season, there is quite a complex of factors which determine the local microclimatic and ecological conditions which either favour the formation of forest communities or restrict their development. A m o n g the factors restricting the vitality of trees and limiting their northern boundary the following m u s t be mentioned : meso- a n d microclimate, permafrost and the initial process of soil formation, biocenotic factors (the influence of different autotrophic and heterotrophic plants and animals), biology and ecology of trees (biornorphs, fruiting process, the vitality of seeds, the survival of seedlings, etc.). Finally w e must mention the influence of m a n a n d domestic animals, this latter proving to b e a decisive factor in s o m e districts. Examining the structure of forest-tundra c o m m u nities in the old a n d the n e w world as well, one m u s t not forget that the m a i n feature which distinguishes t h e m from the northern taiga is the disturbance of cenotic entity or the loss of the role of edscators by trees. O n e must bear in mind that the loss of determinating role by the tree cover leads to s o m e “phytocenotic unsaturation” of habitats which leads, together with the solifluction a n d spot-forming processes (although weakened in the forest-tundra), to the introduction of boreal and arctic elements and to the formation of “mixed” forest-tundra flora. Despite the climatic deterioration in the foresttundra as compared with the taiga, it also possesses s o m e factors favourable to plant life. T h e absence of a dense tree canopy in forest-tundra enables the development of relatively high productive lichen communities which form the m a i n basis of the reindeer industry. It is k n o w n that the predominant part of lesotundru territory (85-90per cent) in the U.S.S.R.is covered with reindeer pastures. There are s o m e data showing that the fruit productivity by the berryshrubs is s o m e w h a t higher in the forest-tundra as compared with the taiga. 1. T h e concept of lesotundro as well as the views of Russian and Soviet scientists on its boundaries have been expresaed b y B. N. Norin (1961) in his paper. “Whet is the ‘lesotundra’?”.Therefore w e will not dwell upon these problems here. Many Soviet scientists are n o w inclined to distinguish forest-tundra as a separate “forest-tundra zone”, which includes the belt of northern parkland forests from the south and the belt of shruh-tundra from the north.
Forest limits as the most important biogeographical boundary in the North
Also the n e w data of Rodin and Basilevitch (Basilevich and Rodin, 1964; Rodin and Basilevich, 1965) reveal the considerable reserves of organic matter in the litter of forest-tundra communities. It is a good source of nitrogen. H o w e v e r the forest-tundra belt differs sufficiently from the treeless tundra in its life condition complex. W i n d s of great velocity (often the storms m o v e at 35 m/s) prevail in the tundra while in forest-tundra they are abated by separately standing trees. T h e m e a n annual wind velocity in the forest-tundra is 6 m/s, increasing in the tundra to 8 m/s. In the tundra the poor s n o w cover is thickened by the strong winds and reaches a compactness which restricts the use of under-snow fodder by reindeer. In the forest-tundra the s n o w cover, although thicker, is friable, making fodder m o r e accessible. T h e air temperature is higher in the forest-tundra. W h e r e s u m m e r thawing of the ground layer is thicker, the microbiological processes are m o r e active and result in the formation of comparatively m o r e fertile soil. T h e latter brings about a m o r e vigorous vegetation. Owing to the above-mentioned differences between the natural conditions of tundra a n d foresttundra the life conditions for plants in t h e m differ considerably. Also the conditions for animals are rather unfavourable in winter whereas the forest-tundra in winter is the life area of numerous animals. It must not b e forgotten that open-ground agriculis connected to a certain extent ture in the U.S.S.R. with the northern forest and parkland boundary. So it is of importance to emphasize that the forest limits and the forest-tundra belt itself represent a very significant biogeographical boundary a n d is also of paramount importance for the people’s economy. It is k n o w n that the life a n d e c o n o m y of m o r e than twenty aboriginal peoples are closely connected with the northern taiga parklands, the forest-tundra a n d its outpost. O n e of the lesotundra characteristics is the peculiarity of w o o d y plant biomorphs, this being circumpolar also. In the north the trees lose their usual form with a n orthotropically growing trunk, it being replaced there by the different modifications of this biomorph (such as trees with plagiotropically or semi-plagiotropically oriented stems, trees with curved tops, flag-shaped forms, bow-trunked ones, khodylni, “trees in skirts”, semi-prostrate or prostrate trees, table-shaped a n d trellis-shaped forms, literally growing into the moss cover (Larix dahurica,for instance). T h e c o m m u n i t y of biomorphs all over the circumpolar area of forest-tundraindicates the approximately similar natural influences o n the different tree species and nearly equal response of organisms to these influences in all parts of it. Considering the forest-tundra as a world-wide botanic-geographical p h e n o m e n o n w e must note the
various cenomorphological structures of its c o m m u nities in the Old and the N e w World, which are rather alike w h e n compared with each other. T h e parklands (redkoles’ja),forest islets, well-spaced trees with their various morphological forms, all are extremely alike in the forest-tundras of both the Old and the N e w World. There is n o doubt that a careful and detailed analysis of the structure and composition of plant communities would result in the revelation of m a n y features both of similarity and of difference. T h e latter depends first and foremost o n the fact that the forest limit is formed by different tree species in different parts of the world. In the Atlantic Arctic the meadow-forests of Betula tortuosa are widely distributed; o n the sand soils without permafrosts Pinus silvestris forms the forest limit. In the north of the European part of the U.S.S.R., side by side with birch, the following c o m munities grow in the forest-tundra: Picea abies (western section), Picea o bovata and Larix sukaczevii (eastern section), a n d in western Siberia, Larix sibirica which is replaced by Larix dahurica east of the Jenissey. T h e last species forms the northern forest limit throughout all eastern Siberia. H o w e v e r in Translenian Siberia the pattern of the northern forest is interspersed with Pinus pumila in the mountain habitats and with Chosenia macrolepis and Populus suaveolens in the valleys. These trees of angaroberingian origin give distinction to the forest communities of the north-eastern U.S.S.R.In the north of the n e w world w e meet with that specific group of trees which forms the northern forest limits. Picea glauca m o v e s northward along the well drained habitats and, as w a s recently shown, does not lose its normal shape up to the northern limits. Picea mariana is distributed over the wet habitats, the bogs; in the north it has a shrubshape. T h e northern limits of w o o d y vegetation are also reached by Larix laricina. It has a very wide ecological range, inhabiting s w a m p s a n d sand dunes as well, where it forms the lichen-larix communities greatly resembling the North Siberian larix forests (bory) composed of Larix sibirica. Proceeding from the analysis of the above-mentioned data, the necessity arises for dividing the forest-tundra belt of the Earth into sectors or segments according to the trees which form the forest limit of the areas. T h e forest limit as well as the width of the forest-tundra belt did not remain stable during the Quaternary period, the Post-glacial in particular. T h e dynamics of the northern biogeographical limits has been discussed repeatedly (Griggs, 1934 ; Tyulina, 1936, 1937 ; Tikhomirov, 1953 ; Andreyev, 1954, 1956 ; Hopkins, 1959, and others) and there is n o need to dwell o n this question. Let us point out however, that all the data concerning several decades of forest advance into the tundra must be carefully studied
37
B. A. Tikhomirov
a n d used to elaborate the theory of the forest-tundra biogeocenoses transformation under the climatic amelioration. T h e necessity of forest plantations in the forest-tundra a n d in the southern tundra has b e c o m e imminent. T h e theoretical possibility of overcoming the absence of trees in the tundra is being translated into reality by the creation of tundra shelter belts. I would like to end the present report with a brief enumeration of the m a i n tasks confronting those investigating the problems of forest-tundra a n d forest a n d tundra interrelations: 1. T h e determination of “forest-tundra’’ (lesotundra) and “parkland” (redkoles )je) conceptions and the geographical boundaries of these types of vegetation. T h e detailed geobotanical characteristics of the different parts of the forest-tundra. 2. T h e biological peculiarities of the m a i n northern trees at their northern limits and defining exactly their t a x o n o m y and distribution limits. 3. T h e characteristics of the structural-biocenotical connexions in different plant communities of forest-tundra and forest boundary (insular treegroves, groups of shrubs, mosses, lichens, herbaceous plants). In particular the bioecological, phytocenological a n d climatological characteristics of forest-margins o n the northernmost outposts of forest with the a i m of revealing their role in the conquest of tundra by forest. 4. T h e role of animals in the life of forest-tundra,in the dynamics of the northern limits of forests and the study of this role (organization of complex biogeocenological investigations). 5. T h e elucidation of the ecologic-physiological, biological and cenological reasons for the absence
of trees from the tundra with extensive experiments at the northern forest limits. 6. T h e elucidation of man’s role in the dynamics of the northern limits of trees and forest-communities. 7. T h e elucidation of the history of the forest-tundra vegetation on the basis of palaeobotanical data. 8. T h e study of forest-tundra peculiarities in connexion with the origin of hypoarctic elements and forest limits’ m o v e m e n t in post-glacial times. 9. T h e plant resources of forest-tundra and the w a y s of making rational use of them. 10. T h e problem of forest preservation at the extreme limit. T h e precise fixation of separate forest islands in the tundra with a view to their preservation and observation. T h e organization of strictly reasoned out biological “bench-marks”, and of a n u m b e r of reservations a n d forbidden territories at the forest limits. 11. T h e problem of reafforestation, raising the biological productivity a n d increasing trees, growth rate o n the forest-limit line as well as the artificial advancement of forest-tundra limits to the north (creation of shelter belts). 12. Botanic-geographical subdivision of the foresttundra territory. 13. T h e elaboration of measures o n the rational use of the forest-tundranatural resources. T h e outlined range of problems is so extensive that it can be achieved only by the combined efforts of m a n y scientists. This calls for the organization of stations for long-term biogeocenological investigations. Most desirable is the organization of a network of stations at the northern forest limits in order to put the study of all the above-mentioned problems o n a strictly experimental basis.
Résumé Importance des limites forestières en tant que frontière biogéographique dans le Nord (B.A. Tikhomirov)
Les limites forestières constituent la frontière biogéographique la plus importante dans le Nord. L’étude de ce problème nécessite un e x a m e n approfondi des caractéristiques de la zone de transition presque circumpolaire entre la taïga septentrionale et la toundra. D e n o m b r e u x spécialistes soviétiques considèrent cette zone (la toundra forestière) c o m m e u n e zone de toundra forestière distincte, qui comprend la zone
38
de la forêt-parc en venant du sud et la zone de la toundra à arbrisseaux en venant du nord. Les facteurs climatiques jouent le rôle le plus important dans le développement de la toundra forestière. Mais il faut tenir compte de la structure et de la composition des c o m m u n a u t é s végétales. L a zone de la toundra forestière doit être divisée e n secteurs selon les arbres qui constituent la limite forestière des territoires. L’étude de ce problème pourra progresser dans l’avenir si l’on entreprend dans des stations des recherches biogéocénologiques de longue haleine.
Forest limits as the most important biogeographical boundary in the North
Bibliography
/ Bibliographie
AKIMOV, A. T.; BRATZEV, L. A. 1957. K voprosu o nastuplenii lesa ha tundru (On the problem of forest advancement into the tundra). Izvestija K o m i $liala Vsesoijusnogo Geograficheskogoo btshestva. ANDREYEV, V. N. 1954. Prodvizhenie drevesnoi rastitelnosti v tundru v svyasi s zatshitnymi svoistvami lesoposadok na severe (Extension of arboreous plants into the tundra as related with the sheltering properties of tree stands in the north). Botanichesky Zhurnal, vol. 39, no. 1. . 1956. Zaselenie tundry lesom v sovremennoyu epokhu (The contemporary afforestation of the tundra). Zbornike: Rastitelnost’ Krainego severa i ee osvoyenie,vyp. I. (Symposium on: the vegetation of the far north of the U.S.S.R. and its utilization,part I). BASILEVICH,N. I.; RODIN, I,. E. 1964. Krugovorot zolnykh elementov i azota v osnovnykh tipakh rastitelnosti severnogo polusharia Starogo Sveta (The turnover of the minerals and the nitrogen in the main types of vegetation in the northern hemisphere of the Old World). Doklady Akademii Nauk, vol. 156, no. 1. DYLIS, N. V. 1947. Sibirskaya listvennitsa (Siberian larch). Materialy k poznani’u ffauny i jlory SSSR, novaia seria, otdel botaniki, 2 (10) (Papers to the knowledge of the fauna and flora of U.S.S.R. New ser. Bot. division,vol. 2 no. 10). GORODKOV, B. N. 1929. Bezles’e tundry (Botanico-geograficheskie nabludenia v Zapadnoi Sibiri) (The treelessness of tundra (Botanico-geographical observation in the Western Siberia) ) . Priroda, vol. 3. GOVORUKHIN, V. S. 1947. Dinamika landshaftov i klimaticheskie kolebania na krainem severe. (Landscapes dynamics and climate fluctuations in the Far North). Izvestia Geogr.obtshestva SSSR,vol. 79, no. 3. . 1956. Redkoles’ya na poljarnom predele lesov Malogo Yamala i v Priobskoi lesotundre. Zemlevedenie. (Forest parklands at the northern forest limit in the Yamal Peninsula and the forest tundra of the lower reaches of O b river. Land Science, New Ser., vol. 64). . 1963. Lesotundra kak fisiko-geograficheskaya zonal’ naya oblast’ (Forest-tundra as a physico-geographical zonal region). Problemy Severa,vol. 7. GRIEZEBAKH, A. 1874. Rastitel’nost’ Zemnogo shara. (Earth vegetation), vol. I. GRIGGS,R. 1934. The edge of the forest in Alaska and reason for its position. Ecology, vol. 15, no. 2. GRIGOR’EV,A. A. 1924. Poljarnaya granitsa drevesnoi rastitelnosti v Bolshezemelskoi i nekotorykh drugikh tundrakh (faktory ee obuslovliva’utshie i kolebania ee v blizhaiishuju k n a m epokhu) (The polar tree limits in the Bolshecemelskaja tundra and some other tundras (the factors influencing it and its recent fluctuations)). Zemlevedenie, vol. 26, no. 1-2. . 1946. Subarktika (Subarctic). HARE, F. K . 1950. Climate and zonal divisions of the boreal forest formation in eastern Canada. Geogr. Rev., New York, vol. 40, no. 4. HEIN, L. 1932. Die polare Waldgrenze in Europa. Beih. z. Bot. Central. blatt, vol. 49, abt. 2. HOPKINS, D.M. 1959. Some characteristics of the climate in forest and tundra regions in Alaska. Arctic, vol. 12, no. 4
--
__
__
-_
HUSTICH, I. 1939. Notes on the coniferous forest and tree limit on the coast of Newfoundland-Labrador.Acta geogr., vol. 7, no. 1. HUSTICH, I. 1950. Notes of the forests of the east coast of Hudson Bay and James Bay. Acta geogr., vol. 10, no. 2. . 1953. The boreal limits of conifers.Arctic, vol. 6, no. 2. -. 1966. O n the forest-tundra and the northern tree lines. Ann. Univers. Turku, ser. A, II, vol. 36. JOHANSEN, H. 1963. Zoogeographical aspects of the birds of subarctic. Proc. 13 Int. Ornitol. Congr. KAMINSKY, A. A. 1924. Klimaticheskie oblasti Vostochnoi Evropy v sv’asi s rasprostraneniem lesov (East European climatic regions as related to the forest limits). Trudy po lesnomu opytnomu delu v Rossii, vol. 44. KIHLMAN, A. C. 1890. Pflanzenbiologische Studien aus Russisch Lapland. Acta Soc. pro fauna et flora fennica, vol. 6, no. 3. KR’UCHKOV, V. V. 1963.O vosmozhnosti oblesenia iuzhnykh kustarnichkovykhtundr. (Possibilities of southern shrubby tundras afforestation). Problemy Severa, vol. 7. LEONT’EV, F. S. 1948. Severnyi predel rasprostranenia listvennykh porod na Omolono-Tshukotskom Dalnem Vostoke. (The northern tree limit in the Omolon-Chukotsk far east). B’ulleten Moskovskogo Obtshestva Ispytatelei Prirody, Otdel Biologii, vol. 4. LESKOV, A. I. 1940. Fitotsenologicheskii ocherk redkolesii basseina reki Poluia (Phytocoenologicalessay of parklands in the Polui river basin). Trudy Botanicheskogo instituta im.Komarova A N SSSR, ser. 3, Geobotanika, vol. 4. MAIDEL, G. 1894. Puteshestvie po krainemu severo-vostoku Yakutskoi oblasti v 1868-1870 g.g. (The journey through the far north-east of the Yakutsk region in 1868-1870). MEDVEDEV, P.M . 1943. O predele lesa na severo-vostochnom poberezh’e Okhotskogo mor’a (Forest limits on the northeastern coast of the Okhotsk sea). Botanicheskij Zhurnal, vol. 2. . 1952. O predele lesa i prichinakh beslesia tundr Asii (Northern forest limit and the causes of the absence of trees in the Asiatic tundras). Isvestia Vsesoiusnogo Geographicheskogo obtshestva,vol. 84, no. 3. MIDDENDORF, A. F. 1867. Puteshestvie na sever i Vostok Sibiri, vol. I, part 4, Rastitel’nost’ Sibiri (Reise in den aussersten Norden und Osten Sibiriens während der Jahre 1843 und 1844, vol. I, part 4, Sibirishe Vegetation). NAUMENKO, Z. M. 1964. Picea obovata (Led.) na severovostochnom predele ee areala. (Picea obovata (Led.) on the north-eastern limits of its area). Botanicheskij Zhurnal, vol. 49, no. 7. . 1966. Severnyi predel rasprostranenia daurskoi listvennitsy Larix gmelini v Prikolymskikh raionakh (The northern limits of the dahurian larch Larix gmelini in the Kolyma districts). Botanicheskij Zhurnal,vol. 51, no. 1. NORDENSKJOLD, A. E. 1882. Die Umsegelung Asiens und Europas auf d. Vega. NORIN, B. N. 1961. Chto takoe lesotundra? (What is the “forest tundra”? Botanicheskij Zhurnal, vol. 46, no. 1. . 1962. O kompleksnosti i mozaichnosti rastitelnogo pokrova lesotundry (Complex and mosaic pattern of the forest-tundra vegetation). Problemy botaniki, vol. 6.
-
_-
__
39
B. A. Tikhomirov
POHLE, R. 1903. Pflanzengeographische Studiens uber die Halbinsel Kanin und das angrenzende Waldgebiet. Trans. St. Peters. Bot. Gard., vol. 21. . 1917. Wald- und Baumgrenze in Nord-Russland. Zeitsch. der Geselschaft fur Erdkunde zu Berlin,vol. 4. RODER, K . 1895.Die polar Waldgrenze. Dresden. RODIN, L.E.;BASILEVICH,N. I. 1965. Dinamika organicheskogo vetshestva i biologicheskii krugovorot v osnovnykh tipakh rastitelnosti (The dynamics of organic matter and the biological turnover in the main vegetation types). ROUSSEAU, J. 1952. Les zones biologiques de la péninsule Québec-Labrador et l’hémiarctique. Canad.J. Bot., vol. 30. ROWE, I. S., 1959. Forest regions of Canada. Ottawa. SAMBUK, F.V. 1937. Predely lesov na Taimyre (Forest limits on the Taimyr peninsula). Botanicheskij Zhurnal, vol. 22, no. 2. SAVILE, D. B. O. 1963. Factors limiting the advance at Great Whale River, Quebec. The Canadian Field Naturalist, vol. 77, no. 2. SCHRENK, A. G.1855. Puteshestviepo severo-vostokuevropeiskoi Rossii cherez tundru samojedov k severnym Ural’skim goram, predprin’yatoe w 1937 godu, vol. I (Reise nach d e m Nordosten des europaischen Russlands durch die Tundren der Samojeden z u m Arktischen Uralgebirge im Jahre 1837 ausgefubrt, vol. I). SIGAFOOS, R. S. 1958. Vegetation of northwestern North America as an aid in interpretation of geological data. (U.S. Geol. Surv. Bull.) SJORS,H . 1959. Bogs and fens in the Hudson Bay lowlands. Arctic, vol. 12, no. 1. . 1963. Amphi-atlantic zonation, nemoral to arctic. In: North Atlantic biota and their history. Oxford. SOCHAVA, V. B. 1940. O beslesii tundr. (On the absence of trees in tundras). Trudy Leningradskogo obtshestva estestuoispytatelei, vol. 68, no. 3. TANFIL’EV,G. I. 1911. Predely lesov v polarnoi Rossii (The timber line in polar Russia). Odessa. TIKHOMIROV,B. A. 1953. Besles’e tundry i ego preodolenie (The absence of trees in tundras, its causes and h o w to overcome it). Botanicheskij Zhurnal, vol. 4. . 1956. Les i tundra (The forest and the tundra). Priroda, vol. 7. . 1962. Besles’e tundry, ego prichiny i peti preodolenia (The absence of trees in tundras, its causes and the ways of overcoming it).
-_
--
-
40
--. 1967. Environment and the mode of plant adaptation to it in the far north of the U.S.S.R.Aquilo, ser. Botanica, vol. 6, p. 104-112. (Oulou, Finland, Societas Amicorum Naturae.) TOLMACHEV, A. I. 1931. O rasprostranenii drevesnykh porod i o severnoi granitse lesov v oblasti mezhdu Eniseem i Khatangoi (On the trees distribution and the northern forest limits in the region between Yenissei river and Khatanga river). Trudy Polarnoi Komissii, vol. 5. TSINZERLING,Y. D. 1932, Geogra6ya rastitei’nogo pokrova severozapada Evropeiskoi chasti SSSR. (The vegetation geography in the north-western part of the European Russia). Trudy SOPS AN SSSR,Kolskaya seria, vol. 10. TYRTIKOV, A. P. 1954. Rost kornei derev’ev v dlinu na severnom predele lesov (The longitudinal increment of roots on the northern timber line). Bulleten’ Moscovskogo Obtshestva Ispytatelei Prirody, Otdel Biologii, vol. 9, no. 1. TYULINA, L. N. 1936. O lesnoi rastitel’nosti Anadyrskogo kraya i ee vsaimootnosheniyakh s tundroi (On the forest vegetation of the Anadyr Territory and its relation to the tundra). Trudy Arkticheskogo Instituta, vol. 40. . 1937. Lesnaya rastitel’nost’ Khatangskogo rayona u ee severnogo predela (Forest vegetation of the Khatanga district at its northern limit). Trudy Arkticheskogo Instituta,vol. 43. VASSKOVSKY,A. P. 1958. Novye dannye o granitsakh rasprostraneniya derev’ev i kustarnikov-tsenosoobrasovateleina krainem severe-vostoke SSSR (Some new data on trees and coenoso-forming shrubs in the far north-east of U.S.S.R.). Materialy PO geologii i poleznym iskopayemym Severo-Vostoka SSSR, n. 13. (Materials on the geology and useful fossils of the north-eastern U.S.S.R., no. 13, Magadan.) WIGGE. 1927. Die Tundra als Landschaftsform. Koln. YURTSEV, B. A. 1962. Botaniko-geograficheskienabl’udenia u severnogo predela rasprostraneniya listvennitsy (Botanico-geographical survey of the northern limits of the larch). Problemy Botaniki, vol. 6. . 1966. Gipoarkticheskii botaniko-geograficheskiipoyas i proiskhozhdenie ego flor (Hypoarctic botanico-geographical belt and the origin of its floras). Komarovskie chtenia,vol. 19. ZUBOV, N.N. 1945. L’dy Polarnogo basseina (Polar basin ices).
__
__
Subarctic meteorology 1. M. Dolgin
F r o m the standpoint of making the most efficient use of available natural resources in agricultural development, cattle-breeding and fishery in the subarctic regions,the study of the meteorological régime of this region is of special importance in so far as it affects the activity of animal and plant life as well as the economic activity of man. The Subarctic is a region with a unique nature, the climate of which forms mainly under the influence of the air of the arctic and of temperate latitudes. During recent years and especially in connexion with the realization of a wide complex of research associated with the IGY and IQSY programmes, new data have appeared which w ill be helpful in extending our knowledge of the meteorological régime of the region under consideration. Unfortunately, the microclimate of this region remains nearly unknown. It should be hoped that in realizing the recommendations of the symposium on the ecology of the Subarctic ill be filled up. the existing gap w While preparing this paper w e have come across some difficulties connected with some uncertainty as regards the subarctic boundaries. S o m e definitions of this region m a y be found in references, but they considerably differ from each other. It is true that it must be taken into account that to m a k e such a division into regions is a complicated problem because it can be done only in relation to all the complex of natural conditions which remain to be studied fully. W e have taken the boundaries of the Subarctic as suggested by the academician Grigor’yev (1956) and think that these are the most appropriate (Fig.1). In view of the great longitudinal extension of the Eurasian Subarctic, the underlying surface and circulation conditions and related weather conditions widely differ. Therefore, the meteorological régime is considered in three individual areas: the western area, which includes the territory as far as Taimyr Penin-
sula, the central area covering the territory from Taimyr Peninsula to Kolyma and the eastern area including Chukchi Peninsula and north-easternSiberia.
PECULIARITIES OF THE ATMOSPHERE CIRCULATION The Subarctic is the zone of the most frequent encounters between the arctic air and the air masses of temperate latitudes. That is the reason for the occurrence of the arctic front with different positions in summer and in winter. Using the synoptic data obtained for the last years the front average winter and summer position has been specified in the Arctic and Antarctic Research Institute. In Eurasia its Atlantic European branch approximately coincides with the position determined earlier by Peterson (1961) and Khromov (1948). T h e Asian branch considerably differs. Its summer and winter positions are given in Figure 2. As far as the difference in thermal properties of the continental air of Siberia and of the arctic air is not great in the winter, then the Asian branch of the front in January can be hardly traced. At the arctic front cyclonic activity of varying intensity develops. It is with the Atlantic European sector that the occurrence and active development of cyclones is associated, and within the Asian sector the regeneration of the polar frontal cyclones in the subarctic region (Chukanin,1965) occurs more frequently. In the western area the more frequent recurrence, from four to five per month, of cyclones is typical of the cold period of the year. Their prevailing direction is towards the western area. Over the central and eastern areas their number decreases to two to three per month. In summer over the western area there are three to four per month, but the intensity and velocity of their movements is considerably less than in winter.
41
I. M. Dolgin
Arctic deserts ond siaciers
0
D A r c t i ct u n d r a s Typical ( m o s s and
lichen)
~ ~ n h ~ ' h i l l o c ktundras
Foot-hills' foresttundras. sparsely w o o d e d
stole O 200 400 600 800 km
FIG.1. Subarctic boundaries.
FIG.2. Average position of frontal zones.
42
Subarctic meteorology
In the central and eastern areas of the Subarctic the number of cyclones somewhat increases (four to five per month). They penetrate into the eastern area primarily from the northern part of the Pacific Ocean and into the central area from the south-west,from the central areas of Yakutiya. In the eastern area the anticyclonic circulation predominates over the whole territory in winter, with as m a n y as three anticyclones with low movement and considerable stability. As a consequence of the high recurrence frequency of cyclonic activity in the Subarctic there occur frequent intrusions of cold air masses into the rear of cyclones (in the region of the Subarctic), sharp changes of temperature and winds, frequent occurrence of lower clouds and precipitation (Ragozin and Chukanin,
1961). Peculiarities of circulation processes over the Subarctic are determined by macrosynoptic processes over the Northern Hemisphere. With the change of the type of a macro process the circulation patterns should change considerably over both areas. Apart from average conditions, some aspects of the atmosphere régime are discussed according to the circulation types. For this purpose w e have used theclassification suggested by the late Professor Vangengeim (1961, see also Girs, 1960a). Taking into account the largest features of the atmosphere circulation, Vangengeim established three forms or types
eastern of circulation over Eurasia: the western (W), meridional (C). Each of the three types is characterized by special conditions of the thermal pressure field,by the distribution and value of temperature contrasts, by the distribution of the surface and high altitude pressure fields, etc. During the W type of circulation waves of small amplitude, moving quickly from west to east, are observed at heights. The shifts of cyclones from the Atlantic Ocean to the east are observed at the surface. During the C and E types of circulation the west to east transfer ceases, because of the development of the high-amplitude stationary waves in the troposphere,the geographical position of which with the E and C types of circulation is nearly reversible. There, where with the C type ill be a stationthere is a ridge,with the E type there w ary trough (Fig.3). D u e to this the regions of positive and negative pressure and temperature anomalies during the W type have a zonal distribution and with the E and C types a meridional one. Studies made by Girs (1960b) have shown the existence of the similar circulation types in the Pacific American sector of the Northern Hemisphere and their relationship with the processes over Eurasia. The relation of circulation conditions over the Pacific Ocean to the considered main circulation forms over Eurasia is of primary importance for the eastern Subarctic.
(E)and
- --- -
FIG. 3. Position of ridges and troughs: -a (W), (E), (Cl. (After Gir~.)
-.
43
I. M . Dolgin
T H E RADIATION CONDITIONS O F T H E SUBARCTIC
TABLE1. Total radiation in the Subarctic at northern (N) and southern (S) boundaries of each area (kcai/ cm2/mo)
Nearly the whole of the Subarctic, except for its easternmost regions, is situated to the north of the Arctic Circle. In the south of the western area of the Subarctic the length of the polar d a y period is about a month, with the sun's m a x i m u m altitude at 400-450 at n o o n in June. In the north, in the central part of the T a i m y r Peninsula, the length of the polar day period increases up to 102 days (late April to midAugust). There the sun's altitude is not higher than 38030'. T h e length of the polar night period is about three months (early N o v e m b e r to early February). A s total radiation (Q)is determined by the length of the d a y time, then in January and December it is near O. In the period of the change of the d a y and night the distribution of Q is mainly of latitudinal character. During the polar d a y from M a y to August, Q is different at the s a m e latitude, due to the irregular distribution of cloudiness. As seen from Figure 4, Q increases to the east with m a x i m u m between h = 150~-170~ E, where cloudiness is usually small. T h e annual variation of Q for all the regions is given in Table 1, and in Figure 5 for the eastern area. T h e increase of Q to the north in M a y and June is due to the increase of diffuse radiation caused by the persistence of s n o w cover. T h e northern part of the western area receives about 70 kcal/cm'/yr, the southern one about 75 kcal/cmZ/yr a n d the eastern one 75 to 80 kcal/cm2/yr (Fig.5). Because of the high frequency
Area
January February March April May June July August September October November December
TOTAL
o.1 0.8 4.7 10.0 13.5 13.4 11.9 7.4 3.6 1.3 0.3 0.0
67.0 78.4
0.0 0.3 3.4 9.7 15.5 15.2 12.6 7.6 2.8 0.7 0.0 0.0
0.0 0.7 4.5 10.4 14.8 14.8 13.0 7.9 3.6 1.2 0.1 0.0
--
67.8 71.0
0.1 1.3 5.6 11.0 14.6 14.9 12.0 8.4 4.1 1.8 0.4 0.0
0.7 2.2 7.1 11.5 14.3 14.5 11.7 8.8 5.0 2.9 1.0 0.3
- -
74.2 80.0
of cloudiness and its relatively small vertical thickness a n d due to a long persistence of s n o w cover diffuse radiation amounts to over. 50 per cent of total radiation. In spite of the northern position of the subarctic region the income of total radiation for spring and s u m m e r is rather high. In M a y a n d J u n e it exceeds the monthly values at m o r e southern latitudes.
FIG.4. Total radiation (Kcal/cm2/mo),July.
44
0.1 1.2 5.4 11.4 14.1 14.6 15.5 9.7 3.7 1.8 0.8 0.1
Subarctic meteorology
.I -1 . \ ' 3"'.-.-. ././ ---:y\ -2. ./-./---
Western area of the Subarctic Eastern area of the Subarctic Leningrad Vladivostok
radiation is about O, then in s u m m e r it amounts to 9-10kcal/cm2/mo (Table 3, Fig. 7).
May
June
13.8 14.5 11.6 13.2
14.0 14.7 13.3 11.6
F o r the subarctic ecology the absorbed radiation, determined by the reflective power (albedo) of the underlyingsurface is of primary importance. During the persistence of s n o w cover (7-8months) the albedo of tundra is 85-75 per cent. In June a n d September the value of albedo from year to year can sharply change depending u p o n the dates of the disappearance and appearance of persistent s n o w cover. During the snowless period the albedo of tundra is rather stable: in July and August over the whole of the Subarctic it varies in the range of 9-23 per cent depending o n the character of the underlying surface (Table 2,
Fig. 6). On the
average, for extensive territories, the most important for estimating albedo will be the area of lakes, the degree of tundra wetness and s n o w precipitation.
A n n u a l values of absorbed radiation increase from west to east at the northern boundary of the area from 34 to 38 kcal/cm2/yr, a n d at the southern boundary from 37 to 46 kcal/cm2/yr. Thus, the Subarctic utilizes little of the incoming heat a n d that is one of the m a i n reasons for the severity of the climate of this region. T h e loss of heat by the underlying surface as a result of radiation is, o n the average, about 32 per cent of total radiation for the year. F o r the considered areas of the Subarctic as a result of various combinations of the a m o u n t of cloudiness a n d inversions the annual value of effective radiation differs rather slightly according to regions, e.g., in the west of the area higher frequency of cloudiness a n d lower frequency of inversions than over the central a n d eastern areas are observed. In the north effective radiation ranges from 23 kcal/cmz/yr in the west to 24 kcal/cm2/yr in the east. A t the southern boundary in the east it amounts to 26 kcal/cm2/yr.
45
I. M. Dolgin
FIG. 6. Total radiation (kcal/cmz/year).
FIG.7. Absorbed radiation (kcal/cm2/rno), July. Within the annual variation due to the recurrence of cloudiness, minimum effective radiation is observed in the s u m m e r months. T h e listed components enable us to calculate the radiation balance for the region under study. T h e annual radiation balance for the whole of the S u b arctic is positive a n d amounts at the northern boundaries to 10 kcal/cm2/yr in the west and 14 kcal/cm2/yr
46
in the east. At the southern boundaries from the central to the eastern areas it increases from 12 to 20 kcal/cmz/yr. Table 4 gives the annual variation of the radiation balance (see also Fig. 8). It follows from the table that a positive balance over the whole territory is observed from M a y to September and only in the south-east from April. T h e m a x i m u m balance values occur in July and are 8-9 kcal/cmZ/yr.
Subarctic meteorology
FIG.8. Radiation balance (kcal/cm2/year). TABLE4. Radiation balance of the Subarctic at northern (N) and southern (S) boundaries of each area (kcal/cm2/mo) Area
January February March April May June July August September October November December
-2.0 -1.6 -1.3 -0.1 1.5 6.1 1.7 4.4 1.2 -1.7 -2.2 -2.2 -,
TOTAL
10.4
-2.4 -2.1 -1.7 -0.2 1.2 6.3 8.4 4.5 0.1 -1.9 -2.3 -2.5
7.4
-2.5 -2.1 -2.0 0.0
1.7 8.0 9.3 4.8 0.6 -1.6 -1.7 -2.4 _ j .
12.1
-2.5 -1.8 -1.3 -0.1 2.6 1.5 8.5 5.4 1.4 -1.2 -1.8 -2.2
14.5
-2.5 -1.7 -0.4 1.3 3.9 8.8 1.7 5.4 2.1 -0.5 -1.5 -2.5 ~
20.1
T h e given results o n the radiation régime of the Subarctic are obtained from research data of the Arctic and Antarctic Research Institute (Chernigovskiy and Marshunova, 1965) a n d the M a i n G e o physical Observatory (Barashkova et al., 1961) in Leningrad. T h e albedo characteristics include materials obtained b y the Flying Meteorological Observatory in the Arctic (Kopter, 1961). Radiation balance, as w e k n o w , determines to a considerable extent the heat budget of the given area. T h e latter is also determined by the turbulent flux a n d evaporation value. T h e annual variation of the
components of the heat balance is s h o w n for the northern and southern boundaries of the western area in Figure 9.As seen from this figure, during the period w h e n the radiation balance is negative the turbulent flow is directed to the underlying surface. On the average for the whole of the Subarctic this is observed from October to March. M a x i m u m values of turbulent heat flow occur during J u n e a n d July ( w h e n radiation balance is the most) a n d amount, as is s h o w n by M. I. B u d y k o (1956),to 1.5-2.5 kcal/cm2/mo,with a slight height change. Only a slight increase of annual values of turbulent heat flow to the south is found: from 3 kcal/cm*/yr in the north, to 5 kcal/cm2/yrin the south. T h e evaporation heat loss is O during the period of negative radiation balance (from October to April). A m a x i m u m value of the evaporation heat loss occurs in July, 3.5 kcal/cm2/moin the north of the area a n d 5.5 kcal/cmz/mo of the southern boundary. These values change with longitude slightly. A n n u a l values of the evaporation heat loss in the north of the area a m o u n t to 8-9 kcal/cmZ/yr a n d in the south to 1315 kcal/cm2/yr. Girs considered the change of radiation conditions of the Arctic and Subarctic in relation to the change of the m a i n forms of the atmosphere circulation. Here is a brief outline of his results. W i t h the W type of circulation as stated above, a weakening of interlatitudinal exchange occurs, the activization of cyclonic activity at temperate latitudes is characteristic. W i t h the W type during the w a r m period of the year, a lowered background of pressure is observed mainly over the western a n d eastern part of the Subarctic.
47
L.M. Dolgin
9.0
FIG.9. Annual variation of heat budget components.
b
8.0 -
7.06.0. 5.0-
4.03.02.0-
1.00-1.0-
2.0'
'
'
,-2.'o.
J F M A M J J A S O N D
-
-. -. -. Radiation b a l a n c e -Evaporation ..-..-----_ Turbulent heat e x c h a n g e
" '
I
'
""
I
(a) Northern b o u n d a r y of the Subarctic (b) S o u t h e r n b o u n d a r y of the Subarctic
This leads to the stability of cloudiness here and, as a consequence, to the decrease of total radiation. T h e central area of the Subarctic is characterized by a positive anomaly of pressure a n d anticyclogenesis contributing to the advection of a greater a m o u n t of direct radiation and, consequently, to the increase of total radiation. W i t h the E a n d C types of circulation the meridional distribution of the areas of pressure a n d temperature anomalies are characteristic. T h e development of a stationary anticyclone over the European continent blocks the west-east m o v e m e n t of cyclones, which is typical of the W - t y p e processes. This leads to the advection of w a r m air into high latitudes. Thus, for example, with the E-type process in the west and east of the Subarctic a n anti-cyclonic régime and negative air temperature anomaly prevail. In these areas with the E type total radiation is considerably greater than with the W type due to the increase of the a m o u n t of direct radiation. During the transformation of the circulation type W - t E , east of Taimyr, total radiation decreases by m o r e than 4 kcal/cm2/mo, and over Chukchi Peninsula increases by 2.7 kcal/cm2/mo. In small areas under the influence of local factors microclimatic features m a y be observed which differ from the climate of the whole region. A s has been shown, depending u p o n the nature of soil a n d plant cover within the territory of the s a m e region, different values of albedo are obtained. Elements of radiation balance in the s a m e region with different relief m a y change even more. Let us give s o m e characteristics of total radiation for slopes which depend, first of all, u p o n their steepness. T h e greatest difference exists between their northern and southern sides and this is especially true of the region under consideration. W i t h the persistence of a well-defined diurnal variation absolute values of
48
'
J F M A M J J A S O N D
total radiation vary only depending o n the situation and steepness of a slope. T h e y change depending o n the angle of incidence of solar rays o n the slope. Table 5 gives diurnal totals of solar radiation for northern and southern slopes with different inclination. For southern slopes these values increase and for northern ones they decrease with steepness (Zakharova, 1959).
TABLE5. Diurnal totals of solar radiation for northern and southern slopes ('p
= 700) in kcal/cma Percentages
Inclination
+ 230
Northern slopes 400 300 200 1O0 Horizontal surface Southern slopes
460 544 599 636 663 681 703 714 706
1O0 2O0 3O0 400
00
+ 230
00
O
8 79 157
52 80 90 96 100
O 5 50 100
225 293 346 400
103 106 108 107
143 187 220 255
O
O
According to changes in totals of heat obtained the differences in exposure of slopes m a y greatly overlap in the geographic latitude. If one compares the diurnal totals of solar radiation at the horizontal surface at a latitude of 420 with the diurnal totals of Q at southern slopes at a latitudes of 700, then with the gun's declination being + 230 and + 130,steep southern slopes, in spite of a latitudinal difference of 280, almost fully componsate for this difference (Table 6).
Subarctic meteorology
TABLE6. Diurnal totals of solar radiation at southern slopes
= 700 as a
percentage of the totals at the horizontal surface at rp = 42O at rp
Declination of B u n
.
+23O +13O
O0
-130 -230
Indination of southern slopes
100
200
300
400
92 76 50 16
95 86 64 26
O
O
96 92 76 40 O
95 96 88 47 O
In June with the sun’s altitude of 400-460at noon the slopes facing the sun and having an inclination of 450-500receive from the sun 4-5 times more heat than that received by the horizontalsurface, additional heat received by southern slopes provides for a short period of vegetation of agricultural plants. Western and eastern slopes receive as m u c h heat as the horizontal surface.
SOME METEOROLOGICAL FEATURES OF THE SUBARCTIC WIND
REGIME
In winter a considerable negative radiation balance contributes to a severe cooling of the earth surface and lower atmosphere, but intense cyclonic activity often sharply breaks the severity of the winter régime directly influencing the meteorological conditions of the western and eastern areas of the Subarctic. As a result mean pressure fields over them have low values and high pressure gradients. The western area is covered by the southern part of the Icelandic Trough directed to the north-eastinto the arctic seas and the eastern area is covered by one of the troughs of the Aleutian L o w directed to the Chukchi Sea. The cyclonic activity, which is especially, intense in the west, determines a great variety ofmain meteorological elements. The m a x i m u m variability of mean monthly pressure, in winter particularly, is observed in the northern part of western Siberia and is equal to k6-8mb.A large extension of the Asian High has a direct influence upon climatic peculiarities of the central area of the Subarctic. This position of pressure fields determines essential differences in the wind régime of the Subarctic (Prik,1964~). In winter over the western area of the Subarctic, south-westerly and southarly winds of considerable velocities are observed. Over the northern coast of the Kola Peninsula and over the Y a m a l and Gydanskiy peninsulas average velocities are 7-9m/s.Winds with a velocity of 6-7 m/s are mostly observed (20-25per cent) and for 5-10days in a month there are storms.
Over the central area and the western part of the eastern area and due to prevalence of the anticyclonic weather with low pressure gradients, south westerly winds also predominate, veering into westerly winds but with low velocities (1-3 m/s),and calm periods often occur. Storms are rare in winter, on the average, as m a n y as 3-5times in 10 years. Only in the easternmost areas over Chukchi Peninsula, and more to the south does an air flow directed from the north-east with high wind velocities (6-8m/s) prevail. T h e number of storms increases. On the average 5-8 days with storms are numbered in a month. Orographic conditions of the country have a signscant influence on the direction and velocity of wind. A distortion of the direction of the air flow, greater or, on the contrary, lesser velocities often occur. In the valleys of the rivers Ob’, Yenisey, Khatanga, Kolyma, etc., or in narrow bays-for instance in Ob’ Inlet and Khatanga Bay- the direction of the prevailing air flow is distorted and the flow runs in the direction of the river valleys and bays. In summer, as a result of the development of the cyclonic recurrence, which is especially intense over the north of eastern Siberia, the mean pressure field is characterized by an extensive but shallow depression over Siberia and increased pressure at higher latitudes, which determines the prevalence of winds with, primarily, the northern components of north-easterly ones. Only in the extreme east of the Subarctic do they transfer into south-easterlyand southerly winds. Thus, winds opposite to winter ones (i.e. of the type of the summer monsoon) predominate there. Slight pressure gradients in summer determine a lesser recurrence of prevailing directions than in winter. Only under the influence of orographic conditions, as happens in winter, is the direction of local winds more stable, especially if this wind coincides with the wind associated with the pressure field. In the river valleys, as in other seasons, winds prevail along the valleys, but in summer their direction is opposite to that of winter. For example,if in the valley of Yenisey the southern wind prevails in winter, then the northern wind prevails in summer. Wind velocities in the western and eastern areas of the Subarctic decrease considerably,and the number of days with storms over the month is reduced (1 or 2). In north-eastern Siberia, due to the cyclogenesis in summer, wind velocities on the contrary increase and reach a m a x i m u m in the annual variation, but still they do not exceed, on the average, 4.5-5.5m/s, and days with storms run to 1-2in a month. Only under the influence of orographic conditions m a y wind velocities considerably increase in some places. The relation of wind régime to ecology is of primary importance. With a weakening of the wind, the cooling of a plant tissue decreases and,if the wind is intended the cooling increases. With increased wind velocity, evaporation on the slopes increases and these are in a
49 4
I. M. Dolgin
worse condition of wetting as compared to the flat country. Therefore it is necessary to take into account the wind distortions due to the relief of the country. It is k n o w n that with winds blowing along a valley one can observe their intensification over the bottom of the valley and with winds directed across the valley s o m e decrease invelocity is possible. W i n d s are stronger at the tops ofhills, and weaker behind a n obstacle. T h e m e c h a nical influence of the relief u p o n wind depends o n the atmosphere stability. U n d e r stable conditions over elevations the increase of wind velocities is m o r e than that found under unstable ones. Therefore, at the top the wind velocity conditions under stable air m a y be twice as high as that at the foot of a slope. U n d e r unstable stratification wind velocity at the top m a y be approximately 1 m/shigher than that at the foot. Air flow in the dissected country has also a thermal influence contributing to the occurrence of local circulation. D u e to the non-homogeneous cooling over a slope and the adjacent atmosphere layers, slope winds occur, resulting in the accumulation of cold air at the foot of the slope and a temperature difference of about 100 C m a y appear between the top and foot. No special studies have been carried out o n the influence of exposure and topographic forms u p o n meteorological conditions in the Subarctic. However, s o m e quantitative characteristics obtained for t e m perate latitudes by the M a i n Geophysical Observatory of the U.S.S.R. m a y be applied to the subarctic regions. These are the coefficients of wind velocity variations in relation to topographic forms in a hilly country as compared to a flat country obtained Ky Gol’tsberg (1961) (Table 7) from extensive experimental data and special literature.
TEMPERATURE
REGIME
T h e Subarctic is characterized by rigorous long winters a n d short cool summers. T h e air temperature of the coldest m o n t h s is nearly the same. According to m e a n multi-annual data the January m i n i m u m is slightly pronounced. S u c h a temperature variation is accounted for b y heat advection in mid-winter in s o m e years; as a result, the m e a n temperature of one of the winter months appears to be higher than that of the adjacent months (a w a r m core) (Dolgin, 1964). Thus, for example, in Salekhard in 1960 the temperature in February appeared to be w a r m e r than that in January and M a r c h by Il0 C. In K y u s y u r in 1933 January w a s w a r m e r than December by 120 C a n d w a r m e r than February by 40 C. A s Rubinshstein (1962)has s h o w n this p h e n o m e n a is observed in the whole of the Subarctic. In its western area the frequency of the w a r m cores amounts to 35-45 per cent, in the central area it decreases to 10 per cent and in the extreme east it increases u p to 50-60per cent. T h e diurnal temperature variation in winter is negligible, its amplitude is 0.20-0.30 C. It is irregular a n d sometimes the temperature reverses w h e n maxim u m falls o n night hours and m i n i m u m o n d a y hours (Rubinshtein, 1958). T h e temperature field in winter in the Subarctic is non-homogeneous (Fig.10). T h e western area is the warmest one affected by Atlantic cyclones and by one of the branches of the Gulf Stream. Thus, at the northern coast of Kola Peninsula the m e a n January temperature amounts to -60 C, -80 C, southward it drops to -120 C, -140 C. Eastward the air temperature also decreases reaching -240 C, -260 C in the west Siberian part of the Subarctic. In the central
TABLE7. Coe5cients of wind velocity variations in a hilly country in relation to topographic forms (for wind velocities from 2 to 7-8 m/s)as compared to an exposed flat country Coefficients with winds blowing towards the axis of a ridge or a valley Topographic forma
in parallels
perpendicularly
at an angle of 450
Exposed flat country Tops of exposed elevations (Ah > 50 m , slope > 100). The upper one-third of windward slopes of the same elevations Tops of small gentle elevations (Ah < 50 m , slope i 100). The upper one-third of the same elevations Middle parts of windward slopes (exposed) Windward slopes of small elevations facing valleys Leeward slopes of elevations.(Ah > 50 m , slope > 100)
1.0
1.0
1.0
1.3-1.4
1.4-1.5
1.4-1.5
Leeward slopes of elevations (Ah < 50 m, slope < 100) Bottoms of valleys, hollows, ravines exposed to winds if wind blows up the valley Bottom of valleys, hollows, parallel to wind if wind blows from below up the valley Bottom of small closed depressions
50
1.0-1.1 1.0-1.1
1.1-1.2 1.1-1.2 1.0-1.1 1.0-1.1 0.9-1.9 0.9-1.0 F r o m 0.9-1.0 in the upper part to 0.6 in the lower part of the slope 0.6-0.7 0.6-0.7
__
1.o-1.2
0.4-0.5
0.6-0.7
-
0.7-0.8
__
Subarctic meteorology
FIG. 10. Mean air temperature,January. area under the influence of anticyclonic circulation temperature drops to -360 C, -400 C. In the easternmost part affected by Pacific cyclones the mean.temperature quickly increases reaching -270 C, -280 C in the far east tundra a n d -200 C or -220 C near the Bering Strait (Girs, 19606; Shcherbakova, 1961). Frequent changes of pressure systems, especially in the western and eastern areas of the Subarctic, produce large periodic temperature variations. Interdiurnal variability in the west is, on the average, 50-60 C, its m a x i m u m values reach 200-250 C. Extreme temperature values in winter a m o u n t to -270 C, -330 C and +50 C, -70 C at the coast of Kola Peninsula; between the lower course of the Ob' and Yenisey rivers they a m o u n t to -450 C, -500 C and -20 C, -30 C, and at the coast of the Bering Sea -350 C, -450 C a n d +30 C, $50 C. T h e largest amplitudes in the winter observed in the central area range from -500 C, -600 C to +20 C, -40 C. T h e given minimum temperature m a y considerably change depending o n the nature of relief.
Mishchenko (1962) obtained corrections to the average from absolute annual m i n i m a of the air tempefor the Subarctic, where three areas rature (Tmin) with different corrections are selected; they are given in Table 8. In s u m m e r a continuous flux of solar radiation contributes to the quick w a r m i n g of the tundra surface. T h e surface layer of the ground in July in the south of the subarctic gets w a r m e d up to 140-160C, and near the coasts of seas a n d on T a i m y r Peninsula UP to 70-8"C. T h e active surface temperatures QN are good evidence of a m o r e complete estimation of daytime heat resources. A t the M a i n Geophyeical Observatory, QN are calculated for the day o n the basis of the use of the heat budget m e t h o d not only for a flat country, but also for the northern and southern slopes. It is k n o w n that vegetation behaves differently on slopes with different exposure. T h e differences in the duration of plant development phases m a y b e from 5 to 18 days o n the northern a n d southern slopes,
TABLE8. Tminchanges under the influences of relief as compared to a flat country (in 0C)I Arei characteristics
Levelled areas with poor relief (Ah = 20-50 m) Hilly relief (Ah = 50-150 m) Low and middle mountainous relief (Ah = 150-300m) 1.
Broad valley
TOP
-2 -3, -4,
-4 -5
1.5 +2, 3
Narrow vaiieyhollow
+39
$3
+49
+5,
+4 +5 +6
Top-hollow .
5-6 7-9 9-11
(-) and (+)designate Tmi, increase and Tmindecrease respectively, in comparison with an exponed flat country.
51
I. M . Dolgin
respectively. Therefore, accomplished calculations enable us to m a k e a m o r e correct approach to the economic estimation of the differences in thermal régime which appear o n slopes with different exposures. Calculation data obtained for Khibin, Verkhoyansk and Turukhansk are given as a n example in Table 9 (see Mishchenko, 1965). T h e transfer to the prevailing positive air temperatures takes place from early M a y at Kola Peninsula to late June in the north of T a i m y r Peninsula. T h e warmest m o n t h everywhere is July (Fig. 11) being w a r m e r than June by 60-70 C a n d w a r m e r than August by 20-40 C. In the north this difference decreases. T h e field temperature in s u m m e r throughout the whole of the Subarctic is m o r e homogeneous than in winter. T h e highest temperature of 130-140 C is characteristic of the southern part of the western S u b arctic and its easternmost part. A sharp temperature fall of 40-60 C in July occurs near the shores of cold seas, the horizontal temperature gradient here is considerable and, in places, amounts to 6-100/100km.
TABLE9. Differences in thermal régime calculated of three stations (these corrections m a y also be considered tentative values for the Subarctic) QN variations at daytime in comparison with the horiaontal surface Southern slope
Station
Month
100
ZOO
100
200
Khibiny
May June July Aug. Sept.
-1.5 -0.8 -0.8 -1.4 -1.4
-2.6 -1.6 -1.5 -2.6 -2.6
2.2 1.0 1.4 2.4 2.2
1.1 0.6 0.9 1.1 1.2
Verkhoyansk
May June July Aug. Sept.
-1.1 -1.1 -1.1 -1.8 -2.3
-2.2 -1.7 -2.2 -3.6 -3.1
1.8 1.5 2.1 3.2 2.5
1.0 0.7 0.7 1.4 1.4
Turukhansk
May June July Aug. Sept.
-1.4 -1.1 -1.1 -1.6 -2.0
-1.9 -1.7 -2.1 -3.3 -3.1
1.6 1.5 1.8 3.2 2.2
0.7 0.8 0.7 1.7 1.2
A s a n example, m e a n multi-annual air temperatures for the stations situated at different distances from the sea over the western area are given below: January
52
TABLE10. Changes of At depending on location in comparison with an exposed flat country (in OC)' Corrections to Location
M e a n multi-annual
Tops and upper parts of slopes Bottom of brood valleys (more than 1 k m across) Closed valleys or hollows Banks of lakes and large rivers 1.
(-) and (+)are the with a tlat country.
A, (MayAugust)
-10
-1.50
+1 -2 1.5 -2.5 +1 -2
With clear skies
-20
-30
-2 -3
-4 -5 -4
+2
At decrease and increase respectively. in comparison
T h e change of At has a n essential influence on the plant growth (Mishchenko, 1965). These corrections are given for temperate latitudes. For the Subarctic these will be s o m e w h a t less because the diurnal temperature amplitudes are less there. T h e diminution of circulation intensity in s u m m e r leads to a decrease of interdiurnal temperature change, its m e a n value being only 10-30C a n d m a x i m u m value
100-150 C.
Northern slope
Lower Pesha, Vzglav'ye Lower Pesha, village Murmansk Kola
T h e amplitude of the diurnal temperature variation a n d it is very sensible to microrelief (Table 10).
(At)is 40-50 C
-14.4 -14.8 -10.0 -11.3
July
11.6 12.5 12.9 12.9
Extreme temperature variations are notable in July w h e n the absolute m a x i m u m everywhere is 300-320 C, except for T a i m y r Peninsula. Absolute m a x i m u m will drop to between-lo C and 30 C, but in the southern part of the Subarctic negative temperatures are very seldom observed (Prik, 1964b). A stable frost-free period ( w h e n frosts are absent for at least a month) m a y not be distinguished in all areas. T h e last frost is observed in the southern areas of the western Subarctic, o n the average, at the end of the second ten days, and in the eastern Subarctic in the third ten days of June. A t T a i m y r Peninsula a n d near the coasts of eastern arctic seas light frosts of -lo C m a y occur even in July, ground frosts disappearing one or t w o weeks later. T h e first frost in the northern areas is observed in mid-August, in the southern areas in the northern part of Arkhangelskiy region in early September, and o n the northern coast of Kola Peninsula in late September. A s a result, a frost-free period (Fig.12) of about a m o n t h is observed near the arctic seas and at T a i m y r Peninsula, but it rapidly lengthens with greater distance from the sea. Its longest duration of 3 to 3.5 m o n t h s is observed in the north-west of the Arkhangelskiy region and o n the northern coast of Kola Peninsula. Local relief has an essential influence u p o n the duration of the frost-free period. In the area of Y u g o r a n d T a i m y r peninsulas (excluding the areas of highest
Subarctic meteorology
FIG.11. Mean air temperature, July.
FIG.12. Duration of frost-free period.
elevations) with elevations of about 300 m at the concave topographic forms, the frost-free period decreases by 20 days and, at the convex topographic forms, it increases by 20 days in comparison with its period in a flat country. In the northern part of W e s t Siberian Lowland, with a n average length of the
frost-free period of 60-90 days o n hills 100-150 m height, the frost-free period decreases o n the average by 15 days, and in the Yenisey valley under the influence of the w a r m waters transported from the south this period increases by 15 days. In the central area between the Anabar and Olenek
53
I. M. Dolgin
rivers, in the valleys of Pronchishchev Ridge, the frost-freeperiod is reduced by 20-25days, and on the tops and upper parts of slopes it increases by 20 days. On the knolls (of 150-300 m height) lying south of Tiksi Bay, the length of the frost-free period ranges from 45 days in valleys to 80 days on hills (Anon.,
1962). Depending on relief frost intensity also changes. If, as was mentioned above, minimum temperatures in summer in the southern areas of the Subarctic reached -lo C, -30 C in a flat country, then over a rough country these temperatures m a y essentially vary. On hill tops minimum temperaturewill be higher by 20 C and in valleys lower by 1.50-40C than in a flat country.These values depend on weather conditions. In individual years summer frosts as a result of advection of cold air masses from the north are possible throughout the whole of the Subarctic. In the northern areas the frost-free period is practically absent. On the ground the frost-freeperiod is shorter by 0.51.5 months than in the air. T h e earliest autumn transit of air temperature over O0 C occurs in the north of Taimyr Peninsula during the first ten days of September and the latest one is observed on the coast of Kola Peninsula during the second ten days of October. Thus, the period of prevailing positive diurnal temperature lasts for 5060 days in the north of Taimyr Peninsula and increases to 160 days at Kola Peninsula. The total of positive temperatures for this period in the north of the subArctic is 100~-200~ C,in the south of the western area it rises up to 800~-1,000~ C, and westward from the
Urals up to 1,2000 C (Fig.13). The totals of temperatures for the period with a steady temperature of over $50 C,even in the south of the western and far C; in eastern Subarctic, do not exceed 800~-1,000~ the northern part of the western and eastern areas C. of the Subarctic these drop to 100~-200~ O n e of the most characteristicfeatures of the thermal régime of the Subarctic is the existence of temperature inversions, both surface and upper throughout most of the year. The inversions, to a considerable extent, determine the nature of the Earth’s surface radiation,cloudiness and precipitation. It is k n o w that the upper boundary of the surface inversion coincides with the lower boundary of the lower clouds, that steady fogs are associated with the inversion, etc. The most frequent recurrence of the surface inversions throughout the whole of the Subarctic falls in February and March. Over the western area days with inversions amount to 40-60per cent of the period, over the central and eastern areas to 60-90 per cent. By July the recurrence of the surface inversions diminished to 10-20per cent. The thickness of the inversion k m ,in the central layer over the western area is 0.6-0.8 and eastern areas it is 0.8-1.1 km; in summer, it is 0.4-0.6 k m everywhere. The intensity of inversions in winter is 30-60 and 70-130 respectively, and in summer it is 20-40 throughout the Subarctic. As a result of such a stratification of the atmosphere the difference of mean monthly air temperatures (according to multiannual data) of the stations Yukskor (altitude 902 m) and Appatite Mountain (altitude
FIG.13. Totals of air temperature for the period of stable temperature above O0 C.
54
Subarctic meteorology
360 m) in January is only -0.80 C and in July, when the inversion frequency is negligible it is -3.70 C (Anon, 1965). The temperature differences of the Appatite Mountain station and Khibiny station (altitude 134 m) in January amount to $1.30 C, and in July -1.00 C (Anon., 1965). In the central and eastern areas, in winter at a height of 1 k m , temperature m a y be higher than at the surface by 9-100 C. In the snow melting period and in summer, inversions begin from the height of 200-300m. Such inversions are rather frequent but of lesser intensity and thickness. Surface inversions of long duration have an unhealthy influence on the organism of m a n and animals because reduced air mixing results in an accumulation of aerosols and CO in the lowest layer of the atmosphere. WETTING
REGIME
Precipitation
The most intense cyclonic activity in the western and eastern areas of the Subarctic is associated with a greater amount of precipitation occurring in those areas. At Kola Peninsula and in the southern part of the western area, as well as in the extreme east, over 350-400 m m of precipitation occurs in the course of a year (Fig.14). Under the influence of relief, the amount of precipitation m a y rise in s o m e places up to 500-600mm. The least amount of precipitation, i.e. 150-200 mm, falls on the central area where in winter anticyclonic activity predominates. In the annual trend the least amount of precipitation, within almost all of the territory under consideration, occurs in February and March: in the west and east 15-20 mm, in the central area 7-10m m in a month (Prik, 1965). M a x i m u m precipitation occurs at the end of summer, in August and September (Fig. 15). The greatest amount of precipitation occurs at this time in the western area, in the south as much as 45-50 m m in a month, but in the north the amount decreases to 30 m m in a month. In the central area the amount of precipitation is less: 40 m m in the south and 25 m m at the northern boundary of the area. The increase of precipitation is observed in the eastern area, in the north it amounts to 30-35m m monthly, and in the Far East, as a result of advection of wet sea-air masses from the Pacific Ocean, the quantity of precipitation exceeds 50 mm. Precipitation in the whole of the Subarctic during summer months is frequent: from 13 to 15 days in a month. This is a period with mainly drizzling rain and occasional wet snow in the northern areas. The number of days with precipitation of no less than 2 m m per day does not exceed one-third of the total
number of days with precipitation but sometimes the daily amount m a y reach 25-40mm. The Subarctic is typical of an exclusively long period of persisting snow cover. In the extreme west it is about 200 days, in the eastern part of the western area as well as in the central and eastern areas this period increases to 230-250 days. The snow cover, settling for the entire winter, appears in the northern part of Taimyr Peninsula and in the north of eastern Siberia in late September, and in the W e s t Siberian tundra in the first ten days of October (Fig.16). Later still, in early November, the snow cover settles in the extreme western and eastern areas affected by w a r m air and water masses. The lapse of time from the appearance of the first unstable snow cover to its settling for the whole winter is from 10 to 12 days in the central area and from 20 to 30 days in the western and eastern areas. The earliest disappearance of the snow cover occurs at Kola Peninsula at the beginning of the second ten days of May, and in the north of Arkhangelskiy region in the third ten days of M a y (Fig. 17). In western Siberia the disappearance of the snow cover occurs from the first ten days of June in the south of the area to the third ten days of June in the North Taimyr Peninsula, and in eastern Siberia primarily in the first ten days of June. Only in the lower course of the rivers Indigirka, Kolyma and in the valley of the Anadyr does the snow cover melt in late May. The height of the snow cover before melting in the western area is 60-70cm, in the central area 30-40cm and in the east up to 50-60 cm. The snow cover is very irregular, especially in sites where strong winds prevail. Snow is blown away from exposed and particularly elevated sites but accumulates in sites of low relief where it remains throughout the summer season. Very low values of absolute humidity, as a result of low temperatures in winter, are especially characteristic of the Subarctic. The hoar-frost and rime formation due to radiation cooling over snow contribute to the decrease of humidity content. Only on the coast of Kola Peninsula does the mean monthly value of absolute humidity amount to 3.5 mb, while moving to the east it decreases to 0.3-0.5 m b in the central area, but in the extreme east it increases again to 1.5-2 mb. The surface temperature inversions produce a slight increase of absolute humidity with height in the lower atmosphere. In summer absolute humidity rises and in the southern area of the Subarctic it amounts to 11-12 mb and in the northern area 7-8mb. The relative humidity in winter is about 85 per cent in the west and 75 per cent in the central area; in the east it rises slightly. W h e n temperature is very low, the air is often oversaturated with water vapour, but there is neither condensation nor sublimation because of an exceptional purity of the air and the lack of condensation
55
I. M. Dolgin
FIG.14. Amount of precipitation (rnmlyear).
FIG.15. Monthly amount of precipitation (mm), August. nuclei. W h e n the condensation nuclei are inserted into the air, freezing fog and cloudiness appear. During this season of the year in the lower 1-kmlayer relative humidity varies little with height. In summer, in the southern area of the Subarctic, relative humidity is 75-78per cent, in the northern area it increases to
56
80-85 per cent. With height the relative humidity decreases. The diurnal variations of relative humidity in the northern area is very slight in summer, the diurnal amplitude is 4-6per cent. In the southern part of the western Subarctic it rises to 12-16 per cent, and in
Subarctic meteorology
FIG.16. Average dates of the formation of persistent snow cover.
FIG.17. Average dates of destruction of persistent snow cover.
the southern part of the central area it becomes still higher. Humidity as low as 30 per cent and less occurs very seldom and only during the advection of w a r m air from the south in some places where the w a r m air moves over elevations lying close to some areas of the
Subarctic (foehns), reducing relative humidity sometimes to 23-25 per cent. This is observed at Kola and Taimyr peninsulas, near the Lena Delta and in northeastern Siberia. At the foot of slopes and in valleys, due to temperature decrease at night-time,relative humidity essen-
57
I.M. Dolgin
tially increases resulting in more frequent formations of fog and dew. With the increase of wind velocities, heavy cloud cover and high humidity, the differences in humidity depending on reliefform smooth down. Owing to frequent non-intensiveprecipitation,high relative humidity and low air temperature, the evaporation process in the Subarctic proceeds slowly and the evaporation is slight. In the northern part of the central area less than 50 mm of moisture evaporates over the course of a year; to the west, east and south evaporation increases and in the southern part of the western Subarctic reaches its highest value of 150-200 mm. Thus, in the Subarctic, precipitation exceeds evaporation resulting in surplus wetting of grounds. The most intense evaporation is observed in July when 20-30 per cent of the annual total evaporates. It is necessary to take into account the macro- and microclimatological conditions given above for the distribution of agricultural plants within individual forms. According to studies made by Gol'tsberg, agriculture is possible in the Subarctic only in hot-houses. The long days of the polar summer favour plant growth in hot-houses with heating of different kinds, which should be arranged,if possible,in sites protected from northern winds. If the site is hilly, it is recommended that the hot-houses be installed in the middle of the southern slopes. As already mentioned, in some areas of the Subarctic the length of the frost-freeperiod in the air is from 60 to 90 days, and on the surface of the ground and vegetation cover from 30 to 60 days. Agriculture in these areas is possible primarily on slopes and in broad valleys with a well-drainedsoil and on the banks of large rivers and lakes where the frost-free period increases by 15-20 days and soil is not overwet. The bottom of broad deep valleys is very dangerous in respect of frost. It is recommended to use natural or to create artificial protections from northern winds. In the area of permafrost, when the moss cover is removed, the soil drained and ploughed, the ground receives more heat and the layer of soil melting during the summer gradually increases. For faster and better warming of the soil in the north a ridge-like surface of the fields is recommended. It should be indicated that for the solving of a number of problems related to the ecology of the Subarctic a more thorough study of this region is necessary, in a number of cases one can make only shortterm itinerary microclimatologic observations which will characterize, first of all, the distribution of minim u m temperatures and wind velocities (Anon., 1960; Gol'tsberg, 1961).
considerable changes. When studying them scientists, for several reasons, often limit themselves to studies of air-temperaturevariations which can sharply affect the development of plant and animal life. A detailed analysis of temperaturevariations is given by Rubinshtein (1956)and Polozova and Rubinshtein (1963).On the basis of sliding 10-year mean monthly temperatures the author has shown that regular m a x i m u m temperature variations are observed to the north of 400 N. and more distinctly pronounced during the cold period of the year. Thus, for instance, in Salekhard in January the air temperature for the period 1944-53 appeared to be higher by 90 C than that of 1885-94.In Anadyr the air temperature for the period 1945-54was higher by 60 C than that of 1931-40.In summer the variations of 10-year means are a little less but the amplitude still exceeds 30 C. The curves of sliding temperatures constructed by Rubinshtein show that the temperature changes have periods of over 10 and 35 years (Fig.18).The horizontal line in the graph shown mean multi-annual temperatures reduced to the period 1881-1960. The curves of Upernivik and Salekhard are examples of temperature variations for 35 years: at the latter station the temperature in January was 220 C for the period of 1891-1925, and -170 C for the period
1924-58. In the latitudinalzone of 600-700N.the synchronous variation of 10-year mean temperatures from Torgshavn to Salekhard and a nearly inverse temperature variation at the Chukchi Peninsula and Alaska are observed. The amplitude of variations in the western area of the Subarctic is miich greater than that in the central and eastern areas;this is due to the fact that the temperaturevariation is more distinctly pronounced in the areas of intense cyclonic activity. In our opinion a comparison between the temperatures for the periods
Climatic changes T h e climatic conditions of any region are subject to
58
FIG.18. Sliding 10-year mean monthly air temperatures.
Subarctic meteorology
1941-60 and 1921-40 (Rubinshtein, 1956) are of unquestionable interest (Table 11). TABLE11. Mean temperature differences in
OC
for the
periods 1941-60and 1921-40 Stations
Time
Dikson
Salekhard
Turukhsnsk
K?z:kuly
April May June July August September Year
2.7 0.8 0.3 o.1 -0.1
o.o 0.3
1.2 0.6 0.8 0.3 0.7 -0.2 -0.2
2.1 2.1 0.6 0.0 0.2 -0.2 0.1
2.2 1.6 1.3 0.9 -0.2 0.2 0.2
Table 11 gives temperature variations only from April to September. In winter an inverse relation is observed for this period, that is, cooling began. In s u m m e r warming u p continued and this affected even m e a n yearly values. As indicated by Rubinshtein the change of m e a n temperature by 0.50-10C already has practical significance. Uspenskiy has reported that o n the territory of Eurasia lying to the north of the forest tundra, due to climatic variations of the last 40-50 years n o less than forty species of birds and m a m m a l s have appeared or considerably increased in population. The extent by which these species m o v e d into the heart of the tundra is estimated by the author to be 0.5010 N. in comparison with the beginning of the century. In Yamal, where the degree of warming w a s highest, this m o v e m e n t amounts to 20 N. Moose which were practically non-existent to the north of the forest tundra in the 1920s have n o w become accustomed to the tundra, and inhabit it u p to the coast of the northern seas. At present it is established that the periods of temperature variations are not strict because of
changes in the nature of the atmosphere circulation associated with changes in solar activity (Dzerdzeevskiy, 1956; Girs, 19606). Vangengeim and Girs have s h o w n the existence of the epochs of the forms of circulation when an a n o m a lous development of the processes of one type (or two) and weakening of the processes of the other types h a d taken place for a long time. Thus, for instance from 1900 to 1928 there existed a n abnormal development of the processes of the W-circulation type producing over the area under consideration, except for the western and eastern peripheries, positive temperature anomalies with the highest values over western Siberia. F r o m 1929 to 1931 there existed the epoch of the E-circulationtype and a positive temperature anomaly covered the entire Subarctic at that time. The highest values of anomaly fell o n the western and easternareas and that w a s in agreement with the conditions of the given circulation type. This epoch of circulation then w a s replaced by the epoch of the C-circulation type (1940-48) when a positive temperature anomaly w a s observed only in the region adjacent to the K a r a Sea. During recent years (1949-64)an abnormal development w a s taken by the E and C types of circulation over Eurasia with an abnormal development of the .W-circulation type over the Pacific Ocean. Under the influence of the latter there was a rearrangement of the E and C types of circulation into the W type over Eurasia which started in the 1950s. This rearrangement reached the highest intensity by 1964 resulting in, at that time, a greater anomaly of temperature over the western and eastern areas of the Subarctic. Over the central area the temperature anomaly appeared to be negative. T h e probability of positive anomalies increased over the western area by 20 per cent and in the east by 23 per cent. At the present time rearrangement has not yet been completed. Taking into account this situation, a further warming u p should be expected in the next few years and, consequently, a m o v e m e n t farther north of a n u m b e r of animals from these regions should also be expected.
Résumé Météorologie subarctique (I. M. Dolgin) L’étude d u régime météorologique des régions subarctiques présente un intérêt particulier si l’on veut tirer tout le parti possible des ressources de ces régions pour le développement de l’agriculture, de l’élevage et- de la pêche. Selon l’académicien A. A. Grigoriev, la frontière méridionale de la région subarctique est située à la limite entre la toundra et la taïga.
D a n s une portion longitudinale considérable de la région subarctique, on constate de grandes différences en ce qui concerne les conditions de la circulation atmosphérique et le régime d u rayonnement. C’est pourquoi l’auteur examine le régime météorologique des régions occidentale, centrale et orientale. D’après l’analyse des composantes d u bilan de rayonnement, o n constate que, dans l’ensemble d u territoire considéré, ce bilan a une valeur annuelle positive qui varie selon la latitude de 10 à 14 kcal/cm2
59
I. M. Dolgin
par an. L’auteur examine les composantes du bilan thermique. Se fondant sur la classification de G. Y a . V a n g e n heim et A. A. Guirs, l’auteur indique les variations de rayonnement dues à l’alternance des principales formes de la circulation atmosphérique et démontre que, dans les régions occidentale et orientale, le rayonn e m e n t total est beaucoup plus grand avec les formes E et C de circulation qu’avec la forme W. L’auteur montre les variations du bilan de rayonnem e n t sous l’effet des facteurs locaux. Les variations des quantités globales de chaleur reçue en fonction de l’exposition des pentes peuvent être supérieures a u x différences dues à la latitude géographique. L’auteur donne des renseignements plus précis sur la position du front subarctique e n été et e n hiver, et sur les particularités de la circulation atmosphérique dans les régions subarctiques.
-
Le climat des régions subarctiques est sujet à des fluctuations considérables. L’auteur indique les caractéristiques des variations de température selon Rubinstein et leurs relations avec les formes de circulation d’après A. A. Guirs. En 1960-1965,un accroissement considérable de la température a été constaté dans les parties occidentale et centrale de la région subarctique, ce qui peut avoir pour effet d’attirer davantage d’oiseaux et de mammifères dans les régions septentrionales de la toundra. L e mésorelief et le microrelief ont une influence considérable sur la répartition des températures, d u vent, de l’humidité, des précipitations et de l’enneigement. L’auteur indique certaines caractéristiques microclimatiques de la région subarctique. En analysant le régime de la température et de l’humidité, l’auteur tient compte de la stratification des couches inférieures de l’atmosphère.
Discussion F. E. ECKARDT. Pourriez-vousindiquer le type d’instrument utilisé par vous pour la mesure du rayonnement? Pour quelle raison exprimez-vous le rayonnement en kcal/cma/mois au lieu d’adopter le système MKSA rationnalisé recommandé par l’organisation internationale de standardisation? I. M. DOLGIN. In measuring radiation the standard equipment was used which is in use at all actinometric stations in the U.S.S.R.The description of these instruments is widely known. The measurements were made in kcal/cma/mo since the MKSA system has not yet been accepted by the Hydrometeorological Service of the U.S.S.R.
J. MALAURIE.(1) I don’t believe that the distinction between Arctic and Subarctic is very valid from a geomorphological point of view. I a m afraid that a thorough analysis will determine a fragmentation in too m a n y aspects of this distinction for it to have any real meaning. In the high Arctic (Inglefield land) on the maritime border of a plateau, the
conditions are, if the sea ice is broken to 40 per cent, subarctic, according to climatologicaland geomorphic processes. (2) In Arctic, journal of the Arctic Institute of North America, it is stated, in a rather old issue, that the temperature measurements of the coastal stations of the northwest coast prove a decline of temperature since 1940. H o w does this compare to north-eastern Siberia, in Tchoukotkaoblast or Kamchatka? I. M.DOLGIN. (1) W e can speak of two ways of distinguishing the Arctic and the Subarctic. One is based on the general circulation, distribution of air masses and radiation, the second on the dependency of climate on local conditions. I agree with Professor Malaurie that w e are able to find such small areas, like the coastal belt, where the influence of sea water creates subarctic conditions in the arctic zone. (2) In eastern Siberia there has been a real amelioration (warming up) of climate in the last twenty years. In Alaska there has been a slight cooling according to m y map. The climatic changes in this period, as I said, are not the same in the whole of the Subarctic.
Bibliography / Bibliographie ANON. 1960. Atlas sel’skogo khozyaistva SSSR. Moscow, Glavnoye Upravleniye Geodezii i Kartografii Ministerstva Geologii i okhrany nedr SSSR. . 1962. Mikroklimatologiya kholmistogo relyefa i yego uliyaniye na sel’skokhozyaistvennyye kul’tury. Leningrad, Gidrometeoizdat. . 1965. Spravochnik PO klimatu SSSR. Vyp. I, chast’ II, Temperatura uozdukha i pochvy. Leningrad, Gidrometeoizdat. BARASHKOVA, Ye. P.;GAYEVSKIY, V. L.;D’YACHENKO, L.N.
--
-_
60
1961. Radiatsionniy rezhim terriiorii SSSR. Leningrad, Gidrometeoizdat. BUDYKO, M . I. 1956. Teplovoy balans zemnoy pouerkhnosti. Gidrometeoizdat. CHERNIGOVSKIY, N . I.; MARSHUNOVA, M . S. 1965. Klimat Sovietskoy Arktiki (radiatsionniy rezhim). Leningrad, Gidrometeoizdat. CHUKANIN, K . I. 1965. Nekotoryye sinoptiko-klimaticheskiye kharakteristiki Arktiki. (Trudy AANII, tom 273.) DZERDZEEVSKIY, B. L. 1956. Problemy kolebaniy obshchey
Subarctic meteorology
tsirkulyatsiiatmosfery i klimata.D.I. Voyeikov i sovremennyye pro blemy klimatologii. Leningrad, Gidrometeoizdat. DOLGIN, I. M . 1964. Bezyadernyye zimy v Arktike. (Trudy AANII, t o m 266.) GIRS, A . A. 1960a. Tipovaya kharakteristika osnovnykh raznovidnostey form atmosfernoy tsirkulyatsii v teploye vremya goda. Sbornik Problemy Arktiki i Antarktiki, vyp. 2, izd. “Morskoy transport”. . 1960b. Osnovy dolgosrochnykh prognozov pogody. Leningrad, Gidrometeoizdat. GOL’TSBERG, I. A. 1961. Agroklirnaticheskaya kharakteristika zamorozkov v SSSR i metody bor’by s nimi. Leningrad, Gidrometeoizdat. GRIGOR’YEV, A. A . 1956. Subarktika. Moskva, Gosudarstvennoye izdatei’stvo geograíicheskoy literatury. KHROMOV, S. P. 1948. Osnovy sinopticheskoy meteorologii. Leningrad, Gidrometeoizdat. KOPTEV, A. P. 1961. Al’bedo oblakov,vody i snezhno-vodyanoi poverkhnosti. (Trudy AANII, t o m 239.) MISHCHENKO, 2. A. 1962. Sutochayy khod temperatury vozdukha i yego agroklimaticheskiye Snacheniya. Leningrad, Gidrometeoizdat. . 1965. O temperature deyatel’noy poverkhnosti v mikroklirnaticheskikh issledovaniyakh. (Trudy G G O , Vyp. 180.) ORLOVA,V. V. 1962. Zapadnaya Sibir’. Klimat SSSR,Vyp. 4, Gidrometeoizdat. PETERSON, S. 1961. Analiz i prognoz pogody. Perevod s angliyskogo. Moscow, Gidrometeoizdat. POLOZOVA, L. G. ; RUBINSHTEIN, Ye. S. 1963. Sovremennoye izmeneniye klimata. (Izvestiya AN SSSR. Seriya Geograficheskaya, No. 5.)
_-
__
PRIK, Z.M . 1964a. Sredneye polozheniye prizemnykh baricheskikh i termicheskikh poley v Arktike. (Trudy AANII, t o m 217.) . 1964b. Baricheskiye i termicheskiye usloviya v Arktike v period MGSS. (Trudy AANII, tom 266.) _- . 1965. Osadki v Arktike. (Trudy AANII, t o m 273.) RAGOZIN, A. I.; CHUKANIN, K . I. 1961. Napravleniye i skorosti peremeshcheniya tsiklonov i antitsiklonov v Arktike. (Trudy AANII, t o m 235.) RUBINSHTEIN, Ye. S. 1956. Ob izmenenii klimata SSSR za posledniye desyatiletiya. D. I. Veyeikov i sovremennyye problemy klimatologii. Leningrad, Gidrometeoizdat. . 1958. Obrashcheniye sutochnogo khoda ternperatury vo vremya polyarnoy nochi i nochnyye povysheniya temperatury zimoy v urnerennykh shirotakh. Moscow, Izvestiya AN SSSR, seriya Geograficheskaya, No. 3. . 1962. Teployadernyye i besyadernyye zimy. Moscow, Izvestiya AN SSSR, Seriya Geograficheskaya, No. 4. SHCHERBAKOVA, Ye. Ya. 1961. Vostochnaya Sibir’. Klirnat SSSR, Vyp. 5, Gidrometeoizdat. VANGENGEIM, G. Ya. 1961. Predskazaniye sezonnykh raspredeleniy meteorologicheskihk elernentov. Izvestiya AN SSSR. Seriya geograficheskaya, No. 3. ZAKHAROVA, A. F. 1959. Radiatsionnyy rezhim severnykh i yuzhnykh sklonov v zavisimosti ot geogra$cheskoy shiroty. (Uchenyye zapiski Leningradskogo Gosudarstvennogo Universiteta No 269, seriya geograíicheskava.)
__
__
__
61
Some features of the microclimate within hilly regions in Finland S. Huovila
T h e weather in Finland varies greatly from day to day and from year to year. This is due to the battle between maritime air masses from the Atlantic Ocean and the continental air masses from Europe and Asia. The annual temperature range can therefore be quite large in s o m e years. As an example, the m i n i m u m temperature in Ivalo w a s -48.60 C o n 1 February 1966 and the m a x i m u m temperature 31.70 C on 20 June 1966. T h e annual variation of screen t e m perature can thus exceed 800 C in northern Finland and 700 C in southern Finland. The rate of incoming solar radiation is highly variable from season to season in different parts of Finland. In Table 1 w e see that practically no radiation is received from the sun at latitude 700 N.in December and January. At this time the earth surface loses heat continuously owing to long-wave terrestrial radiation. The effective heat loss reaches its m a x i m u m during cloudless weather. The coldest winter temperatures are thus observed during clear, calm days under the influence of cold continental air masses. Near to the ground we can often read temperatures that are m o r e than 100 C colder than the simultaneous reading in the thermometer screen. In June, northern Finland receives considerably m o r e solar radiation than southern Finland during cloudless days. If the w a r m continental air from eastern Europe enters northern Finland, we can expect very w a r m days at those latitudes. Table 1 also reveals large differences in the a m o u n t of solar radiation received by surfaces of different slopes and directions. In the springtime walls and slopes facing south receive m u c h m o r e solar radiation than horizontal surfaces and about ten times more than the north-facing walls and slopes. This unequal distribution of short-wave radiation results in the early melting of s n o w and awakening of life forms o n the southern slopes of hills and mountains during
sunny days. D u e to the midnight sun the distribution of solar radiation is m u c h more uniform in the middle of the s u m m e r , especially in northern Finland. Actually, northern slopes have been found to be safer than southern slopes from d a m a g e caused by radiation frost in northern Finland. T h e reason for this paradox is that the northern slopes also receive solar radiation during clear nights or in a situation that is favourable to radiation frost. With dense vegetation in the middle of the s u m m e r , too, the height and the density of the vegetation are more important from a microclimatological viewpoint than the slope and direction of the surface. It must be pointed out that this rule is only valid for s u m m e r days, not for nights. At night, the relative temperature of a given point in a hilly region is determined mostly by the relative height of this point compared with its surroundings a n d the thermal conductivity of the soil at this point. Since cold air is heavier than w a r m air, the coldest air is liable to flow iato deep pits and valleys and the warmest air is found at the hilltops. This kind of t e m perature distribution is most evident during calm, cloudless nights in both s u m m e r and winter. Figure 1 shows the temperature distribution measured at a height of 1.7 m above the ground along an old main road in central Finland during four clear and calm nights in August 1963. Temperature readings were taken at 91 different measuring points along the road during outward and return journeys m a d e in a car. T h e length of the route w a s 33.2 km. The correlation between relative temperature and relative height is quite obvious. This means that the temperature on a hilltop is generally w a r m e r than the temperature in the neighbouring valley at the s a m e height above the ground. T h e best correlation, however, is to be found between temperature curves recorded during different nights. The conclusion to be drawn is that every point within a certain small area has a fixed relative
63
S. Huovila
TABLE1. Diurnal amount of solar radiation (direct solar radiation plus sky radiation) on various surfaces during clear days, measured on the fifteenth day of each month. Unit: kcal/m2. (Based on data from Lunelund, 1940) Surface
Jan.
Feb.
Mer.
Apr.
390 400 2 060 55 1470 60 1710 95 425 1260 95
1160 1040 3 585 105 2 640 175 3 265 190 1230 2 550 260
2 500 1935 4 380 165 3 505 515 4 730 305 2 415 3 975 880
4 290 2 840 4 120 350 3 935 1220 5 690 740 3 905 5 220 2 140
5 720 3 435 3 540 930 3 835 1990 6 045 2 330 4 990 5 900 3 330
-
320 420 2 O00 55 1430 55 1595 90 420 1220 90
1480 1535 3 975 135 3 115 385 3 750 240 1685 3 050 530
3 440 2 935 4 420 440 4 150 1340 5 300 710 3 525 4 850 1800
5 360 3 935 4 220 1800 4 315 2 610 6 150 2 400 5 050 5 960 3 355
May
July
Aug.
Sept.
Oct.
6 350 3 590 3 i50 1400 3 595 2 360 6 045 3 385 5 405 6 015 3 945
5 930 3 385 3 185 1150 3 560 2 120 5 920 2 860 5 060 5 785 3 585
4 730 2 900 3 600 535 3 650 1470 5 590 1275 4 150 5 255 2 535
3 180 2 215 4 080 205 3 525 780 4 920 390 2 915 4 310 1345
1660 1330 3 820 135 2 875 27 5 3 755 245 1625 3 035 460
600 575 2 570 75 1840 80 2 170 130 655 1650 130
6 470 4 330 3 965 3 170 4 225 3 630 6 415 4 170 5 865 6 330 4 745
5 920 4 065 3 965 2 650 4 130 3 205 6 195 3 435 5 395 6 030 4 165
4 240 3 300 4 160 880 4 140 1795 5 555 1335 4 130 5 250 2 440
2 280 2 085 4 190 180 3 540 700 4 415 335 2 415 3 800 985
740 860 2 950 90 2 155 135 2 540 155 895 1960 215
25 50 430 10 340 10 325 15 45 270 20
June
Nov.
Dee.
Latitude 600 N.
Horizontal surface W.and E.wall
S. wall
N.wall S W . and SE.wall NW.and NE.wall S.slope 450 N.slope 450 W . and E. slope 450 SW.and SE. slope 450 N W . and NE. slope 450 Latitude 700 N.
Horizontal surface
W.and E.wall
s.wall
N.wall SW.and SE.wall N W . and NE.wall S. slope 450 N. slope 450 W.and E.slope 450 SW.and SE.slope 450 N W . and NE. slope 450
-
-
minimum temperature durine calm, clear nights, v provided that the vegetation and thermal conductivity of the soil do not change greatly and that all the measuring points lie within the same air mass. In other words :if point A is warmer than point B a clear, ill be warmer than B during every calm, calm night,A w clear night as long as A and B are situated at the same height above the ground and the conditions above are fulfilled. Figure 1 also tells us that serious mistakes can be made under certain conditions if ordinary methods are used for the reduction of meteorological data. An ordinary method of adjusting temperature observations is to subtract 0.650 C for an increase of 100 m in the altitude. This rule would lead to completely false results in regions where the hills are some tens of metres high, like those in Figure 1. For example, point No.35 was situated at a relative height of 40 m and point 36 at a relative height of 71 m. The temperature difference between those points was negligible o n clear days whereas point 35 was 7.10 C colder o n clear, calm nights, as a mean of the trips 2, 3 and 4. The daily mean temperature at point 35 is also several degrees colder than at point 36 but, point 36 would be colder if w e used the rule which is valid for big differences in the altitudes within high mountains. The effectof hills on the distribution of precipitation is not yet completely understood, but it is being
64
250 280 1615 40 1150 40 1265 70 280 940 70
-
-
-
-
-
studied both in Sweden and in Finland. As pointed out earlier, the distribution of solar radiation is unequal on different surfaces in spring and the rate of decrease of the snow cover is thus considerably higher on southern than on northern slopes. Cooling during clear nights is a reason for the ample formation of dew and fog in valleys but no data are available on the amount of d e w in Finland. The wind speed in Finland is generally very low during summer nights. Several studies have revealed that a wind speed of 0.5 m/s or less is very typical in central Finland on clear nights. This value refers to the screen height (2 m above the ground), while 1-2 m/s can be observed at the anemometer height during a night. This calmness of clear nights often results in damage by radiation frost. The experience of farmers from olden times agrees fully with the results shown in Figure 1 and, for this reason, farms in northern and north-eastern Finland are situated mostly on hilltops. T o s u m up w e can say that microclimatological conditions play an important part in Finland. A knowledge of these conditions is most important in hilly regions,especially in northern and north-eastern Finland. The influence of solar radiation, the distribution of temperature during clear nights and the properties of the snow cover are among the most important microclimatological factors.
S o m e features of the microclimate within hilly regions in Finland
"""
I""""'I""""'I""""'~""""'~""' 10
lo
20
30
50
40
60
'
'
I
70
'
'
"
'
1
~
~
'
l ' ' ' ' ' i I r i l '
80
90
IO
3
'C
Fig. 1. Temperature curves during four nights (1-4) and the mean curve of nights (2-4). Horizontal scale: measuring points. Bottom curve (h): relative altitude.
Résumé Quelques caractéristiques du microclimat dans les régions accidentées de Finlande (S.Huovila)
L'influence des accidents de terrain sur la distribution des températures pendant l'année est manifeste. Pendant le jour, la pente d u terrain joue un rôle déterminant dans l'intensité du rayonnement solaire et la température du sol. Pendant les nuits claires et calmes, d'autre part, la descente de masses d'air froid et lourd dans les vallées et les dépressions peuvent provoquer d'importantes différences entre les températures minimales. Supposons que des mesures soient effectuées en deux points d'un m ê m e village situés à des altitudes différentes au-dessus d u niveau de la mer, mais à une
certaine hauteur au-dessus du sol. I1 n'est pas du tout exceptionnel de trouver des différences de l'ordre de 10 O C entre les températures minimales de ces points pendant une nuit claire et calme. P a r m i les effets secondaires de ces phénomènes, signalons l'abondance de rosée et de brouillard que l'on observe en m ê m e temps dans les vallées. L'orographie et la pente du terrain influent égalem e n t sur les précipitations, ainsi que sur l'accumulation et la vitesse de diminution de la couche de neige. L a distribution des vents, en direction et en vitesse, dépend aussimdu relief, et les nuits sont souvent très calmes dans le centre de la Finlande.
Discussion F. E. ECKARDT. Pourriez-vousindiquer le type d'instrument utilisé pour la mesure du rayonnement?
S. HUOVILA. Table 1 is based on data of Lunelund (1940). H e carried out a long series of observations in the years 1922
to 1939 by using Gorcaynskipyrheliagraphs, Angström pyranometers, Bechstein photometers and several other instruments. At that time, Lunelund was one of the best experts in the study of short-waveradiation and he published m a n y papers on solar radiation and illumination.
65 5
S. Huovile
Bibliography / Bibliographie HUOVILA, S. 1963. Some features of the microclimate of a luxuriant turnip rape field on a clear day. 20 p. Geopfiysica, vol. 6, no. 3-4. . 1964. O n precautions against crop damage due to radiation frost within hilly regions. 22 p. Soe. Scient. Fenn. Comm. Pfiys.-Matfi.,vol. XXIX,no. 4. LUNELUND, H. 1940. Bestrahlung verschieden orientierter Flächen in Finnland durch Sonne und Himmel.27 p. Soc. Scient. Fenn. Comm. Pfiys.-Matfi., vol. X,no. 13.
--
66
SEPPANEN,M. 1963. On the influence of the amount of snow, slope of terrain and position of trees on the rate of decrease of the snow in pine-dominated forest. Geopfiysica, vol. 8, no. 3, p. 213-224. VALMARI, A. 1966. O n night frost research in Finland. Acta Agralia Fennica, no. 107, p. 191-214.
Radiation measurements near the forest limit in northern Sweden Hans Odin and Kurth Perttu
OBJECTS O F T H E INVESTIGATION T h e climate at the forest limit, i.e., in the boundary regions between mountain and forest country, constitutes a limiting factor o n the expansion and regeneration of the forest. Radiation is one of the primary factors affecting the exchange of heat a n d moisture between the atmosphere and the surface of the ground with-its vegetation. T h e objects of the present investigation were : 1. T o determine the albedo of different types of terrain such as treeless mountain plateaux, clearings and forest, thereby obtaining data on the quantity of absorbed radiant energy. 2. T o compare the total radiation exchange of the ground at a single locality o n a cloudless s u m m e r day, both in a clearing and infjeld forest. 3. T o estimate the difference in available energy between clearing and fjeld forest with respect to energy other than radiation, e.g., in the form of evaporation and turbulent heat transfer. T w o approaches to the problem were adopted, namely measurements m a d e from aircraft, this m e t h o d permitting large areas to b e surveyed, and measurements at ground level. T h e measurements from the air gave clear readings, as did those m a d e o n level, treeless terrain, whereas the measurements m a d e in forest country were extremely difficult to interpret owing to the complex system of radiation.
MEASUREMENTS M A D E F R O M AIRCRAFT T H E O R Y A N D PRINCIPLES
Incident short-wave radiation and short-wave radiation reflected from the ground, i.e., energy in the w a v e -
length interval from 0.3 to 3 p m , were measured from a n aircraft. T h e energy w a s measured in gcal/cma/min or in ly/min. Figure 1 represents conditions o n a cloudless s u m m e r day. Diffuse sky radiation (D)comprises about 20 per cent of the total incident short-wave radiation (global radiation). T h e ratio between G, a n d G is called the albedo a n d is designated A. This expression thus indicates the proportion of incident short-wave radiation reflected from the surface of the earth. T h e radiation varies with the fligh alttitude (page 68). T h e albedo can however be regarded as constant with the height above the ground, since the increase in global radiation with height above the ground is balanced by the increase in radiation reflected from the atmospheric layer between the ground a n d the aircraft. This is supported by the results.
G
D
. I sin h
Short-wave radiation 0 . 3 - 3 ~
FIG.^. Components of short-waveradiationat asmoothsurface. I sin h = direct solar radiation (h = sun-altitude) D = diffuse sky radiation G = global radiation Gr = reflected global radiation.
67
H.Odin and K.Perttu
INSTRUMENTS A N D MEASURING TECHNIQUE
T h e instruments used consisted of t w o Moll-Gorczynski Solarimeters, one directed u p w a r d a n d the other d o w n w a r d . These measured radiationin the w a v e length interval from 0.3 to 3 p m . T h e Solarimeters were m o u n t e d at the wingtips in place of navigation lights o n a n aircraft of type Sk 16 provided by the Swedish Air Force. Cables running along the wings and into the cabin were connected to a millivoltmeter. Either Solarimeter could be connected to the voltmeter as desired by simple manipulation of a switch. T h e orientation of the Solarimeters w a s almost exactly horizontal w h e n the aircraft w a s in flight, In the following description, any deviations of the meters from the horizontal position have been ignored. Changes in the altitude of the aircraft with variations in air speed are similarly disregarded. T h e instruments w h e n in operation are influenced (Fig. 2) by radiation over the entire aperture angle of 1800. Radiation from the area near the horizon is however small compared to radiation from the area about the zenith, provided that the sun is not low in the sky. T h e ground area from which the radiation is received is therefore assumed in the following to have a magnitude of x(H/tg20)2 (Fig.2). Measurements from the air were m a d e over different types of ground, principally clearings, forests and treeless mountain plateaux. Owing to the reaction time of the meters (about 20 seconda from zero to full deflection), the course of the aircraft w a s laid so as to traverse similar ground for fairly long periods. In measurements m a d e at low altitude, repeated passes were m a d e over the s a m e terrain in order to obtain data for calculation of the m e a n albedo of the terrain.
Instrument
---
Horizontal plane
Ground //rt////////////////////////y/ /
TI.
["I2
tg
20
FIG. 2. Relation between the height of the instrument above ground surface from which the principal part of the measured radiation is received.
T h e flights covered a region extending from Östersund in the south to Gällivare in the north and from Enafors in western Jämtland in the west to Luleå
in the east. T h e y were m a d e at different altitudes above ground level, from 100 m to 1,500 m.T h e speed of the aircraft w a s 180 k m / h r at 100 m and 230 km/hr at other altitudes. MEASUREMENTS
T h e flights were m a d e o n 15, 16 and 19 July 1965 and are illustrated in Figures 3-8.In these, the time is laid off along the abscissa and the radiant energy in ly/min along the ordinate. Percentages for albedo are plotted at the top of each graph. T h e first measurement (Fig.3), which w a s a trial run, w a s m a d e over the Luleå area. T h e sky w a s cloudless while the measurements were being m a d e ,
O
20- Co 15 *
0 Moo.
-
10 5-
F
o.m. 0.10-
till
I
;,
1l;l III' 'III
I
111; IlII
I
0.50-
I
0.40-
I I
0.10
I
I
I I I
I I
o 20-
I
Ilil IIlI III :III 1111
I
0.60-
0.30
------------3
._._._._._._._._._.
100.
0.m
8
I I I I
I I I ' I I I I
;I;
I I I
I i I
I I
1 ; '
I
i¿i
pl, 111
; :I";i I l
A
o m
- --
FIG. 3. Flight near Luleå, 15 July 1965. . global radiation, i. e., incoming short-waveradiation measured global radiation O measured reflected radiation albedo B: bog; Co: concrete; F: forest;
Radiation measurements near the forest limit in northern Sweden
FIG.4. Flight Luleå-Gällivare, 16 July 1965. Code as for Figure 3. C: clearing; T:treeless mountain plateau.
16- Albedo i % 1514-
13-
1211-
10-
' F
0.900.80-
0.700.600.500.40.
0.300.200.10
0-
FIG. 5. Linalompolo, 21 July 1965. Global radiation ( ), reflected radiation (---) and albedo (top), at 1.6 m above ground.
.T.
lo
I
0.30
except o n t w o occasions w h e n small cumulus clouds h and 10.40affected the radiation input (10.05-10.07 10.41 h). T h e dot-dash line represents the radiation that would have prevailed in clear weather conditions. Flight altitude w a s 200 m except in the case of the first reading, which w a s taken over concrete prior to take-off. T h e second measurement (Fig 4) w a s m a d e between Luleå a n d Gällivare at a height of 500 m, and over Linalompolo, 25 k m north-west of Gällivare, at a height of 100 m. During the flight the global radiation rose from 0.80 ly/min (or gcal/cmz/min) at 08.15 h
to 0.87 ly/min a n hour later. Unfortunately, cumulus h a d begun to build up over Linalompolo, so that the global radiation is assumed to correspond to the dotdash line except at 09.06 h, w h e n both the aircraft and the ground were overshadowed by the s a m e cloud. On this occasion, however, the upward-facing meter did not have time to adjust to the true global radiation, and the albedo consequently c a m e out too low. Other measurements showed that the albedo increased in the shade (Fig.5), 12.06 h-12.35 h). T h e third flight w a s m a d e between Luleå and Enafors, with touchdowns at ûstersund o n both the
69
H.Odim and K. Perttu
Albedo i %
6. Flight Luleå-Ostersund, ' ' FIG. tude 500 m, 19 July 1965. global radiation
e
:
F e
L
O! 0945 0950 0455 Lat 65.3 Lon2
20.9
1600
1005
1090
10)s
65.0
64.7
20.2
19.1
t:
1020
1025
10'30
64.5 18.6
10'75
13'40
alti-
0
reflected radiation albedo B: bog; C: clearing; F: forest; L:lake.
M.E.T.
1Ó"5 64.2
17.3
FIG. 7. Flight Östersund-Enafors-ÖsterSund, 19 July 1965. F: forest; T:treeless mountain plateau.
M.E.T. 600 rn altitude
outward and return trips. Figure.6 shows the measurements m a d e between Luleå a n d Ostersund at a height of 500 m. In Figure 7, which is a record of the flight from Östersund to Enafors, it can be clearly seen h o w the global radiation varies with the height of the meter. A t a n altitude of 550 m the radiation w a s 0.96 ly/min, while twenty minutes later, at 100 m, it w a s 0.92 ly/ min, and fifteen minutes after that, at 600 m, it w a s 0.94 ly/min. T h e fact that the incident radiation
70
during this last part of the flight did not reach the s a m e figure as during the first part w a s due to the general fall-offin incoming radiation during the afternoon. As in Figures 3 and 4, the dot-dash lines in Figure 7 indicate the global radiation that would have prevailed in clear weather. T h e albedo over different types of terrain is computed from the corrected global radiation and from the reflected radiation over sunlit areas. T h e Hight from Östersund to Luleå w a s m a d e at a n
Radiation measurements near the forest limit in northern Sweden
FIG.8. Flight Ostersund-Luleå,19 July 1965.Global radiation ( radiation
-
), reflected .
(--) and albedo (top).
ecIia 16 14
12 10 Ob0
0.70 0.60
0.50 OM
o.sa 0.M
1625
16”
163,5
lbk5
16”
1705M.E.T.
1,500rn altitude
altitude of 1,500 m. T h e decline in incident radiation during the afternoon is clearly apparent from Figure 8. T h e small variations in the curve are due to cirrus clouds which partly obscured the sun from time to time. The incoming radiation showed a m u c h greater percentage drop than the reflected radiation, with the result that the albedo rose during the afternoon. The sharp rise in albedo, e.g., between 16.20 and 16.30 h, w a s caused by a fall in incoming radiation due to clouds which concealed the sun, while the reflected radiation remained constant or even rose slightly. T h e surface of the ground viewed from an altitude of 1,500 m presented a fairly uniform appearance over the whole route, so that no division w a s made. It consisted for the most part of forest, bogs, clearings and small lakes. With a 1400 aperture angle, reflection readings are obtained from a n area with a diameter of over 8 km. R E S U L T S A N D D I S C U S S I O N (see T a b l e
1)
The
albedo over coniferous forest varied from 10 to 13 per cent. Over Linalompolo, near Gällivare, the albedo w a s 10-11per cent, whereas over Enafors, four degrees further south, it w a s 13 per cent. The albedo figures for clearings were 14-16per cent. Bogs, moors and treeless mountain plateaux showed the s a m e albedo as clearings or s o m e w h a t higher. W a t e r surfaces (the Gulf of Bothnia and L a k e Ctorsjön) had a low albedo and showed great variations at different times; this is because the albedo over water is dependent o n both solar altitude and the smoothness of the surface.
MEASUREMENTS M A D E AT G R O U N D LEVEL Meteorological phenomena, including radiation, are continuously recorded at a micrometeorological station at Linalompolo. For the purpose of studying radiation conditions in forest as compared to clearings, measurements were m a d e with portable instruments as a supplement to the measurements m a d e from the air. The measurements described below were m a d e o n 18 July 1965 between 10.30 and 16.55 h. T H E O R Y A N D PRINCIPLES
The ground
measurements, like those m a d e from the air, comprised the short-wave interval (0.3-3 pm). Measurements were also m a d e of the radiation balance, i.e., the difference between incoming and outgoing radiation. This comprises both short-wave and longw a v e radiation (0.3-60pm). The components of the short-wave radiation o n level, treeless ground are identical with those s h o w n in Figure 1. Figure 9 also shows the long-wave components (wavelength interval 3-60pm). The long-wave radiation from the atmosphere (Ea) is of the order of %E,. o n a cloudless day. The radiation balance Eb is expressed by the formula (1 - A)G - Eeff,where (Liljequist, 1962): Eb (1 A)G = global radiation absorbed by ground G,/G albedo, and Eeff effective surface, A long-wave radiation from the ground. The effective long-wave radiation from the ground can for example be computed from Angetröm’s formula = cTo4f(eo), where f(eo) = (Liljequist, 1962): Eeff a + b + 10-ce0, To= temperature in OK at instrument
-
=
=
=
=
71
H.Odin and K.Perttu
TABLE1. Albedo, from measurements by aircraft over different types of surface Location rind altitude
Near Luleå 200 rn
Surface
Concrete Meadow Bog Coniferous forest Sea Coniferous forest and bog Clearings Treeless mountain plateau Treeless moor Lakes Forest with bare patches
221 13 14 12 3-5
Luleå-Gallirare 500 rn
Linalompolo 100 m
Luleå Ontersund 500 m
10-11
10-11
15 10-11
12 14-15 15-16
OsteruundEnafors-Ontersund 500-600 m
12-13 4-8a
Enefors 100 rn
13
14 14-16 16-18
17 6-7 14-16
1. Measured on the ground. 2. Lake Storsjön.
altitude, eo = vapour pressure in millibars at instrum e n t altitude, and a,
c
greatly from place to place, the measured incident radiations cannot be taken as representative of a n y area larger than that of the instrument itself. T h e downward-facing meters, however, indicate the m e a n radiation from a n area of a size which is independent of the a m o u n t of shade. By taking readings of incoming radiation at a large n u m b e r of points close to each other at each site of measurement, it should be possible to obtain a m e a n value for incoming radiation that is comparable with the a m o u n t of radiation from the surface of the ground. T h e m e a n albedo a n d the m e a n radiation balance can then b e computed for individual places in the forest. In the present investigation, with only one reading per site as a rule, the m e a n values have been computed for the forest as a whole (from 31 measuring sites a n d 42 readings). INSTRUMENTS A N D M E T H O D
7 D
I
sin h
Sliori-xm~ i n d i d o n 0.3-3p (SB- Fig. I)
G.
E.
.E
L o n g - w a n radiotion 3-6Op
FIG. 9. Components of short- and long-wave radiation at a smooth and treeless surface. E, = long-waveradiation from the atmosphere E , = long-wave radiation from the ground =
72
T h e instruments consisted of t w o Moll-Gorczynski Solarimeters (0.3-3 pm), one facing u p w a r d a n d the other facing the ground, and a Middleton net radiometer (0.3-60 pm), all permanently m o u n t e d in a clearing at a height of 1.6 m, plus t w o Solarimeters (as above) a n d a Schenk net radiometer (0.3-60um), which were movable. T h e measurements were carried out in such a w a y that the movable instruments were m o v e d to selected points in the clearing, such as in hollows and over different types of ground vegetation; the measuring points in the forest lay in a straight line and were 20 m apart. This w a s designed to determine the m e a n conditions in the Linalompolo forest, which is sparse and consists of pine a n d mountain birch. T h e tallest pines are about 15 m high.
Radiation measurements near the forest limit in northern Sweden
FIG. 10. Sketch-map of experimental area, Linalompolo.
i + *
c
c
c c
' 4 +
clearing I
forest
*
+
+ !I I
o
FIG.11. Linalompolo, 18 July 1965.
50
100
150
+
4
+
L
t
* *
L
+ c Gällivare
200 m
1.101 iy/min
Global radiation, measured by fixed ), and instrumentsin the clearing ( by movable instruments (---) in both clearing and forest. C: clearing; F : forest.
T h e measurement area is situated at a height of about 475 m above sea level. Figure 10 shows a sketch of the measurement area and the positions of the measuring points. In order to be able to compare the radiation conditions at the various measuring points with the fixed measurements in the clearing, one must eliminate the daily rise and fall of radiation. T h e relative difference in global radiation is therefore formulated in the expression :
where Geix is the global radiation measured with the fixed instrument in the clearing and G is the global radiation measured at the s a m e time with the movable instrument (either in the clearing or in the forest). In the s a m e m a n n e r as for global radiation, the relative percentage difference for radiation balance is expressed by the formula :
=
(
(Eb)fix- Eb )mean (Eb) fix
where (Eó)eixis the radiation balance measured with the fixed instrument in the clearing and Eb is the radiation balance measured with the movable instrum e n t (either in the clearing or in the forest). M E A S U R E M E NTS
T h e global radiation at the mast carrying the fixed instruments and at the various points measured by movable instruments is illustrated in Figure 11. T h e numbers of the measuring points are arranged along the abscissa and under these numbers appear the letter F (forest) or C (clearing). T h e measured radiation reading in ly/min is indicated o n the ordinate. T h e continuous line shows the profile of global radiation o n the clearing as measured b y the fixed instrument. This line is very smooth except at point 21, w h e n a small cloud briefly overshadowed the clearing at the
73
H.Odin and K. Perttu
80 Albedo i
1-
10
FIG.12. Linalompolo, 18 July 1965. Albedo, measured by fixed instruments ), and by movable in the clearing ( instruments (O). Open circles (O) indicate points where measured incoming and outgoing radiation are not comparable. C: clearing; F: forest.
O
% O
O O
O
60O
ri0 30 20 -
50
O O O
O O
O
a -a-
10
e
a.
-
a
*+a
.=
..
0
O
FIG. 13. Linalompolo, 18 July 1965.
0.90-
0.80O
O
8.700.60-
O.b&
o .40 0.30-
.
O
O
100- ly/min
0.20.
O
O
O
O O
.
Radiation balance, measured by fixed instrumentsin the clearing ( ), and by movable instruments in both clearing and forest (0); radiation balance calculated by Angström's formula (x). Open circles indicate points where measured incoming and outgoing radiation are not comparable. C: clearing; F: forest.
0.10-
mast. T h e global radiation measured by the movable instrument is indicated by the broken line. T h e variations were small in the clearing,but the global radiation w a s greater in the t w o hollows (points 8 and 45) than o n level ground, a n d greater at a reading height of 10 c m above ground (point 4) than at points farther from the ground. T h e global radiation in the forest varied a great deal from one measuring point to another but w a s nowhere greater than in the clearing. Figure 12 shows the albedo in the clearing measured with the fixed instruments (continuous curve) and the albedo measured with the movable instruments (dotted curve). O p e n circles denote points where the measurements of reflected radiation a n d of global
74
radiation from above cannot b e compared. These are included only for the purpose of indicating the great variations in the forest. T h e albedo of the clearing measured with fixed instruments w a s constant through the forenoon, but rose a few per cent during the afternoon. T h e continuous radiation balance curve (Fig. 13) measured by the fixed instrument has a s m o o t h profile T h e value at the end of the measurement period wan half that during the forenoon. T h e readings obtained with the movable instrument (dotted curve) s h o w little variation over the clearing. T h e measurement at 10 c m above ground (point 4), however, gave a lower reading than those m a d e at greater heights. In the forest, a radiation balance higher than that
Radiation measurements near the forest limit in northern Sweden
Wind direction and velocity in the clearing. Air temperature in the clearing (mast) and in the forest (point 26).
- --- *-u
o
9o
180-
270360-
.
.- . ...-
-
--o’
o
o-
0
. --• .
zt LH S ~W
* .
N
64.
tsaM2-
10 m altitude
OJ
15
r
Air temperature
19-
I
l
O c
-
in the clearing w a s found in clearings that showed n o reduction in global radiation (points 12, 15, 19 and 37). T h e meters gave lower readings in shaded areas (open circles) than in the clearing. In these areas the m e a surements of incoming a n d outgoing radiation are not comparable. Negative readings were obtained at points completely in the shade, i.e., the radiation coming from below w a s greater than that coming from above. T h e negative readings are in agreement with the temperature inversion that takes place during the daytime at the ground near and under tree tops (Odin, 1964). T h e crosses in Figure 13 s h o w radiation balance values computed from Angström’s formula. Temperature measurements were m a d e at the mast in the clearing and at one point in the forest (Fig.14). T h e thermometer bulbs were placed in ventilated radiation screens. T h e variation w a s great at both points. The station in the forest (broken line) had a higher temperature than the clearing during the forenoon, w h e n the area w a s sunlit, but a lower t e m perature in the afternoon, w h e n it w a s in the shade.
R E S U L T S A N D DISCUSSION
Clearing
On the average there w a s n o difference in the global radiation measured by the fixed and movable instruments (Table 2). T w o measurements in t w o deep hollows have however been excluded. T h e m a x i m u m relative difference lay between -3 a n d +3 per cent a n d is d u e to uncertainty in the readings a n d to the fact that the movable meter w a s not always perfectly horizontal. In the case of radiation balance the m e a n difference w a s +2 per cent in the clearing ( m a x i m u m relative difference -1 to +6 per cent), which m a y be d u e to calibration error. N o differences were obtained in radiation balance between points above different types of ground. T h e albedo of the clearing as measured by the movable instruments w a s 15 per cent. According to 1 per cent the fixed instrument, the albedo w a s 14 in the forenoon a n d 16 1 per cent in the afternoon.
&
TABLE2. Comparison between fixed and movable instrumentsin the clearing Relative difference global radiation
ii- W o ) Average M a x i m u m variation Total
O
-3-+3 11
Albedo
radiation balance
q
movable instrumente
fixed instruments
10.3912.54 h
(“lo)
2 -+1-+6 9
p/J
15 14-17 10
14
*1
13-16 23
14.3916.54 b
16 i- 1 16-18 23
75
H.Odin and K. Perttu
All these figures are in good agreement with the results obtained by measurement from the air, T h e albedo w a s similar (16 per cent) for t w o types of ground vegetation, namely lichen and grass. T h e albedo reading w a s not affected w h e n the instruments were lowered from 1.6 m to 0.5 m above the ground, but diminished with further reduction of the reading height. Forest
Table
3 illustrates the reduction of global radiation
in fjeld forest as compared to clearing, i.e. the relative difference between global radiation in the clearing (fixed instrument) and in the forest (movable instrument) according to the formula o n pages 71-3. T h e distribution reflects the low density of the fjeld forest, which is in the nature of open ground with individual trees or groups of trees. W h e n the instruments were exposed to sunlight, little or n o reduction w a s noted even w h e n they stood close to trees, but as soon as they were in the shade, a large reduction w a s
TABLE3. Distribution
of the measuring points in respect of reduction of global radiation in the forest
Reduction of global radiation
-3-+ 3 4-10 11-17 18-24 25-31 32-38 39-45 46-52 53-59 60-66 67-73 74-80 81-87 88-94 95-100
Measuring point
Total
7
10, 11, 12, 15, 19, 33, 37 18, 20, 27, 30, 41, 44
6 O O
O O 26 29, 40 13
1 2 1
found. In a denser forest a lower percentage of cases is found in the first t w o classes. Table 4 gives the m e a n radiation conditions in the forest-global radiation a n d radiation balance are s h o w n in relation to the clearing in the form of relative differences. T h e relative reduction of global radiation in the forest w a s 44 per cent, i.e., the ground in the forest along the lines of measurement received o n the average a short-wave radiation that w a s 56 per cent of the global radiation in the clearing, which is the s a m e as at tree-top height. T h e short-wave radiation from the ground in the forest averaged 29 per cent of the shortw a v e radiation coming from above. A t points in unshaded glades the albedo w a s nearly equal to that of the clearing. T h e ground vegetation in forest and clearings at Linalompolo thus has approximately the s a m e albedo. Measurements m a d e over forest from the air gave a n albedo of 10-11 per cent. T h e ratio of short-wave radiation coming from below to incident global radiation at treep-top level has been computed for each measuring point in the forest, i.e., the influence of shadows has been taken into account. T h e average figure for the measuring points w a s 10 per cent. T h u s the ground shadows in the sparse fjeld forest were a strong contributory factor in making the albedo over the forest at Linalompolo (measured f r o m the air) lower than the albedo over treeless ground. T h e ground in the forest h a d o n the average a positive radiation balance 50 per cent lower than that in the clearing, i.e., the net incident radiation in the forest w a s half that in the clearing. However, the glades in the sparse forest gave higher figures for radiation balance than did the clearing. T h e effective emitted long-wave radiation (with k n o w n ) w a s about 60 per cent lower from the and forest than from the clearing.
7:
O 9, 36, 38 43 16, 34. 39 14, 17, 28, 31, 32, 35, 42
3 1 3
Conclusions concerning the energy balance in Clearing and forest at Linalompolo on 18 July 1965
7 O
T h e energy balance can be expressed by the formula
(1
TABLE4. Relative differences of global radiation
(z)and
radiation balance (E) between clearing and forest, plus albedo and estimated area of shaded ground in the forest Percentages
K
Average
76
+ 44
8
+ 50
Albedo
29
Shaded ground
40
- A)G - (E, - Ea) + H = L(ETA)+ Q i- S
where
(1
- A)G - (E, - E a ) = Eb.
T h e energy absorbed by the ground in the clearing (1 A)G,which is derived from the global radiation, w a s nearly twice as m u c h as that absorbed by the ground in the forest. T h e absorbed energy, which causes a rise in temperature, is partly propagated d o w n into the ground by molecular thermal conduction (S).Most of the energy, however, is re-emitted to the atmosphere in a n u m b e r of different forms, the three principal ones being long-wave radiation from evaporation from the surface the surface layer (Enz),
-
Radiation measurements near the forest limit in northern Sweden
layer, L(ETA),and turbulent heat transfer (Q)from the ground upwards. T h e average ground temperatures in clearing and forest can be assumed to be equal, since the ground in the forest had an average shade factor of 40 per cent (estimated from photographs taken from each measuring point), and the sunlit areas in the forest had a higher temperature than the clearing, while the shaded areas had a lower temperature. Long-wave radiation from the ground in clearing and forest was thus equal on the average. Since the effective long-wave radiation emission from the ground was lower in the forest than in the clearing, more long-wave radiation fell upon the ground in the forest than in the clearing. Let us assume that the quantities of energy stored in the ground in the clearing and in the sparse forest were of equal magnitude (same average temperature) and that the transfer of energy by horizontal advection
(H) can be neglected in both localities.Since the ground in the clearing received twice as m u c h radiant energy (the positive radiation equilibrium Eb was 50 per cent greater) as the ground in the forest,the above assumptions lead to the conclusion that the amount of energy accounted for by evaporation and turbulent heat transfer from the ground together was twice as m u c h in the clearing as in the forest. This state of affairs is reflected by the nature of the ground vegetation in the two areas. T h e vegetation in the clearing consists largely of lichens and dry heather, and the humus layer is very thin. In the forest, the humus layer is considerably thicker and the ground vegetation consists predominantly of bilberry and mosses. In the forest, ground vegetation similar to that of the clearing is found in very large glades as well as in the zone extending 30-50m into the forest from the edge of the clearing.
Résumé Mesure du rayonnement aux limites de la forêt dans la Suède septentrionale (H. Odin et K. Perttu)
L’étude avait pour objet de fournir des données sur l’albédo de différents types de surface et de comparer l’échange total de rayonnement à la surface du sol dans une clairière et dans une forêt de fjeld. Les mesures de l’albédo ont été faites dans de vastes régions, mais les autres questions n’ont été étudiées qu’en un seul lieu par un jour d’été où le ciel était sans nuages. Un avion a été utilisé pour obtenir certainesmesures ; les autres ont été effectuées au lieu mentionné. On a découvert que les terrains dépourvus d’arbres c o m m e les clairières,les plateaux montagneux dénudés et les marécages avaient un albédo plus élevé (14-18y0) que les forêts de conifères (10-13y0)et que les forêts denses avaient un albédo plus élevé (12-13y0)que les forêts clairsemées (lO-ll~o).L a présence d’ombres sur le sol est une des raisons pour lesquelles les forêts clairsemées ont un albédo inférieur à celui des terrains découverts et des forêts denses. A Linalompolo,en Laponie du Nord (67’20nord 20,50 est), dans une clairière, on a trouvé la m ê m e valeur pour l’albédo mesuré au sol et pour l’albédo mesuré d’une certaine altitude. En outre, l’albédo
du tapis végétal de la forêt voisine était à peu près le m ê m e que celui de la clairière. Dans la forêt, le rayonnement incident à faible longueur d’ondes (0.3-3 p) était en moyenne d’environ la moitié de celui de la clairière. L e fait qu’en 42% des points où les mesures ont été faites o n n’a guère enregistré de diminution de ce rayonnement montre que la densité des arbres était faible. L e bilan du rayonnement, c’est-à-direla différence entre le rayonnement incident global et le rayonnement réfléchi global (0,3-60p), était en moyenne positif aussi bien pour la forêt que pour la clairière, mais dans la forêt il était deux fois moins important que dans la clairière. On a évalué à 40% la proportion ombragée du sol de la forêt. Les mesures de la température, à un point situé dans la forêt et à un autre dans la clairière, ont confirmé l’hypothèse selon laquelle la température à la surface du sol et l’énergie emmagasinée dans le sol étaient en moyenne les m ê m e s dans la clairière que dans la forêt. Ces observations,ainsi que la mesure du bilan du rayonnement, donnent à penser que la quantité d’énergie disponible pour l’évaporation totale et les échanges turbulents de chaleur sont deux fois plus élevés au niveau du sol dans la clairière que dans la forêt.
H.Odin and K. Perttu
Discussion S. HUOVILA. The Eppley pyrheliometer has been used, in the
H. ODIN.Yes, perhaps, and we know nothing about the
same way as described by Mr. Odin, at the University of Wisconsin. The results were not m u c h different from those presented by Mr. Odin. The albedo of the Earth has also been measured from meteorological satellites (TIROS)and values of about 35 per cent for the whole earth surface have been received.
effect on the turbulence patterns of the instruments’ being mounted on the wings of the aeroplane. There are also
F. E. ECHARDT. J’ai écouté avec le plus vif intérêt votre excellent exposé sur les échanges d‘énergie rayonnante s’effectuant au niveau de la basse atmosphère. Je vous félicite pour avoir si bien et si clairement indiqué non seulement le type d’instruments utilisé, mais aussi le mode d‘installation de ces derniers. Vous avez utilisé, pour vos mesures du rayonnement de courte longueur d’onde,deux solarimètres de Moll-Gorczinsky fixés respectivement à la face supérieure et à la face inférieure d’une aile d’avion. Les deux appareils sont donc exposés à des régimes tourbillonnaires très différents. Ne pensez-vous pas qu’il peut en résulter des erreurs? E n effet, le signal électrique fourni par ce type de solarimètre dépend du gradient de température existant entre la surface réceptrice et un drain de chaleur ou heat sink qui est constitué par la base du boîtier de l’instrument.
other sources of errors as you mentioned both in the instruments, and in the meaning technique-some were mentioned in m y paper. We chose this type of instrument (a solarimeter by Kipp u. Zonen) for measuring shortwave radiation because it is generally used at Swedish meteorological stations, and for the practical reason that it was easy to mount on the wings of the aeroplane. The instruments were calibrated against Angström’s pyrheliometer. The solarimeter takes about twenty seconds to react, in moving from zero to full deflection, but the variation in solar radiation is small during clear weather. T o eliminate the error due to this reaction time, the measurements were made over extensive areas of any one type, such as clearfelled areas, coniferous forests, etc. Moreover, the readings were repeated over each type of terrain. Despite these sources of error, w e obtained the same albedo from aerial measurements as from those m a d e under more closely controlled conditions on the ground at Linalompolo. There was only a small variation in the albedo of any one type of terrain when it was measured from the air.
Bibliography / Bibliographie DAVIES, J. A. 1963. Albedo investigations in LabradorUngava. Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B, Wien. FRITZ,S. 1948. The albedo of the ground and atmosphere. Bull. Amer. Met. Soc. GEIGER, R. 1961. Das Klima der bodennahen Luftschicht. HOHNE, W. 1965. Ein Beitrag zur Strahlungsbilanz-Messtechnik. Abh. des Met. Diensier der DDR,no. 74 (Band X), Akademie-Verlag. Berlin. JACKSON C. 1961. Estimates of total radiation and albedo in subarctic Canada. Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B, Wien.
78
LILJEQUIST,H. 1962. Meteorologi. Stockholm. ODIN, H.1964. Skogsmeteorologiska undersökningar i höjdlägen. Kungl. Skogs- och lantbruksakademiens Tidskrift, no. 2-3. ROBINSON,G. D. 1959. S o m e observations from aircraft of surface albedo and the albedo and absorption of cloud. Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B, Wien. ANGSTROM-WALLEN.1937. O n the illumination in stands of different character and density. S o m m . met. agricole, Tag. Bet. Salzburg.
Temperature sum as a restricting factor in the development of forest in the Subarctic R. Sarvas
There are several climatic factors o n the basis of which the northern timber line is formed, but temperature is undoubtedly the most important. It has, however, turned out to be difficult to describe exactly the influence of temperature. Apparently, it is manifold a n d scarcely describable with one regress only. T h e most important of the adaptive characteristics is probably the adaptation of a sub-population to the length of the local growing season. In another connexion (Carvas, 1965)the author has described investigations and their results pertaining to adaptation to the length of the growing season by s o m e tree species growing in Finland. As the publication is still in print, the m a i n points are briefly outlined here. T h e subject w a s “ T h e Annual Period of Developm e n t of Forest Trees”. T h e annual period consists of numerous sequences of physiological events, which can only occur in a certain intrasequence; they are, therefore, given a n exact position in the period. These sequences form a closed cycle, inasmuch as they end every a u t u m n in the s a m e physiological state from which they started in the spring. Events making up the annual period include flushing, leaf-fall, annual diameter growth, height growth, flowering, maturing of seed, and so forth. Important activities that are apparently not tied to the time-table of the annual period include, for instance. CO2-assimilation,assimilation of mineral nutrients, respiration, and so forth. It w a s supposed that the time-table of the annual period w a s under genetic control. It w a s further supposed that this control system does not interfere with the internal mechanism of the events, but only regulates the time-table. It w a s established that the control system does not measure time but the thermal factor. As for the calculation of temperature sums, out of m a n y alternatives tried so far (cf. Schnelle, 1955), the principle defined by the following equation has proved to be the most useful:
T+50
c
=Y In=
(tm 1
-5).
where T+50 c indicates the temperature s u m , the threshold value of which is +50 C, n the total of days with a m e a n temperature higher than +50 C a n d tm the m e a n temperature of those days. As in m a n y previous investigations, d.d. (degree days) w a s used as the symbol for the unit of the temperature sum arrived at in this way. In investigations carried out during several consecutive years, w h e n the temperature readings were taken at the height of the tree canopies of dominant trees, certain events of the annual period, for instance, the average temperature s u m of anthesis, always occurred, if the entire subpopulation w a s flowering, at almost the s a m e t e m perature sum. For instance, in Tuusula (about 25 k m north of Helsinki) the average temperature s u m s of the anthesis of Picea abies were as follows: 139.0 d.d. (in 1951), 140.0 d.d. (1956), 141.5 d.d. (1962) and 139.3 d.d. (in 1964) or, o n a n average, 140.0 d.d. It further proved that in the m a i n part of the range of a species the average temperature s u m of a certain physiological event of the annual period, as in the case of anthesis, is the s a m e percentage of the local average annual temperature sum. For instance, the above-mentioned average temperature s u m of the anthesis of P. abies, viz., 140.0 d.d., is 10.5 per cent of the average annual temperature s u m of 1,333 d.d. in the Tuusula area. On the basis of the measuring results published by Ccamoni (1955) it can b e calculated that the average temperature sum of the anthesis of P. abies is in Eberswalde (25 k m north-east of Berlin) about 199 d.d. (in 1936), which is about 9.9 per cent of the average annual temperature sum (2,013 d.d.) in the Eberswalde area. T h e percentage, or the relative temperature s u m of the anthesis, as it has been called in the present investigations, is thus almost the s a m e in Tuusula as in Eberswalde.
79
R. Servas
On the basis of the measuring results published by Chalupa (1964)it can be calculated,similarly,that the average temperature sum of the anthesis of P. abies is in Bedovice 180 d.d. (in 1958), which is about 10.1 per cent of the average annualtemperatures u m (1,772d.d.) in the Bedovice area. The relative temperature sum of anthesis is thus almost the same as in Tuusula and Eberswalde. The present investigations further supported the concept that the annual period of forest trees has become mainly adapted to the local average annual temperaturesum,and not to other factors,for instance, day length. However, it is evident that the annual period cannot adapt itselfto a very short growing season.It is evident in the case of Betula verrucosa, for instance, that the flushing, flowering, maturing of seed, dispersal of seed and leaf-fall cannot take place in the course of one day. It is almost equally evident that even a week is too short a period. There must exist a minimum time to which, for instance, the annual period of B. verrucosa can become adapted. This apparently applies to all forest trees. In view of subarctic and subalpine forests, forest-genetic as well as treeimprovement investigations give rise to an important and interesting question: what is the minimum length of time required? In Finland, detailed studies endeavouring to shed more light on the subject have been concentrated on Pinus silvestris.
In the spring of 1964, when P. silvestris flowered abundantly in all parts of Finland, anthesis measurements were carried out, using recording pollen samplers, (cf. Sarvas, 1952, p. 4-7;1962, p. 27) in several places in southern Finland and in one sample stand in northern Finland. In southern Finland,for instance in Tuusula, the mean obtained was 230.0 d.d., which is 17.2 per cent of the local average annual temperature s u m (1,333d.d.). This correspondsto the measurements carried out by Scamoni (1955) in Eberswalde: in 1936, the average temperature sum of the anthesis of P. silvestris was about 344 d.d. or 17.1 per cent of the average annual temperature s u m (2,013 d.d.) in Eberswalde. The relative temperature s u m of the anthesis of P.silvestris is thus in Eberswalde almost the same as in Tuusula. Measurements carried out in northern Finland in the neighbourhood of Rovaniemi (66021.4‘ N., 26044.2’ E., 118 m above sea level) gave 201.3 d.d. as the average temperature sum of the anthesis of P. silvestris,which is 22.2 per cent of the local average annual temperature s u m (904 d.d.). The relative temperaturesum thus rose from 17.2 per cent (Tuusula, southern Finland) to 22.2 per cent (in northern Finland). In the summer of 1965 the measurements were repeated in southern as well as in northern Finland. In northern Finland, the number of sample-standswas 80
increased to four. The n e w sample-standsare located as follows: two at Kittilä (68001.5’N.,24009.3‘ E., 280 m , and 68001.5‘ N.,24007.6’ E., 330 m) and one at Utsjoki (69045’N.,27002‘ E., 80 m). All the new sample-standsare almost at the arctic tree line. Measurements concerning the anthesis of P.silvestris that were carried out in southern Finland in 1965 gave practically the same results as in 1964: the average temperature s u m of anthesis at Tuusula was 230 d.d. In the sample-stand in the neighbourhood of Rovaniemi, the result obtained in 1965, namely 198.5 d.d., was slightly below that obtained the year before (201.3d.d.). The difference is,however, so small that it is hardly significant (as a time unit, the difference would amount to about three hours). In the two sample-stands at Kittilä, the average temperature sum of the anthesis was 195 d.d., which is 28.4 per cent of the local average annual temperature s u m (686 d.d.). In the Utsjoki sample-stand,the result was 196 d.d. (32.4 per cent of 605 d.d.). The relative temperatures u m of the anthesis of P.silvestris thus continues to increase as w e m o v e upward from Rovaniemi (66021’N.)toward the arctic timber line (69045’N.). This indicates that the further north w e get the worse the annual period of the P. silvestris subpopulations are adapted to the local average annual temperature sum. The present measurements also indicate the reason that the relative temperature sum of the anthesis of P. silvestris abruptly increases as w e approach the arctic tree limit. A s w e shift from Eberswalde to Tuusula,the numerator (the average temperature s u m of the anthesis) as well as the denominator (the local average annual temperature sum) decrease in the same proportion, leaving the quotient unchanged. However, as w e approach the arctic tree limit the decrease in the numerator (the average temperature s u m of the anthesis) stops but in the denominator (the local average annual temperature sum) the decrease continues and, consequently,the quotient increases. On the basis of obtained measuring results and by means of interpolation, it can be calculated that in a zone with an average annual temperature sum of about 950 d.d. the decrease in the averagetemperature sum of anthesis almost stops. Thus, 950 d.d. seems to be about the smallest average annual temperature s u m to which the annual period of a P. silvestris subpopulation has become in northern Europe genetically adapted. The result is surprising. The general presumption has probably been that the line is m u c h further north, somewhere close to the arctic tree line. In Finland, for instance, around 40 per cent of the forested area falls north of this “critical line”. North of this line, the annual period of P.silvestris sub-populationsstops, in most years, prematurely. The consequences are fatal. The best k n o w n is probably that the seed matures only in warmer-than-average summers.
Temperature sum as a restricting factor in the development of forest in the subarctic
Actually, from this last-mentioned generally k n o w n fact (e.g. Kujala, 1927), it could have been concluded long ago that the P. silvestris sub-populations in this zone have not b e c o m e fully adapted genetically to the local climate. North of this critical line, the existence of P. silvestris sub-populations is, o n account of their genetic non-adaptability, largely dependent o n warmer-than-average growing seasons and modificative variation. Let us p a y further attention to another interesting characteristic: as the arctic tree line is approached, the selection pressure for a short annual period increases m o r e and more, but this increase has n o genetic effect. W h y not? Because all the existing biological potentials of the sub-populations concerned, the genetic ones included, have evidently been exhausted for the shortening of the annual period. It is, however, most interesting to ascertain that
the existence of the nordic forests does not necessarily require that they have in every respect accommodated themselves to the climate where they grow; o n the contrary, for example the forests of the Nordic Fennoscandia offer a n example of a region where the yearly period of the tree population has not adapted itself to the length of the growing season. And apparently forests like these can be found o n huge areas of the globe o n the southern and lower part of the arctic a n d alpine forest line. There does, however, exist a n absolute limit which the forest and, n o doubt, a n y other vegetation cannot cross: the limit beyond which the nutritive balance remains negative during a n u m b e r of years. A temperature s u m of the d.d. type offers hardly a n y assistance in the climate description of this limit. Assimilation and growth depend essentially in a different w a y o n the temperature than yearly period.
Résumé L a temph-ature globale en tant que facteur restrictif dans le développement des for6ts subarctiques (R.Sarvas)
L’auteur expose les résultats de recherches qui montrent l’effet du facteur thermique sur la période annuelle des arbres des peuplements sylvestres. Plusieurs méthodes ont été essayées pour calculer la température globale, mais c’est l’équation ci-après qui a permis d’obtenir le chiffre le plus utile: n
Ti6 OC
=
(tm - 5). In= 1
où T+5 indique la température globale (nombre de jours-degrés) dont le seuil se situe à + 5 OC, n le n o m b r e total des jours pendant lesquels la température m o y e n n e a été supérieure à + 5 O C et ,t la température m o y e n n e enregistrée pendant les jours en question. Les recherches actuelles montrent qu’un certain
événement physiologique de la période annuelle d’une essence, l’anthèse par exemple, se produit à la m ê m e température globale relative dans sa principale manifestation (la température globale relative est le quotient de la température globale m o y e n n e de l’événem e n t par la température globale locale [en m o y e n n e annuelle]). Cependant, près de la limite de la forêt arctique (ou alpine), la température globale relative augmente, ce qui montre que la période annuelle des sous-peuplements d’arbres considérés n e s’est pas adaptée génétiquement au climat local. L a ligne critique a u nord de laquelle la période annuelle n’est plus génétiquement adaptée a u climat local se situe plus a u sud qu’on n e s’y attendrait; dans le cas, par exemple, du Pinus silvestris, elle se situe dans u n e zone où la température globale locale ( m o y e n n e annuelle) est de 950 jours-degrés (contre 600 jours-degrés à la limite de la forêt).
Discussion J. BL~THGEN. Professor Sarvas’ critical line for degree days sum is one for flowering, m y reproduction line (line for ripening seeds) thus must lie somewhat south of the flowering temperature sum.
R. SARVAS.In m y opinion, m y critical line is a theoretical concept to such an extent that there is no use comparing it with the reproduction line. C. O. TAMM. I agree with Professor Sarvas that temperature
sums are useful climatic parameters, and I only wish that there could be some international agreement on the use of one or a few temperature parameters instead of the present mores of terms like “vegetation periods”, “temperature ums", “tetraterms”, etc. However, I disagree with what has been suggested in the discussion, viz, that the critical line is where pine reproduction has a 50 per cent chance. Considering trees with a lifetime of several centuries, a good reproduction year once in 50 or 100 years will be enough to warrant the persistence of
81 6
R . Sarvas
the forest. Professor Hustich’s “climatic hazard factor” will provide such opportunities even where average summer temperatures are considerably below the physiological needs for flower initiation, seed ripening, etc.
R. SARVAS. I want to emphasize the fact that the concept,
R . SARVAS. 1. W h e n I talk about genetic behaviour, I refer to only one characteristic, as Professor Sirén mentioned, namely the annual period. 2. I still want to stress that m y viewpoint is theoretical to a large extent while Professor Tamm’s is of a more practical nature.
critical line, as defined by m e is only theoretical. M y critical line does not necessarily mean anything critical from the point of view of the persistence of the forest. M y critical line is critical only so far that the forest populations of the tree species in question have adapted themselves genetically to the average annual temperature s u m of the locality, i.e., the length of the vegetation period, only as far as this line but not to the north of it.
species, the spruce (Picea abies) and the birch (Betula pubescens), could you guess where the “critical line” of these species is situated in Finland?
G. SIREN.1. Regarding genetic homogeneity of pines at the tree line it should be observed that Professor Sarvas is discussing the genetic homogeneity of the pine population considering only one characteristic. In reality the genetic variation of fine especially at the tree line is extremely wide due to the lack of interindividual competition.Then of course quite a number of different tree characteristics contribute to the accumulated effect of variation. 2. Taking into account the longer rotation period in the northern forests the discrepancy between the viewpoints of Professor T a m m and the proposals of Professor Sarvas seems to be reduced considerably even if not enough.
F.E.ECKARDT. Pourriez-vousindiquerla raison pour laquelle vous avez choisi le seuil de 5 O C figurant dans votre formule?
T. AHTI.Although you have not studied the other tree
R. SARVAS. For the present I do not want to express m y opinion concerning other tree species than Pinus silvestris, because data of sufficient reliability are available for this species only. Certain observations, however, indicate that, as far as Finnish tree species are concerned, the critical line lies farther to the north for Betula pubescens than for other species.
R . SARVAS.I have tried several different methods for the calculation of the temperature s u m and I have chosen the one which during several consecutive years has given a certain phenological process with the least diverging values of the temperature sum.
Bibliography / Bibliographie CHALUPA, V. 1964. Dynamika kvetení lesních drevin. Práce vyzkumn$ch icstavú lesnick$ch CSSR,svazek 28, p. 141-173. KUJALA, V. 1927. Untersuchungen über den Bau und die Keimfähigkeit von Kiefern- und Fichtensamen in Finnland. ( C o m m . Inst. Quaest. forest. Finl. 12.) SARVAS, R.1952. On thefloweringof birchandthequality ofseed crop. ( C o m m . Inst. forest. fenn. 40.) __ . 1962. Investigations on the flowering and seed crop of
Pinus silvestris. (Comm. Inst. forest. fenn. 53.) SARVAS, R. 1966. T h e annual period of development of forest trees. Proc. Finn. Acad. Sci.Lett. (1965). SCAMONI, A. 1955. Beobachtungen über den Pollenflug der Waldbäume in Eberswalde. Zeitschr. f. Forstgenetik u. Forstpflanzenzücht,vol. 4, p. 113-122. SCHNELLE, F. 1955. Pflanzen-Phanologie.(Probleme der Bioklimatologie 3.)
Some ecological aspects of snow W.O.Pruitt,Jr.
INTRODUCTION
Although some of the early naturalists (e.g. Seton,
1909 ; Dugmore, 1913) appreciated the ecological role W h e n Good Icing Wenceslaus looked out,the snow that he saw m a y have been deep but it almost certainly was not crisp and even. These latter two properties are very unusual for snow covers in the Northern Hemisphere, where about 53 per cent of the land area possesses a snow cover at some time during the year. Generations of graduate students would doubtless have received degrees for research into the properties and potential uses of snow, if snowflakes were rare objects that were obtainable only by expensive and involved laboratory processes. As it is, some tens of billions of these beautiful crystals of frozen watervapour pile up in each square metre of snow. As w e will seelater,there is a basic differencebetween the ecological relations of mammals and birds which inhabit environments having a snow cover that lasts all winter and those wherein the cover is intermittent or lacking. At the same time there is also a basic difference between those h u m a n cultures which are adapted to a permanent winter snow cover and those which are not. The reactions of various h u m a n cultures to snow vary from a welcoming acceptance to near hysteria when it falls. T h e culture of western Europe and those derived from it usually look upon snow only as something to be got rid of as quickly as possible. Vast sums of money are spent every winter in removing snow from roads,sidewalks and air strips. Recognition of the importance of snow by agriculturists and investigation of ways to retain the snow cover on semi-arid cropland are comparatively recent innovations in Western technology. The very abundance of snow seems to have suppressed almost all but the negative aspects of getting rid of it as quickly as [possible. In the literature of the sciences that ought to be most concerned there is even yet little to suggest that snow is a major element in the environment of life.
of snow the later preoccupation with synecology which accompanied the rise of the separate discipline of ecology overshadowed the rather personal contact with the environment which the earlier workers enjoyed. The modern period of appreciation of the role of snow as an ecological factor must date from 1946 when Formozov published his classic work on the subject. It is indeed fitting that this symposium should convene on the twentieth anniversary of the publication of Formozov’s paper. Certainly not the least of the contributions in Formozov’s paper was his classification of animals based on their ecological reactions to snow. “Snow cover, for m a n y species,is the most important element of environmental resistance and the struggle against this particular element is almost beyond some species’ ability. Such species do not inhabit snowy regions and w e can unite them into a group that avoids snow, or ‘chionophobes’-the small cats, steppe antelope,steppe sand grouse, black partridge, m a n y small terrestrial birds, etc. This group is connected by a number of gradations with the species than can withstand winters with considerable snow. These species w e can call ‘chioneuphores’ (moose, reindeer, wolverine, wolf, fox, m a n y voles, moles, shrews, etc.). Finally there are forms which have characteristic adaptations (winter-white coloration, winter peculiarities of footcoverings,etc.) which were undoubtedly perfected by snow cover taking part in selection. The ranges of these forms lie completely or almost completely in regions of hard and continuous winters with m u c h snow (willow ptarmigan, rock ptarmigan, varying hare, arctic fox, collared lemming, etc.). ‘Chionees’ (snowy) or ‘chioniophiles’ (snGw-lovers) are quite appropriate names for these forms ...” (Formozov,
1946).
83
W.O.Pruitt,Jr.
If the role of the snow cover in the ecology of animal (and plants) has been so poorly understood, consider the situation concerning man. One of the significant inventions of boreal m a n was the snowshoe. Davidson (1937) in his monographic work on snow-shoes made the flat statement, when discussing the spread of various snow-shoetypes,“It is important to note that the type of snow is not a barrier”. H e did indeed acknowledge that the presence of snow is a pre-requisitefor the presence of snow-shoes,but could not bring himself to acknowledgefurtherrelationships. If, however, w e compare his maps of the distribution of various characteristicsof snowshoe types in:northern North America with, for example, the m a p (Fig.1) of the major snow regions of Canada (Gold and Williams, 1957) w e find a number of striking,and significant, correlations. For example, the Freeze-Thaw region of eastern Canada, with its dense, yet thick, cover, has several traits restricted to it, namely, most of the native occurrences of the “bearpaw” type, the “Naskapi frame”, and the occurrence of one or two cross-pieces.The “bent toe” and “square toe” traits are restricted to the Freeze-Thaw and Prairie regions, both of which are, of course, characterized by dense snow. The so-called Athapaskan type of two-piece frame and the “turned-uptoe” trait are both lacking from the Freeze-Thaw region. The “hexagonal” type of netting weave is found over most of the Northern
Forest region (characterized by light, fluffy snow cover) while the “rectangular” weave is restricted to those parts of the taiga with the lightest, fluffiest snow cover. The Northern Forest region,with a snow cover which offers little support to the netting, is characterized by snowshoes with three supporting crosspieces. Snowshoes with no toe-hole are found only in some regions of hard, dense snow or infrequent snow. Davidson commented on the pride of workmanship and fine appearance of the snow-shoesin the Mackenzie Basin. H e noted that other traits just happened to be found in the western Mackenzie Basin. W e , as boreal ecologists, k n o w that the Mackenzie Basin is a region characterized by very light, fluffy snow which is little moved by wind. This is clearly a region where selection pressures are rigorous (i.e.,fine workmanship the usual condition) and close conformation to a successful pattern is vital to the survival of the user. Most of the traits peculiar to show-shoe types have a clear and immediate relation with the characteristics of the regional snow cover where they were used. Thus Occam’s Razor compels us to consider these relationships the more likely explanation, instead of the rather nebulous “peripheral spread” of traits invoked by Davidson. In a consideration of such flotation mechanisms as snow-shoes,the prime data should be the surface area of the frame and netting
FIG.1. Map of
m a j o r snow regions in Canada.( F r o m National Research Council Publication4389.)
84
Some ecological aspects of snow
complex as well as the average weights of the users. Unfortunately these data were not presented. The culture which was without doubt the most successfully adapted to snow was that of the Eskimo of the central North American Arctic. In this culture snow was frequently used as a building material that had admirable properties and ease of handling. Cultural adaptations had been evolved so that the eightor nine-month snow cover was utilized in lieu of a wheel. Nothing having the same qualities of ease of pulling, lightness, ruggedness and economy of manufacture and maintainance has surpassed the Eskimo komatik as the perfect vehicle for traversing windhardened tundra snow. Within the regions of the Northern Hemisphere characterized by the boreal coniferous forest or taiga the snow cover lies quite different from that of the tundra regions. Here in the forest the snow cover lies thick, fluffy and soft and it m a y last nearly as long as on the tundra. The Eskimo komatik bogs down helpless here and the snow is too soft and light to be cut into blocks for building purposes. T h e culture of the northern Indians-the Dinje of Alaska, the Loucheux of Yukon, the Dogribs, Hares, Yellowknives and Chipewyans of the North-west Territories and adjacent parts of the Canadian Provinces-were all adapted in varying degrees to cope with a permanent winter snow cover. In addition to snow-shoes, other adaptations to soft, fluffy forest snow are the woodland toboggan and the tandem dog hitch. In former days the Dinje (Athapaskans) of Alaska made temporary snow-houses (quin-zee) by heaping up a huge mound of snow, allowing it to “set” for several hours, and then hollowing out a cavern inside the mound. This same principle of disturbing taiga snow and letting it “set” has been used in the construction of winter air strips in the north. More recently, m y colleagues and I ( E h e r and Pruitt, 1959) used the same principle in modernizing the technique of the Dinje to construct emergency survival shelters in the subarctic forests.
TAIGA S N O W As a result of experience with snow at varied places in the northern parts of North America I have come to look on the snow of the subarctic taiga in interior Alaska as “typical” snow. That is,it is snow that is the least modified by external factors. There, in the great topographic bowl north of the Alaska Range and south of the Brooks Range, the snow arrives early in the autumn and remainsuntillatespring (Pruitt,1957).For a good part of the winter it is virtually unaffected by incoming solar radiation (List, 1953), since this region is only two degrees south of the Arctic Circle. The peculiar meteorological conditions prevailing there cause a standing inversion to be c o m m o n
(Johnson, 1953), thus the snow is little affected by wind. The result is a snow cover that piles up loose and fluffy, modified only be the beat and moisture which rise through it from the soil below. T h e thermal inversion is also responsible for another phenomenon which has ecological importance-the deep cold of the calm, dense, air that puddles in the great, flat, valleys of the Tanana and Y u k o n rivers, and is also responsible for the “ice-fog’’that at these times rises from any source of free moisture. The snow piles up on the tree branches and the surface is rough with crystals, lying just as they fell and undisturbed by
wind.
Let us n o w examine a taiga snow cover in some detail. Snow, as w e all know, is but frozen water vapour. All gradations between the various kinds of frozen hydrometeors occur. e.g.,snow grades graduaIly into graupel depending upon the degree of crushing to which it has been subjected while falling. In temperate and cold-temperateregions snow usually occurs as six-armed stars or plates which have diameters of 1-5 mm and thicknesses of about 0.1 mm. (In the Arctic and Subarctic snow occurs more frequently as needles or tiny prisms, ranging in length from 0.1 to 3 mm and in diameter from 0.01 to 0.2 mm) T h e stars or plates, although rather small and simple when first formed, usually increase in size as they fall through the air and become extremely complex through accumulation of additional frozen vapour on to the tips of the arms and through aggregation with other flakes. F r o m the m o m e n t they come to rest on the earth or on other flakes which had preceded them on to our snow study plot, the delicate stars and flakes change their shape. In its undisturbed, uncompacted state a taiga snow cover m a y best be visualized as an emulsion of air and ice, with the amount of air far exceeding the amount of ice. For example, new-fallen taiga snow m a y have densities of 0.05. or even less. As a rule, even the coldest and dryest soil is markedly warmer and moister than the air through which the flakes have fallen. Therefore heat and moisture flow upward from the soil through the snow cover. These temperature and vapour pressure gradients are of extreme importance in the developmental history of the cover. Through the process of sublimation,niolecules of water leave the attenuated tips of the rays of each flake and attach themselves to other flakes which are further from the soil and therefore colder. Thus the larger and more massive flakes grow at the expense of the smaller or more delicate flakes and eventually the basal layer of the cover assumes a granular structure,the pukak, or what is erroneously called “depth hoar”. T h e snow particles next to the substrate m a y in time be eroded completely away, leaving,just above the soil, a vacant space with fragile, lattice-likewalls and roof. As the cover thickens with successive falls, s o m 85
W.O.Pruitt,Jr.
pression takes place. Each fall of snow, having originated through a different sequence of meteorological events is somewhat different from all others. Therefore the snow cover builds up by an accumulationof layers, each of a different thickness, hardness, density and type of flake or grain. Possibly a warm, moist air mass moves across our snow study plot, bringing the temperatureto thawing, or perhaps even a light rain falls. The upper centimetre or so of the cover n o w has more moisture in it, increasing its density. W h e n the w a r m air is replaced by colder air this additional moisture freezes. Depending on h o w m u c h moisture was absorbed, the snow cover n o w possesses a hard, dense upper layer or perhaps even a crust of ice. This layer is buried by subsequent snow-falls and, if dense enough, m a y cause a change in the steepness of the moisture and heat flow gradients through the cover because it is relatively impervious t3 the passage of air and water vapour. If the air temperature fluctuates markedly, quite complex moisture and heat gradients m a y occur, resulting in ever-changing densities and hardnesses through the cover. Later during the winter a wind storm m a y sweep over our plot. Although our plot is situated in an open glade surrounded by spruce forest the wind m a y be strong enough to reach d o w n through the trees, pick up the surface flakes and move them about. Such jumbling and tossing of the flakes abrades the delicate rays, reducing the flakes to shattered remnants. Because of their n o w smaller size and simpler outlines the particles nest more closely together,resulting in a hard, dense layer. Ifthis layer is only on the surface it is known as “wind slab”, but if all the cover has been reworked by the wind, as happens regularly on the tundra, it is transformed into a hard, dense mass (upsik) capable of supporting fox, wolf, caribou or m a n on its surface. Thus the snow cover undergoes a maturation process until the days lengthen in the spring and the increased solar radiation strikes through the trees and acts upon the snow surface. Melting m a y take place in the top layer during the warmest part of the day. At night the soggy surface freezes hard. W h e n it freezes it becomes tough, forming sipoptoap or a “sun crust”. Because the heat came from above, the surface is melted more than the snow immediately beneath the surface. In high latitudes and altitudes where long periods of intense radiation occur even though the air temperature is below freezing, these inverted temperature gradients cause a redistribution of water molecules. T h e surface flakes are transformed into long, vertical spicules called qulu or “ablation needles”. Taiga snow is the interface between the warm, moist soil and the dry, very cold, subarctic air; it is in fact an ecotone between drastically different environments. Although the instruments in a standard Stevenson
86
screen m a y measure an air temperature of, say, -400 or -450 C, the air immediately above the snow m a y be as low as -55” C (Johnson, 1954).But once underneath the surface of the snow the temperature rises dramatically, until at the base of a fully-developed taiga snow cover the temperaturehovers not far below freezing and with the addition of a moss layer under the snow the temperature m a y be only slightly below freezing (Fig. 2). Not only is the subnivean environment relatively w a r m and moist, it is also markedly stable (Pruitt,1958). The presence of such a warm, moist and stable environment is of extreme importance to small m a m mals, plants and invertebrates (Koskimies, 1958; Mezhzherin, 1964). It has been shown by m a n y experiments (Scholander et al., 1950a, 6, c) that small m a m mals (mice, voles and shrews) cannot withstand the supranivean environment of the taiga winter. They are physiologically incapable of producing enough heat to offset the loss to the cold, dry, air and their mass-surface relationship does not allow them sufficient insulation.Thus, if it were not for the snow cover large areas of the northern taiga would be lacking in small mammals. This would have far-reaching effects, since the small mammals are the main herbivorous base of the ecosystem, and m a n y carnivorous mammals and birds depend on themfor food.Nevertheless because the thickness of the snow cover fluctuates from year to year (Pruitt, 1957) there are yearly differences in the severity of the bioclimate in which the small mammals live (Fig.3) and so there are yearly differences in the numbers of small mammals (Formozov, 1948). The autumnal decline in air temperature is a fairly
5-
l
A-
i
3-
i
2-
i
- -
soil
n
,
I
-70
-60
-50
-40
-30
-20
-10
O
¡J
. _
-
20
FIG.2. Generalized temperature gradient through supranivean air and the snow cover to the soil surface. (From Johnson, 1954.)
Some ecological aspects of snow
O F
3s 35
30
25
20
15
10
5
9’ bslow moss surface
at moss surface
1954-55
0 1
O
-5
-10
FIG.3. Climographs of the bioclimate of small mammals beneath a taiga snow cover. Note the effects of different thicknesses of snow cover in successive years. (From Pruitt, 1960.)
regular event, governed by the regular decrease in incoming solar energy as the days shorten. T h e onset of a snow cover is more fortuitous, being governed by the precise succession of meteorological events that bring the correct mass of moist air into contact with cold air. Thus, a snow cover m a y arrive quite early in October one year but m a y not arrive until late November the next year (Fig.4) (Pruitt, 1957;Bider, 1961). W h e n there are only a few centimetres of snow on the forest floor in the subarctic taiga, one m a y see m a n y signs of small m a m m a l activity but when the snow cover reaches a thickness of 15-20 c m there is a dramatic decrease in activity on the surface (Formozov, 1961). Such a thickness of snow is sufficient to insulate the soil against fluctuations in air temperature. When this thickness arrives I call it the “hiemal threshold”, for its arrival is the true beginning of winter for the small creatures of the forest floor. In the spring there is a great variation in the actual rate of disappearance of the snow cover, although the sequences are quite regular. Just as the period between the autumnal thermal overturn and the hiemal threshold is a critical period for the small mammals of the taiga floor, so also is the period between the time the snow cover loses its insulating power and the onset of the vernal thermal overturn. I urge that boreal ecologists concentrate research on these two periods and I suggest that such work will clarify m a n y aspects of population fluctuations in small mammals. The snow cover varies not only in time but it also varies in space. As the snow flakes fall, m a n y of them are caught on the needled branches of the coniferous trees of the taiga. Thus there is a “snow shadow” formed around the base of each spruce tree. As the
FIG.4. Comparison of thickness of snow cover in two successive years. Note the variation in time of arrival of the hiemql threshold. (From Pruitt 1957.) 87
W.O.Pruitt, Jr.
TABLE1. Comparison of soil temperatures within and beyond qarnaniq Base of trunk
Date
Air Temperature
Temperature
(“C)
(“C)
Beyond qamaniq
Snow thickness
Temperature
(“C)
(4 22 November 1954 22 December 1954 28 January 1955 21 February 1955
winter progresses the snow shadow continues and increases in sharpness. T h e forest Eskimo of the K o b u k Valley in north-western Alaska (Giddings, 1961) call these bo-rvl-shapeddepressions in the snow at the base of trees qamaniq (Pruitt, 1959a). W e have seen h o w small mammals prosper with deep snow and decline in numbers with little snow, so it is no surprise to learn that individual red-backedvoles (Clethrionomys) avoid the frigid qamaniq (Table 1) and more often frequent the parts of their h o m e ranges that have thicker snow cover (Pruitt, 1959b). Their h o m e ranges have vacuoles, as it were, and these vacuoles are the qamaniq. A thick protective blanket of snow is not without dangers, however. The relatively w a r m subnivean environment allows a certain amount of bacterial action to continue,even in mid-winter.This results in the production of carbon dioxide under the snow. If the snow cover is thick enough, and particularly if there are dense, relativeIy impervious layers within it, there m a y accumulate in certain low spots enough carbon dioxide to be harmful to the small mammals (Bashenina, 1956). T h e voles Clethrionomys and Microtus have developed a behavioural adaptation to counteract this situation. They construct ventilator shafts up through the snow cover to the surface. In regions of especially thick snow cover the small m a m m a l s are virtually immune to predation by carnivorous birds during the time when the snow cover is present. T h e only time that they are exposed to predation is when they come up their ventilator shafts to the upper air. Formozov (1946; 1962) has .shown that it is such activity on the part of the small mammals which allows certain species of small owls to winter over in parts of the Eurasian taiga. If it were not for voles caught at their ventilator shafts the owls would have insufficient food for survival in the region. S o m e kinds of mammals, and m a n y birds, have evolved an indirect but very effective behavioural adaptation to the ecological challenge of a winter snow cover. They simply move away from the snowy region. Insectivorous and ground-feeding birds are especialy susceptible to having their food supply m a d e inacces-
88
-12.5 -21.5
-16.5 -31.5 -24 -36
-19 -16.5
trace 8 8 13
Snow thickness
(4
-10
-12.5 -10.5 -12
1 17 20 35
sible by a snow cover. Certain mammals, such as the caribou (discussed later in more detail) and the wapiti (Cervus canadensis) in North America and the roe deer (Capreohs capreolus) in Eurasia, perform spectacular migrations which are closely correlated with the arrival of snow conditions at critical thresholds of hardness, density or thickness. T h e general pattern of all these migrations is a spring or summer reproductive period spent in a region that is ideally suited for that purpose during the spring and summer but which has winter snow conditions inimicable to survival. In the autumn there occurs a mass exodus of animals to a region of more suitable snow or even no snow at all. Investigation has shown that, in most cases, the animals, if provided with sufficient food, are able to withstand the prevailing winter temperatures of the abandoned region. Therefore a shortage of food caused by their inability to procure it from under the snow seems to be the underlying reason for these migrations. There are two main classes of mammals in the taiga -those that are large and live above the snow cover and those that are small and forced to live beneath the protecting blanket. There is one mammal, the red squirrel (Tarniasciurus hudsonicus) that lies just on the borderline between the two size-groups of m a m mals. R e d squirrels are active above the snow most of the time, but when the air temperature falls below 300 C, they vanish from the scene (Pruitt and Lucier, 1958). This is a critical temperature for them and when it arrives they leave the supranivean environment of the moose and the hare and join the voles in their w a r m subnivean environment. Thus because of the snow cover they have the best of two worlds
-
(Fig.5). There are three main ways in which the resident or non-migratory mammals adapt to snow. One is by using the snow as a blanket to avoid the deep cold. A second adaptation is that of the moose-the possession of long legs or stilts; even this adaptation is not perfect, since the snow m a y sometimes become too thick for stilts. The places where moose winter over aye also places where the snow cover is thinnest (Nasimovich, 1955 ; Edwards, 1956; Edwards and
Some ecological aspects of m o w
Ritcey, 1956). Pulliainen (1965) has shown that wolves (Canis lupus) in the taiga of Finland clearly prefer areas of thin and hard snow cover. T h e third w a y of adapting to snow is to float over it. T h e lynx and the snow-shoe hare are the best examples of this method. A lynx appears to be a large animal,but when skinned out it is seen to be not much larger than a small dog. Most of the animal is fluffy insulating fur and large snow-shoe feet. These large feet enable it to float on the surface of the snow, even when galloping. T h e snow-shoehare is the classic example of a floater. In closely related species in Eurasia, Lepus tirnidus and L. europaeus, the northern snow-shoe hare possesses more bearing surface on the hind feet, even though it is smaller in body weight than its southern relative (Formozov, 1946). Sometimes even flotation fails. The snow m a y be so light and fluffy that even the snow-shoe hares sink into it. Then they change theit behaviour and become “trailers”, following each other’s trails. Each time an animal passes the snow gets packed a little bit more. Before long a hard trail is formed-along which the animals can move freely. Ciivonen (1952,1956,1962)and Sulkava (1964)have shown some of the intricate relationships possible between thickness of the snow cover and population densities of animals. Ingles (1949)discussed the effects of snow-cover on local and seasonal distribution of pocket gophers in a montane forest region. For small m a m m a l s with a comparatively short life span one cannot afford to ignore such a significant 10095. 90. 85. 80. 75. 7065. 6055. 50-
O
O
O.
O
O O
- 0
A
O
FIG.5. Relationship between supranivean activity, subnivean activity and ambient air temperatures in Tamiasciurus hudsoniciis. Closed circles indicate winter 1954-55; closed triangles, winter 1955-56; open circles, winter 1956-57. (From Pruitt and Lucier, 1958.)
portion of it that is spent beneath the snow cover. Any attempt to study the winter activity of m a m m a l s under soft, fluffy taiga snow immediately runs into several practical difficulties. T h e h u m a n organism, itself, being a tropical primate, must make numerous adjustments in clothing, equipment and behaviour patterns. Next comes the realization that a two-footed creature lumbering through a snow cover effectively alters that snow cover not only for the present and immediate future but for the remainder of the winter as well. Finally there is the problem of contact between the investigator and the animals. I found that a satisfactory method is to find or make some sort of contact tunnels into the subnivean habitat of the animals (Pruitt, 19596). I fully realize that this method disturbs the habitat somewhat, but, with care, the disturbance is minimal. Continued presence of well-weathered trap chimneys, which I devised, probably disturbs the animals’ activities no more than does the introduction of live traps into any habitat. I made a number of chimneys 19 x 30 x 116 c m , with hinged lids, of 1.5 c m plywood. These were screwed to stakes which had been driven into the ground at regular intervals on a study plot. One side of the chimney was about 10 c m shorter at the bottom than the other three. The chimneys were placed so that the three long sides were below the moss surface and the short side slightly above it. This allowed surface entrance to the chimney while preventing snow from completely filling the base. The anticipated thickness of the snow cover governed the height of the chimneys. These chimneys were put in place in the summer and allowed to become thoroughly weathered. In the fall the interior of the chimney bases were prebaited with a supply of oat flakes. In order to disturb the snow cover as little as possible, only one continuous snow-shoe trail was made from trap to trap. Preservation of the fragile, lattice-likepÜkak layer is necessary so that free movement of the animals is not hindered. Previous investigation had revealed that when a snow cover with an average density of 0.20 reached a thickness of over 60 c m it would support a carefully-madesnowshoe trail without collapse of the pukak. T h e animals were caught in masonite live traps which fitted inside the chimneys. T w o 1.5 c m cup hooks were screwed into the top of each trap. A piece of welding rod was bent into a hook and used to engage the cup hooks and thus raise and lower the traps so that they could be checked for captures (Fig. 6). With ambient supranivean temperatures of -35 OC or thereabouts the traps had to be checked at least every eight hours. The only voles lost were those left in the traps for ten hours. No nesting material was used in the traps. Experience has shown that at temperatures consistently below freezing the animals’urine freezes in the nesting
89
W.O.Pruitt,Jr.
material and creates problems such as jamming of the door or treadle. The traps were kept heavily stocked with oat flakes. I feel that a supply of food in the traps and constant attention so that no animal remains in the traps too long is better than relying on nesting material for survival of the animals. Equipment needed for checking the traps consisted of a small bag of oat flakes for replenishing the bait, the hook for pulling up the traps, a pack of record cards, scissors for clipping toes, a small cloth sack for receiving the animals from the traps, a screwdriver (for emergency repairs to traps) and a gasoline lantern. About sixty or seventy traps m a y be handled each hour. Another species that is closely associated with snow is the caribou (Rangifer tarandus) (Bergerud, 1963). In 1957-58 I studied their ecology for the Canadian Wildlife Service (Pruitt, 1959b). That winter w e flew m a n y hundreds of kilometres in light aircraft at low elevations, carefully plotting on maps the locations of caribou wintering over, and also where the animals were not present. W e found that most of the area had no caribou. Then there were areas with scattered caribou;within these areas there were regions of heavy concentrations of caribou. At this time I set out a series of snow stations. I measured the thickness, hardness and density of each layer of the snow, and also the grain size and type, and the temperature of each layer. W h e n I plotted the results on maps and compared them to the distribution of caribou I found that the areas Hasp Hinged
lid
Ch i mney
c
p ! i
,.
h o o k s s Live trap
Stake
Moss sur fa ce
FIG.6. Equipment used for subnivean live-trapping(From Pruitt,1959.)
90
with heavy caribou concentration had snow that was light, soft, and thin. Hardness ranged from 6.5 to 60 gm/cm2for forest stations and from 50 to 700 gm/cm2 for lake stations,while density varied from 0.13 to C.20 for forest stations and 0.13 to 0.32 for lake Stations. Thickness varied from 19 to 59 cm. For the areas without caribou the snow could sometimes be soft but it could also be very hard, dense and thick. Hardness varied from 35 to 7,000 gm/cm2 for forest stations and 150 to 9,000 gm/cm2 for lake stations, while density varied from 0.16 to 0.92 for forest stations and 0.17 to' 0.92 for lake stations. Thickness varied from 19 to 82 c m (Table 2). Remembering that the density of freshwater ice is 0.92 it is clear that the non-caribou areas had ice layers in the snow cover, whereas the caribou areas did not. Thus w e see that caribouhave thresholds of sensitivity to the hardness, thickness and density of the snow cover. The threshold of hardness sensitivity is approximately 50 gm/cm2 for forest snow and 500 gm/cm2 for lake snow. The threshold of sensitivity for density is approximately 0.19 or 0.20 for forest snow and 0.25 or 0.30 for lake snow. The threshold of sensitivity for thickness is approximately 60 cm. W h e n these thresholds are exceeded the caribou react by exhibiting a migratory type of appetitive behaviour until they encounter snow oflesser hardness,thickness or density. The areas of suitable snow cover wherein caribou winter over are surrounded by fences of unsuitable snow. In years when the thickness,hardness and density gradients are well-developed,they are as effective as pasture fences in restricting the animals' movements. Ifthe areas of suitable snow conditions coincide with areas of good food supply the caribou can pass the snow season in good condition. If, however, the areas of suitable snow conditions coincide only with areas of poor food supplies (e.g. burns) then the caribou m a y survive only with difficulty. W e m a y extend our reasoning into time and consider the situation during a period of climatic warming. As the temperate zone snow conditions intrude poleward into the taiga, the caribou are literally squeezed between their advance and the treeless tundra where the winds still rework and harden the snow cover. The optimum wintering over environment for caribou seems to be subarctic taiga-a coniferlichen open forest-with relatively thin snow cover that is not modified by near-thawing atmospheric conditions, and with little wind. Departure from these optimum conditions, especially as regards thickening, hardening and densification of the snow cover leads toward an environmental Pessimum. Thus, changing nival conditions is just as reasonable an explanation for the extirpation of such groups as Rangifer t. dawsoni on the Queen Charlotte Islands (Banfield, 1961) and R. t. eogroenlandicus in East Greenland (Degerbd, 1957) as are more abstract explanations based o n genetic drift.
Some ecological aspects of snow
TABLE2. Morphological characteristics of snow cover in relation to caribou distribution (from Pruitt 19596). (Hardness of hardest snow layers in gm/cm2;speciíic density of densest snow layers;thickness of snow in cm.) ~
~
~~
Snow hardness (grn/cm9) Density of caribou
Mean
Range
-
Stations
Mean
Snow density Range
Snow thickness (ern) Stations
Mean
Range
Stations
Forest stations
Concentration Occasional None
34 469 993
6.5-60 60-3 O00 35-7 O00
19 12 23
0.17 0.25 0.31
0.13-0.20 0.16-0.48 0.16-0.92
19 12 23
45 45 56
19-59 32-62 31-82
20 11 23
291 1954 3016
50-700 50-6000 150-9000
16 12 25
0.19 0.35 0.49
0.13-0.32 0.18-0.92 0.17-0.92
18 10 25
33 30 41
21-43 20-38 19-46
16 12 25
Lake stations
Concentration Occasional None
There is a clear need for further research on this problem of thresholds of sensitivity to snow morphology. For example, Nasimovich (1955) consistently reported sensitivity thresholds for thickness and hardness in old world reindeer well above the ones I noted for North American caribou. This is puzzling, in view of the fact that Henshaw (1964)found,in the taiga and forest-tundra of north-western Alaska, thresholds of sensitivity in caribou that were quite comparable to those I found in the central Canadian taiga. Moreover, Makridin (1963)reported thresholds, on Taimyr Peninsula, quite comparable to those of myself (Pruitt, 19596) and Henshaw (1964). O n e wonders if there are behavioural differences between Rangifer in the old world and the new, or if the variation in results is only an artifact caused by different instrumentation. In my work in the north I have found that the ‘‘official’’meteorological words for snow are woefully inadequate to describe its phases. Consequently I have turned to the languages of the native peoples of the north, the Eskimoes and the Indians,for words which represent those phrases of snow which are important to animals and plants (Table 3). Of all the native languages I have examined, that of the K o b u k Valley Eskimo (the “Forest Eskimo”, Giddings, 1961) of north-western Alaska appears richest in snow terms (Pruitt, 1960). Their word for the snow that collects on trees is qali, for the snow that collects on the ground it is api. One must be careful to differentiate between true qali and rime (kanik). Qali forms under cold, windless conditions, while kanik shows m a x i m u m farmation slightly below freezing and with slight to moderate winds. Kanik shows best formation when relatively warm, moist air passes over cold objects such as vegetation, wires or towers. The phenomenon m a y be complicated by later qali accumulation on the kanik. In a series of papers, Miller (1961~;1961b; 1962; 1964) has examined a number of aspects of the phenomena of qali and kanik. H e has calculated (Miller,
1961b; 1964) the meteorological parameters that govern kanik accumulation (Fig.7).It should be nated that Miller’s excellent analyses of qali and kanik have been in the temperate zone montane region. I should point out here that the understanding of such a complex phenomenon as qali, which has meteorological, biological and chronological facets would undoubtedly be simplified and hastened by first analysing these component facets in the conditions of the subarctic taiga. Here, as w e have nated earlier, snow phenomena and their metamorphosis are reduced to their simplest components. Let us examine some of the ecological aspects of the snow which collects on trees. Qali accumulation is exceptionally difficult to measure numerically. I once devised a “qalimeter” to do this job. It enabled m e to make objective comparisons of qali accumulation, both geographically and chronologically. None the less, there is a need for better ways of measuring qali. In the windless taiga of central Alaska, qali assumes great ecological importance (Pruitt, 1958). It is one of the agents initiating forest succession. If a spruce departs from the vertical it is doomed to breakage, ill accumulate qali. W h e n a some day, because it w tree breaks, adjacent trees become susceptible to qalibreakage and the “glade” grows until it is sufficiently large that wind circulation prevents massive accumu lation of qali. In the glade the broken spruces die and the rain of dead needles chokes out the feather-mosses on the forest floor. Thus seeds have a good site for germination. Deciduous trees invade-alders, birches, aspens and willows. These trees mature and die and in their leaf litter young spruce can germinate. They mature and the spruce forest is eventually restored at the site, to await further qali-breakage. Qali is also of direct economic importance to man’s activities. Power lines are frequently snapped either by qali-brokentrees or by a heavy accumulation of qali on the lines themselves. Near Fairbanks, Alaska, the local Rural Electrification Co-operative has met 91
W.O. Pruitt, Jr.
TABLE3. Specialized snow terms of some northern peoples (from Pruitt, 1960)
English
Kobuk Valley Eskimo (Alsska)
Dindye (Fort Yukan, Alaska)
Chipewyan (northern Saskatchewan)
Snow
anniu
ia
sil(ch)
Snow that collects on trees
qalí
dé-ia
de-chén-kay-síl(ch)
Snow on the ground
api
non-kót-za
sil(ch)-de-trán
Depth hoar
pukak
gai-ya
yath(k)óna
Wind-beaten snow
upsik
seth(ch)
sil(ch)- t( ch)rán-al
theh-ni-zee
yath-they-yé-ree-la y
za-he-áh-tree
nil(ch)-see-ni-(k)oth
Fluffy taiga Snow Smoky snow or drifting snow
siqóq
Smooth snow surface of very fine particles
salumá roaq
Rough snow surface of large particles
natatgónaq
Sun crust
siqoqtoaq
ia-es-(i.h)a
na-hó-t(ch)ran
Drift
kimoaqruk
za-ké-an-é-hae
yath-neé-eus
Space formed between drift and obstruction causing it
agamapa
Sharply etched wind-eroded snow surface (zastrugi or skavler)
kaioglaq
Irregular surface caused by differential erosion of hard and soft layers
tumarínyiq
(zh)e-quin-zee
day-chen-yath-dó-dee
Snow deep enough to need snow-shoes
det-thlo(k)
yath-thay-t(r)án-ai(ch)há
Spot blown bare of snow
si(ch)
oh-béh
Area of deep snow that persists perhaps all summer
ea-kay-tak-kok
yath-thay-(án)
Bowl-shaped depression in snow around base of
qámaniq
the challenge by flying a helicopter at a low altitude along the most vulnerable lines so that the rotor blast cleans the qali from the power wires. In s u m m e r the taiga m a y be a mass of greeneryprimarily alders and y o u n g birches. In winter these trees and shrubs are bent over by qali accumulation. This is their w a y of adapting to the presence of qali. T h e spruce stands straight a n d tall and resists qali; alders a n d birches are limber and b e n d with the qali and recover in the spring. W h e n the trees are bent over by the qali their tender growing-tips are brought within range of the snow-shoe hares, which feed extensively o n them. This is a very important source
92
of food for the hares. W h e n the trees are bent over, snow-caves form under them. In very cold weather, w h e n even the hares avoid the infinite heat-sink of the night sky, these caves are Iefuges for the hares. In the spring the alders and birches spring back vertical again, with the hare-barked twigs high in the air and at this time of the year one m a y see m a n y signs of the winter-feeding by hares. T h e relationship of hares and shrubs is reciprocal: the plants furnish the hares with food but the hares return the food to the soil by their faecal pellets, which accumulate in quantities around the shrubs utilized most heavily. In a winter w h e n the s n o w comes early and a c c u m u -
S o m e ecological aspects of snow
/
‘-A
’ /
/
Negligible load
/ / FIG.7. Meteorological parameters governing kanik accumulation, showing relative weight of snow load in tree crown as a function of snowstorm temperature and wind speed (unscaled). (From Miller, 1964.) lates gradually all winter long, the hares are constantly elevated o n top of the s n o w surface to reach fresh supplies of food. B u t in a winter w h e n the s n o w remains at a constant thickness or w h e n it even settles and decreases in thickness, the hares are unable to reach fresh sources of food higher u p and they turn to unpalatable foods, such as spruce. S u c h s n o w conditions m a y result in a decline in the hare population by spring. Bider (1961) has s h o w n the close relationship between thickness of s n o w cover and hare feeding activities. In January the birches shed their seeds o n to the s n o w surface and m a n y resident birds, such as redpolls, utilize this food source. Eventually, however, another snowfall covers this seed layer. Mice a n d shrews then tunnel up through the s n o w and mine out t.he seed layer.
TUNDRA SNOW So
far w e have concerned ourselves with taiga snow. Let us n o w examine the ecology of the hard, windm o v e d tundra snow. T u n d r a s n o w is characterized primarily by this factor of having been m o v e d by the
wind. There are t w o phases, in the physical sense, to the s n o w in a n Arctic tundra region. T h e s n o w cover proper (api) consists of those particles which are not picked up by ‘the wind a n d of wind-worked particles which have b e c o m e consolidated into a hard mass (upsik). A b o v e the s n o w cover, in the air, o n the surface of the s n o w cover, and sometimes incorporating even the top layers of the s n o w cover itself, is another phase. This is the moving s n o w or siqoq which, depending o n the wind direction and force, is either consolidated
into a succession of drift forms or m o v e s along and above the s n o w surface (Orlov, 1961). Because of variations in the force of the wind transporting t h e m , a n d because of nivographic details, the particles comprising the siqoq b e c o m e stabilized for varying periods of time and form drifts. T h e sequence of drift types appears to be as follows. S n o w particles are released from suspension in the air whenever the speed of air m o v e m e n t is not sufficient to support them. Thus, s n o w accumulates in microtopographic depressions, stream-beds a n d behind obstructions (which m a y themselves be nivographic details). Later winds of greater force or different direction m a y scour these spots a n d redeposit the particles elsewhere. O n a flat, relatively unobstructed surface m a n y of these particles advance in groups. A group assumes a characteristic arrow-head shape with the point up-wind, a gradually sloping up-wind face a n d a lee slope which is abrupt and concave laterally. A t the tang of the arrow-head the thickness of the drift is greatest. These drifts are k n o w n popularly as barkhans but m o r e accurately, in Eskimo, as kalutoganiq. (A barkhan is technically a sand drift of the s a m e shape while kalutoganip refers to this precise s n o w drift type.) Kalutoganiq migrate d o w n wind as the particles are exposed o n the windward face, are m o v e d over the surface of the drift and then are temporarily immobilized o n the steep lee slope. W h e n e v e r the wind slackens the kalutoganiq b e c o m e consolidated through the processes of sublimation a n d re-crystallization. Later winds, if of sufficient force, will erode a w a y the kalutoganiq, producing sculptured forms which have great beauty but which are exceedingly difficult to traverse. T h e sculpturings are widely k n o w n by the terms zastrugi (Russian) or skavler (Norwegian), but are m o r e accurately k n o w n as kaioglaq (Eskimo). Zastrugi or skavler refer to surface sculpturings in general. Kaioglaq refers to large, hard sculpturings while the w o r d turnarinyiq (Eskimo) refers to small zastrugi or “ripple marks” which are the last remains of kaioglaq. Kaioglaq eventually m a y be eroded a w a y completely a n d the particles regrouped d o w n w i n d again into kalutoganiq. A late stage of kaioglaq is the formation of overhanging drifts or mapsuk. T h e windward point of a ridge of kaioglaq is eroded fastet at base level than above it, thus forming the characteristic anvil tip which points upward. This succession of drift forms m a y be diagrammed as in Figure 8. Valleys of small streams b e c o m e completely filled with a thick mass of wind-blown s n o w and these drifts m a y not melt until late in the following s u m m e r . S u c h a drift is k n o w n by the Russian w o r d zaboi a n d m a y be of considerable ecological effect. T h e y retard plant growth and, in those spots where they d o not melt until late in the s u m m e r , their presence m a y
93
W.O.Pruitt,Jr.
prevent certain species from living. In extreme cases they m a y prevent all plants from growing on the site where they form. These bare spots are then subject to intense cryopedological processes. Zabois also m a y be regulators of mesic habitats in an expanse of otherwise rather xeric conditions (Cooke,1955). Zabois have another, and little known, ecological role. They act as concentrators of radio-activefall-out contaminants in the ecosystem. The basic research on this problem was done in the high alpine tundra of the temperate zone (Osburn, 1963) but the zabois of the arctic tundra undoubtedly play a similar role. Here is a fruitful area for research,particularly investigation of the mechanisms ofreleaseand dissemination of the contaminants from the zaboi site. In some spots on the tundra where snow has completely filled a small stream valley, eddy currents m a y scour out the snow and produce a cavity that m a y assume tremendous size. This is one of the “traffic hazards” of tundra travel. These scoured spots are k n o w n to the Eskimo as aqmaqa. Such scoured spots can be seen even in the taiga when there is a light wind. In the folk-knowledge of northern peoples aqmaga are k n o w n as good places to set traps or snares, since here the subnivean vegetation or soil is exposed. Sulkava (1964)has shown h o w aqmaga in edificarian habitats are important in the ecology of partridge (Perdix perdix) and hare (Lepus. europaeus) in Finland. N o w , h o w do animals react to the varied types of tundra snow? S o m e animals, such as the caribou, react by moving. Most caribou leave the tundra during the snow season and migrate to the taiga where the snow is softer and less dense. S o m e caribou remain o n the tundra and these groups m a y be found in either of two situations: 1.There m a y be caribou where,because of topography, even on the open tundra there are regions of relatively soft snow. 2. There m a y be caribou in those regions of the tundra where the winds are so strong that virtually all the
*pi
1t1
Upsik
FIG.8. Succession of drift forms.
94
snow is blown away from the vegetation; these spots are k n o w n as good caribou hunting grounds for they regularly support a population of animals wintering over. One of the most striking features of the snow cover of the tundra is the remarkable similarity between different winters. Because of the overwhelming influence of wind, which is relatively constant in its moulding and shaping effects, tundra zabois and vyduu (respectively, deep drifts that form in early winter and last long into the following summer; and spots blasted clean of snow by the wind) occur in the same sites year after year. Such regularity in subnival environmental conditions results in the presence of strikingly characteristic vegetation in these sites (Gjaervoll, 1956). I have also noted comparable regularities in the distribution of some tundra mammals (Pruitt,
1966).
An increase in snow thickness,within limits, tends, as w e have seen,to improve living conditions for small mammals as regards temperature. A decrease in the density of the snow works similarly (Johnson, 1954). Consequently,w e m a y express a rough approximation of the relative ecological values of total amount of cover, its thickness and its density in the following way : C(XTD), where C = snow cover of a plot expressed as a percentage of the area of the plot covered by it, T = thickness (cm) and D = density of each discrete layer of snow as measured when exposed in a vertical profile. As a temporary working n a m e I have called the resulting number a “Snow Index”.Thus :
SI = C(ZTD). In a study of the ecology of the terrestrial mammals of an area of low arctic tundra in north-western Alaska (Pruitt, 1966) I calculated the SI of my m a m mal sampling plots (Table 4). F r o m even a casual perusal of this table it is clear that: (a) the plots fall into three groups, viz., those with SI above 10,those below 1 and those in between; (b) the grouping of the plots into these three divisions is remarkably constant for the two successive winters sampled; and (c) in the middle group, although the relative rank m a y have been greatly changed, the actual numerical indices are remarkably constant. U p o n more detailed study of other data presented in that paper (Table 5) w e could also conclude that: (a) in the study region Microtus gregalis was limited in distribution to an area having an SIof more than 10; (b) plots with SI of less than 1 were virtually unutilized by small mammals; (e) plots with SI over 10 are not generally favourable for small mámmals other than M. gregalis (perhaps in such restricted spots of excessive snow accumulation high levels of CO, concentration m a y be a factor restricting small m a m m a l
S o m e ecological aspects of snow
TABLE4. Snow indices for small m a m m a l plots 16-21 April 1960 and 31 March to 1 Apri 11961;Cape Thompson region, north-western Alaska (from Pruitt, 1966) Plot
1960
ANG-1
41.401 39.160 15.400 12.976 9.920 7.500 6.816 6.202 6.175 5.130 4.644 4.122 3.920 3.654 3.516 2.800 2.720 2.772 2.364 2.185 0.984 0.738 0.522 0.070
1961
Plot
~
CT-19 47 15 32 28 5 4 34 46 9 36 49 42 1 35 45 12 7 48 25 13 40 33
ANG-1 CT-19 47
15 32 34 42 48 25 4 46 36 45 49
7 1 28 9 12 35 5 33 40 13
90.850 57.670 34.230 21.308 8.325 7.334 5.382 4.997 4.263 3.850 3.808 2.865 2.352 2.265 1.728 1.725 1.350 1.320 0.905 0.770 0.640 0.090 0.080 0.075
distribution, as w a s described by Bashenina (1956)for certain forest areas) ; and (d) differences in SI between subsequent winters were reflected in changes in kinds of small m a m m a l s found o n a plot the following summer. T h u s it appears that the s n o w index, or s o m e similar statistic, is a valid quantitative expression of the relative ecological effect of s n o w cover morphology o n small m a m m a l s . W i t h m o r e refined observations w e can perhaps include time as a term in our SI formula, because the details of the onset, duration and disappearance of the s n o w cover are also powerful ecological factors. S o m e resident birds, such as the ptarmigan, are able to find, even o n the tundra, small pockets of soft s n o w and dive into t h e m and use t h e m as a blanket for protection from the cold winds. S o m e m a m m a l s , such as the arctic ground squirrel, that hibernate, choose their hibernation sites where thick zabois form. Through t h e m they m a k e “escape holes”, and the tunnel d o w n through the s n o w to the ground m a y be up to 5 ar 6 m long. T h e animal thus spends the winter where it is relatively w a r m . F o r example, M a y e r (unpublished research report) in a n ingenious a n d little-known work, established that the temperature in the hibernation burrows of the arctic ground squirrel (Citellus parryi) in the Alaska R a n g e alpine tundra varied through a range of only 11 O C (-80 to 30) while the ambient air varied through a range of 550 C (-410 to 14.50). T h e tundra
TABLE5. Distribution of summer small m a m m a l catch in relation to previous winter’s snow indices, Cape Thompson region, north-westernAlaska (from Pruitt, 1966)
1960 catch
Species
Clethrionomys Non-Clethrionomys M.œconomus N o n - M . œconomus Lemmus Non-Lemmus Dicrostonyx Non-Dicrostonyx Sorex arcticus Non-S. arcticus S.cinereus Non-S. cinereus Microtus gregalis N o n - M . gregalis Average of the snow indices
1961 catch
No. of
Total
Average
No. of
Total
plots
SI
SI
plots
SI
2 22 4 20 2 22 4 20 8 16 3 21 1 23
10.324 174.967 26.684 159.007 13.436 172.255 19.720 165.971 76.737 108.954 12.906 172.785 41.401 144.290
5.162 7.953 6.671 7.950 6.718 7.829 4.930 8.298 9.592 6.809 4.302 8.227
1 23 1 23
3.850 254.332 2.265 255.917
6.273 7.737
Average
SI
11.057 11.126
-
-
-
24 1 23 3 21 6 18 1 23
258.182 1.320 246.411 10.130 248.052 19.075 239.107 90.850 167.332
10.757 10.757 3.376 11.812 3.179 13.283 7.275 10.757
95
w.O. Pruitt, Jr. snow cover also protects small mammals, such as voles and lemmings, from predation. T h e ecological relations of animals and tundra snow are particularly difficult to study. These difficulties are undoubtedly the reason why most North American workers (e.g. Krebs, 1964) have dismissed weather phenomena as important controls in L e m m u s and Dicrostonyx cyclic fluctuations. (There have been a few noteworthy exceptions, e.g. Shelford (1943).)Yet these phenomena m a y be precisely the critical factors involved. For example, Fuller (1966, in press) has reworked Krebs’ data, as published, and has derived from them a perfectly logical and reasonable explanation for the population changes. This subject is so germane to our general thesis that I would like to discuss it in greater detail.1 Fuller emphasizes the importance of the autumnal critical period (that is, the time between the thermal overturn and the hiemal threshold) and the vernal critical period when flooding can occur in low-lyingareas. If the autumnal critical period is short and dry, vegetation is quick-frozen and remains relatively nutritious and the animals are not exposed to great thermal stress. If, on the other hand, the autumnal critical period is extended or wet the vegetation is repeatedly soaked,frozen and thawed and loses m u c h of its nutritive value. The animals,lacking a protective blanket, are exposed to thermal stress. Ifthe snow cover is of sufficientthickness,the subnivean temperatures do not fall m u c h below O O C and winter breeding can occur. If the snow cover is thin the subnivean temperatures m a y be low enough for (u) thermal stress and (b) inhibition of ovulation. If the vernal critical period is short and dry the snow disappears mainly by sublimation. If, on the other hand, spring comes late or is wet the animals m a y be in dire straits because of flooding and delayed onset of plant growth. Kreb’s weather data, replotted by Fuller,show that in the study region the theoretically proper conditions actually prevailed at the correct seasons for the predicted phenomena to occur. Thus, September and October 1958 were wet, showing a prolonged and intense autumnal critical period. October, November and December subnivean temperatures approached the Pessimum for thermoregulation and ovulation. M a y and June 1959 were cold. Thus, Fuller’s theory leads us to a predict a resulting low population. This, in fact, occurred. September 1959 was dry, October was cool and dry. S n o w cover built up in December to twice the hiemal threshold level. Fuller’s theory would predict winter breeding; this, in fact, occurred. M a y and June 1960 were warm. Fuller’s theory would predict rapid spring growth of the animals and early beginning of summer reproduction. This, in fact, occurred. Rather than prolong this discussion of h o w Fuller’s nival theory adequately explains the population
96
fluctuations considered, I refer you to Professor Fuller’s excellent paper and urge that the population fluctuations of boreal animals be studied from this point of view. S n o w cover is certainly not the only factor governing such phenomena but it is clearly m u c h more important than generally realized. F a y (1960) has described a technique for live-trapping small mammals beneath a tundra snow cover. In Alaska the distribution of contemporary annual accumulations of snow follows very closely the k n o w n distribution of Wisconsin glaciation. Formozov (1946) noted a similar correspondence between regions of heavy snowfallin the U.S.S.R.and the region occupied by the Dnieper and D o n lobes of the W i u r m glacial advance. Undoubtedly the snow-producingconditions which increased in intensity to bring on the Pleistocene had a very considerable role in the wholesale extermination of m a m m a l species at the end of the Pliocene. It is also reasonable to suppose that the relatively recent extermination of such kinds of m a m mals as the m a m m o t h (Mummuthus), mastodon (Mastodon), antelope (Suigu), muskox (Ovibos spp., symbos and bootherium) and moosedeer (Cervulces) in North America, the woolly rhinoceros (Coelondonta) and the “Irish elk” (Meguceros) in Eurasia might well have occurred because these animals were chionophobes and were morphologically and behaviourally tied to an environment which, albeit cold, had little snow. (See, for example, Semlren, Miller and Stevens (1964)). As their semi-arid arctic prairie environment retreated poleward before the advance of the forest with its deep, soft snows the doomed chionophobe species possibly became trapped behind such barriers as Glacial Lake Agassiz, the post-Wisconsin Great Lakes (Hibbard, 1951) or in cul-de-sacsof suitable environmental conditions surrounded by ever-constricting fences of unsuitable snow. T o sum up this discussion of the ecology of snow w e can do no better than to quote Professor Formozov. “Analysis of the described factors leads to the conclusion that, in order to study the winter ecology of mammals and birds in regions with snowy winters, zoologists and biogeographers must have available, without fail, in addition to the data furnished by meteorological stations,numerous specialized measurements and descriptions of the snow cover and its structurein various habitats and environmentaltypes. The study of snow cover is necessarily conducted at the same time as systematic calculations of distribution,numbers and characteristicsof the vital activities of animals, since elucidation of their varied reactions allows one to judge the positive or negative influence of the snow cover of a certain strength and structure on the winter conditions of existence of the fauna. “The study of the ecological role of the snow cover 1. I a m deeply grateful to Professor W.A. Fuller of the University of Alberta for permission to discuss his work in advance of its actual publication,
S o m e ecological aspects of snow
and its structure requires long-term observations which will give the answers to a series of questions that are important for protection of valuable animals and rational planning for their utilization, to w o r k out predictions of the numbers of harmful rodents, predictions of the probability of d a m a g e to winter
crops, fruit trees and cultivated shelter belts, etc. These observations are best carried out at specialized stations, which conduct, in addition, simple experiments on animals, carried out in large pens arranged under the open sky.”
Résumé Quelques aspects écologiques de la neige (W.O. Pruitt Jr.)
L’auteur étudie la couverture de neige de la partie intérieure de l’Alaska qui est le type de neige des régions de taiga, puisqu’elle n’est pas sensiblement altérée par les agents extérieurs si ce n’est la chaleur et l’humidité d u sous-sol. L’auteur considère la neige qui recouvre la taiga en tant qu’agent de liaison (écotone) entre des milieux très différenciés. Les mammifères qui peuplent la taiga seront étudiés sous d e u x points de vue: leur activité et leur adaptation à la couverture de neige. “Qali”, la neige qui s’amasse sur les arbres, étant u n facteur écologique très important dans certaines parties de la taiga, o n observe les résultats de cette action. On
étudie aussi les relations entre la neige de la taiga sous ses différents aspects et l’activité humaine. L a neige qui recouvre la toundra est différente de celle que l’on trouve sur la taiga, ce fait résultant d u vent qui déplace la neige et la remodele. L’auteur observe les effets du vent sur la neige et le développement chronologique des congères. I1 étudie aussi les effets du point de v u e écologique de la neige de toundra. I1 formule spécialement un “index de la neige”, essai préliminaire sur l’importance du rôle de la couverture de neige sur les petits mammifères qui peuplent la toundra. Enfin l’auteur aborde quelques méthodes d’étude des relations entre les mammifères et la couverture de neige.
Discussion F. E.ECHARDT. Pourriez-vousbrièvement indiquer les types d‘instruments utilisés par vous pour la mesure des précipitations en hiver ainsi que l’ordre de grandeur des erreurs sur ces mesures?
W.O. PRUITT. I use the Snow Test Kit furnished by the Committee on Snow and Soil Mechanics of the National Research Council of Canada. The instruments have a high degree of accuracy but large errors can creep in because of the great spatial variations in snow morphology. Thus a very large number of measurements must be made in each layer at each site.
J. BLUTHGEN. Does any snow crust occur because of thawing periods at least during the first half of winter and what is its consequence for all kinds of animals? W . O. PRUITT. Crusting decreases in frequency as one approaches the northward limit of trees. In subarctic taiga it is relatively infrequent. Crusting can be of extreme importance to animals. For example; (a) it m a y cause caribou to shift their wintering grounds; (b) it m a y protect small m a m m a l s so effectively that carnivores cannot obtain food; (c) it m a y interfere with subnivean air circulation so that excessive CO, accumulates; and (d) it m a y severely abrade the legs of large m a m m a l s such as moose (Alces).
A. G. LOUGHREY. Have you considered the importance of water content of snow as an ecological factor for large and small mammals, and h o w would you measure it? W.O.PRUITT. Water content is very important ecologically. Simple density measurements do not tell whether the water is frozen or not, which is the important aspect. For example, during the spring migration, caribou usually move during the warmer parts of the day when the dense, granular snow is melting and soft. At night or during “cold snaps” the migration stops. The spring critical period, for small m a m mals, occurs during the time when the snow is dense and water-logged. During these times, hardness of the snow, especially when frozen at night, is a more meaningful measurement than density.
J. MALAURIE. J e voudrais signaler que c’est une
des premières fois que l’on souligne, dans u n congrès de géographie physique et d’écologie, l’importance qu’il y a à collaborer étroitement avec les Indiens et les Esquimaux, parfaits écologues. J’ai moi-même rassemblé à Igloolik (nord du bassin de Foxe) et à Thuli (Groenland du Nord-Ouest) les termes esquimaux relatifs à la neige et à la glace. J’aimerais savoir si ces termes ont été systématiquement rassemblés et où, éventuellement, le Dr Pruitt a l’intention de les publier. Cc
97 7
W.O. Pruitt, Jr.
serait la première fois, à m a connaissance, qu’un tel travail aurait été assuré par un biologiste.
W.O. PRUITT. A table comparing the Eskimo, Athapaskan and Chipewyau names for a number of snow phenomena w a s published in 1960. I have made active use of such t e r m s in papers published in 1957, 1958, 1959, 1960 and subsequently. I refer you to the bibliography of the present paper for complete citations.
T.AHTI.Would you think that when the carrying capacities for caribou-or even reindeer in the fairly undeveloped
reindeer industry-are calculated, the snow conditions are more important than the distribution and production of the wintering ground (winter pastures) which are often regarded as the most critical factors? W . O. PRUITT. Snow conditions are an integral part of the winter range. They must indeed be taken into account when calculating carrying capacity of a winter range or even when delimiting it. They should also be givcn high consideration when determining priorities of areas of winter ranges for protection from fire or destructive exploitation.
Bibliography / Bibliographie BANFIELD,A. W.F. 1961. A revision of the reindeer and caribou, genus Rangifer. 137 p. (National Museum of Canada, Bulletin 177, Biological series no. 66.) BASHENINA,N.V. 1956. Influence of the quality of subnivean air on the distribution of winter nests of voles. Zool. Zhurn., vol. 35, no. 6, p. 940-942. BERGERUD, A. T. 1963. Aerial winter census of caribou. J. wildlife Mgmt.,vol. 27, no. 3, p. 438-449. BIDER,J. R. 1961. A n ecological study of the hare Lepus americanus. Canad. J. Zool.,vol. 39, p. 81-103. COOKE, W.B.1955. Subalpine fungi and snowbanks.Ecology, vol. 36, no. 1, p. 124-130. DAVIDSON,D. S. 1937. Snowshoes. Mem. Amer. Phil. Soc., vol. 6, p. 1-207. DEGERBUL, M. 1957. The extinct reindeer of East Greenland (Rangifer tarandus eogroenlandicus subsp. nov.) compared with reindeer f r o m other arctic regions. Acta Arctica, Vol. x,57 p. DUGMORE, A. R. 1913. The romance of the Newfoundland caribou.Philadelphia,J. B.Eippincott Co. 191 p. EDWARDS, R.Y.1956. Snow depths and ungulate abundance in the mountains of western Canada. J. wildlife Mgmt., vol. 20, no. 2, p. 159-168. ; RITCEY, R. W.1956. The migrations of a moose herd, J. Mummalogy, vol. 37, no. 4, p. 486-494. ELSNER, R. W.;PRUITT, W.O.,Jr. 1959. Some structural and thermal characteristics of snow shelters. Arctic, vol. 12, no. 1, p. 20-27. FAY, F. H. 1960. Technique for trapping small tundra mammals in winter. J. Mammalogy, vol. 41, p. 141-142. FORMOZOV,A. N. 1946. The snow cover as an environmental factor and its importance in the life of mammals and birds. Moscow Society of Naturalists, Materials for Fauna and Flora USSR, Zoology Section, New Series, vol. 5, p. 1-152.(Translationby W.Prychodko and W.O. Pruitt, Jr. published 1963 as Occas. pap. no. 1, Boreal Institute, University of Alberta.) FORMOZOV, A.N. 1948. Small rodents and insectivores of Shariinski District, Kostroma Oblast’ during the period 1930-1940. Moscow Society of Naturalists, Materials for Fauna and Flora USSR, Zoology Section, New Series, vol. 17, no. 32, p. 3-110. . 1961. O n the significance of the structure of the snow cover in the ecology and geography of mammals and birds. In: M. I. Iveronova (ed.),Role of the snow cover in natural
_-
__ 98
processes. Commemorative volume on 60th birthday oj G. D. Rikhter, p. 166-209.Moscow, Academy of Sciences of the U.S.S.R. 272 p. . 1962. A study of mammalian ecology under the conditions of snowy and cold winters in northern Eurasia. Symposium Theriologicum (Brno), 1960, p. 102-111. --; BIRULYA, N. B. 1937. Supplementary data on the question of interrelationshipsof raptorial birds and rodents. “Student Notes” Moscow State University,vol. 13, p. 71-84. FULLER, W . A. 1966. Winter ecology and population fluctuations of lemmings. Arctic. GIDDINGS,J. L. 1961. Kobuk river people. 159 p. (University of Alaska, Studies of northern peoples, no. 1.) GJAERVOLL,O. 1956. Plant communities of the Scandinavian alpine snow-beds.Trondheim.405 p. (Norske Videnskabers Selskab, Skrifter, no. 1.) GOLD, L. W.;WILLIAMS,G.P. 1957. Some results of the snow survey of Canada. 15 p. (Canada, National Research Council, Div. Building Research, Res. pap. no. 38; NRC 4389.) HENSHAW, J. 1964. A n environmental study of wintering caribou in northwestern Alaska. Thesis, Institute of Biology, London. 154 p., illus.,mimeo. (University Microfilms, A n n Arbor, Michigan, no. M-746.) HIBBARD, C. W . 1951. Animal life in Michigan during the Ice Age. Mich. Alum. quart. Rev.,vol. LVII, no. 18, p. 200-208. INGLES, L. G. 1949. Ground water and snow as factors affecting the seasonal distribution of pocket gophers Thomomys monticola. J. Mammalogy, vol. 30, no. 4, p. 343-350. JOHNSON, H. M. 1953. Preliminary ecological studies of microclimates inhabited by the smaller arctic and subarctic mammals. Proceedings Alaska Science Conference, 1951, p. 125-131. . 1954. Winter microclimates of importance to Alaskan small mammals and birds. Unpubl. Ph.D.thesis, Cornel1 University, Ithaca, New York. KOSKIMIES, J. 1958. Lumipeitteen merkityksesta elainten lamposuojana (Snow cover as heat insulator). Suornen Riista, p. 137-140. KREBS, C. J. 1964. The lemming cycle at Baker Lake, Northwest Territories, during 1959-62, 104 p. (Arctic Institute of North America, Tech. pap. no. 16.) LIST, R. J. (ed.) 1953. Smithsonian meteorological tabZe8.
_-
_-
S o m e ecological aspects of snow
(Sixth revised edition, second reprint.) Washington, D.C., Government Printing Office, 527 p. (Smithsonian Misc. Coll., vol. 114;publ. 4014.) MAKRIDIN, L. N. 1963. Distribution and migration of wild reindeer on Taimyr Peninsula. All-Union Agricultural Institute, Extension Education, Trudy, vol. 15, p. 150-159. MAYER, W . V. Effect of environmental influences upon the reproductive cycle of the Arctic ground squirrel Spermophilus undulatus. Final report submitted to Arctic Aeromedical Laboratory, Alaska, 562 p. (Unpublished.) MEZHZHERIN, V. A. 1964. Dehnel’s phenomenon and its possible explanation. Acta Theriologica, vol. 8, no. 6, p. 95-114. MILLER, D. H.1961a. Folklore about snowfall interception. J. geophys.Res., vol. 66, no. 2547,p. 1-5. . 1961b. Meteorological influences on interception of falling snow. Abstract. Bull. Amer.Met.Soc.,vol. 42,p. 289. .1962.Snow in the trees-where does it go? Proceedings 1962 Western Snow Conference, p. 21-27. . 1964. Interception processes during snowstorms. 24 p. (U.S.Forest Service res. pap. PSW-18.) NASIMOVICH,A. A. 1955. Role of the snow cover regime in the life of ungulates in the USSR.Moscow, Academy of Sciences of the U.S.S.R., 403 p. ORLOV, N. I. 1961. A new method of measuring wind-blown snow. In: M. I. Iveronova (ed.), Role of snow cover in natural processes. Commemorative volume on 60th birthday of G.D.Rikhter, p. 258-264.Moscow, Academy of Sciences of the U.S.S.R. 272 p. OSBURN, W.S.,Jr. 1963.The dynamics of fallout distribution in a Colorado alpine tundra snow accumulation ecosystem. In: V. Schultz and A. W.Klement, Jr. (eds.) Radioecology p. 51-71.Reinhold Publ. Corp. and Amer. Inst. Biol. Sci. (Institute of Arctic and Alpine Research, University of Colorado, contrib. no. 8.) PRUITT, W.O.,Jr. 1957. Observations on the bioclimate of some taiga mammals. Arctic,vol. 10,no. 3, p. 130-138. . 1958. Qali, a taiga snow formation of ecological importance. Ecology,vol. 39,no. 1,p. 169-172. .1959a.A method of live-trapping small taiga m a m m a l s in winter. J. Mammalogy, vol. 40,no. 1,p. 139-143. . 1959b. Snow as factor in the winter ecology of the barren ground caribou (Rangifer arcticus). Arctic, vol. 12, no. 3,p. 158-179.
__ __ -_
__ __ __
_- . 1960.Animals in the snow. Sci. Amer.,vol. 202, no. 1, p. 60-68.
__ . 1966. Ecology
of terrestrial mammals. Chap. 20 in: Environment of the Cape Thompson region, northwestern Alaska, p. 519-564.Washington, D.C.,U.S. Government Printing Office. ; LUCIER, C. V. 1958. Winter activity of red squirrels in interior Alaska. J. Mammalogy, vol. 39,no. 3,p. 443-444. PULLIAINEN, E. 1965. Studies on the wolf (Canis lupus) in Finland. Ann. Zool. Fenn.,vol. 2,no. 4,p. 214-259. SCHOLANDER, P. F.; HOCK, R.;WALTERS, V.; IRVING, L. 1950a.Adaptation to cold in arctic and tropical m a m m a l s and birds, in reference to body temperature, insulation and basal metabolism. Bid.Bull., vol. 99,p. 259-271. _-. -. ; JOHNSON, F.;IRVING, L. 1950b. Heat regulation in some arctic and tropical m a m m a l s and birds. Biol. Bull., vol. 99,p. 237-258. __ .; WALTERS, V.; HOCK, R.;IRVING, L. 1950c. Body insulation of some arctic and tropical m a m m a l s and birds. Biol. Bull., vol. 99,p. 225-236. SEMKEN, H.A.; MILLER, B. B.; STEVENS, J. B. 1964. Late Wisconsin woodland musk oxen in association with pollen and invertebrates from Michigan. J. Paleont., vol. 38, no. 5, p. 823-835. SETON, E.T. 1909. Life histories of northern animals. Vol. 1 and 2. New York, Charles Scribner’s Sons. SHELFORD, V. E. 1943. The abundance of the collared lemming (Dicrostonyx groenlandicus (Tr.)var. richardsoni Mer.)in the Churchill area, 1929 to 1940.Ecology,vol. 24, no. 4,p. 472-484. SIIVONEN,L.1952.O n the influence of the climatic variations of recent decades on the game ecology. Fennia, vol. 75; p. 77-88. . 1956. The correlation between the fluctuations of partridge and European hare populations and the climatic conditions of winters in southwest Finland during the last thirty years. Papers on game research, no. 17,p. 1-30. . 1962.Die Schneemenge als uberwinterungsökologischer Faktor. Sitzunger. Finn. Akad. Wiss.,vol. 1962,p. 111-125. SULKAVA, S. 1964. O n the living conditions of the partridge (Perdix perdiz L.)and the brown hare (Lepus europaeus L.) in Ostrobothnia. Aquilo, Ser. Zoologica, vol. 2, p. 17-24.
__
__
-
99
Bioecological aspects of the snow plant communities of Cape Spring,Argentine Antarctica Alfredo Corte
INTRODUCTION Based on samples brought by the Belgian Antarctic Expedition (1897-1899)Wildeman (1934)reported on the presence of snow micro algae. Fritsch (1912) reported from samples collected by Rudmose Brown that the yellow and red snow from the South Orkneys are caused by algae. The species responsible for the colour of the green and red snow were reported by Gain (1912). Kühnemann (1966) reported on the species which m a k e up the nival communities of the Antarctic. Kol (1944) made comparative studies of the snow flora of the Arctic and Antarctic; pointing out that several species of the genus Stichococcus are responsible for the green colour of the antarctic snow, and that the green colour of the arctic snow is due to the genera Rhaphidonema, Carteria, and Chlamydomonas, and the fungus Chionaster nivalis. Steinböck (1934)pointed out the frost hardiness of Chlamydomonas at -360 C, that however dies at above lo C. Palibin (1925) and our o w n studies show that the metabolic activity of the snow biological components, due to the absorption of radiation, produces an increase of the snow melting. T h e influence of temperature, the physical and chemical properties of the snow and the solar radiation on the nival biotope, were studied by Kol (1934) for the French Alps. Chodat (1917) showed that the cryoplankton at a stable temperature of O O C fulfils the biological cycle and stores carotene xantophyll in an oily solution; also pointing out the influence of the high content of carbonic acid (CO,H,) in the assimilation process of fats and sugars.
FIELD STUDIES During the antarctic summer seasons of 1957-1958 and 1960-1961(Instituto Antártico Argentino) at Cape
Spring (64010’S., 60057’W.), Danco Coast, observations were made on the biological components of the nival flora. In this survey the coloured snow areas were observed from the beginning until the end of the summer (Fig.1).
BIOECOLOGY The cryoplankton appears by mid-December and grows with the increase of the photo-period. In its full development it produces an increase of snow melting, with an irregular surface having holes and cracks. A similar process has been observed by Gerdel and Drouet (1958) for the area of Thule, Greenland. At Cape Spring cryoplankton develops in particular areas. The same areas that showed cryoplankton in 1957-1958also displayed it in 1960-1961. FROST HARDINESS EVIDENCE
In November 1960 test pits dug in the snow indicated that below 2 m there was cryoplankton in an ice layer 30 c m thick, over the rock surface. Near the rock surface there were hard reddish crust and also reddishbrown spots of a mushy consistence produced by Chlamydomonas, Scotiella, and Chlorosphaera which were covered by a thick reddish-yellow envelope which protects the cells. With the increase of temperature and the photo-period the cryoplankton also grows. N o forms or special cases of sexual reproduction were seen. In Chlamydomonas and in melting snow, zoospores and zygospores were observed. The pigment in all forms is pale at the beginning, and later on becomes more intense. Scotiella antarctica presents at the beginning a green chromatophore which fills the whole cell. Afterwards the pigment is more centralized and is surrounded by an orange-red pigment, which 101
A. Corte
at the end of the s u m m e r is changed into reddishm a u v e covering the whole cell a n d giving a n akinetes aspect of 60 p m . S u c h would-be akinetes have a verrucose m e m b r a n e without mucilage and without winged expansions. Chlorosphaera’s pigment is yellow-green and covers the whole cell, which at the end of the s u m m e r is covered by a sheath of mucilage. T h e fungus filaments of Hirano (1959) were not observed in a n y stage of development of this alga. Other algae, such as Chodatella, Oocystis, Trochiscia, a n d Mycacanthococcus, which appear later in the nival biotope, would be present as forms of resting-cells; for this reason it would be difficult to identify t h e m at this stage. It is probable that the biotope should be previously prepared by the above-mentioned species at the beginning before the development of a m o r e complete flora takes place. T h e cryoplankton is developed at a few millimetres below the s n o w surface. It is found that coloured layers are at a depth d o w n to 20 c m ; in these layers the m o r e intense colour is in the upper part. T h e depth of the coloured layer depends o n various factors such as: s n o w cover, temperature (Table 1) solar radiation and the nature of the substratum.
FIG.1. Coloured snow areas 0 = green snow;
=
at Cape Spring, Danco Coast.
red snow;
A = experimental seedling of cryoplankton; A = non-contaminated snow.
by W a t a n a b e et
al.
(1961)for several mixed algae of
the Antarctic.
REPRODUCTION
S N O W COLOUR
T h e dominance of the vegetative or asexual forms of cryoplankton in the reproduction process is outstanding. Also the presence of cells of akinete types is remarkable. This casts the doubt whether or not certain types of cells which founded the species of the genera Mycacanthococcus, Scotiella, Trochiscia, Chlorosphaera, a n d others, are intermediate stages of the reproduction cycle of one or different algae species. Unialgae experiments in the laboratory with special nutrients a n d under temperature-controlled conditions, could solve this problem. Such study w a s started
T h e s n o w colour depends o n the pigment of the chromatophores of the dominant species. For the special case of Scotiella changes in pigment from green to orange and red were observed. As a consequence the s n o w colour in a certain area can change if Scotiella is dominant, and therefore it is not safe to state what species of algae is responsible for colour. During the end of the s u m m e r of 1957-1958coloured s n o w w a s seeded in non-coloured areas. S u c h places did not s h o w growth of coloured algae at the end of the s u m m e r of 1960-1961
TABLE1. Temperature data for Cape Spring for the summers of 1957-1958 (A)and 1960-1961 (B) December Measurement
Mean M a x i m u m mean Minimum mean Absolute m a x i m u m Almolute minimum
Period
A B A B A B A B A B
102
Mesn vslues
Temp.
No. of
Temp.
No. of
Temp.
No. of
Temp.
No. of
(“C)
days
(“C)
days
(“C)
days
(“C)
days
0.9 1.3 5.5 4.6 -1.6 -0.6 10.2 6.6 -5.0 --1.7 -
February
January
25 17 25 17 25 17 25 17 25 17
1.3 2.4 5.0 5.3 -0.1 0.5 8.2 10.4 -3.0 -2.5
31 31 31 31 31 31 31 31 31 31
1.4 1.1 4.9 3.4 0.0 -0.6 7.5 6.3 -1.2 -4.5
11 20 11 20 11 20 11 20 11 20
1.2 1.9 5.2 4.5 -0.7 -0.1 10.2 10.4 -5.0
-4.5
67 68 67 68 67 68 67 68 67 68
Bioecological aspects of the snow plant communities of Cape Spring
THE S N O W LAYER
O T H E R FIELD O B S E R V A T I O N S
P H Y S I C A L A N D C H E M I C A L ASPECTS
The analyses were made in the field with samples taken in the test sites. Samples were left to settle before analysis. The total alkalinity is shown in mg/ litre of carbonic acid and the organic matter in grammes per 100litres (Table 2). Cations and inorganic ions were determined qualitatively. The melt water is clear, odourless and insipid. In some cases there is some turbidity produced by organic and inorganic material. Low p H measurements were obtained for non-contaminated snow. For red snow a p H reading between 4.5 and 5.6 was obtained; for green snow p H values were between 5.8 and 6.2 The total alkalinity (CO,€€,)and the organic content are greater in the samples of green snow and somewhat smaller for the red snow (Table 2). The analyses of the non-contaminated snow show a m u c h smaller amount of C0,H2 and organic matter, and only traces of chIorides. Qualitative analyses of the inorganic components found in samples C, D,E and F (Table 2) indicate the presence of chlorides, nitrites,sulphates,phosphates, ammonia, calcium and magnesium.
The author has also observed the snow flora of other antarctic areas: Hope B a y (63023’ S., 57000’ W.), Deception Island (62059’ S., 60043‘ W.), Paradise Harbour (64053’ S., 62053’ W.),Anvers Island and Petermann Island (65011’S., (64046’S.,64039’w.) 64010’W.). Such observations show the presence of Chlamydomonas and Scotiella for red snow and Stichococcus, Rhaphidonema, Ankistrodesmus, Chlorosphaera and Chlorella for green snow. The filamentous forms are very scarce and all species found are Chlorophyceae.
Similar observations were made by K ü h n e m a n n with samples obtained from T a u Islet, Melchior ATchipelago. T h e components of the nival flora Cyanophyceae would be related to contaminated snow from penguin rookeries; and the snows which show zygnematales, unicellular or pluricellular, would belong to fresh water ponds.
ACKNOWLEDGEMENTS The author is indebted to Dr. O. R. Kühnemann (Centro de Investigación de Biología Marina) for the analysis of the algae and to Mr. N. Bienati (Instituto Antártico Argentino) for the chemical analyses of melt-water samples.
TABLE2. Chemical analysis of melt waters of different snows (++= positive, + = traces, Snow at 150 m Sample A (16.12.1960)
Volume (mi) Colour Turbidity Odour
PH Alkalinity total (mg/l) Organic matter (di00 i) Nitrates Nitrites Chlorides Sulphates Sulphides Phosphates Ammonia Calcium Magnesium
1 O00
Sample B
(29.12.1960)
Green snow
R e d snow
Sample C (21.12.1960)
Sample D
1 O00
Sample E (22.12.1960)
Sample F
1 O00 + -
1 O00
1 O00
t
-
+ -
5.8 295 7.90
6.2 279 7.90
(28.1.1961)
+ -
1 O00
+ -
+ -
6.6 62 0.47
6.2 0.57 0.15
4.7 189 3.60
5.6 170 3.72
-
+ ++ + + + + +
++
-
+ + -
+ +
+
+ -
+ +
-= negative)
-
+ + + + + +
+ t
+ +
t
+ + +
(27.1.1961)
+ + t
-
+ + + +
103
A. Corte
Résumé Aspects bio-écologiques des communautés de plantes des neiges au cap Spring, Argentine antarctique (A.Corte)
L’auteur expose les résultats des travaux effectués pendant les campagnes d’été organisées dans l’dntarctique en 1957-1958 et 1960-1961 (Institute Antártico Argentine) à Cape Spring (lat. 64010’C., long. 60057‘), D a n c e Coast. A C a p e Spring, l’oryoplancton apparaît vers le milieu de décembre et sa croissance est fonction du photopériodisme. On y a observé des Chlamychomonas, Scotiella et Chlorosphaera. Les Chodatella, Oocystis,
Trochiscia et Mycacanthococcus apparaissent plus tard dans le biotope nival. En ce qui concerne la reproduction, il y a prédominance des formes végétatives ou asexuées du cryoplancton. D’autres endroits des régions côtières ont été visités. Les observations ont révélé la présence de Chlamydomonas et Scotiella pour la neige réelle, et de Stichococcus, Rhaphidonema, Aakistrodesmus, Chlorosphaera et Chlorella pour la neige verte. Les formes filamenteuses sont très rares et toutes les espèces que l’on a trouvées sont des Chlorophyceae.
Discussion P. KALLIO. I wonder whether Coccomyxa is represented in antarctic regions. I ask this because in subarctic Finland these algae are important components in some symbiotic organisms. Coccomyxa has adapted to low temperatures and m a y have great importance in the flow of energy there in the “northern boundary” of life.
A. CORTE.It is probable that coccomixe are found in Antarctica. I have observed microfungus filaments,perhaps related with the simbiotic activity of algae. Hirano reported the Presence ofthe fungifilamentin ChlorosPhaera.Your question is very interesting for future studies in Antarctica.
Bibliography CHODAT, R. 1917. Les neiges colorées. Rev. gén. sci., vol. 28, p. 12. FRITSCH, F.E. 1912. Freshwater algae collected in the South Orkneys by Mr. R. N. Rudmose Brown. Scottish National Antarctic Expedition 1902-04, vol. 3: Botany. London. GAIN, L. 1912. L a flore algologique d‘eau douce de l’Antarctide. Thèses présentées à la Faculté des sciences de Paris, 3e partie, p. 156. DROUET, F. 1958. The cryoconite ofthe Thule GERDEL, R.W.; area. U.S. A r m y Snow Ice and Permafrost Research Establishment. (Research report no. 50.) HIRANO, M . 1959. Notes on some algae from the Antarctic collected by the Japanese Antarctic Research Expedition. Special Publications from the Seto Marine Biological Laboratory, Japan. KOL, E. 1934. Biologie de la cryovégétation des Alpes valaisannes et du massif du mont Blanc. Bull. Soc. Botan., Genève, ser. 2, vol. 25, p. 287.
104
At the Arctic and Alpine Institute of Colorado P. KILBURN. considerable work is going on into the study of the ecology of snow algae. Would Dr. Corte care to comment on first, On top Of the mow (‘.g. how they the OriginOf begin growth afresh each season); and second, the immediate environment of these algae in terms of temperature and light. A. CORTE. First, I a m not sure about the reason for the vertical movement, but I suppose that it is due to the production during the metabolic activity of these cells. Concerning the second question I have no data on the effects of temperature and light on the snow layer.
/ Bibliographie KOL, E.1944. Comparison between the cryovegetation of the Northern and Southern Hemispheres. Arch. Hidrobiol. Illinois, vol. 40, p. 835-846. K ~ H N E M A N N , O. 1966. Criovegetación de ‘la Antártida. Boletín de la Soc. Arg. de BotQnica,Buenos Aires. PALIBIN, I. V. 1925. Mykroorganismy, kok razrushitel: poliarnykh l’dov.Leningrad, Izvestiia Tsentral’nogo Gidrometeorologicheskogo Biuro. STEINBOCK, O. 1934. Die Tierwelt der Glets-Chergewässer. 2.Deut. u. Osterr. Alpenver, vol. 65, p. 263. WATANABE, A.;FUKUSHIMA, H.; FUJITA,Y.; KIYOHARA, T.; ISHIKAMA, M. 1961. S o m e remarks, on the cultivation of microalgae collected in the Ongul Islands and adjacent areas. Antarctic Record, no. 11, p. 153. (Reports of the Japanese Antarctic Research Expedition, Tokyo.) WILDEMAN, E. DE. 1934. Observations sur des algues rapportées par l’expédition antarctique de la “Belgica”. Expédition antarctique belge, 1897-1899.Anvers.
Some geomorphological processes in cold climates A. Rapp
INTRODUCTION Recognition and description of existing landforms as well as explanation of their genesis is the task of geomorphology. In order to understand the genesis of a landform it is necessary to k n o w not only its shape and occurrence, but also the sculpturing processes. In other words, geomorphology must deal both with geomorphological processes and their resulting landforms and draw conclusions as to the history of a landscape both from studies of form, material and process. Although the title of this review only mentions processes, w e will also consider the corresponding landforms. If w e define the Subarctic as a transitional region between the Arctic and the Boreal, we m a y also characterize its geomorphological processes as transitional in intensity between those of the arctic and the boreal regions. D u e to this and to the fact that in m a n y cases the processes and landforms have been more studied in the Arctic. I shall base this review on research and experience in both arctic and subarctic conditions,with particular reference to Fennoscandia. I readily admit that there are large gaps in my list of references, particularly as to Finnish and Russian papers. In all cold climates the geomorphological processes are to a large extent controlled by the seasonal shift from frozen to non-frozen ground and vice versa. O n e of the first systematical studies on this subject was that of B.H ö g b o m (1914),Über die geologische Bedeutung des Frostes. It was based on studies in the Arctic, particularly Spitsbergen. H ö g b o m (1926) published another paper dealing with contemporary frost processes in subarctic areas, Beobachtungen aus NordSchweden iìber den Frost als geologischer Faktor. In both his papers H ö g b o m made interpretations of patterned ground, solifluction features and other
forms and tried to reconstruct the type and rate of the frost-controlled processes and evaluate their role in sculpturing the landscape. His conclusion was that the contemporary rate of weathering and other geomorphological processes is very rapid in the Arctic and also, although to a lesser degree, in the Subarctic. ". ..my intention is to stress the until n o w underestimated geomorphological action of frost also in the subarctic areas." (Translated from Högbom, 1926,
p. 244.) Comprehensive reviews of the scientific literature on landforms and processes controlled by frost action have been published later, e.g. by Troll (1944),Büdel (1948),Callieux and Taylor (1954), Washburn (1956) and Tricart (1963). F r o m Fennoscandia two of the most recent and comprehensive papers in this field are those by J. Lundqvist (1962)and Ohlson (1964). In the following review I prefer to select some groups of processes and landforms, which I consider to be of particular interest to geomorphologists and ecologists working in the Subarctic. I have mainly chosen such features which can be used as indicators of contemporary or former climatic conditions.
WEATHERING OF ROCKS T h e weathering of rocks is the first link in the chain of processes, which by and by create a mantle of loose deposits, soil and landforms-small or large. If the rock did not weather, the valleys, hills and plains would look very different from what they actually do. W e distinguish between mechanical and chemical weathering. It is obvious that the climate is a controlling factor of weathering, together with the type of rock. If w e compare the conditions on cold, equatorial mountains with the conditions on high latitudes, frost
105
A. Rapp
is an effective weathering agent in both cases, but in entirely different ways, as has been clearly demonstrated by Troll and others. The frost-bursting is caused by the expansion of water when it is freezing. Near the Equator there is no great difference between the seasons of the year. At certain altitudes there is frost practically every night and thaw every day. But the freeze-thaw action does not penetrate more than a few centimetres into the ground, in contrast to the deep reaching seasonal freeze-thaw of the high latitudes. There w e m a y distinguish between a more maritime type of freeze-thaw cycle (short, frequent and shallow) and a more continental type. Freezethaw cycles can be divided according to frequency and duration in the following way: 1. Short cycles. Freeze-thaw several times per day. 2. Daily cycle. Freezing at night, thawing next day. 3. Several-dayscycle. 4. Annual cycle. Freezing in winter, thawing next summer. 5. Several-yearscycle. Permafrost fol1:wed by thaw. Frost-weatheringof rocks is one of the geomorphological processes which can be studied experimentally in laboratories. This was done by Tricart (1956)and W i m a n (1963).Tricart made his experiments on sedimentary rocks from France with high porosity and low resistance; W i m a n used metamorphic and igneous rocks from Sweden, with low porosity and higher resistance to frost-bursting. T w o different types of freeze-thaw cycles were used, the Icelandic and the Siberian. The former were of 24 hours duration, with
x
90 60 70
70
60
60
so
50
LO
LO
30
10
70
10
10
10
o
O 30
60 70
so
60
50
50
IO
O'
30
10
m
20
O
O
O
O
c
FIG.1. Grain-size distributions calculated on ten rock pieces, Unshaded piles : Icelandic temperature conditions. Shaded piles : Siberian temperature conditions. (From Wiman, 1963.) 106
an amplitude from 6 O C to -7 OC. The Siberian type had an amplitude from 15 O C to -30 OC, each cycle covering 4 days. The experiment lasted 36 days. Altogether thirty-sixIcelandic and nine Siberian cycles were run. Fifteen pieces of each rock type were placed in three plastic boxes, of which one was dry. In the other two about half of the rock pieces were under water. The rocks in the dry boxes were not weathered at all during the experiments, which proves the inefficiency of dry freeze-thaw cycles as regards frcstbursting. Water is the necessary dynamite. T h e wet pieces were all slightly weathered,although the broken fragments were small in Wiman's case. The weathering products were only about 2 per cent or less of the total initial weight of the rocks, in contrast to Tricart's experiments, where some rccks were broken to 100 per cent (Fiman, 1963, p. 3C4). In Wiman's experiments the weathering was greater in the Icelandic cycles than in the Siberian, which indicates that for weathering into small grains (Fig.i), the number of freeze-thaw cycles are probably of greater importance than their amplitude below zero. Figure 1 shows the grain-sizesproduced in this experiment. Tricart received stronger fragmentation in the Siberian cycles than in the Icelandic. These experiments clearly demonstrate the importance of water in frost-burstingand the great difference in resistance of different types of rocks. In nature this means that the water-soakedrocks can be quickly attacked and fractured, in contrast to nearby dry sites which are intact, e.g. with their glacial striations remaining after thousands of years. The local site and the type of material is thus of greater importance than the gross climate. The influence of local factors like those mentioned makes it difficult to evaluate in a general w a y the efficiency of rock-weathering in a certain subarctic climate as compared to an arctic or a boreal type. A certain basis for such comparisons can be the size of talus cones in similar rocks in both areas, as an indication of the total post-glacialfrost-weathering of steep rock walls. However, local variations in structure, jointing etc. of the rock can cause marked differences between nearby sites in the same rock. Another method for the evaluation of weathering of horizontal rock surfaces is to analyse the extent of destruction of glacial roches moutonnées. Probably the most important type of freeze-thaw cycles in high latitudes are the annual cycles or the several-daycycles, at least for the bursting of rockfalls from cliffs or walls. This is indicated by Figure 2, which also shows that the rockfalls tend to occur with a certain lag in time during the thaw period in spring (Rapp 1961, p. 105). T h e conclusion concerning the contemporary rate of frost-weathering of rocks in the Subarctic is, that detailed and direct evidence for an evaluation in this
Some geomorphological processes in cold climates
respect is on the whole missing, but indirect evidence, (talus slopes, block fields, destruction of roches m o u tonnées observed) indicates a weathering of the same strength as in arctic climates as far as wet sites are concerned. ill not be discussed here, Chemical weathering w although it is active also in the subarctic zone. In particular, limestone weathering is active in the Subarctic, judging from m a n y indications, such as salt content in rivers and existence of karst features in limestone areas.
OC
.15
2 i10
FROST-HEAVING AND CREEP
f *5 L
!O
4
The upward movement of boulders in the soil, caused by frost-heave,is a well-knownphenomenon, occurring as far south as there is a marked seasonal frost in the ground. W h e n the ground is freezing in autumn it swells, due to the addition of moisture from below and the expansion of water when freezing into ice lenses. Figure 3 gives an example of the frost-heaving in a sorted polygon-as m u c h as 10-20c m in the finegrained centre and 2-5 c m in the fringe of coarse material. T h e individual stones and boulders are frozen and heaved with the soil, leaving an open void at their lower end. This is either filled by material falling in from the sides or by ice. In this w a y the boulder does not sink back to its former position when the thaw occurs in spring;it stops in a slightly higher position than before. A number of freeze-thaw cycles will eventually heave the boulder to the surface. Thus, in the long run there w ill be a concentration of boulders on the ground. If the ground is sloping, the frost-heavecan cause the individual boulders to move d o w n slope (gliding boulders). These are generally orientated with their long axes in the direction of movement. An additional characteristic is a cushion of pushed soil at the down-slope end of the block with a furrow behind marking the line of movement. Another type of creep is caused by the formation of needle-ice or pipkrake. They consist of narrow ice needles formed at the beginning of frost penetration, when only a thin layer of the ground surface is frozen and when water is moving up by capillary action from the unfrozen substratum towards the cold layer. The ice needles grow as a dense mat and lift up particles as m u c h as several centimetres above the ground. In this w a y they can break the vegetation cover and, by repeated growth and collapse, cause a measurable movement d o w n slope of the whole surface layer. Particles as large as pebbles m a y be moved in this manner. Needle-ice formation is a c o m m o n feature all through the cold zones, including the Boreal, and is favoured by silty soils with high capillarity, high moisture content and slow freezing.
--
-5
rnrn 15
2 10 P
::5 L
O
FIG.2. Correlation between rockfalls, air temperature and precipitation in May and June 1953. Rockfalls recorded by inventories on snow all days except those marked by a minus (-)on top of the graph. The diagram shows highest frequency of rockfalls at thawing. Each case of rockfall recorded is marked by a dot (pebble-fall),a small triangle (small boulder-fall)or a large triangle (big boulder-fall). Air temperature at 10.00h and 13.00hfromrecordingsinKärkevagge at 820 m altitude. Precipitation at Riksgränsen weather station. (From Rapp, 1961.)
h i .
Oct.
... I
DEC.
Nov.
hiw
Jm. Feb.
May
Api.
Juri*
July
plug
I
8
I 6 5 4 3 2 1 O 1 2 3 d
5 6
I
I
I
I
I
I
I
I
I
I \ I
7
FIG.3. Soil surface movement in a sorted circle, from September 1957 to August 1958. Frost-heavingin the finegrained centre is shown by the continuousline.The less pronounced frost-heavingin the coarse-grainedmargin is shown by the dashed line.(After Z.Czeppe,from Jahn,1961.)
BOULDER DEPRESSIONS AND B L O C K FIELDS Frost-heavingand creep are considered to have caused the formation of boulder depressions and block fields. Both of these types occur above the timber line but they are perhaps more characteristic of the subarctic
107
A. Rapp
and northern boreal forests. It is an open question if they are still in formation on these low levels, or if they are fossil forms,indicating an earlier period with colder climate. The boulder depressionsin Sweden have been studied and discussed particularly by G.Lundqvist (1951)and J. Lundqvist (1962). They are “flat, barren fields of pure boulder material, situated in shallow depressions in the landscape.They occur also above the timber line, but are most typical in the forest region. T h e size of a boulder depression m a y vary from a couple of metres to large complexes, several hundred metres wide. Actually there are transitions downwards to typical stone pits. ... T h e largest boulders are those at the surface, downwards the material is gradually finer. T h e bottom often consists of a thick,fine-grainedsoil”. (J. Lundqvist, 1962, p..73-74.) Boulder depressions are interpreted as a result of frost heaving and re-sorting of till in wet sites. Most of the boulder depressions in forest have lichen-covered boulders and m a n y are partly overgrown by vegetation at the sides, which indicates that they are no longer active. But perhaps they are re-activated in extreme winters. The block fields (Blockfelder) described by H ö g b o m (1926)from the northern coniferous forest of Sweden are quite similar to the boulder depressions,but seem to differ in some respects. The block fieldsrest “directly upon the bedrock and the rock types in the boulders change strictly with those of the substratum. These conditions clearly indicate that the boulders are formed by direct disintegration of the bedrock” (J. Lundqvist, 1962, p. 76). Also in this form, the particles decrease in size downwards and it is often difficult to judge if they rest upon bedrock or not. H ö g b o m interpreted the block fields described by him as proof of a strong, contemporary frost-bursting and of boulder creep, effective even on almost flat ground in wet sites of the northern forests. Figure 4
illustrates his opinion. In some cases w e have tried to check his interpretation by examining the block fields photographed and described by him. T h e photographs and comparisons of the details 40 years later did not show any traces of either frost-heaveor lateral creep. This is an argument for stability. T h e formation of boulder depressions and block fields can probably be studied on high elevations in the mountains, where they seem to be forming today. If it is correct that these forms are mainly active on high altitudes above the forest limit, which some observations seem to indicate, this is another argument for the opinion that the boulder depressions and block fields of the forest zone were formed during m u c h more severe freeze-thaw conditions than those of today. It is particularly difficultto share Högbom’s opinion of a considerable boulder creep, with sidewise movements, on almost flat areas under forest cover. Therefore in the opinion of the present author the elongated block fields as well as the stone stripes found under forest cover are probably fossil features from a colder climate and forest-free conditions. They could, for example, have been formed soon after deglaciation, when the ground was still naked and unprotected by forest. Block fields and stone stripes are useful and resistant indicators of former cold climates also in the areas surrounding the W ü r m ice sheets, e.g. middle Europe and parts of the United States of America (cf. Büdel, 1937 ; Peltier, 1949 ; H.T.U.Smith, 1953).
SORTED POLYGONS As
concerns the boulder depressions,block fields and stone stripes of the subarctic forest lands, w e concluded that there are no proving observations of their recent formation. On the other hand, there are m a n y clear proofs of strong frost-heaving today creating other types of patterned ground, in the northern forests, such as the sorted polygons occurring on lake shores with fluctuating water level (J. Lundqvist, 1962). A recent detailed description of active, sorted polygons on the shores of Lake Gardiken, Wästerbotten, northern Sweden, is given by Wassén (1966,
p. 14-17,53-66). Small sorted polygons, so-called micropolygons,
with a diameter of 1-2d m , are actively forming today as far south as the island of dland (Rydqvist, 1960).
SOLIFLUCTION FIG. 4.A subarcticblock field below a rock wall at Mt.Skansberget, northern Sweden. The boulders are supposed to be creeping towards the lake to the right. The width of the field is about 50 m.(From Högbom, 1926.) 108
The term “solifluction” was created by J. G.Andersson (1906,p. 95), w h o defined it as the “slow flowing from higher to lower ground of masses of waste, saturated with water” and considered it as a chief agent of denudation in cold climates. T h e term “solifluction”
Some geomorphological processes in cold climates
was later used in a wider sense also for slow “soilflowing” in temperate and tropical regions. S o m e authors unfortunately also include rapid movements, such as obvious mudflows and slides in the term. T o avoid confusion, n e w terms have been proposed for the cold-climate solifluction, e.g. “congelifluction” (Dylik, 1951), gelifluction (Hamelin, 1960; cf. 1961, p. 63). As these n e w terms are perhaps not well established yet, w e prefer in this report to use the old term solifluction. The action of frost in the ground is in m a n y cases indicated on the surface by the different types of patterned ground (Washburn, 1956). On horizontal ground these develop as, e.g. circles, nets and polygons of different types but on sloping ground they are elongated d o w n slope (cf. Troll, 1944, Fig. 29) or are represented by different types of steps, terraces, lobes and stripes. The down-slopemovements seem to occur from minimum inclinations of about 2-30 (Rudberg, 1962, p. 313; Büdel, 1960, p. 61; and others). T h e transition from flat-ground patterns to down-slope solifluction forms is often quite obvious in the field and there seems to be a close relationship between frost-heaving and similar cryoturbation processes on flat ground and solifluction on slopes. Probably most of the ground above the forest limit is affected by movements caused by frost, although there are widespread areas where no structural pattern is developed. Below the forest limit actual solifluction seems to be absent, or at least very slow, except on open, forestfree patches with sloping ground, silty soil and poor protection from snow and vegetation. Well-drained ground is not affected by frost action to any appreciable degree either above or below the forest limit. Solifluction, as well as frost-heaving,is facilitated by silty soil and an abundant supply of water (cf. Beskow, 1930; Williams, 1957, p. 47). F r o m field experiments on solifluction in the Colorado Front Range, United States, Benedict (1966) concludes: “Annual measurements in a variety of micro-sites show that present rates of downslope movement.. . are determined almost entirely by soil-moistureconditions. ... Rates of movements greater than 20 mm per year have been measured only where the water table lies within 10 c m of the ground surface at the beginning of the fall freeze-up.’’
MEASUREMENTS OF SOLIFLUCTION MOVEMENTS In m a n y localities in the Arctic, or above the timber line in mountains of lower latitudes, the actual rate of solifluction movements have been measured. These movements mainly occur during the thaw period in spring or early summer (Williams, 1959, p: 484) but also during freeze-up in autumn (J. Smith, 1960; Jahn, 1961).
The movement of the surface layer can be measured
by annual checking of the position of short wooden stakes, or other markings in the ground. They can be checked by theodolite or steel tape readings from fixed points. This method, combined with recording of the vertical gradient of the moving layer by means of vertically incised test pillars (Rudberg, 1962,p. 318) has been used by the present author in Kärkevagge, Swedish Lapland (Rapp, 1961). The rate of actual solifluction in the period 1953-1960 varied between O and 8 c m per year, in extreme cases being as m u c h as 30 c m per year on ground with fine-grainedtill and abundant water supply. T h e gradients of the slopes were 10-300. The movement was most rapid at the surface,decreasing rapidly with depth, there being no movement at all at about 50-60 c m depth in most of the cases recorded. M a n y of these solifluction slopes had a clos-e cover of grass which was not broken by the flow. Most recordings of solifluction in Swedish mountains indicate the same magnitude as those reported from measurements by other authors, e.g., Williams (1957),in Norway, Büdel (1960) and Jahn (1961) in Spitsbergen, J. Smith (1960) in South Georgia, Benedict (1966) in Colorado, etc., and by Washburn in 1963 in Greenland. These studies indicate that solifluction on favourable sites above the forest limit causes a surficial movement of some centimetres or at a m a x i m u m some decimetres per year. All the recordings mentioned above concern the actual movements. An attempt to measure the total solifluctionmovement in post-glacialtime was made by T. Lindell in a study of ancient ice-lake shore-lines at Lake Grövelsjön, Dalecarlia, Sweden (Lindell,1962). The shore-lineswere formed at deglaciation,probably about 8,000 years ago. Most of the shore-lineshad not moved to any measurable extent, but in the wet sites they were deflected downslope by solifluction for about 1-9 m. T h e soil material is a wave-washed till with only about 10-20per cent of silt and finer grains and is consequently rather resistant to creep and solifluction. The inclination of the slope is as high as 25-330and it is situated in the upper part of the birch forest belt. The movements are probably fossil as there is a well-developed podzol profile all over this slope now. Another promising method of measuring the movement by solifluction over a longer period of time is by means of Cl,-dating of a buried organic layer beneath solifluction lobes. Benedict (1966) dated and reconstructed the movement of a solifluction terrace in the Colorado Front Range in this way: “In their buried organic layers, solifluction lobes and terraces preserve a record of climatic change. Reading this record involves the expense of a closelyspaced series of radiocarbon dates and requires a knowledge of the factors controlling present rates of downslope movement. “I have dated the entire thickness of the buried A
109
A. Rapp
horizon. This has introduced m a n y sources of error that would not have been present if I had been able to limit my sampling to the remains of plants actually growing on the ground surface at the time of burial. “. .. The terrace originated 3,000-2,500years ago, during the Tempel Lake time. T w o periods of rapid downslope movement occurred each during the latter part of a minor glacial episode, and each reflecting an interval in which the soil was saturated,but snowfree, at the beginning of the fall freeze-up.Optimum conditions for rapid movement probably occurred as the terraces first emerged from perennial snowbanks that covered this slope during glacial maxima.” (Benedict,
1966.)
PERMAFROST Permafrost or perennially frozen groundAis typically restricted to areas with very cold winters and a thin snow cover,which cannot prevent the deep penetration of frost into the ground. Northern Siberia with its extremely cold winters is the heartland of the Eurasian zone of continuous permafrost. F r o m there a zone of discontinuous permafrost extends westwards along the coast of the Arctic Ocean to northernmost Scandinavia, with a lobe towards the south along the Scandinavian mountains and the rather continental areas to the east of them. The general trend of the permafrost zones in northern Europe can be seen on small-scale maps, e.g. those compiled by Brown and Johnston (1964) and Maarleveld (1965). P E R M A F R O S T IN PALSAS
T h e palsas are the most conspicuous indicators of permafrost in marginal areas. The interior of a palsa consists of frozen peat with or without ice layers. In m a n y areas it also includes frozen updomed parts of the non-organic substratum such as silty till or glaciofluvial material (Wramner, 1965, p. 498). Figure 5 shows the approximate area of discontinuous permafrost in northern Sweden and adjacent parts of Finland. It is mainly based on the distribution of palsa bogs (after J. Lundqvist, 1962, p. 15; Hoppe and Blake, 1963, p. 167; Ohlson, 1964, p. 152). “Observations have shown that the mean annual air temperature is less than the mean annual ground temperature by about 20 C to 50 C, depending on local conditions: the over-all average is about 30 C.”(Brown and Johnston, 1964, p. 67.) The outer limit of palsa bogs in Sweden, as shown in Figure 5, roughly coincides with the annual isotherm for -20 C (period 1901-1930). However, both these lines are broad generalizations, so the actual relationship between air and ground temperature must be judged from measurements at the individual permafrost localities, taking into consideration not only air
110
temperature but also local relief, snow cover, soil type, vegetation, and drainage. These factors have a great influence upon the existence and depth of permafrost. It can be supposed that permafrost lenses also occur in bed-rock or minerogenic loose deposits outside the palsa mires. PERMAFROST INDICATED B Y T U N D R A POLYGONS
As concerns northern Fennoscandia very little is so far k n o w n about permafrost types other than palsas. A compilation by E k m a n (1957, p. 34 f.) summarizes some reports of permafrost found in drilling or excavating operations in northern Sweden. F r o m northern Norway Svensson (1963) reports that permafrost is not encountered in road construction, except in peat bogs. Ohlsson (1964)reports from the Enontekiö area in northern Finland that permafrost has not been found there, outside the palsa bogs. As mentioned above, the palsas are regarded as indicators of rather weak permafrost and seem to require a mean annual air temperature of -20 C or colder to develop and persist. Other surface indicators of permafrost are ice-wedgepolygons (or frost-fissure polygons/thermal contraction polygons/tundra polygons) but in order to develop they require a m u c h more severe climate. Actively growing ice wedges in Alaska “occur, for the most part, in the continuous permafrost zone of northern Alaska, where mean annual air temperature ranges from -60 to -120. .. Inactive ice wedges, no longer growing, occur in the northern part of the discontinuous permafrost zone of central Alaska,where mean annual air temperatures range from -20 to -80 C: they have been found only in fine-grained silty sediments” (Péwé 1963,
p. 129). Recently some localities with crack-like patterns on the ground have been discovered in mountain valleys in northern Sweden within the zone of discontinuous permafrost (Rapp, Gustafsson, Jobs 1962; Fig.5). The patterns are similar to true tundra polygons, either in a fresh or fossil form. In a current project w e are making closer investigations of these forms, with regard to their origin and age, and the climatic and ground conditions they indicate. The tundra polygons studied by us are very like those investigated by Svensson (1962, 1963) in northern Norway, particularly the forms in the upland areas. Three areas with tundra polygons have so far been discovered in the Padjelanta upland, west of the Sarek mountains, and one on the shores of Lake Satisjaure in the birch forest belt, east of the main ridge of the Caledonides. This locality is particularly interesting.The altitude is not more than about 440 m , but the cooling effect of the lake and the strong winds
Some geomorphological processes in cold climates
FIG. 5. The approximate southern limit of palsas in Sweden and part of Finland is indicated by the continuous line on this map. P = permafrost localities with tundra polygons in Padjelanta.S = the same, at Lake Satisjaure. L = permafrost in bed-rock at Mt. Låktatjakko. Dotted lines = two isotherms for mean annual temperature. (From Rapp and Annersten.)
prevent the forest from growing and probably keeps the snow cover very thin in the locality.Similartundra patches occur in valley bottoms and along shores of other lakes in Lapland, e.g. Lake Torneträsk, 342 m. The polygons of the Puolejokk field have been investigated in some detail (Rapp and Annersten, 1966, in press). It is clear from ground temperature measurements by thermistors d o w n to 4 m depth at ten sites in the polygon area, that permafrost occurs below the polygon patterns. The active layer is about 1-2m thick. On the other hand there is no permafrost below the gullies which dissect the plateaux of silty ice-lakesediments or till upon which the tundra polygons are sitting. Thus the occurrence of permafrost in this locality is very m u c h influenced by snow. On the ridges the snow cover is very thin, even in the coldest part of the winter, due to strong snow-drifting. There the frost penetrates very deep down. In the gullies where the drifting snow is trapped and accumulated to drifts of several metres thickness, no permafrost exists. The tundra polygons have four or five sides and a diameter of about 10-40m. They are rather irregular and in m a n y cases one side is open. They occur on air photographs as dark lines, surrounding a lighter central area. T h e polygon pattern is very conspicuous when seen from the air, due to the contrast between the low and dense vegetation (Betula nana, E m p e t r u m and Salix sp.) growing in the wind-shelteredtrenches and the partly naked ground close by, which has only lichens, mosses and some few Empetrum stands.
The width of the trenches in these tundra polygons varies from about 0.2 to 2.8 m on the upper rim and 0.1 to 1.8 m at the bottom. The depth of the trenches is from 0.05 to 0.4 m.They have a thick humus layer and also a thickening of the podzol profile in the bottom. Below the surficial trenches are very tiny, fossil ice wedges and narrow, vertical ice veins, probably in the active layer. The rather wide surficial polygon trenches were probably initiated by frost cracking and infilling, but were probably also widened through secondary processes -running water at the edges ofsloping plateaux, the growth and wedging action of roots (?),frostheave (?)and possibly wind action (?). Even if the permafrost still remains below an active layer of 1-2 m,the large polygons are probably fossil features or have been intermittentlyrevived to a slight extent on rare occasions of very cold Conditions. I have reported these observations in rather m u c h detail, as I consider them to be of particular interest for the continued research in subarctic and arctic areas. Probably tundra polygons will be found in m a n y localities in northern Fennoscandia and if so they will be interesting indicators of present and past climatic and ecological conditions. P E R M A F R O S T MOUNDS UNDER FOREST COVER
A recent paper by Viereck (1965)gives an interesting example of the possible influence of small perma€rost
111
R. Rapp
lenses on forest growth and also the relations between snow cover and permafrost in a marginal area. T h e paper has the title, Relationship of white spruce to lenses of perennially frozen ground, M o u n t McKinley
Frozen lens under mature white spruce
National Park, Alaska.
T h e area is within the zone of discontinuous permafrost. The study reports an investigationof permafrost lenses beneath individual white spruce (Picea glauca (Moench) Voss) trees growing on a terrace of the McKinley River in Alaska, at an elevation of 550 m. This in close to the altitudinal limit of timber in the region, which reaches 750 m on south-facing slopes. The climate of the area is continental although the temperatureextremes are not as great as in the interior lowland forests north of the Alaska Range. The mean annual temperature is approximately -4.60 C, January is the coldest month with a mean temperature of -19.00 C, July the warmest with 11.50 C. Annual precipitation is about 515 mm and snow accumulation in the forest during the winter varies from 0.5 to 1.5 m. All these data suggest a climate which is only slightly colder than that of the continental north of Sweden, Finland and Norway, where Karesuando, Siccajaure and Kautokeino had a mean annual temperature of -1.40 C to -2.30 C in the period 1921-1950 and still lower in 1901-1930(Ohlson,1964,p. 25, 27). Perennially frozen mounds have developed in the forest studied by Viereck under white spruce growing in silty clay. The frozen lens is thought to result from the insulating effect in summer of a thickened moss mat and from soil cooling in winter as a result of a thin snow layer under the trees. The mound is created through expansion of the silty clay caused by incorporation of water into the lens as thin layers of clear ice. Disturbance of the moss m a t results in a melting of the lens, a collapse of the mound, and often the death of the tree. As n e w trees develop, n e w mounds and frozen lenses develop in the soil beneath them (Fig.6).
bouldery levées are built out in lakes,creating barriers which cut off lagoons (F.Hjulström, personal communication). The effect of ice erosion on the river banks varies in degree from year to year depending on the discharge and water level. As an average the erosion and sedimentation in subarctic rivers is rather intense due to the great differences in discharge during the year and the strong effect of ice break up.
REMARKS ON RIVER ACTIVITY IN THE SUBARCTIC
CONCLUSION AND SUGGESTIONS FOR FURTHER RESEARCH
The hydrology of subarctic rivers is of course marked
The distribution and frequency of frost controlled landforms are probably less well k n o w n in the Subarctic than in parts of the arctic areas and so also is the contemporary activity of processes such as frostheaving, solifluction, creep, formation and decay of permafrost. T h e climatic and palaeoclimatic significance of patterned ground and permafrost forms in the subarctic areas is a fruitful field for future research, e.g. by comparisons with contemporary tundra areas and the fossil tundra zones south of the Pleistocene ice sheets. Most of our observations of frost-controlled forms and processes in the subarctic areas have been performed during the shift towards warmer climate until
by a seasonal variation: low discharge during winter, spring flood, sometimes a second flood in summer or early autumn due to heavy rains, marked decrease in discharge after freeze-up and during the winter. Only a few remarks concerning subarctic rivers will be included here. The subject is a very large and interesting field for future studies. Ice break-up is often a rather sudden and catastrophic event. The ice is then pushed up in high ridges which can cause damage and efosion on the shores (and their vegetation cover) as well as on the river beds. Boulders on the shores are moved by the ice and in the long run,ice-pushedridges of rock debris can be formed on gently sloping shores. In m a n y cases such
112
Frost scar and seedling
Moss and decayed organic material
W
Silty clay Fraxen silty ciay
FIG.6. Cycle of development and collapse of tree mounds. Seedlings germinate on frost scar or bare surface created by melting of the frozen core in the mound. (From Viereck, 1965.)
Some geomorphological processes in cold climates
the 1950s. If the series of cold winters and cool summers in recent years continues this m a y be a turning point towards increased activity of frost processes and the extension of the permafrost features in the Subarctic, with considerable effects on the general
ecology and life in the north. Thus there are several reasons for maintaining and widening our knowledge of the cold climate processes, their dynamical balance, and response to climatic shifts.
Resumé lhude de certains processus géornorphologiques dans les régions à climat froid (A.Rapp)
L’auteur passe en revue différents processus géomorphologiques et les formes qu’ils donnent au relief, en accordant une attention particulière aux phénomènes cryergiques qui se produisent dans le nord de la Fennoscandie. L a question de la gélifraction des roches et de ses rapports avec la nature de la roche de fond,les cycles de gel-dégelet la teneur en eau est illustrée au moyen d’exemplestirés d’expériences et d’observations faites sur le terrain. L’auteur étudie la poussée de gel verticale et le déplacement des particules sous l’action du gel et il conclut que ces phénomènes jouent un rôle important à notre époque dans la région subarctique, notamment sur les rives exposées des lacs dont le niveau varie, c o m m e l’indique l’existencede polygones de matériaux récemment triés. Selon lui, la plupart des grandes dépressions à gros blocs et des champs de pierre qu’on trouve dans les régions subarctiques boisées se sont formés à une époque où le climat était plus froid que pendant la période 1920-1950.
Les mouvements de solifluxion sont sans doute exceptionnels ou très limités dans les terrains boisés. L a zone subarctique du nord de la Fennoscandie est située à la limite de la zone eurasienne de pergélisol permafrost discontinu. En surface, les régions de pergélisol se caractérisent principalement par l’existence de fondrières à palses. L’auteur décrit certaines grosses lentilles dues au pergélisol que l’on trouve parfois en Laponie suédoise dans les dépôts glaciaires. On peut déceler leur présence à la surface du sol par suite de la disposition en pclygones de la toundra, qui apparaît sur des photographies aériennes. Mais, tous ces polygones ne sont pas situés sur la zone de pergélisol. Les petits monticules de pergélisol qui gênent la croissance des arbres en Alaska fournissentun exemple de la formation et de la disparition du pergélisol dans les zones marginales à l’époque contemporaine. L’auteur recommande que l’on étudie de façon permanente l’évolution et le développement des phénomènes cryopédologiques dans la région subarctique de la Fennoscandieà la suite de la série d’annéesfroides enregistrées depuis 1960.
Discussion A. JAHN. Does solifluctionexist in the subarcticarea? There are examples of solifluctionin the subarcticzone for example in Seward peninsula (Alaska) described by Hopkins and Sigafoos, and in Norway, examined by Williams. This kind of solifluction exists under the vegetation cover.
A.RAPP.I think that solifluctioncan occur on open patches in the forest,with sloping ground, silty soil and poor protection from snow and vegetation. Thus it can occur but on a smaller scale and at a lower rate than in favourable localities above the forest limit.That a slight movement of the soil may take place even on forested slopes is in some localities indicated by the bending or curvature of tree trunks just above the ground. J. MALAURIE. Quelles preuves avez-vous de la contemporanéité de l’érosion mécanique dans les pierres du relief
analysé? Quelle est, en ce cas, la durée appréciée des phénomènes d’érosion mécanique? Connaissez-vous l’existence de sols structuraux sousmarins? Quelle conclusion en tirez-vouspour les sols structuraux présentés?
A. RAPP.Proofs of contemporary mechanical erosion. Evidence of fresh weathering,etc. is, for example,on talus slopes, occurrence of boulders with fresh surfaces,not covered by lichens or other vegetation,in contrast to the other vegetation-covered surfaces nearby. The same evidence can be applied to patterned ground: stone circles, etc., with fresh and not vegetation-coveredmineral soil in the centre, are active; those with vegetation cover or with an undisturbed soil profile are inactive. The best way to prove if a feature is active or not is to make continuous observations from year to year, using
113 8
A. Rapp
.
bench-marks, painted lines or photographs as a basis for comparison. Sorted polygons and similar forms of patterned ground are c o m m o n on the shores of some lakes and ponds, but these observations m a d e in summer do not show if the polygons were formed under water. I think they were formed above
Bibliography ANDERSSON, J. G. 1906. Solifluction, a compoment of subaerial denudation. J. Geol.,Chicago, vol. XIV. BENEDICT, J. B. 1966. Radiocarbon dates from a stonebanked terrace in the Colorado Rocky Mountains, U.S.A. Geograf. Annaler. Stockholm, vol. 48 A, p. 1. BESKOW, G. 1930.Erdfliessen und Strukturböden der Hochgebirge im Licht der Frosthebung. Geolog.Fören. Förhandl. Stockholm,vol. 52,p. 4. BROWN, R. J. E.;JOHNSTON, G. H. 1964. Permafrost and related engineering problems.Endeavour,vol. XXIII,p. 89. BUDEL, J. 1937. Eizeitliche und rezente Verwitterung und Abtragung in ehemals nicht vereisten Teil. Mitteleuropas. Peterm. Geogr. Mitt. Erg.-H.,no. 229. . 1948. Die klima-morphologischen Zonen der Polarlander. Erdkunde, no. 2,p. 1-3. _- . 1960. Die Frostschutt-Zone Südost-Spitsbergens. Colloquium Geograph.,Bonn, vol. 6. CALLIEUX, A.;TAYLOR, G. 1954. Cryopédologie. Exp. Pol. Franc., Miss. P. E. Victor. Paris, Hermann &Cie. DYLIK, J. 1951. S o m e periglacial structures in Pleistocene deposits of Middle Poland. Bull. Soc. Sci. Lett. Lodz. EKMAN, S. 1957. Die Gewässer des Abisko-Gebietesund ihre Bedingungen. Kungl. Sv. Vet. Akad.: s Handl. 4th ser., Stockholm, vol. 6,no. 6. HAMELIN, L.-E. 1961. Périglaciaire du Canada. Coll. de Géograph. de Québec, vol. 10. HOPPE, G. ; BLAKE, I. 1963. Palsmyar och flygbilder : Einer, Stockholm, p. 165-168. HÖGBOM, B. 1914. Über die geologische Bedeutung des Frostes. Bull. Geol. Inst. Upsala, vol. XII. . 1926. Beobachtungen aus Nord-Schweden über den Frost als geologischen Faktor. Bull. Geol. Inst. Upsala, vol. xx. JAHN, A. 1961. Quantitative analysis of some periglacial processes in Spitsbergen Nauka o. Ziemi, Warszawa, vol. II, Ser. B, no. 5. LINDELL, T. 1962. D e n postglaciala deformationen av issjöstrandlinjer vid Grövelsjön. Uppsala, Dept. of Geography (unpublished). LUNDQVIST, G. 1951. Blocksänkor och några andra frostfenomen. Geol. Fören. Förhandl. Stockholm, vol. 73. LUNDQVIST, J. 1962. Patterned ground and related frost phenomena in Sweden. Sveriges Geol. Unders. Stockholm, vol. C, p. 583. MAARLEVELD, G. C. 1965. Frost mounds. A summary of the literature of the past decade Meded. v. Geol. Sticht. Nieuwe Serie, Maastricht, vol. 17. OHLSON, B. 1964. Frostaktivität, Verwitterung und Bodenbildung in den Fjeldgegenden von Enontekiö, FinnischLappland. Fennia, Helsinki, vol. 89, no. 3. PELTIER, L. 1949. Pleistocene terraces of the Susquehanna
_-
_-
114
the lake surface, but that they m a y have been drowned by rising water level in summer. Several studies have been made and published showing that the sorted types of patterned ground do develop on ridges or slopes where sub-aquatic situations are topographically impossible.
I Bibliographie river. Pennsylvania. Harrisburg. (Penna. Geol. Surv. Bull. G.23.) P É w É , T. L. 1963. Ice wedges in Alaska-classification, distribution, and climatic significance. Abstract, 1963 Annual Meeting, Geological Society of America. RAPP, A. 1961. Recent development of mountain slopes in Kärkevagge and surroundings, northern Scandinavia. Geogr. Annaler, Stockholm, vol. 42,p. 2-3. __ ,. ANNERSTEN, L. (In press.) Permafrost and tundra polygons in northern Sweden. Proceedings of the INQUA Symposium in Alaska, 1965. ; GUSTAFSSON,K.;JOBS, P. 1962. Iskilar i Padjelanta? (Ice-wedge polygons in Padjelanta, Swedish Lapland?). Ymer,Stockholm,vol. 82. RUDBERG, S. 1962. A report on some field observations concerning periglacial morphology and mass movement on slopes in Sweden. Biul. Peryglac.,Lodz, vol. 11. RYDQUIST, F. 1960. Studier inom öländska polygonmarker. Ymer,Stockholm, vol. 80. SMITH, H . T.U. 1953.The Hickory R u n boulder field. Amer. J. Sci., vol. 251. SMITH, J. 1960. Cryoturbation data from South Georgia. Bid. Peryglac., Lodz, vol. 8. SVENSSON, H. 1962. Note on a type of patterned ground on the Varanger Peninsula, Norway. Geogr. Annaler, Stockholm,vol. 44. . 1963. Tundra polygons. Photographic interpretation and field studies in North-Norwegian polygon areas. Norges Geol. Unders. Arsbok 1962, Oslo. TRICART,J. 1956. Études expérimentales du problème de la gélivation. Biul. Peryglac, Lodz, vol. 4. . 1963. Géomorphologie des régionsfroides. Paris, Presses Universitaires de France. TROLL, C. 1944. Strukturböden, Solifluktion und Frostklimate der Erde. Geol. Rundschau, Stuttgart, vol. 34. VIERECK, L. A. 1965. Relationship of white spruce to lenses of perennially frozen ground, Mount McKinley National Park, Alaska. Arctic, vol. 18, no. 4. WASHBURN, A. L. 1956. Classification of patterned ground and review of suggested origins. Bull. Geol. Soc. Amer., New York,vol. 67. WASSEN, G. 1966. Gardiken. Vegetation und Flora eines lappländischen Seeufers. Kungl. SV. Vet. Akad. Avhandl. i Naturskydd, Stockholm, vol. 20. WILLIAMS,P. J. 1957. S o m e investigations into solifluction features in Norway. Geogr. J.,London, vol. 123,p. 1. . 1959. A n investigation into processes occurring in solifluction. Amer. J. Sci., vol. 257. W I M A N , S. 1963. A preliminary study of experimental frost weathering. Geograf. Annaler, Stockholm, vol. 45, p. 2-3. W R A M N E R , P. 1965. Fynd av palsar m e d mineraljordkärna i Sverige. Geol. Fören. Förhandl., Stockholm,vol. 86,p. 4.
__
__
--
-_
Denudation and river erosion in the “zone of pronounced valley formation”on South-east Spitsbergen J. Büdel
Vallèys, as a product of linear river erosion, are the most important elements of the exogenous formation of the relief. It is above all through valleys that the relief is formed in all hilly and mountainous areas in the tropics as well as in the entire extra tropical zone. Research w o r k of the last few years has revealed that even with the s a m e tectonic-epeirogeneous and petrographic conditions (i.e. with the s a m e situation concerning epeiro-variance and petro-variance) valleys s h o w great quantitative and qualitative variations according to the climatic conditions during their formation. Quantitative variations. T h e longitudinal and the vertical evolution of valleys vary in speed. Qualitative variations. Within a given time span valleys develop either a narrow cross-profile plus a short long-profile with m a n y knickpoints or a wide cross-profile plus a long, parabolic, a n d graded longprofile (in German: durchhangend b‘concave-upwards”). T h e following t w o p h e n o m e n a have been discovered: 1. On the one hand, large, currently formed erosion surfaces are predominant (apart from deposition plains) in the lowlands of the tropical a n d subtropical zones. Valleys are confined to areas of strong uplift, i.e. to hilly and mountainous regions. E v e n there the quantitative process of valley formation is often only very slow. As regards the qualitative process narrow cross-profiles and long-profiles with knickpoints (often even with waterfalls) are predominant, according to the varying erodibility of rocks. 2. On the other hand, all parts of the relief are controlled by valley formation in the extra-tropical zones (even in lowlands, except very y o u n g deposition plains). Wide, level-bottomed valleys with rapid headward erosion are dominant everywhere in hilly regions up to areas with mountains of moderate height. Long-profiles are m u c h less influenced by
varying rock resistance; rather smooth, parabolic long-profiles without distinct knickpoints are prevailing. Detailed investigations have s h o w n that these wide a n d rather smoothly sloping valleys in the middle latitudes have not been shaped under present climatic conditions but under those which existed during the Pleistocene cold periods in the ice-free region. It m e a n s that today the processes forming these valleys can n o longer be observed. In order to gather knowledge on this complex of morphogenous processes w e have to visit areas with present climatic conditions which are similar to those which existed during the W ü r m glaciation period in the moderate climatic areas of Europe, North America and northern Asia beyond the ice cover. T o this end the G e r m a n “Stauferland-Expedition” to South-east Spitsbergen took place during the s u m m e r m o n t h s of 1959 a n d 1960. Subpolar climatic conditions are very distinct there; in large parts of this archipelago glaciers and ice sheets are still lacking. T h e scientific yield of the expedition is the fact that the most active valley formation o n the Earth takes place in this region, both from the quantitative a n d qualitative points of view. This induces us to create the term “zone of pronounced valley formation” for this climatic zone, according to the system of climatomorphological zones of the Earth. There are three essential causes accounting for the extreme intensity of valley formation characterized b y graded profiles and wide valley-bottoms-even with small rivers: 1. Large a m o u n t s of coarse, mechanically weathered debris are transported from the slopes to the valleybottoms by a special kind of erosion process. 2. T h o u g h the absolute a m o u n t of precipitation is only small in most subpolar regions of the world, a n extraordinary high run-off coefficient is obtained during a few weeks every year, i.e. during the spring
115
J. Biidel
melting period. So the rivers are able to transport the high bed load and to form thereby wide riverbottoms through strong lateral erosion. 3. Within the reach of permafrost, which always dominates in these subpolar zones, vertical erosion and retrogressive erosion of rivers are facilitated to a degree which is not reached in any other climatic zone of the earth. The scientific results of these three causes follow in more detail.
EROSION The greater part of the islands of South-eastSpitsbergen consists of ice-free,almost vegetationless plateaux, 250-400 m above sea level. T h e steeper parts of the slopes of these plateaux have gradients of 30-350; channel wash predominates here. Erosion works by a combination of sheet wash and solifluction on the rather gentle, convex upper slopes and the also gentle, concave lower slopes. The two processes take place within the uppermost 20-30 c m , where the ground freezes only seasonally; 400 m of permafrost lie below that level. The surface of the seasonally frozen ground shows frost structures in the form of striped ground. These are the result of a combination of sixteen different processes.Even with the same bed-rock and gradient there m a y occur different kinds of striped ground according to the position on the convex upper slopes or the concave lower slopes respectively. This is because sheet wash dominates over solifluction on the upper slopes, whereas the effect of solifluction is predominant on the lower slopes. The annual speed of solifluction was accurately measured on slopes with all gradients. During the 15,000 years of the postglacial period 6 m of ground have been removed by all these denudation processes on upper slopes with a gradient of 60; steep slopes with more than 350 have retreated 15 m at most. The same occurred on the concave lower slopes although here it is the result of pronounced down-cutting of the rivers. These results of measurement allow us to estimate more accurately the amount of debris entering the rivers in early summer every year.
LATERAL EROSION The rivers of the subpolar zone are able to transport the entire load even when precipitation is small. This takes place during the flood period of spring thaw. Almost the entire amount of winter precipitation melts at the same time and runs off in the form of a high flood, which lasts several weeks. Loss by evaporation is extraordinarily small. A complete cover of vegetation with high water consumption is mostly absent. Striped ground and rills on the slopes lead the meltwater stiaight d o w n into the valleys and,what is more
116
important, there is no seepage as permafrost lies below. Thus the rivers have the extraordinary runoff coefficient which is necessary to transport the entire load, to cause intense lateral erosion, and to form the wide valley-bottoms. It often occurs that river beds are already 60-150m wide after a course of only 6-10km.
DOWNCUTTING AND H E A D W A R D EROSION In subtropical and tropical climates river erosion has two effects: (a) it loosens the solid rack and disintegrates it into such small pieces that these can be carried away, and (b) it slowly abrades the solid rock below using the coarse material as tools. This is an extremely slow process, because each outcrop of hard rock is an element of resistance. All these phenomena result in longitudinal profiles with m a n y knickpoints, often even with waterfalls, especially each outcrop of resistant rock causing a knickpoint. In the subarctic region the permafrost is not even interrupted beneath the rivers. The entire upper layer consists of parts of ground ice; the solid rock in and between these parts is already completely shattered. The 0.5-1m thick upper zone of the permafrost is the so-called “ice rind” (German: Eisrinde). In this uppermost zone of the permafrost the temperature change from summer to winter results in a great change in volume. T h e ice contracts and thus disintegrates the solid rock, when temperatures are low in winter. The resulting fissures become filled with ice which expands in summer with the temperatures rising to almost 00 C. The solid rock beneath the ice rind consequentlyis completely disintegrated and ready to be carried away. The river’s task is only to melt the ground and to add the debris to its load. It is therefore able to carry it away immediately. This means that the most difficult role of river erosion has already taken place within the ice rind. Therefore the speed of river erosion is m u c h higher here than anywhere else. In the longitudinal profile knickpoints are m u c h less obvious, because there is no solid rock which could resist frost-wedging.Graded, parabolic longitudinalprofiles with wide river-bottoms are the result. Here, in the subpolar zone of pronounced valley formation they can even be faund in the uppermost branches of the valley systems and in areas with relatively bold relief. As already indicated, the whole process is also very important for the formation of the relief in the middle latitudes, because the graded longitudinal profiles and the wide valley-bottomsof all rivers from the subpolar region to the regions near the subtropical zone (450 N. and 450 S.)are caused by the frost climate and the permafrost during the glacial periods of the Pleistocene.
Denudation and river erosion in the “zone of pronounced valley formation”
Résumé Dénudation et érosion jluviale dans la zone de formation accusée de vallées au Spitzberg du Sud-Est (J. Büdel)
L’auteur examine les résultats de l’expédition allemande Stauferland au Spitzberg du Sud-Est(étés 1959 et 1960). Dénudation
L a région comprend des plateaux sans glace, presque sans végétation, situés à 250-400 mètres au-dessus du niveau de la mer. Au milieu des pentes de ces plateaux, on trouve des gradients de 30 à 350; leruissellement en rigoles y prédomine. Sur les pentes supérieures convexes et les pentes inférieures concaves, la dénudation est due à l’effet combiné du ruissellement et de la solifluxion.Les deux phénomènes se produisent jusqu’à une profondeur de 20 à 30 c m de la surface, là où le sol n’est gelé que saisonnièrement; au-dessous de ce niveau se trouvent 400 mètres de pergélisol. Sur les pentes, la surface du sol gelé saisonnièrement présente un agencement en forme de stries. Celles-ci résultent de l’action combinée de seize phénomènes différents. M ê m e lorsque la roche en place et le gradient sont identiques, on peut trouver différentes sortes de sols striés, selon qu’il s’agit des pentes supérieures convexes ou des pentes inférieures concaves, parce que le ruissellementen nappe est plus important que la solifluxion sur les pentes supérieures, tandis que l’effet de la solifluxion prédomine sur les pentes inférieures. L a vitesse annuelle de la solifluxion a été mesurée avec exactitude sur des pentes de divers gradients. Pendant les quinze mille ans de la période post-glaciaire,6 mètres de sol ont été éliminés par ces phénomènes de dénudation sur les pentes supérieures avec un gradient de 60. Les pentes raides ayant des gradients de plus de 350 ont reculé de 15 mètres au maximum. Des chiffres analogues ont été enregistrés sur les pentes inférieuresconcaves,bien qu’ils résultent aussi d’une forte érosion fluviale.
Érosion jZuviaZe
Sous les climats subtropicaux et tropicaux, l’érosion fluviale a les deux effets suivants: a) elle ameublit la roche en place et la désagrège en particules assez petites pour être transportées (ce phénomène est extrêmement lent) ; b) elle emporte les particules résultant de la désagrégation de la roche. I1 en résulte des profils longitudinaux qui présentent des brisures, souvent m ê m e des chutes d’eau; en particulier, toute veine résistante provoque une brisure. Dans la région subarctique,le pergélisol n’est m ê m e pas interrompu au-dessous des cours d’eau. Sa couche supérieure est constituée en totalité de morceaux de glace ;la roche en place dans ces morceaux et entre eux est déjà entièrement fragmentée. Cette couche supérieure du pergélisol, épaisse de 0,5à 1 mètre constitue ce qu’on appelle la “croûte de glace” (en allemand: Eis-Rinde). L a roche en place y est complètement désagrégée et transportable. L e cours d’eau n’a plus qu’à faire fondre le sous-sol et à ajouter ces débris à sa charge. I1 peut ainsi emporter ces débris i m m é diatement. Autrement dit, la première partie du processus d’érosion s’est déjà produite dans la “croûte de glace”. L a vitesse de l’érosion fluviale est donc beaucoup plus élevée dans ces régions qu’ailleurs. C c m m e il n’y a pas de roche en place capable de résister à l’éclatement dû au gel, les brisures intervenant dans le profil longitudinal sont beaucoup moins apparentes. I1 en résulte une formation remarquable de profils longitudinaux paraboliques et de vallées à fond large. L’ensemble du phénomène revêt aussi une grande importance pour la formation du relief aux latitudes moyennes, car les profils longitudinaux régularisés et le fond large des vallées de tous les cours d’eau depuis la région subpolaire jusqu’aux régions situées près de la zone subtropicale (450 N et 350 S) ont été formés par le gel et le pergélisol pendant les périodes glaciaires du pléistocène,
117
Soil movements under the influence of freezing Alfred Jahn
A Finnish explorer of soil-freezingprocesses, P.K o k k o -
in the Soviet Union, in the United States, a n d in
n e n (1926) wrote in one of his papers (as translated into English): “The freezing of the soil, due to the action of frost, can be massive or stratified. Massive or porous freezing is less harmful to plants than stratified freezing. During the processes of freezing, the soil rises upwards to the height of 10 to 22 per cent of the layer affected by frost. “Far m o r e noxious to vegetation appears to b e another property of soil freezing, i.e. the tendency to lift the objects within the soil to the height which is threefold or even fivefold that of the soil uplift d u e to freezing. This p h e n o m e n o n is of particular importance in springtime.” Further, the author goes o n to say that frost m o v e ments of the soil destroy the roots of plants and thus contribute to a ruin of vegetation cover. These effects of frost action are therefore ecologically significant. K o k k o n e n is one of those pioneer explorers w h o determined the methods of measurement of the soil m o v e m e n t s due to frost. T h e results of Kokkonen’s investigation were particularly useful for the cultivation of cereals in cold-zone countries. Kokkonen’s w o r k w a s well k n o w n to Stanislaw Bac, of Poland, w h o w a s carrying out extensive research (1931-1943)in frost m o v e m e n t s of the soil, and w a s investigating their significance for agriculture (Baranowski, 1966). B a c found that in cold weather, the soil level is lifted upwards several centimetres in Poland, i.e., in the temperate-climate zone. Foy his measurements B a c used a simple contrivance consisting of a fixed iron frame with movable rods a n d plates. T h e frame w a s stuck deep into the soil a n d the system of plates and rods m a d e it possible for him to measure soil m o v e m e n t s o n diffeFent levels. T h e investigations into the nature of the soil structure, its form a n d m o v e m e n t , as d u e to freezing, have a long history; they have developed particularly well
Sweden. It is superfluous to cite such familiar n a m e s as Sumgin, Cytowicz, Taber or Beskow. T h e results of their research w o r k are universally k n o w n . I have dwelt longer o n the w o r k undertaken by K o k k o n e n and Bac, because they investigated frost processes in soil o n account of their impact o n vegetation. F o r soil freezing is responsible for breaking plant roots. B a c also noted another interesting fact, namely that, owing to stratified freezing, s o m e plant roots are uplifted, and they are not heavy enough to return to their original position during the spring thaw, w h e n the ground subsides. T h e action of frost has, therefore, a triple effect o n the root system: roots are uplifted, broken, a n d displaced. T h e B a c motometer h a d proved to be a convenient device in our extensive measurement w o r k in different regions of Poland. W e have been carrying measurements in W r o c € a w (510 N., altitude 200 m), in the Sudetic Mountains (altitude 1,400 m), a n d in the arctic zone in Spitsbergen (780 N.).O u r observations extending over m a n y years indicate that the range of frost m o v e m e n t s of the soil surface in W r o c l a w amounts to s o m e millimetres. In the Sudetic M o u n tains, above the timber line, the m a x i m u m range of frost m o v e m e n t s c o m e s u p to 5.5 c m . In east Poland, in Puiawy, w h e n the winter is cold, the soil surface m o v e m e n t m a y reach the amplitude of 6.5 c m (Bac, 1952). In the Spitsbergen tundra, in 1957-1958, our measurements indicated that the range of the soil surface m o v e m e n t s w a s 15 c m . In s o m e places, in m o r e favourable conditions, the annual amplitude of the soil surface m o v e m e n t s m a y c o m e up to as m u c h as 0.5 m T h e latter figure amounts to a good m a n y times as m u c h as the range of the soil m o v e m e n t s in temperate climate zones. This fact accounts for the paramount importance of frost as a morphological a n d ecological factor in arctic regions.
119
A. Jahn
It is worth trying then to have a closer look at the work we had done in Spitsbergen (Czeppe, 1961; Jahn, 1961, 1963). Measurements were taken on the surface of the terrace with fully-developed patterned ground in the shape of sorted circles. A full year’s series of measurements displayed the following regularities: 1. The curve representing the movement is asymmetrical. In the course of autumn the soil is lifted successively, following repeated freeze-and-thaw cycles. In spring the soil subsides below its autumn level. An analogous instance of the curve is cited by Andrews (1963)w h o took measurements in a subarctic station, Schefferville,Labrador. 2. In a longitudinalprofile,soil movements are uneven: in the centre of a clay field the range of movements amounts to 15 c m while the lateral sections of the circle show a range of about 5 cm. It follows therefrom that the soil movement due to freezing is a variable process, depending on time and space. It is contingent upon the range of annual temperatures as well as varying horizontal conditions, chiefly hydrological. W e shall concentrate for a while on these two major parameters. Air temperaturein Spitsbergenin the year 1957-1958 varied from 70 C to -230 C. More important than the absolute value is a transition through the O0 C point. It has been computed that in the course of a single year the air near the soil surface reached the O0 C point 120 times, i.e., the number of gelivation cycles amounted to 60 (Czeppe, 1961; Baranowski, 1966). In one of my previous papers (Jahn,1963)I mentioned that the number of frost cycles in Spitsbergen m a y come up to 120. This is obviously the amount of temperature shifts through 00 C point. A freeze-and-thaw cycle implies a double transition through O0 C point, i.e., the number of freeze-and-thawcycles in Spitsbergen in 1957-1958amounted to 60. The number of gelivation cycles rapidly decreases as w e reach deeper into the soil: only 20 cycles were recorded at a depth of 10 c m , and at a depth of 50 c m differences in temperature practically disappear. Soil movements and the consequent upfreezing of stones, i.e.,the process of sorting out of materials, do not reach deeper than 30 cm. Thus a 30 c m soil horizon is most affected by frost processes. This horizon is almost free from stones which had been lifted to the surface. The problems of the number of gelivation cycles and of their impact on soil movements are still under discussion. In Schefferville, Labrador, only 30 gelivation cycles, i.e., half the number of those registered in Spitsbergen, were recorded in 1959-1960.Andrews believes that the process of soil movements is principally determined by a single cycle: “It would appear”, he writes, “that as far as this site is concerned, there is only one important freeze-thaw fluctuation, the annual cycle.” A similar view is held by Cook (1963; Cook and 120
Raiche,1962)w h o was watching the course of temperature fluctuations in soils of Resolute in Arctic Canada. Multicyclical temperature fluctuations around (30 C principally affect the upper surface of the soil (down to 10 cm), i.e., the layer which is most formative from the ecological standpoint. For this reason the problem demands closer attention, especially as the number of freeze-and-thawcycles varies considerably in North Canada and in the European section of the Arctic. The second major parameter of the soil movements consistsin spatialvariability. During a gelivation cycle not all of the soil surface is uplifted or subsequently subsides in a uniform way. Differences in movement depend on mechanical composition of the soil, on vegetation cover, and particularly, on various degrees of soil humidity. It is heat conductivity of soil materials as well as the actual amount of water, i.e., the water contained in the soil and the water coming from underneath by w a y of capillary suction,that determine the soil movement. In the profile described above w e could find major differentiation of the movement within a sorted circle where in the diameter of 2 m the differences in soil uplift amounted to 10 cm. Andrews’ measurements taken in the Schefferville station were surface measurements, and not profile measurements. H e could therefore do no more than draw a contour line m a p of his observation field where the range of movements was marked for the period extending from October to December. It can be seen from the m a p that the differences in soil movements amount to 6 cm. T h e soil movement w e are discussing is due to a vertical migration of water and its freezing in the form of segregation ice in the upper soil horizons. W e are faced here with one type of soil movement. It is of supreme importance from the ecological standpoint since it produces major changes in the soil structure. Thus in the soil, a whole network is developed of thin horizontal and vertical fissures which are subsequently filled with ice (ice layers and ice lenses). The formation of these structures is most detrimental to plant roots. The sorting out of materials is a simultaneous concomitant of the soil movement. This process is also harmful to soil stability.The upfreezing coarser stones break through a fine-grained soil mass. T h e shifting of materials in the soil makes it difficult for plants to take root. W e can easily find the places where the processes of the soil movement and sorting out are alive. In such places the soil surface presents bare spots completely stripped of vegetation cover. As soon as the processes of the soil movement and sorting out decline, there follows a rapid succession of plant growth. Patterned ground is soon furnished with a uniform vegetation cover. Apparently there exists an interdependencebetween the soil movement due to frost and the absence of vegetation cover. It is not only that the soil movement prevents the growth of vegetation: conversely,
Soil movements under the influence of freezing
a fully-developed vegetation cover can successfully protect the soil against the action of frost. That is why in places where there occurs a cover of turf and peat, only non-sortedfissure forms (tundra polygons) are found in the absence of patterned ground (sorted circles). Another type of soil swellings, those related to the so-called injection ice, are formed here. Contrary to the soil movements described above, injection ice causes an abrupt rise of a soil hummock. In this w a y are formed such forms as bugors with ice cores and the forms k n o w n in Canada and Alaska as pingos while in Siberia they are called bulkhunyakhy. An ice lens under turf, mostly at the fissure outlet, lifts upwards the surface soil horizon. This is a single movement, unlike the movements connected with segregation ice. The melting of an ice lens causes a negative movement, i.e. the sinking of the soil surface. A sinkhole form then results. Similar to the action of injection ice is the action of vein of ground ice. If an ice wedge is formed in a tundra fissure, then, while expanding, it can exert pressure upon the soil. Thus, owing to lateral pressure, the soil is lifted alongside of ice fissures to form ridges and swellings. Solifluction on slopes represents still another form of the soil movement. This phenomenon as a ecological factor deserves a separate treatment and therefore cannot be dealt with here more fully.It must, however, be observed that solifluction is related to the phenomena of segregated ice and vein ice formation in the soil. An additional factor is provided in gravitation which is responsible for the ice mass flowing d o w n the slope. A number of microrelief forms are due to solifluction-such as terraces and soil tongues,to mention but a couple. In an attempt to determine the ecological significance of the soil movements due to frost, w e shall divide them into two groups, the first comprising reiterated or multiple movements (rhythmical, seasonal), and the second, single movements. T o the first group belong the movements involving the active layer of permafrost or else the soil which is periodically frozen. T h e component members of the group are the autumn upheaving of the soil and its spring sinking. This kind of movement is related to the freezing of the water contained in the soil; it depends not so much on the actual amount of water but rather on its opportunities for migration (potential migration). Water migrates towards the freezing surface and the conditions of migration are determined by mechanical composition of the soil as well as the course of ice crystallization process. Crystallization implicates the upward suction of water together with an increase in the volume of the surface soil horizons. Such upheaving of the soil surface has reference to formation, within the soil, of segregated ice, or, to use a term proposed by P o p o w (1965) migration ice.
This fact must be emphasized because it has failed to receive due consideration. Different soil movements correspond to different types of ground ice. Rhythmical movements exist in places where the circumstances favour the formation of segregated ice. Parallel to, and simultaneous with segregated ice formation operates the process of sorting out the soil materials. The latter process is responsible for a peculiar soil structure as well as some microrelief forms. Rhythmical movements, i.e., annual upheaving and sinking of the soil surface, produce durable morphological effects. These are due to the circumstance that the autumn soil upheaving, cwing to freezing, is not altogether levelled during the spring thaw, for a counteracting force is provided in the internal friction of materials. Hence the convex forms (the elevations of thurfurs and bugors, cupola-like fields of sorted circles, etc.) stay for good. As a rule, they are dependent on segregated ice. T h e second type of soil movements are single vehement movements yielding lasting morphological effects. The movements are connected with the formation of ice lenses, hydrolaccoliths, pingos, and other allied forms are then developed. This type of movement corresponds to injection ice. A single negative movement takes the form of thermokarst processes. S o m e intermediate forms of soil movements, which can be assigned to either group,pesult from the action of the pressure due to freezing of free water in the soil. They manifest themselves on the soil surface as bulges (swellings) arising from the action of the water since they are bound, from above and from below, by the two layers of freezing water. In addition to bulging out of the soil surface,w e can also notice the breaking up of the upper soil horizons and pouring out of a muddy mass (tundra craters). A lateral expansion of ice wedges brings forth other forms of pressure: some ridges appear along the edges of ice wedges. The ridges can heave rhythmically or non-rhythmically. Soil movements connected with the formation of injection ice can be allied to rhythmicalmovements. It often happens that the surface of the bugors which have been formed on an ice lens is subject to rhythmical movements. Clear evidence is provided in the presence of segregated ice within the soil horizon which has been exposed to annual freeze and thaw. T h e above-mentioned types of soil movement in arctic and subarctic conditions m a y further be classified according to their impact on vegetation cover. Without doubt,rhythmical seasonalmovements appear to be most detrimental. They are destructive to vegetation because they constantly loosen (disjoint) the soil and produce cracks, in addition to the process of upfreezing of stones (mechanical sorting). In places where violent rhythmical movements occur there is no vegetation cover or else the existing cover fails to produce certain species.
121
A. Jahn
In principle, single m o v e m e n t s are not noxious to vegetation. Luxuriantly overgrown hillocks of palsen and pingos are of c o m m o n occurrence. Still, these m o v e m e n t s w e a k e n the soil (cracks) and frequently leave it unprotected to destructive action of other factors such as, for example, wind and water operating o n exposed hillocks. T h e problem of soil stability in the arctic and subarctic regions, in respect to soil dynamism, has so far
escaped full recognition. T h e present attempt has been intended to arrest attention to this important question which can apparently be solved by a combined effort of geomorphologists and botanists. It therefore seems desirable that m o r e care should in future be devoted to sponsoring specialized t e a m w o r k to deal with the tundra problem. T h e present s y m p o s i u m seems particularly well-suited to serve this purpose.
Résumé Les
mouvements du sol sous l’injluence du gel
(A.Jahn)
Les m o u v e m e n t s d u sol sont u n facteur important du milieu écologique. D a n s l’Arctique et les régions subarctiques, les m o u v e m e n t s provoqués par le gel et le dégel du sol ont u n e importance particulière. Là où ces m o u v e m e n t s ont lieu, la végétation fait défaut. Les sols structurés “vivants ”, en développement, les et donc les formes où la ségrégation sols polygonaux des matériaux minéraux et les fractures du sol sont liées a u m o u v e m e n t de ce sol -tous ces sols sont en principe privés de végétation. C e n’est qu’avec la disparition progressive de ce processus qu’on voit graduellement apparaître la végétation à la surface de ces sols. On distingue trois types principaux d e m o u v e m e n t du sol, dus a u x processus suivants: a) l’action directe de la température (gel et dégel); b) l’action de la pression du sol gelé sur le sol n o n gelé; c) l’action de la gravitation. L e m o u v e m e n t de la surface du sol dû à l’action de la température dépend d u n o m b r e de cycles de geldégel a u cours d’une année. L e plus important pour ce m o u v e m e n t est évidemment le cycle d’hiver de plusieurs mois, qui est, en principe, unique et atteint une épaisse couche de sol (la couche superficielle de pergélisol). Les cycles d’automne et de printemps, cycles d’un o u de plusieurs jours, atteignent une pro-
-
fondeur de quelques centimètres, une vingtaine tout au plus. Ces cycles peuvent être n o m b r e u x et atteindre m ê m e le n o m b r e de 120 a u cours d’une année (îles Spitzberg). L e n o m b r e des cycles d’automne dépasse celui des cycles de printemps. N o u s avons déjà réuni beaucoup de données, ce qui permet d’étudier plus précisément le phénomène. On a effectué de simples mesures du m o u v e m e n t d u sol dans divers pays: a u Spitzberg, a u Canada, en Europe centrale (Pologne), et cela nous autorise à estimer que les m o u v e m e n t s d u sol dus a u gel-dégel résultent n o n seulement du climat mais aussi des conditions locales des sols, telles que leur morphologie, leur géologie et leur hydrologie. P a r m i celles-ci les plus importantes sont les conditions hydrologiques. L’influence des m o u v e m e n t s d u sol sur la végétation se manifeste de d e u x façons: a) une action mécanique arrachement du $01, fractures, arrachement directe la quantité des m o u v e m e n t s (des des racines cycles) décidant de l’importance des destructions ; b) u n e action indirecte, entraînant le développement des cristaux de glace et u n e ségrégation du matériau du sol ( m o u v e m e n t des éléments plus volumineux vers la surface gelée). L a destruction de’la végétation qui en résulte est souvent liée à l’action intensivement destructive d u vent sur la végétation affaiblie par les m o u v e m e n t s du sol dus a u x cycles de gel-dégel.
-
-
Discussion L. AARIO.One of the most interesting things in Professor Jahn’s lecture was the thermokarst phenomenon. It would be interesting to hear, moreover, what evidence there is to show that even the large water-covered hollows are caused by the melting ice.
A. JAHN. Thermokarst phenomena are widely known in the southern part of the subarctic zone, i.e., in the taiga. I had
122
occasion to examine this type of area at Jakotsk (Siberia). This kind of morphology-large forms of thermokarstpredominates in that region where the large hollows, partially occupied by lakes, are here called alas. The thermokarst forms occur only in the areas where large masses of ground ice exist. The fluctuation of climate is the cause of the thermokarst processes.
Soil movements under the influence of freezing
Bibliography ANDREWS, J. T. 1963. The analysis of frost-heave data collected by B. H.J. Haywood from Schefferville, Labrador-Umgava, Canadian Geographer, vol. VII, no. 4. BAC, S.1952. Soil movements caused by acting offrost. (Pafist. Inst. Geolog. Biuletyn 66, Warszawa.) (In Polish with English summary.) BARANOWSKI, S. 1966. Termika tundry peryglacjalnej, S W Spitsbergen,Wrociaw. (In Polish, unpublished.) COOK, F. A. 1963. Patterned-ground research in Canada. Proceedings, Permafrost International Conference 1963, Lafayette, Indiana. Washington, D.C., National Academy of Sciences, National Research Council. ; RAICEE,V. G. 1962. Freeze-thaw cycles at Resolute, N W T . Geogr. Bull., no. 18.
__
/ Bibliographie CZEPPE, Z. 1961. Annual course of frost ground movements ai Hornsund (Spitsbergen) 1957-1958.Kraków, Zeszyty Naukowe Uniwersytetu Jagiellónskiego. (In Polish with English summary.) JAHN, A. 1961. Quantitative analysis of some periglacial processes in Spitsbergen.Warszawa, Wroclaw, Uniwersytet Wroclawski Zeszyty Naukowe. . 1963. Origin and development of patterned ground in Spitsbergen. Proceedings, Permafrost International Conference 1963, Lafayette, Indiana. Washington, D.C., National Academy of Sciences, National Research Council. KOKKONEN, P. 1926. Beobachtungen über die Struktur des Boden frostes, Acta Forest. Fennica, 33, Helsinki. POPOW,A. J. 1965. (Underground ice.) Moscow, Moscow State University. (In Russian.)
__
123
Complexité des notions de faciès morphologique arctique et subarctique (nord-ouestet centre-ouest du Groenland) Géographie boréale et anthropologie :fondements - physiques des notions de lieu de territoire J. Malaurie
COMPLEXITE DES NOTIONS DE FACIES MORPHOLOGIQUE ARCTIQUE ET SUBARCTIQUE (NORD-OUESTET CENTRE OUEST DU GROENLAND) Arctique sec, Subarctique humide : l’opposition ne peut être retenue sous cette forme simple,si l’on considère qu’en bordure de l’inlandsis ou d’une banquise plus ou moins crevassée, l’Arctique sec de la terre d’hglefield, sur la côte nord-ouest du Groenland, par exemple, offre ici, en bordure de l’inlandsis, et là, en bordure de la banquise, des faciès subarctiques humides. Les facteurs en jeu en géomorphologie boréale vent, hydrométrie, permafrost, etc. sont susceptibles de variations si sensibles dans un m ê m e espace latitudinal,dans u n m ê m e secteur, en une m ê m e roche, selon la topographie,l’exposition,l’épaisseur de neige, sa qualité, son degré de compacité (compte tenu du vent notamment), la nébulosité,le calendrier du dégel, l’altitude, la proximité d’une nappe d’eau, la profondeur de celle-ci,la forme de la pierre, la présence ou non de polissage, l’héritage cryologique du secteur, que l’on en vient à se demander si, dans l’état actuel des recherches, une expression moyenne à l’échelon d’une aire latitudinale, d’un secteur, n’a pas d’autre valeur que d’une grossière et première classification. Formes actuelles et vives, formes anciennes préglaciaires et fossiles mal définies paléoclimatiquement coexistent si intimement dans l’Arctique et le Subarctique qu’il n’est pas toujours possible de discerner ce qui revient à l’un et à l’autre. Seules les formes en matériel meuble sont,dans le nord-ouestdu Groenland, exclusivement postglaciaires. En roche résistante, en ce secteur à tout le moins,le polygénisme (préglaciaire, glaciaire et postglaciaire) du modelé n’est pas douteux. L’identité arctique vient seulement de la localisation géographique des formes. Et la définition d’un “sys-
-
-
tème arctique” est d’autant moins aisée en géomorphologie que l’observateurignore aussi bien l’essentiel de la structure profonde des processus en jeu que de la finalité morphologique des forces en présence. I1 n’y a, pour de nombreuses formes, pas de système mais une pluralité d’épisodes dont nous ignorons l’échelle et la perspective à partir desquelles ils deviennent cohérents. Compte tenu de la difficulté OU l’on se trouve pour décider, devant tel ou tel accident du terrain, de la prédominance ou non des agents arctiques ayant concouru à leur élaboration,il n’est pas étonnant que des spécificités abusives aient été accordées en certains cas par le paléomorphologue à des faciès arctiques ou subarctiques.A fortiori,certains attributs subarctiques se rencontrant dans les régions arctiques et vice-versa, ainsi que nous avons pu le voir, le maniement des notions retenues à première vue est particulièrement délicat. On en devine les conséquences néfastes pour le géologue ou le préhistorien aux latitudes plus tempérées. Les pingos, par exemple, avaient été reconnus c o m m e bien caractéristiques des pays à permafrost1. M. Pissart (1965)vient d’en signaler en Belgique, sur le plateau des Hautes-Fagnes.I1 a ainsi démontré qu’un permafrost n’était pas nécessaire, le bed-rock imperméable le remplaçant. Les conditions paléoclimatiques qui ont présidé à l’élaboration de ces formes étaient beaucoup moins rigoureuses que prévu. Les craquelures de gel (frost-cracking), qui avaient été considérées c o m m e spécifiques des pays de permafrost, ont été décrites par Washburn pendant les hivers 1939/40et 1958159 à Hanover, en N e w H a m p shire (Etats-Unis), par conséquent SOUS un climat tempéré. 1. <‘Lemilieu climatique accompagnant la formation d u pingo n’évoque point de doute. I1 paraít indispensable d’admettre l’existence d u pergélisol et de régime climatique rude et tree froid.” (Dnis, 1964, p. 61.)
125
J. Malaurie
Ou bien une v u e trop grossière est à l’origine de ces erreurs et une meilleure connaissance scientifique de ces phénomènes naturels les corrigera; ou bien et cette hypothèse n’est plus à écarter il y a une erreur fondamentale dans la perspective. N o u s n e connaissons pas, par ailleurs, avec assez de rigueur les seuils critiques des processus en cours, seuils qui, sous ces climats polaires de caractère paroxystique et discontinu, sont plus qu’ailleurs déterminants. D e s travaux comparés sur les processus, les corrélations en jeu, les divers seuils critiques de ces forces sont rares. D e s études interdisciplinaires recensant c o m m e préliminairement tous les seuils critiques géographiques, géologiques, physiques, des facteurs magnétiques, botaniques, zoologiques, hydrologiques, qui pourraient exercer un rôle majeur dans etc. la caractérisation d’un système géographique donné seraient é m i n e m m e n t souhaitables. Elles seraient précieuses : en mettant en valeur des contradictions, pour ce milieu encore m a l élaboré, elles en préciseraient les singularités; et c’est par là également, l’analyse d e ces singularités, q u e l’analyse morphologique pourrait, pensons-nous, devenir plus exacte. On rappellera brièvement, à titre d’exemple, certaines données concernant le nord-ouest (terre d’Inglefield) et le centre ouest (île de Disko) du Groenland. Signalons d’abord l’extrême brièveté des épisodes morphogéniques :huit à dix semaines a u plus de fonte dans l’année dans le nord-ouest du Groenland. L’écoule débit annuel lement principal, brutal, rapide débarrassant varie dans la proportion de 1 à 80 plateaux et versants de leur couverture neigeuse e n moins d’une semaine. E t il suffit que coïncident un printemps chaud, précoce et un vent marin du sud o u de l’ouest pour que l’érosion torrentielle n e soit opérante dans l’année, en terre d’hglefield, que pendant quelques jours. L a cartographie de la neige présente également de très grandes variations. Outre sa faible épaisseur, par sublimation, cette neige est soustraite à la fonte pour moitié en période de gel. L a nivométrie a u m o m e n t de la grande fonte est si variable que, le 18 juillet 1949, le taux d’enneigement moyen, a u sud du plateau littoral de la terre d’hglefield, est réduit à 19OlO de la surface a u sud, à 2‘310 a u nord, et ce sur un espace de moins de 200 km. L e calendrier de la fonte -jour par jour d’avril à juillet -est, compte tenu des variations propres de la couche active au-dessus du permafrost, à suivre avec u n e vive attention. J’ai défini en 1953 un relief de badlands sur matériau sableux a u sud de Disko. Dix degrés de latitude plus a u nord, les processus et les formes afférentes n’ont pas été retrouvés en dehors des talus sarqaq o u adrets, o u tournés vers le sud, enneigés parce que proches du littoral et d’une banquise très crevassée. L a grande fonte, pas assez abondante d’une part, trop étalée d’autre part, se développe avant m ê m e que le sol dégelé ait pris une épaisseur
-
-
-
-
126
-
suffisante pour être en mesure d’être enrayée. Un m ê m e système morphoclimatique, a p p a r e m m e n t s e m si l’on excepte les talus de blable, donne ainsi lieu du fait d’un petit décalage certaines expositions des calendriers et d’une modification d’intensité des agents en présence, à des processus fort différents: ici, ravinements et modelé torrentiel, là (en dehors des talus sarqaq proches d’un littoral de banquise crevassée), glissements et relief informe.
-
D e s précautions du m ê m e ordre, qui opposent une géographie latitudinale classificatoire (en un premier temps, certes, utile, mais nécessairement sommaire) à u n e géographie des processus tout à la fois plus ponctuelle et globale, doivent être observées pour ce qui concerne les éboulis stratifiés et les grèzes litées. Y v e s Guillien (1964) a insisté avec justesse sur la nécessité, pou1 le développement de ces formations cryoclastiques, de bancs de neige et de climats contrastés à saisons sèches prolongées. I1 présuppose une stabilité parfaite des talus, des écoulements brefs suivis de percolation, des t e m p s de dessication suivis et saisonniers. C’est définir le climat actuel nord-groenlandais. Or, c’est dans le centre ouest du Groenland et dans une formation n o n contemporaine, sur la côte méridionale d e l’île de Disko, que nous l’avons d‘abord repéré a u pied d’un profil concave évoluant vers un glacis rectiligne. I1 doit être, du fait d e sa position, antérieur à la régression marine en cours et contemporain du m a x i m u m cryoclastique du début de la déglaciation. C o m p t e tenu que nous ne l’avons retrouvé, sous le régime aride actuel du nord du Groenland, qu’au pied de certains versants sarqaq, il doit correspondre à un système morphoclimatique plus h u m i d e que celui donné par le climat actuel de Disko. D e m ê m e s observations indiquant que les faciès subarctiques peuvent être relevés dans le haut Arctique et vice versa pour des raisons de position par rapport à la mer, par rapport à une banquise plus o u moins crevassée, de topographie, de couverture végétale, d’exposition, seront données à propos des sols polygonaux (péninsule de Boothia et terre d’lnglefield), la cryoplanation (île Ellef Ringnes et terre d’lnglefield), les pentes critiques des éboulis basaltiques, gneissiques, calcaires et gréseux de Disko et de Thulé, la notion d’âge d’éboulis et d’éboulis superposés, les ravinements sur permafrost. Connaissance insuffisante des processus. N o u s voudrions en donner un dernier exemple à propos du processus le plus essentiel de l’Arctique, la gélifraction. J’insisterai particulièrement sur des expériences que j’ai faites en 1955, à la Station d u froid du CNRS (Paris). Elles ont été faites n o t a m m e n t sur des pierres d u Groenland. En résumant très rapidement certains de ces résultats, il sera présenté les observations suivantes : plus de dix alternances de refroidissement
Complexité des notions de facies morphologique. Fondements physiques des notions de territoire
(-L500) et de réchauffement (+900) étaient nécessaires pour réduire en m e n u s fragments u n cube de craie de 7 c m d’arête et saturé d’eau sous vide. D a n s les pierres de grandes dimensions étudiées (70 c m de diamètre), la roche, c o m p t e tenu des temps de propagation thermique dans le nord-ouest d u Groenland et des vitesses d’absorption d’eau, n’est altérable que superficiellement. D a n s des cubes calcaires définis de 7 c m et de 14 c m d’arête, le calcul des temps de propagation de la température préjuge de la possibilité, a u cours d’une m ê m e journée, de la répétition de cycles de réchauffement et de refroidissement, à condition que l’écart de température soit a u moins de 30 et 20°C et q u e les écarts se répètent c o m m e tels dans la journée. D’autre part, les roches à très fins canalicules, à l’intérieur desquelles les pressions sont particulièrement fortes, se trouvent relativement protégées. En effet, les très basses températures (-40oC) n’assurent- pas, théoriquement, le gel de l’eau occluse en surpression et, inversement, n’assurent pas le dégel de l’eau passée à l’état IV ou V de glace ( A m m a g a t et T a m m a n ) dans les canalicules. D e s études de cycles ont été assurées mais, si l’on ajoute que la température de congélation d’une soluet l’eau de la roche est rarement pure se tion situe, d’après des observations concordantes mais avec de grandes variables dues à la variété des solutions, a u x alentours de - 4oC, o n comprendra la variété des observations qu’on pourra être conduit, ici et là, à faire. I1 n’est pas dans nos intentions de nier l’existence de nombreuses roches gélivées à la surface de ces plateaux, d’autant que nous ignorons tout, dans la perspective d u “temps géologique”, des effets réels sur les processus des changements chronologiques et métriques. N o u s souhaitons seulement attirer l’attenprocessus princition sur le fait que la gélifraction palement desquamatoire - n’est probablement que l’aboutissement de n o m b r e u x processus à l’intérieur desquels intervient tout un complexe d’agents dont l’importance respective est encore ignorée. Les gels et dégels répétés n e feraient en s o m m e qu’achever une fragmentation préparée de longue date par de n o m b r e u x facteurs, pour certains encore inconnus.
-
-
-
Autant les processus physiques de la gélifraction prêtent encore à étude et controverse, autant les données microclimatiques qui président à la gélifraction et qui sont indispensables à connaître pour la reconstitution en laboratoire sont encore m a l connues. D’un point à un autre, les micro-conditions, c o m p t e tenu de l’exposition, de l’état du sol, des conditions thermiques, des qualités magnétiques de la roche, de son polissage, etc., peuvent varier considérablement. Un exemple de notre ignorance: des conclusions rapides de l’analyse des données météorologiques
-
-
presque toujours des stations littorales ont été tirées pour l’intelligence de la micromorphogénie des hauts-plateaux. Hâtives ; contrairement à ce qu’on observe à nos latitudes, il n’est pas rare de noter, en terre d’hglefield par exemple, u n e inversion de la température avec l’altitude. L a température, en certaines conditions synoptiques, croît d u littoral vers le s o m m e t d u plateau. Ainsi s’expliquent, à l’amont des rivières, la précocité - responsable n o t a m m e n t la grande densité d u des formes en U des vallées tapis végétal, le stationnement de la faune.
-
GROGRAPHIE ET ANTHROPOLOGIE : FONDEMENTS PHYSIQUES DES NOTIONS DE LIEU ET DE TERRITOIRE Sans cesse sollicité par l’archéologue, l’anthropologue, le sociologue et pas seulement pour des questions spécifiques et isolées de paléoclimat o u de distribution de phénomènes, le géographe, a u moins dans l’Arctique, ne répond guère par l’analyse globale attendue, mais par des introductions de “physiogéographie” s o m maire. L e site, le théâtre, la région, objets d’études passionnées pour l’archéologue et l’ethnographe, restent de la sorte, sur le plan géographique, des cadres inertes ne devenant jamais supports et explications. U n s y m p o s i u m récent à Yale en a donné la preuve. L e géographe n’a pour ainsi dire jamais été en mesure, a u moins à ces hautes latitudes, d’établir par des faits concrets la réalité, la globalité d’un itinéraire, d’un lieu, d’une région. O r la notion de territoire, si riche pour le zoologue en ethnologie animale, appelle de l’observateur u n e réponse concrète. L’unité d’un espace, telle que les bêtes (caribous, cétacés, poissons), les chasseurs s’y soient regroupés, accouplés et, a u moins pour les h o m m e s , maintenus pendant des générations et ce, dans les m ê m e s sites, les m ê m e s orientations, pratiquement les m ê m e s aires d’action, répond de facteurs corrélatifs, n o t a m m e n t de caractères telluriques et atmosphériques. C’est à b o n titre que les disciplines relevant des sciences humaines invitent, avec insistance, les disciplines relevant des sciences naturelles à un effort autrement vigoureux d’élaboration et de complication notionnelles. L a pauvreté d u vocabulaire en usage à cet égard en ethnologie - milieu, environnement, correspond à u n e forêt boréale, toundra, littoral pauvreté des concepts physiques. E t il s’agit moins d’idées pauvres que d’idées appauvries par la division d’un substrat géographique unique, global. Souhaiter que l’expérience géographique, écologique relève du m ê m e régime d’indivision méthodologique que l’expérience anthropologique marquerait un tournant et une date pour les sciences humaines.
-
127
J. Malaurie
Summary Complexity of the terms “Arctic” and “Subarctic” as notions of morphological aspect (North-west and Middlewest Greenland) (J. Malaurie)
F o r a definition of the Arctic and the Subarctic,
criteria of different branches of natural sciences should be taken into account in their complexity and interrelationships. A definition based only o n data from one limited field will never b e sufficient and can lead to unfortunate misunderstandings and mistakes.
Bibliography / Bibliographie BROWN, R.J. E. 1960. Arctic, 13, 163,1960. COOKE, F. A.; RAICHE, V. C. 1962. Freeze-thaw cycles at Resolute, N. W.T. Geog. Bull.,no 18,p. 64-78. DAVIES,W . E. Surface features of permafrost in arid areas. Physical geography of Greenland Symposium. Folia Geographica Danica, t. IX,p. 48-57. GUILLIEN,Y.1964. Grèzes litées et bancs de neige. Géologie en Mijnbouw, t. 45,no 3,p. 103-112. IVES, J. D. 1958. Mountain-top detritus and the extent of the last glaciation in northeastern Labrador-Ungava. Canadian Geographer,no 12,p. 25-31. MALAURIE,J. 1953.Le modelé cryonival des versants meubles de Skansen (Groenland). Bulletin de la Société Géologique de France, 1953,p. 707-721. . 1966a. Problèmes de géographie arctique en Amérique du Nord et au Groenland. (Bull. de l’Association des géographes français, janv.-fév.1966.) . 1966b. Thèmes de recherches géomorphologiques dans le nord-ouestdu Groenland.650 p., 79 pl.,200fig.,cartes h. t. (cc Mémoires et Documentsa, CNRS). . Igloulik, bassin de Foxe (Canada) :étude géographique
--
__
-_
128
et socio-économique. Paris, S E V P E N ,
13, rue du Four.
(A paraître.) __ ; NAT, D.1966. Préhistoire nord-américaine:observations sur u n symposium. Inter-Nord,p. 223-227,Paris, S E V P E N , 13,rue du Four. MATVEEV, N. N. (Évolution et âge des éboulis et des ‘‘m e r s de pierres” dans la zone à galets de l’Oural du Nord, d‘après l’exemple du massif de Denezkin kamen.) Problemi severa, no 7, p. 211-216.(En russe.) PISSART,A. 1965.Les pingos des Hautes-Fagnes:les problèmes de leur genèse.(Extrait des Annales de la Société géologique de Belgique, t. 88,1964-1965,Bull. no 5-6.) RAPP, A. 1957. Studien über Schutthalden in Lappland und auf Spitzbergen. Zeitschrift für Geomorphologie, Bd. I, Heft 2,p. 179-200. SAINT-ONGE, D. L a géomorphologie de l’île Ellef Ringnes. 46 p. (TNO,Ottawa, Canada, Etude géographique no 38.) SHVETSOV, P.F. 1959.(Pub. par) Principes de la géocryologie (permafrost). Moscou, Académie des Sciences de l’URSS. W A S H B U R N , A. L.;SMITH, D. A.; GODDARD, R. H. Frostcracking in a middle latitude climate. Biuletyn Peryglacjalny, no 12,p. 175-189.
Permafrost as an ecological factor in the Subarctic R. J. E.Brown
Permafrost is a natural phenomenon of subarctic regions. Its occurrence and distribution are determined by the close and complex interaction of a large number of climatic and terrain factors. In turn, these factors are influenced by the presence of permafrost. For this symposium it is useful to describe the distribution and occurrence of permafrost relative to the limits of the Subarctic. These limits are not fixed by a single universally accepted criterion but vary from one scientific discipline to another. The southern limit of the Subarctic is particularly debatable. In the consideration of permafrost, it is convenient to state arbitrarily that the southern limit of the Subarctic coincides with the southern limit of permafrost.A convenient criterion for the northern limit of the Subarctic would be the division between the discontinuous and continuous permafrost zones. M a n y scientific disciplines, however, accept the tree line, despite its debatable definition and location,as the northern limit of the Subarctic. Consequently,in this paper, the tree line is considered as the northern limit of the subarctic permafrost region although this region includes parts of the continuous permafrost zone.
metres thick. Thus the English term, climafrost, and the Russian,pereletok, are part of permafrost. At the other end of the scale, in the continuous zone,permafrost is thousandsof years old and hundreds of metres thick. The mode of formation cf such old and thick permafrost is identical to that of permafrost only one year old and a few centimetres thick. In the case of the former, even a small negative heat imbalance each year results in a thin layer being added annually to the permafrast. After several thousands of years have elapsed, this annually repeated process can produce a layer of permafrost hundreds of metres thick. This process does not cause the permafrost to increase in thickness indefinitely. Rather, a quasiequilibrium is reached whereby the downward penetration of frozen ground is balanced by heat from the unfrozen ground below. Permafrost is not permanent. This is particularly true in the Subarctic where changes in climate and terrain can cause the permafrast to thaw and disappear. Thus the English term, perennially frozen ground, and the Russian, mnogoletnemerzlyi grunt, are used to denote permafrast.
DEFINITION OF PERMAFROST
NATURE AND DISTRIBUTION OF PERMAFROST IN THE SUBARCTIC
Permafrost is defined exclusively on the basis of temperature, and refers to the thermal condition of earth materials such as soil and rock when their temperature remains below 00 C continuously for a number of years (Muller, 1945; Pihlainen and Johnston, 1963; Shvetsov, 1959; Sumgin et al., 1940). Permafrost includes ground that freezes in one winter, and remains frozen through the following summer and into the next winter. This is the minimum limit for the duration of permafrost; it m a y be only a few centi-
The Subarctic includes all of the discontinuous permafrost zone and the southern part of the continuous zone. The -50 C isotherm of mean annual ground temperature measured just below the zone of seasonal variation was chosen arbitrarily by Russian permafrost investigators as the division between the discontinuous and continuous permafrost zones (Bondarev, 1959). This criterion has been adopted in North America.
129 9
R. J.'E. Brown
DISCONTINUOUS PERMAFROST Z O N E
In the discontinuous permafrost zone, there are areas and layers of unfrozen ground. In the southern fringe of the Subarctic,permafrost occurs in scattered islands a few square metres to several hectares in area, and it is restricted to certain types of terrain, mainly peatlands. Other occurrence sare associated either with the north-facing slopes of east-west oriented valleys, or isolated patches in forested stream banks apparently associated with increased shading from summer thawing and reduced snow cover. Northward it becomes increasingly widespread and is associated with a greater variety of,terraintypes. It varies in thickness from a few centimetres or metres at the southern limit to several hundred metres at the boundary of the continuous zone-up to 250 m in eastern Siberia. Unfrozen layers or taliks m a y occur between layers of permafrost. The depth to the permafTost table is extremely variable ranging from about 50 c m to several metres. The active layer, which freezes in winter and thaws in summer, does not always extend to the permafrost table. The temperature of the permafrost at the level of zero annual amplitude ranges from a few tenths of a degree below O0 C at the southern limit to -50 C at the boundary of the continuous zone.
problem of the differences between slopes of different orientation.Permafrost is more widespread and thicker on north-facing slopes than o n south-facing slopes. S n o w cover, which has considerable influence on permafrost, is heavier on windward than leeward slopes. Other terrain factors also vary from one slope to another to complicate the distribution of permafrost. RELIC P E R M A F R O S T
Along the southern limit of permafrost the k n o w n occurrences seem to be in reasonable equilibrium with the present environment. To date no relic occurrences in North America south of the permafrost region which represent radically different conditions have been described. There are a few random reports, however, of isolated bodies of permafrost lying at depth beneath the ground surface. If such permafrost does exist, it probably formed during a previous period of cooler climate and lies at a depth below that affected by the current climate. There is no evidence on the ground surface of these bodies of permafrost and they would be detected only by mining operations or ground temperature measurements. Relic occurrences have possibly been found and described in the eastern hemisphere but no references have been found in the literature. Occurrences of relic permafrost at depth are k n o w n farther north in the permafrost region.
CONTINUOUS PERMAFROST ZONE C I R C U M P O L A R S O U T H E R N LIMIT O F P E R M A F R O S T
In the continuous zone,permafrost occurs everywhere beneath the ground surface except in newly deposited unconsolidated sediments where the climate has just begun to impose its influence on the ground thermal regime. The thickness of permafrost varies from about 250 m at the southern limit of the continuous zone to 500 m at the tree line in Siberia. In North America the range is probably about 100 to 300 m.The active layer generally varies in thickness from about 50 to 100 c m and usually extends to the permafrost table. The temperature of the permafrost at the level of zero annual amplitude ranges from -50 C in the south to about -100 C at the tree line in Siberia and to about -80 C in North America. P E R M A F R O S T DISTRIBUTION IN M O U N T A I N O U S R E GI O NS
Permafrost occurs at high altitudes in mo.untainous regions south of the subarctic region. It is found in the Western Cordillera in North America, and in the Alps and Himalayas in Eurasia. Its distribution varies with altitude as well as latitude. The lower altitudinal limit of permafrost decreases in elevation progressively from south to north. With increasing elevation the distribution of permafrost changes progressively from scattered islands to widespread and finally continuous. In mountainous regions there is the added
130
T h e most southerly extent of permafrost in the Western Hemisphere, excluding the Western Cordillera,is about 510-520 N. around and east of James Bay1 (Fig.1). West of Hudson B a y it extends north-westwards through the northern parts of the provinces and the south-westcorner of Y u k o n Territory. In Alaska the southern limit forms an arc around the Gulf of Alaska (Ferrians, 1965). Greenland lies entirely within the permafrost region. The existence of permafrost in Iceland is uncertain but probably occurs at high elevations in the interior. In Scandinavia the southern limit of permafrost lies at its most northerly location in the Northern Hemisphere (closeto 700N.).Eastward in the Kola Peninsula it extends south-eastwards. Farther east in the U.S.S.R. a tongue extends southwards in the Ural Mountains beyond which the trend is south-eastwardsto the Yenissey River. East of this river the southern limit extends below 500 N. in Manchuria and Outer Mongolia (Zhukov, 1961). Towards the Pacific Ocean it extends north-eastwards to the Sea of Okhotsk and across central Kamchatka (Batanov,1959).
1. R.J. E. Brown, Pernafrmr M a p of Canada, Division of Building Research. National Research Council, Canada and the Geological Survey of Canada. In press.
Permafrost as an ecological factor in the Subarctic
50°
O0
w
O
N
3Oo
6Oo
6Oo
30° O N
Oo E
50°
FIG.1. Northern Hemisphere permafrost distribution. 131
R.J. E.Brown
Table 1. Ecological aspects of permafrost in the Subarctic Ecological factor
Influence of ecologicalfactor o n permafrost
Influence of permafrost o n ecological factor
Climate
Air temperature and insolation influence ground thermal regime Energy-exchange regime at ground surface
Frost heaving in active layer, solifluction, and thermokarst change ground surface configuration which influences microclimate
Relief Microrelief
Degree and orientation of slope influences amount of insolationreceived at ground surface, and snowfall
Frost heaving in active layer causes uneven microrelief. Solifluction and thermokarst change ground surface configuration
Vegetation
Moss and peat insulate ground. Vegetation influences moisture regime. Variations in albedo control net radiation. Trees shade ground surface and intercept snowfall. Roughness of vegetation influences wind velocities
Impedes warming of soil: low temperatures in root zone impede root growth and cause physiological dryness. Impermeability of permafrost causes poor drainage and s w a m p conditions
Drainage
Standing and moving water thaw underlying permafrost. Moving water thermally erodes permafrost
Impermeable to water
Snow cover
Degree of insulation depends on duration, and properties
Retards melting at bottom of snow cover
Soil and rock
Influence depends on albedo, moisture content, thermal properties
Alternating freezing and thawing reduce soil particle size in active layer. Frost fissuring and heaving move large rock blocks
Microclimate
C I R C U M P O L A R N O R T H E R N LIMIT O F DISCONTINUOUS P E R M A F R O S T RELATIVE T O T R E E LINE
Throughout the Northern [Hemisphere,the northern limit of discontinuous permafrost generally lies south of the tree line (Fig.1). Exceptions to this occur in North America east of Hudson Bay, and in Eurasia from the Atlantic Ocean to the Yenissey River. The northern limit of discontinuous permafrost and the tree line are closer together in the Western Hemisphere than in the Eastern Hemisphere and are approximately parallel to each other. In Canada, the tree line is approximately 300 k m south of the continuous permafrost zone east of Hudson B a y and about the same distance north of it in western Canada. In Alaska, the tree line lies about 500 k m north of the discontinuous permafrost zone. In Greenland the division between discontinuousand continuous permafrost varies from about 660 N. to 690 N. in West Greenland and 680 N. to 690 N. in East Green1and.l Continuous permafrost has not been reported in northern Scandinavia and the tree line actually lies close to the southern limit of discontinuous permafrost. In the Soviet Union, the northern limit of discontinuous permafrost extends inland from the Arctic coast at Novaya Zemlya, south-eastwardsacross the Ob Gulf at 700 N.,and reaches 600 N. east of Yakutsk, more than 1,000 k m south of the ttee line. Farther east it trends to the north-eastand reaches the
132
Bering Sea at Anadyr. In contrast to the great southward penetration of the northern limit of discontinuous permafrost in Siberia,the tree line extends eastwards at about 700 N.to the far east where it dips southwards to Kamchatka Peninsula. I N T E R A C T I O N OF P E R M A F R O S T A N D E N V I R O N M E N TA L F A C T O R S
Permafrost in the Subarctic exists in a close and complex interaction with a large number of ecological factors (Table 1). The most important factor is climate which is basic to the formation and existence ofpermafrost and controls the broad pattern of distribution and occurrence. Terrain conditions are responsible for local variations within this broad pattern.
Climate The influence of climate on permafrost is most readily expressed by the temperatwe of the air. This parameter is easily measured and most directly related to ground heat losses and heat gains. The complex energy exchange regime at the ground surface and the snow cover cause the mean annual ground temperature, measured at the level of zero annual amplitude,to be several degrees warmer than the mean annual air temperature. Local microclimates and terrain con1. A. Weidick. private communication.
Permafrost as an ecological factor in the Subarctic
ditions cause minor variations but a value between 30 C and 40 C can be used as an average figure (R.J. E.Brown, 1963a). Present knowledge of the southern limit of permafrost indicates that it coincides roughly with the -10 C mean annual air isotherm. Southward, permafrost occurrences are rare and small in size because the climate is too warm. They are related to unusual insulation conditions caused by thick dry peat or low insolation where summer cloud cover is unusually heavy such as around the Gulf of Alaska. Between the -lo C and -3.50 C mean annual air isotherms, permafrost is restricted mainly to the drier portions of peatlands because of the low thermal conductivity of the peat. Scattered bodies of permafrost occur also on some north-facing slopes and in some heavily shaded areas. In the vicinity of the -3.50 C mean annual air isotherm, the difference of 3.50 C between the mean annual air and ground temperatures produces a mean annual ground temperature of a fraction of a degree below Oo C in most types of terrain. Northward to the boundary of the continuous zone, permafrost becomes increasingly widespread and thicker and the mean annual ground temperature at the level of zero annual amplitude decreases to -50 C. This corresponds to a mean annual air temperature of about -8.50 C. Northward to the tree line, permafrost is continuous and increases in thickness in response to the progressive reduction in the mean annual air temperature. Over a long period of time, a change in the mean annual air temperature can result in a signscant change in the extent and thickness of permafrost. Geothermal gradients in permafrost ranging from about 10 C/20 m to lo C/160 m-depending to some degree on the type of soil or rock-have been observed in Canada, Alaska, and the U.S.S.R.(Brewer, 1958; W.G.Brown et al., 1964; Melnikov, 1959). A change of lo C in the mean annual air temperature can result, over a long period of time, in a change of lo C in the mean annual ground temperature. This would cause a change in permafrost thickness of approximately
20-160 m. Microclimatic factors are also important in influenc-
height which affect air movement at the ground surface. Resulting variations in degree and orientation of slope are small but m a y be sufficient to cause variations in the quantities of insolation received at the ground surface. The depth to the permafrost table influences the types of plants that can grow which thus affect the climate near the ground-air movement, net radiation, and other energy components.
Terrain In the discontinuous zone, variations in terrain conditions are responsible for the patchy occurrence of permafrost, size of permafrost islands, depth to the permafrost table, and thickness of permafrost (Legget et al., 1961) (Fig. 2 to 7). In the continuous zone, the thermal properties of the peat and other terrain factors assume a relatively minor role and the thermal properties of the ground as a whole,together with the climate, become dominant, Terrain factors that affect permafrost conditions include relief,vegetation, drainage, snow cover, soil type, and glacier ice. In its turn,permafrost influencesthese featuresin the subarctic environment.
Relief. Relief influences the amount of solar radiation received by the ground surface. The influence of orientation and degree of slope on permafrost distribution is particularly evident in mountainous regions but smaller scale variations cause similar situations elsewhere in the Subarctic. Permafrost m a y occur, for example,in the north-facing bank of even a small stream but not in the opposite south-facing bank. Similar differences between north- and south-facing slopes can occur even in areas of intensive microrelief such as peat mounds, ridges and plateaux. Permafrost affects the relief or ground surface configuration. Solifluction and other down-slope mass movements of earth material over permafrost surfaces alter the ground surface configuration. Thermokarst is frequently an active process in areas where large masses of ground ice exist. The melting of this ice and resulting subsidenceof the ground produce undulations and hollows which considerably alter the relief.
ing the distribution of permafrost. Net radiation, evaporation (including evapotranspiration), condensation, and conduction-convectionare all elements of the energy-exchange regime at the ground surface (R.J. E. Brown, 1965). Although they are climatic in origin, their contribution to the ground thermal regime is determined by the nature of the ground surface and thus can be considered as terrain factors. Although climate has a profund influence on permafrost, there is little or no direct influence exerted by permafrost on the climate. Indirectly, permafrost m a y influence the microclimate by modifying the ground surface configuration.Frost action in the active layer can produce microrelief features 1 m or more in
Vegetation. Vegetation and permafrost are closely related in the subarctic environment and exert considerable influence on each other (R.J. E. Brown, 1963b; Tyrtikov, 1959). The most obvious effect of vegetation is its role of shielding the permafrost from solar heat. This protection is provided mainly by the insulating properties of the widespread moss and peat cover. Removal or even disturbance of this surface cover causes degradation of the underlyingpermafrost. In the discontinuous zone,this m a y result in the disappearance of bodies of permafrost. In the continuous zone,the permafrost tablewillbe lowered. The predominance ofthe moss and peat in protecting thepermafrost 133
R.J. E.Brown
PEAT PLATEAU IM HIGH 8 20-30M
7SPRUCE-SPHAGNUM AREA
STAGNANT WATER IN HOLLOWS BETWEEN PEAT HUMMOCKS. PERMAFROST IN HUMMOCKS
-- .-
. ... ..
FIG.2. Profile through typical peatland in southern fringe of discontinous zone showing interaction of permafrost and terrain factors. from atmosphericheat is demonstrated by the factthat little change occurs in the depth of the permafrost table when trees and brush are removed provided that the moss and peat are not disturbed. A fire m a y burn trees, brush, and even the surface of the moss without altering the underlying permafrost. The mechanism that causes permafrost to form in peatlands in the southern fringe of the permafrost region south of the -3.50 C mean annual air isotherm appears to be associated with variations through the year of the insulating properties of the moss and peat (Tyrtikov, 1959). During the w a r m season a thin surface layer of peat becomes dry. Its thermal conductivity is low and warming of the underlying soil is impeded. During the cold part of the year the peat becomes saturated from the surface and then freezes, greatly increasing its thermal conductivity. Consequently,the amount of heat transferred in winter from the ground to the atmosphere through the frozen icesaturated peat is greater than the amount transmitted in summer in the opposite direction through the surface layer of dry peat. A considerable amount of heat is also required during the w a r m season to melt the ice and to w a r m and evaporate the water. The net result is a negative imbalance of heat and conditions conducive to the formation and preservation of permafrost. Although the influence of the ground vegetation on permafrost is dominant, trees are of some importance. They shade the ground from solar radiation and intercept some cf the snowfall in winter. The effect of even a single tree in shading the ground in summer and reducing the snow cover at its base in winter appears to influence the heat exchange at the ground surface sufficiently to produce a lens of permafrost beneath the tree (Viereck, 1965). The density and height of trees influence the microclimatic effects of ground-
134
surface wind velocities. Wind speeds are lower in arcas of dense growth than in areas where trees are sparse or absent. T h e movement of air represents the transfer of heat from one area to another. In peatlands the trees are generally stunted and scattered, and there are numerous open areas that permit higher wind speeds and the removal ofheat per unit time.Therefore, the possibility of slightly lower air temperatures and ground temperatures,because of higher wind speeds is greater than in areas of dense tree growth (Johnston et al., 1963). Permafrost exerts considerable influence on the environment in which the subsurface organs of plants have developed (Tyrtikov,1959).The effects of permafrost are mostly detrimental to plant development because of its cold temperatures and impermeability to moisture. Permafrost impedes warming of the soil during the growing season and the temperature of the root zone is considerably below the optimum. Absorption of water by the roots is reduced which leads to physiological dryness of the plants. On the other hand, water is gradually released to the root zone through the summer as the active layer thaws. Root development is retarded and roots are forced to grow laterally because downward penetration is prevented by the permafrost. Large trees cannot be supported by these shallow root systems and a “drunken” forest results. Permafrost forms an impermeable layer which impedes drainage leading to a decline in aeration and impoverishment of nutritive substancesbecause of the weakening of the activity of micro-organisms. Soil movement in the active layer also influences the vegetation. Frost action causes unevenness in the ground surface, solifluction and other down-slopemass movements of earth material disturb the vegetation, and thermokarst changes the surface configuration of
Permafrost as an ecological factor in the Subarctic
FIG. 3. Palsas containing permafrost in uret peatland with no permafrost, located at southern limit of discontinuous zone near south end of James Bay, Canada.
the ground, all producing detrimental influences on the vegetation. These various influences of permafrost on vegetation increase northwards as permafrost becomes increasingly widespread and the active layer becomes thinner.
Drainage. Water greatly influences the distribution and thermal regime of permafrost.Inthe discontinuous zone,the existence of permafrost is inhibited in poorly drained areas. Precipitation influences the depth of thaw and soil temperatures (Shvetsov and Zaporozhtseva, 1963). First, the amount of moisture in the soil immediatelybefore it freezesin the autumn determines the ice content and depth of thaw the following summer. Second, the moisture content of the soil surface and the infiltration of atmospheric water influence the heat transfer to the frozen soil during the thaw period. Moving water is an effective erosive agent of perennially frozen soils. There is almost always an unfrozen zone beneath water bodies that do not freeze to the bottom. The extent of this thawed zone varies with a large number of factors-area and depth of the water body, water temperature, thickness of winter ice and snow cover, general hydrology, and composition and history of accumulation of bottom sediments (Johnston and Brown, 1964). The ocean has an important thermal influence on permafrost causing it to be thinner at the shore than inland (Lachenbruch,1957). Permafrost greatly influences the hydrological regime, Its impermeability to water is responsible for the existence of m a n y small shallow lakes and ponds in the continuous zone and in the northern part of the discontinuous zone where permafrost is widespread. Beaded streams are another indication of the influence of permafrost. Irregular enlargements of stream channels result from the melting of masses of ground ice beneath streams.
FIG.4.Sectionof Royal Canadian Air Force aerialphotograph A 14975-31of terrain in Figure 3.
135
FIG. 5. Permafrost exists in forested peat plateau in background but not in low wet area in foreground. Location in discontinuous permafrost zone near Nelson River, northern Manitoba (Canada).
FIG 6. Aerial view from alt.itude of 150 m of terrain in Figurei 5 show,ing forested peat plateaux with- Pe:rma€r:Ost and low wet treeless are'as without perinafrost.
FIG. 7. Section of Royal Canadiain Air force aerial photograph A 14188-119 of terrain in Figure 5. Light grey i.ough pattern denotes forested peat plaiteaux with permafrost. Medium to dark grey smoother pattern denotes low wet areas without permafrost.
Permafrost a.s an ecological factor in the Subarctic
Snow cover. S n o w cover influences the heat transfer between the air and the ground and hence affects the distribution of permafrost. The snowfall regime and the length of time that snow lies on the ground are criticai factors. A heavy fall of snow in the autumn and early winter inhibits winter frost penetration. On the other hand, a thick snow cover that persists on the ground in the spring delays thawing of the underlying ground. The relation between these two situations determines the net effect of snow cover on the ground thermal regime. In the discontinuous zone, particularly in the southern fringe,it can be a critical factor in the formation and existence of permafrost. In both Western and Eastern Hemispheres,the thickest permafrost in the southern fringe of the permafrost region occurs in palsas on which snow cover is thin because of their exposure to wind (Lindqvist and Mattsson, 1965; Sjörs, 1959). In the continuous zone,snow cover influences the thickness of the active layer. T h e considerable influence of snow cover on the ground thermal regime can best be illustrated by several quantitative examples. At Norilsk, U.S.S.R. (mean annual air temperature -8.40 C), it was shown that a +now cover exceeding 1.5 m completely damped out air temperature influences on heat emission from the ground (Shamshura, 1959). Studies at Schefferville, Canada (mean annual air temperature about -50 C), showed that snow cover is a dominant factor in controlling Permafrost distribution at that site. Variations in snow cover cause temperature variations greater than those resulting from vegetation cover. It has been postulated that a snow depth of about 40 c m can be regarded as the critical snow depth for permafrost to survive. Beneath a greater depth either there is no permafrost or a degrading condition prevails (Annersten, 19ó4). There is little if any signifcant effect of permafrost on the snow cover. T h e low ground temperatures m a y retard melting at the bottom of the snow cover but this possibility does not appear to exert m u c h influence.
Bare soil and rock. Bare soil and rock have considerable influenceon the temperatureof the groundbecause of their ability to reflect solar radiation. Reflectivity values in the range of 12-15 per cent for rock and 15-30per cent for tilled soil have been observed. There will also be different evaporation rates and intakes of precipitation. Variations in thermal properties such as conductivity,diffusivity,and specific heat affect the rate of permafrost accumulation.The thermal conductivity of silt, for example, is about one-halfthat of coarse-grainedsoils and several times less than that of rock. These factors assume their greatest significance in the northern part of the Subarctic where the climate is sufficiently cool to produce permafrost regardless of the type of terrain (Brown and Johnston,
1964).
The influence of permafrost on soil and rock is manifested by such phenomena as clay boils, solifluction, and other down-slope mass movements of earth material. These movements and frost action in the active layer tend to break d o w n coarse soil particles and rock fragmentsinto fine-grainedmaterial. Intensivefrost action is also responsiblefor the heaving of massive blocks of fractured bed-rock. Glacier ice. The growth and regime of glaciers and ice caps is determined by climate but ice is considered as a terrain factor in this paper because like vegetation, water, and snow, it forms a layer on the ground surface between the permafrost and the atmosphere which affects the heat exchange between them. It has been postulated that the bottom temperaturebeneath m u c h of a continental ice sheet is below O0 C. In temperate glacier conditions, the ice bottom temperature is at the pressure melting point. Beneath an ice sheet 2,000 m thick, for example, the temperature of the water film at the bottom of the ice would be about -lo C. In polar glacier conditions,the bottom of the ice is frozen to the underlying ground and the temperature at the ice-groundinterface is below 00 C. Both of these glacier conditions probably occur extensively throughout an ice sheet. Consequently, beneath continental ice sheets in the Northern Hemisphere permafrost was probably widespread but thin because of the proximity of bottom temperatures to O0 C. Permafrost m a y have been somewhat thickerbeneath the margins of ice masses where the effect of cold air temperatures can penetrate to the underlying ground (Shumskiy, 1964). After the ice retreated, permafrost in areas covered by post-glacial inundations was probably dissipated and would not have re-formed until these bodies of water receded several thousand years later. In contrast to areas covered by ice sheets, m u c h colder temperatures were imposed by the periglacial climate on ice-freeareas producing thicker and colder permafrost. Thus, continental glaciation has undoubtedly exerted great inffuence on permafrost conditions from one region of the Subarctic to another. It is notable that ice sheetswere m u c h more extensive in the Western Hemisphere than the Eestern Hemisphere (see Figure 1). All of subarctic North America, excluding western Yukon Territory and central Alaska, were covered with ice sheets during the Pleistocene (Flint, 1957). In contrast,a large part of central and eastern Siberia was not covered and here continuous permafrost n o w extends more than 1,000km south of the tree line, and the thickest permafrost in the Northern Hemisphere has been recorded.
CONCLUSION M a n y fluctuations have occurred through time in the extent, thickness, and temperature of the permafrost
137
R.J. E.Brown
in response to changes in climate and terrain. Since its initial formation,the permafrost in any area m a y have dissipated and re-formedseveral times during periods of climatic warming and coaling. Glacial history has had a marked effect. Changes in vegetation caused by fire,climatic succession,encroachment in water basins, or by the permafrost itself all have pronounced local effects. The regime of the fall and accumulation Gf snow influences the ground thermal regime. T h e geothermal gradient also affects the ground thermal regime. It varies in different types of sail androck, with changes in geologicalstructure and with time. Thus the environment in which permafrost exists is a complex dynamic system,the product of past and
present climate and terrain features,which are in turn influenced by the permafrost. The thermal sensitivity of permafrost is such that even small changes in climate and/or terrain will produce changes in the extent, thickness, and temperature of the permafrost. The interactions of permafrost and other factors inthe Subarctic are varied and very complex. Even a slight change in one factor produces a change in one or several other factors. This paper is a contribution from the Division of
Building Research,National Research Council,Canada, and is published with the approval of the Director of the Division.
Résumé Le pergélisol,facteur (R.J. E.Brown)
écologique de la région subarctique
Dans la région subarctique, le pergélisol est en corrélation étroite et complexe avec de nombreux facteurs climatiques et morphologiques. L a région subarctique est considérée c o m m e comprenant taute la zone discontinue de pergélisol et la partie de la zone continue située au sud de la limite de la forêt. On admet arbitrairement que les deux zones de pergélisol sont séparées par l’isotherme -50 C de la température moyenne annuelle du sol mesurée immédiatement au-dessous de la profondeur où se font sentir les variations saisonnières. L’extension du pergélisol augmente progressivement du sud au nord de la région subarctique, ainsi qu’avec l’altitude dans les secteurs montagneux.
On a constaté l’existence de zones fossiles de pergélisol en profondeur dans la région du pergélisol; il est possible qu’il en existe également:au sud de cette région. Dans l’hémisphère nord, le pergélisol détend plus au sud et les conditions qui y règnent sont plus rigoureuses en Eurasie qu’en Amérique du Nord. L e climat est le plus important des facteurs qui influent sur la formation et l’existence du pergélisol et qui déterminent les formes générales de son apparition et de sa répartition. Les conditions du terrain,c o m m e le relief, la végétation, l’écoulement des eaux, l’enneigement, le type de sol et la glaciation déterminent les variations locales dans ce cadre général. L e pergélisol influe de son côté sur ces facteurs en milieu subarctique. Ce système dynamique peut évoluer avec le temps,une légère modification de l’un des facteurs entraînant une modification d’un ou de plusieurs autres facteurs.
Discussion F. E. ECKARDT. Vous avez montré un nombre impressionnant d‘exemples de permagel dans des paysages extrêmement variés, et il semble à première vue très difficile d’établir des corrélations entre l’existence de ce phenomène et l’action des facteurs physiques et biologiques qui en sont la cause,le nombre de ces facteursétant très élevé. Pourriezvous m e dire si l’on a déjà essayé d’établir de telles corrélations au moyen de techniquesde calcul modernes ? R.J. E.BROWN.Yes, there are correlations between climatic and terrain factors on the one hand, and permafrost on the other hand, which can be used to predict with considerable confidence the existence and distribution of permafrost. . Many of these correlations are discussed in m y paper. \
138
However, much work remains to be done. There is a great need for conducting detailed and complete energy balance measurements in permafrost areas. In the discontinuous zone, it is important to know why permafrost occurs in one area and not in au adjacent area. This situation is due to variations in the thermal contributions of the various environmental components to the permafrost.The problem then arises of precisely what measurements to make and what instrumentation to use. Some studies of this nature have been carried out in North America and the Soviet Union.
A. JAHN. The problem of thickness of permafrost: both Dr. Brown and Dr. Péwé called the very shallow,very thin
Permafrost as an ecological factor in the Subarctic
and extends downward. Let us suppose n o w that thawing continues into the autumn and stops without thawing a few centimetres (perhaps 5 or 10) at the bottom of the frozen layer. At this time, the seasonal frost begins again to penetrate downward from the surface of the ground. The thin layer which persisted through the summer is permafrost, or perennially frozen ground, because it persisted through one year. Perhaps it will thaw during the next summer but while it existed, it was considered as permafrost. At the other extreme, there is the permafrost which persists through thousands of consecutive summers.
(20-30cm) layer of frozen ground “permafrost”. Is it right to call such a layer permafrost? This m a y be seasonal frozen ground, rather than permafrost.
R.J. E. BROWN.In m y paper I defined “permafrost” and stated the minimum and m a x i m u m lengths of time involved. The minimum length of time required for frozen ground to persist in order that we m a y call it permafrost is one year. That is to say, let us consider a situation where seasonal frost begins to form in the late autumn and penetrates to a depth of, say, 1 m through the winter. During the following summer, thawing of the ground proceeds from the surface
Bibliography
/ Bibliographie
ANNERSTEN, L.J. 1964. Investigations of permafrost in the vicinity of K n o b Lake 1961-62. In: J. B. Bird (ed.), Permafrost studies in central Labrador-Ungava. p. 51-137. (McGill Sub-Arctic Research Papers, no. 16.) BARANOV, I. Ya. 1959. Geograficheskoye rasprostraneniye sezonnomerzayushchikh pochv i mnogoletnemërzlykh gornykh porod (Geographical distribution of seasonally frozen ground and permafrost). In: P. F. Shvetsov (ed.). Osnovy geokriologii (Principles of geocryology), vol. I, p. 193-219.Moscow, Academy of Sciences of the U.S.S.R. (Translated into English by the National Research Council of Canada; issued as NRC Tech. Translation No. 1121.) BONDAREV, P. D. 1959. Obshchaya inzhenerno-geokriologicheskaya otsenka oblasti mnogoletnemërzlykh gornykh porod predelakh SSSR i metody stroitel’stva na nikh (A general engineering-geocryologicalsurvey of the permafrost regions of the U.S.S.R. and methods of construction in permafrost areas). Problemy severa, no. 3, p. 24-50. (Translated into English by the National Research Council of Canada.) BREWER, M. C. 1958. Some results of geothermal investigations of permafrost in northern Alaska. Trans. Amer. Geophys. Uni.,vol. 39,no. 1,p. 19-26. BROWN, R. J. E. 1963a. The relation between meun annual air and ground temperatures in the permafrost region of Canada. Presented at the International Conference on Permafrost,Pnrdue University, November 1963. . 19636. The influence of vegetation on permafrost. Presented at the International Conference on Permafrost, Purdue University, November 1963. . 1965.Some observations on the influence of climate and terrain features on permafrost at Norman Wells, N.W.T., Canada. Can. J. Earth Sci.,vol. 2, p. 15-31. ; JOHNSTON, G. H.1964. Permafrost and related engineering problems. Endeavour, vol. XXIII, no. 89,p. 66-72, BROWN, W.G.;JOHNSTON, G. H.;BROWN, R. J. E. 1964. Comparison of observed and calculated ground temperatures with permafrost distribution under a northern lake. Canad. Geotech. J.,vol. 1, no. 3, p. 147-154. FERRIANS,O. J., Jr. 1965. Permafrost m a p of Alaska. (U.S. Geol. Survey Misc. Geological Investigations, M a p 1-445.) FLINT, R. F. 1957. Glacial and Pleistocene geology. New York, John Wiley &Sons Inc., 553 p.
__
__
__
JOHNSTON, G.H.; BROWN, R.J. E. 1964.S o m e observations on permafrost distribution at a lake in the Mackenzie Delta, N.W.T., Canada. Arctic, vol. 17, no. 3, p. 162-175. --. ; PICKERSGILL,D. N. 1963. Permafrost investigations at Thompson, Manitoba. 51 p. (National Research Council of Canada, Division of Building Research Tech. Paper 158 (NRC 7568).) LACHENBRUCH, A. H. 1957. Thermal effects of the oceans on permafrost. Bull. Geol. Soc. Amer., vol. 68, no. 11, p. 1515-1530. LEGGET, R.F.; DICKENS, H.B.; BROWN, R.J. E. 1961. Permafrost investigationsin Canada. Geology of the Arctic, vol. II, p. 956-969. LINDQVIST, S.; MATTSSON, J. O. 1965. Studies on the thermal structure of a pals. Lund studies in geography, Ser. A., Physical Geog. no. 34, p. 38-49. MELNIKOV, R. I. 1959. O zakonomernostyakh rasprostraneniya i razvitiya merzlykh pochv i gornykh porod v basseyne r. Leny (Mechanisms of distribution and development of frozen soil and rock in the Lena river basin). Mezhduvedomstvennoye Soveshchaniye Po Merzlotovedeniyu, 7th (Seventh Interdepartmental Permafrost Conference), p. 91-102.Moscow, Academy of Sciences of the U.S.S.R. MULLER, S. W . 1945.Permafrost or permanently frozen ground and related engineering problems. 231 p. (Strategic Engineering Study no. 62, U.S. Army, Washington, D.C.) PHILAINEN, J. A.; JOHNSTON, G. H.1963. Guide to a Jield description of permafrost. 23 p. (National Research Council of Canada. Assoc. Committee on Soil and Snow Mechs., Tech. Mem. 79. (NRC 7576).) SHAMSHURA, G. Ya. 1959. Vliyaniye snezbnogo pokrova na teplovoy rezhim gruntov v Taymyrskoy tundre (Influence of snow cover on the thermal regime of the ground in the Taymyr tundra). Meshduvedomstvennoye Soveshchaniye P o Merzlotovedeniyu, 7th (Seventh Interdepartmental Permafrost Conference), p. 186-201. Moscow, Academy of Sciences of the U.S.S.R. SHUMSKIY, P. A. 1964. Principles of structural glaciology. New York, Dover PubIications Inc. 497 p. (Translated from the Russian by David Kraus.) SHVETSOV, P. F. (ed.) 1959. Osnovy geokriologii (Principles of geocryology), vol. I, 459 p. Moscow, Academy of Sciences of the U.S.S.R.
__
139
R. J. E.Brown
_- ; ZAPOROZHTSEVA, I. V.
1963. Povtoryayemot’i inzhenerno-geokriologicheskoye znacheniye dvukhtrekhletnikh povy sheniy temperatury gruntov v Subarktike (The recurrent nature and permafrost engineering significance of two to three year soil temperature increases in rhe Subarctic). Problemy severa, no. 7, p. 22-45.(Translated) into English by the National Research Council of Canada. SJÖRS, H. 1959. Forest and peatlands at Hawley Lake, northern Ontario. Contr. Bot., Nat. Mus. Can. Bull., 171, p. 1-31. SUMGIN, M. I.; KACHURIN, S. P.; TOLSTIKHIN, N. I.; TUMEL, V. F. 1940. Obshcheye merzlotouedeniye (General permafrwt studies). Moscow, Academy of Sciences of the U.S.S.R. 337 p.
140
TYF~TIKOV, A. P. 1959. Mnogoletnemerzlyye porody i rastitel’nost (Perennially frozen ground and vegetation). In : P. F. Shvetsov (ed.) Osnouy geokriologii (Principles of geocryology), vol I, p. 399-421. Moscow, Academy of Sciences of the U.S.S.R.(Translated into English by the National Research Council of Canada; issued as NRC Tech. Translation no. 1163.) VIERECK, L. A. 1965. Relationship of white spruce to lenses of perennially frozen ground, Mount McKinley National Park, Alaqka. Arctic. vol. 18, no. 4, p. 262-267. ZHUKOV, V. F. 1961. Sezonnoye i mnogoletnee promerzaniya gruntov v Mongol’skoy Narodnoy Respublike (Seasonally add perennially frozen ground in the Mongolian Peoples Republic). Izv. Alcad. Nauk SSSR, Ser. Geogr. no. 2, p. 61-69.
Permafrost and vegetation on flood-plains of subarctic rivers (Alaska): a summary Troy L. Péwé
Flood-plains of major rivers in the Subarctic can be divided into several clearly marked phases based mostly on drainage patterns and distribution of vegetation. Each phase is a distinct temporary appearance of the surface features produced by a meandering stream during lateral plantation of the valley. The vegetation is quite distinctive in each phase of the flood-plainbecause of differences in drainage-in large part dictated by the presence of and depth to the top of the permafrost, The permafrost table gradually rises in the younger to the older phases of the flood-plain because on the older surfaces, the mosses are better established and thicker, and act as a better insulation. The oldest flood-plain surfaces have the highest permafrost table because of the thick, protective moss carpet.
At least four phases can be recognized on major flood-plains: (a) linear phase (the youngest), distinct linear lakes parallel to the river, no integration of drainage, large deciduous trees, permafrost absent or low, no large masses of ground ice; (b) advanced h e a r phase, distinct linear lakes, some segmentation, large coniferous trees,slightly higher permafrost table, no large masses of ground ice; (c) coalescent phase, coalescing linear lakes, generally at an angle to the river, coniferous trees, some tundra, high permafrost table, and no large masses of ground ice; (d) scalloped phase (the oldest), irregularly shaped lakes with scalloped borders, integrated drainage, stunted coniferous trees, m u c h tundra, very high permafrost table, and much ground ice.
Résumé; Le pergélisol et la végétation dans les plaines des
COUTS
inondables d’eau subarctiques -[Alaska] (Troy L.Péwé)
Dans l’évolution des plaines inondables des principaux cours d’eau subarctiques,on peut distinguer plusieurs phases nettement déterminées en fonction surtout des caractéristiques du drainage et de la répartition de la végétation. A chacune de ces phases correspond un aspect temporaire distinct des Caractéristiques de surface produites par un cours d’eau à méandres pendant la planation latérale de la vallée. Chaque phase est marquée par une végétation très caractéristique, ce qui s’explique par les différences dues en partie à la présence du pergéde drainage lisol et à la profondeur à laquelle il se situe. L a table
-
de pergélisol est d’autant plus élevée que l’évolution de la plaine inondable est plus avancée, car sur les surfaces anciennes la mousse, établie de plus longue date, est plus épaisse et assure,de ce fait,une meilleure isolation. C’est dans les parties les plus anciennes de la plaine inondable que la table de pergélisol est le plus élevée, en raison de l’épaisseur du tapis de mousse protecteur. On peut distinguer au moins quatre phases dans l’évolution des principales plaines inondables: a) une phase linéaire (la première) - lacs linéaires distincts parallèles au cours d’eau, aucune intégration du drainage, grands arbres à feuilles caduques, pergélisol inexistant ou profond, aucun amas important de glace; b) une phase linéaire avancée -lacs linéaires
141
T.L. Péwé
distincts, une certaine segmentation, grands coniferes, table de pergélisol légèrement plus élevée, aucun a m a s important de glace; c) une phase coalescente lacs linéaires coalescents formant généralement u n angle avec le cours d’eau, conifères, début de “toundra”, table de pergélisol élevée, aucun amas important
de glace ; d) une phase échancrée (la dernière) -lacs de forme irrégulière a u x rives échancrées, drainage intégré, conifères rabougris, “toundra” abondante, table de pergélisol très élevée et a m a s important
de glace.
Discussion R. J. E. BROWN.A very important aspect of permafrost is the changes experienced by this phenomenon with the passage of time. M a n y changes in the extent and thickness of permafrost have occurred through time in response to fluctuations in climatic and terrain conditions. At any location in the permafrost region it is important to obtain as m u c h information as w e can on the history of the permafrost -the variations or changes that have occurred in the permafrost. It is possible that, at some particular location, where permafrost exists today, it did not exist at some time in the past. Or, conversely, permafrost was present in an area in the past where it does not occur today. T o trace such changes, w e must examine terrain features which are indicators of previous conditions. Thus Dr. Péwé’s contribution is important because he has traced the development of the vegetation on the flood-plain of subarctic rivers and managed to reconstruct the development of permafrost and its changes with time.
close to the surface under the older vegetation. I have some evidence from a limited area that the first perennial frozen layer forms at a depth of 1 m or less. Do you have any information as to the level at which permafrost first forms in the area reported in your paper?
L. A. VIERECK. M y question relates to the depth at which permafrost first forms on the slip-offsideof the river.As stated by Dr.Péwé there is no permafrost under the rivers but it is
T. L. PÉwÉ. Permafrost is only truly measured by thermal means-temperature cables in borings through the permafrost. Permafrost occurs in both mantle, and bed-rock.
142
T. L. Péwé. The depth
at which the permafrost forms is dependent upon the surface cover of snow and vegetation as well as the climate. In silt the permafrost was first found at 4 or 5 ft depth if the moss cover is absent. In dry gravel the permafrost table m a y be at greater depths if the vegetation is the same as in silt areas.
V. OKKO.I should like to ask two questions. H o w is the true thickness of permafrost measured? Where measured, do the soil and the loose earth form mantles of a thickness corresponding to the figures obtained for the permafrost or does the permafrost extend down into the bed-rock?
Investigationson palsas in Lapland Finnish M. Salmi
INTRODUCTION Palsas, or frost mounds, as they are scmetimes called, represent the northernmost complex type of bogs. In Finland they are found in the north-westernregion of the country,in the area of the Enontekiö and Utsjoki districts, where the annual mean temperature is less than -lo C and the annual precipitation under 400 mm. This is the coldest region of the country, and permafrost occurs there sporadically in palsas and in other hummocks of bogs. It has also been discovered in moraine and sandy soils to a depth of at least 5 m (Mikkola 1941). Most of the palsas in Finland are situated in the subalpine birch zone,some in the alpine zone, and the most southerly ones in the northern parts of the coniferous forest zone. A m o n g the investigators of palsas in this country can be mentioned Kihlman (Kairamo) (1890), Auer (1927), Ruuhijärvi (1960, 1962) and Ohlson (1964). During two years, in both winter and summer, the writer has studied the palsas of Iitto, Aatsajoki and Kelottijärvi, located in Enontekiö. This paper describes the investigations carried out in the Kelottijänkä bog area of Kelottijärvi. This bog is situated about 10 k m north of the pine limit and 2 km north of Kelottijärvi lake, 300 m east of the road leading to Kilpisjärvi. The exact geographical location of this spot is 68036' N., 22000' E. The palsas here are of different sizes and also occur in groups. In this paper, two adjacent palsas have been selected for a detailed description. The purpose of this study was to investigate the external and internal structure of palsas, their relationto the surrounding ground,as well astheir development.
EXTERNAL APPEARANCE OF PALSAS AND THEIR RELATION TO THE SURROUNDINGS Of the two palsas investigated,the larger one is elongated in the direction SW.-NE. The smaller one, nearly circular in shape,is located immediately to the south of the larger palsa The large palsa is about 85 m long. Its width at the broadest point is 42 m but is only about 8 m at the narrowest point in its northern, part. The outer contour of this palsa, as is the case with m a n y of the more extensive palsas, is quite meandering in shape, and thus numerous indentations are formed in its edges, some narrow and some very broad, extending inward even close to the centre of the palsa. The elevation of the base of the palsa at its northern end is about 359 m above sea level and at its southern end 1 m less than this. The summit of the palsa is slightly more than 3 m above the level of the surrounding bog. Both sides of its most narrow point are steep, as are also its eastern and south-easternsides. On the other hand, the southern flank of the palsa iises gently to the summit, and similarly its western and northern parts have a very gentle slope. Since this region has little precipitation and the winds are strong,the summits of palsas become exposed already early in the spring or late winter, while at the same time there m a y be several metres of snow remaining along the edges of the palsas. Ground freezing is naturally most pronounced on the exposed upper surface of the palsa, and as a consequence, the layer of superficial peat which generally thaws in the summer becomes fractured and dissected into blocks. The smallest such fissures thus formed are only a few centimetres wide, while the largest ones m a y have a breadth of nearly a metre and they often extend downward to the upper surface of the permafrost. 143
M. Salmi
FIG.1. M a p sketch of the palsa investigated. The thick lines show the margin of the palsas, the thin ones are elevation contour lines. The investigation section lines A-F are also seen. The crosshatched area indicates the cut made with the motor saw. In the index m a p is shown the location of the palsas, and to the lower right is the habitat of Alnus.
144
Investigationson palsas in Finnish Lapland
In the bigger of the Kelottijänkä palsas, such fissures are found especially at the summits of its northern narrow part and from here further to the north. Along the steep sides of the palsa, there are also slides in m a n y spots, where the peat layer has slipped down. T h e fissures as well as the upper border of the slides are mainly located at the places which already in early spring are free of snow, or at the boundary between these places and the snow. N o other signs of erosion are visible on the palsa; instead, its surface has a continuous cover of vegetation. The vegetation is most sparse in those places where there are the greatest number offissures and slides. The vegetative cover of palsas is comprised to a great extent of shrubs and brush. The most c o m m o n species are Betula nana, L e d u m palustre, Vaccinium vitisidaea, Rubus chamaemorus and Empetrum. Moss species include S p h a g n u m , Polytrichum, Pleurozium and Dicranum. Lichens are also abundant, species of cladonia being the most frequent. Furthermore, some birches (Betula tortuosa) also grow on palsas. The summit of the smaller palsa is flat and it is only about 0.5-1 m higher than the level of the surrounding bog. N o signs of erosion are visible on this palsa. The bog surrounding the investigated palsas is a soft “rimpi” bog, i.e. a treeless, watery bryales-carex bog. It is wettest especially along the southern,eastern and north-westernflanks of the palsas, where there is often a small body of water,the “rimpi” or “pals lagg”. Adjacent to the palsas the bog is very deep,but the peat layer diminishesrapidly in going away from them. Thus on the eastern side of the large palsa, along line A,the peat layer is 3.5 m thick at the edge of the palsa, but at a distance of 15 m it has decreased to only 0.5 m and at 25 m boulders are seen on the surface of the bog. At the north side of the palsa, the layer of peat here is 1.8 m thick,while at a distance of 30 m boulders are visible. Along line C on the western side,the peat is 3.2 m thick adjacent to the palsa but only 90 c m at a distance of 25 m , and at 50 m stones are seen on the surface of the bog. T o the south, the direction in which the bog surface descends,the peat layer does not diminish so rapidly. Adjoining the palsa, the layer is 2.8 m thick, and at a distance of 100 m to the south it still measures 2 m in thickness. Boulders become visible at a distance of 200 m from the palsa. At the bottom ofthe bog under the peat there is a very narrow zone of silt adjacent to the palsa, while further away occurs moraine and finally boulders. The palsas thus are situated at the site of a basin about 500 m long and 100 m wide.
STRUCTURE OF PALSAC AND THEIR RELATION TO THE UNDERLYING GROUND The structure of the palsas and their relation to the underlying ground was studied by means of borings and cuttings. T h e borings were made with both a motor-drivencobra auger and an ordinary carpenter’s auger which had been constructed so that it could be fitted with Hiller auger shafts.The cuttings were made in winter with a motor saw provided with a special blade 80 c m long. For the structural studies, three lateral investigation lines, A, B and C,were drawn across the large palsa as well as the line D extending longitudinally through the palsa. On the smaller palsa there were the two lines E and F (Fig.1). Except for line B,borings were made at intervals of 2-10m along the lines. Both augers were used to penetrate the frozen part of the palsa, but the underlying mineral ground was bored with the cobra model. The cross-hatched area of the large palsa in Figure 1 was cut with the motor saw beginning at the northern edge of the palsa and proceeding in a direction towards line A. In this w a y it was possible to continuously observe the structure of the palsa as the section enlarged. By cutting in this manner, a section 4.5 m long and at its highest point over 3 m deep was made. About 25 m3of frozen peat were removed in the process. Figure 2 shows a profile drawing of the palsa along line A on the basis of the borings. At its surface is a 50-80c m thick layer of peat which is unfrozen in the summer. This superficial peat layer is thinnest on the eastern side of the palsa and thickest in the centre and north-westernside. Just below the surface layer is the upper border of the permanently frozen ground. This frozen portion of the palsa makes up the greatest
m
3627
+22m+20 +I8 +16 +I4 +12 +IO
+E
+6
t4
t2
fC- - 3
-81
361-
360
-
359 -
357 356 355 354 353 358
352 -
3511
FIG.2. Profile A. Key to the symbols in Figure 3.
145 10
M. Salmi
part of the section. It is thickest at the centre of the profile, about 6 m.Both edges of this so-calledice cake are steep, nearly vertical at its north-western end. T h e ice cake is thicker at its northern and northwestern side than at the opposite side. The frozen mass of the palsa consists mainly of peat and ice, with a greater proportion of peat in its upper part thanlower. A t the centre of the palsa, the ice begins to 'distinctly increase already at a depth of 1 m , and along the sides at a depth of 1-2m.The lower part of the ice cake is made up of alternating layers, 5-10 c m thick, of peat and ice. At the north-western side of the palsa on a level with the surface of the adjacent bog. there is a small area of pure ice about 30 c m thick. It was discovered during investigations made in the winter. The central part of the bottom of the approximately 18 m broad ice cake lies in contact with mineral soil over a distance of about 6 m. This mineral soil consists of silt and fine sand. Part of it, too, is frozen, and thus the bottom of the ice cake contains alternating layers of ice and silt. Under the ice mass in certain places lies unfrozen peat, and below this, extending under the entire area of the ice cake is 1-2 m of unfrozen silt and fine sand, which is apparently a sediment layer of an ancient ice lake. Below this is situated moraine and still further d o w n is a hard surface which the cobra auger was not able to penetrate. It m a y be bed-rock or the level of large stone blocks. Along line C the palsa is about 30 m wide (Fig.3). The surface layer of peat which is unfrozen in the summer is 50-80c m thick, being thicker at the southeastern side of the palsa than at the opposite side. In general,the ice cake here is thinner than along line A; its m a x i m u m thickness is about 5 m. The ice mass is distinctly thinner at the left (south-east)than at the right of the profile, just as in the case of line A. Along line C the bottom of the ice cake lies in contact with mineral ground for about 19 m. At its thickest, the mass of frozen silt is over 2 m thick, and it is clearly
FIG,3. Profile C. 1, peat in upper part of palsa which is unfrozen in summer, as well as peat layer in adjacent bog; 2, permanently frozen palsa peat; 3, alternating layers of ice and frozen peat,with large proportion of ice;4,fine sand and silt; 5,frozen silt with ice layers; 6,moraine; 7, stones or bed-rock;8, detritus gyttja. 146
seen to alternate with ice layers. The upper surface of the silt is convex and it rises above the upper edge of the moraine basin of the bog. It is evident that the silt has been lifted from its original position by the action of freezing, just as the peat, too, was raised by the ice. In some places under the ice cake there is less than 0.5 m of silt, and it is generally thinner than along line A. Very little unfrozen peat lies below the ice mass, being more abundant at the right in the drawing than at the left. The longitudinal section of the palsa (Fig.4)shows that at the northern end of the palsa-to the left in the drawing-the surface peat layer is thinner (c. 40 cm) than in the centre and at the southern end. The northern end of the ice cake is thicker and under it is less unfrozen peat than at the southern end. The ice mass, which is 80 m long,lies in contact with mineral ground for a distance of about 35 m , and in some places the silt included in the ice cake is as m u c h as 1.5 m thick. At both ends of the ice cake,water has been encountered. The upper surface of the frozen silt in this profile, as also in the profile through line C,rises above the lowerlying upper edge of the moraine basin. The southern part of the ice cake is thinner than its northern part. On the basis of the previously described profiles A and C,the ice mass at the east and south-eastsides of the palsa is also rather thin,and moreover its formation appears to be slower here than elsewhere-perhaps sometimes even thawing for short times. In contrast, the ice mass toward the west and northwest of the palsa is thicker and constantly increasing. On the basis of these observations, it seems that the palsa is probably gradually shifting toward the north, north-west and west, directions in which the effect of the sun is less than in the opposite directions. This movement is also suggested by the fact that immediately adjacent to the north, north-westand west flanks of the palsa, there are peat embankments about 0.5 m high and 1-2m wide.
351'
FIG.4. Longitudinal section D of large palsa. Key symbols in Figure 3.
to the
Investigations on palsas in Finnish Lapland
T w o profiles of the smaller palsa are shown in Figure
5. Profile E is made in a SW.-NE.direction, with the north-eastpart to the right in the drawing;here the ice mass is slightly thicker and its edge steeper than at the opposite side. Profile F is made perpendicular to the former. At its south-easternside a cavity is visible in the ice which was found in the summer just below the level of the bog; here the mass of ice is slightly thinner than at the opposite side. Similar cavities were evidently found by Ruuhijärvi (1962) in the palsas investigatedby him. The m a x i m u m thickness of the ice cake of the smaller palsa is about 3 m , and below its entire extent is unfrozen peat. The palsa, thus, floats in peat. This is apparently the reason for the fact that the upper surface of the palsa is flat and also that it is elevated only a small amount above the level of the surrounding bog. By comparing the profiles of both palsas and their relation to the underlying ground, it seems that the
E m
359 358 357 356 355 354
-
-10m
-3.5
tio
+3m
-
353
-
351351 i
F
m
359r 350 357 356
-
355354353
-
352351-
FIG.5. Profiles E and F of small palsa. Key to the symbols iniFigure 3.
mineral soil below the bog is not of primary significance in the formation of such peat palsas. Instead, when the growing palsa attains such a thickness that its lower surface comes in contact with the mineral soil, its upper surface at this corresponding point begins to rise, and thus it obtains its characteristic d o m e shape, which is evident in the large palsa. The smaller, low and flat palsa is obviously just in its early stages of development. An important factor for the initiation of palsas appears to be the site in the ancient basin of the bog where the peat layer is thick. Both of the palsas under study were located at such a site. On the basis of the profiles, calculations were made to determine what percentage of the total cross-sectional area of the palsas is elevated above the surface of the surrounding bog. The results were as follows: Profile A approximately 39 per cent. Profile C approximately 41 per cent. Profile D approximately 36 per cent. Profile E and F approximately 16 per cent. It is evident that in the latter two profiles the proportion of the palsa rising above the surface of the bog is m u c h less than in the profiles of the larger palsa. The above ratio in the small palsa is approximately similar to the ratio of above-water ice in a floating iceberg to the mass of ice below the water. A cutting made with the motor saw made it possible to study in more detail the structure of the palsa, Several photographs and drawings illustrate this. Figure 6 shows the major part of the cut. The line drawn indicates the upper limit of the permafrost. The surface peat, containing cavities, as well as the lower frozen peat, appear on the basis of the branches and layerings in them to have become stratifiedwithout disturbance. At the left in the picture is seen the place wherc a sample series was taken for more detailed study. Figure 7 shows this sample series after the ice in it had thawed. It is seen that in the upper part of the cut (samples 1 and 2), the peat is very compact. Sample 3, representing the frozen part of the palsa, contains peat which is considerably more porous, and its porosity increases in going further down. W h e n dry, this peat is extremely light in weight. The crevices seen in the photograph are due to the fact that ice here has melted. In the place where the sample series was taken, ice occurs in the form of lenses,the longest ones being about 10 c m in length and 5-10mm thick. Thick ice layers were not observed in the section where the cut was made, as is the case, for example, in the lower part of the palsa shown in Figure 2. The reason for this is probably that the sample series was taken close to the edge of the palsa (cf.Fig.1). Figure 8is a drawing made on the basis of the abovedescribed saw cutting and extending slightly deeper than the level of the adjacent bog. The position of the lower surface of the ice cake as well as data still lower are based on borings. At the top is seen the layer of
147
M.Salmi
FIG.6. Cut made by motor saw on line A. The location of the sample series is seen ait the left. Explanation in text.
2
3
4
5
6
FIG 7. Sample series cut from the sectioned palsa; six samples of 45-55 c m in length were taken, the upper ones beginning at the left. The depth (in cm) of the lower end of each sample is as follows: 1, 45; 2, 100; 3, 150; 4, 200; 5, 250; 6, 300. Worthy of note is the stratification of the peat as well as its primary origin.
148
Investigationson palsas in Finnish Lapland
surface peat which is unfrozen in the summer. It contains cavities which lead to the upper surface of the palsa, where they are visible as long fissures. The direction of the peat layers in the frozen mass is indicated on the basis of the position of tree trunks, branches, shrubs and variations in peat type. It is consequently seen that in the upper part of the section these layers are slightly curved,following the contour of the upper surface of the palsa,but they become more and more horizontal in going deeper in the section. It is possible that even lower,the peat layers gradually curve downward and follow the lower surface of the ice cake. At the right edge of the palsa ice mass is seen the lense of pure ice mentioned previously.At the left is indicated the place where the sample series was taken with the motor saw. Its lower end extends to a point about half w a y into the peat layer under the lower edge of the ice, and below this is still about 1 m of silt.
t4rn
+3
+2
tl
O
-1
DEVELOPMENT OF PALSAS T w o pollen diagrams,presented in Figure 9,were made from the sample series taken from the palsa and from the adjacent bog. By chance, both diagrams comprise a depth of 3 m , but the sample series taken from the adjacent bog extends to the bottom of the bog. The diagrams appear quite different, but they have enough c o m m o n features to indicate a possible connexion. Starting from the top of each diagram, at a depth of 20 c m occurs a Betula m a x i m u m in each of them which evidently correspond to one another. The similarity continues to a depth of 60-'i0cm. Here the upper surface of the frozen palsa lies at a depth of 65 cm. Consequently,the upper peat layer of the palsa, which thaws in the summer, appears to grow just as rapidly as the surface peat in the adjacent bog. Lower in the palsa diagram, however, a distinct extension begins to be visible. Thus, the point in the palsa 2.2 m deep apparently corresponds to only 1.3 m in the neighbouring bog. Further, the lowest point in the palsa diagram, 3 m , corresponds to 1.8 m in the bogs. In any case, the palsa diagram lacks the entire Betula period seen in the bog diagram and evidently also the peat layer represented by the sharp Pinus pollen peak just above it. By comparing the two diagrams in this manner, their peat layers correspond well in the lower parts. However, in the upper part of the palsa diagram, the peat-containingshrubs,deciduoustrees and uppermost S p h a g n u m are evidences of an independent character of the palsa. The same fact is suggested by the abundance of Fricaceae pollen in the upper part of the palsa peat, as well as by the Betula pollen types, of which some is Betula nana and some Betula tortuosa. It can be further mentioned that the uppermost sample taken from the bog contains 14 per cent Alnus pollen. As an explanation for this, a grove of Alnus incana, consisting of about a dozen trees, was located somewhat over 2 km south-east of the palsa (see
Fig.1).
l
' t4m
t3
t2
+I
O
-1
FIG.8.Drawing made on the basis of the motor saw cut.See also Figures 6 and 7. Explanation in the text.
The writer has earlier performed peat geological studies on the bogs of Kittilä in western Lapland (Salmi, 1963). On the basis of these studies, as well as on the yet unpublished results of C14-datingm a d e later on the bogs of Kittilä, it appears that the depth of 2.4 m in the diagram of the bog corresponds to the time 7800-8000years B.P., or the final phase of the boreal period. Accordingly, the palsa peat investigated by pollen analyses was evidently formed after the above period. The upper limit of the permafrost in the palsa apparently corresponds to the latter half of the sub-borealperiod. Earlier pollen studies of palsas have been made by Auer (1927), Lundqvist (1951) and Ruuhijärvi (1962). According to them, large amounts of peat had been removed from the surface of the palsas by the action of wind erosion, a phenomenon which was not found
149
M. Salmi
AP
NAP
Sphagnum peat
Carex peat
Bryales peat
Equisetum peat
AP
IPpPpPd
Deciduous peat
NAP
A
Picea Pinur
O
Alnus
o
Betula
Nanalignidi peat
x
FIG9.Pollen diagrams of the palsa and the peat layers of the adjacent bog.
on the palsa at Kellottijänkä described presently. On the other hand, the studies of Lundqvist and Ruuhijärvi showed that the peat of the palsas was stratified under uniform conditions, and that during freezing, the layers were greatly enlarged, as m u c h as doubling in thickness. The origin of the palsas has been speculated by m a n y earlier investigators. The conception proposed by Lundqvist (1951)was agreed to in its main parts by Ruuhijärvi (1962), and it corresponds also closely to the ideas of the writer. According to this conception, the ice mass in the palsa after having first been initiated, grows from below. Keinonen (1961) states that as the water in the freezing layers changes to ice, the soil particles attempt to obtain additional water as replacement. In the case of palsas, such a water supply is available in the unfrozen peat below the palsa or in an actual source of water, such as seen for example in Figure 4.Furthermore,also between the ice crystals, empty spaces are formed which are filled with water coming in from below. Moreover, the temperature difference itself between the cclder superficial layers and the deeper warmer areas causes water to flow in the direction of the higher temperature. Thus, the movement of water in the freezing part of the palsa is the result of the combined influence of m a n y mechanisms. It is evident that the ice mass of an active palsa also grows to some extent frum its upper surface as the peat layer becomes thicker. This growth, however, is smaller than that occurring from below. As the ice cake in the palsa grows thicker,it initially
150
remains to a great part within the peat layer of the bog, and only a small fraction rises above the level of the bog surface. This is seen on the basis of the profiles E and F (see Fig. 5). But when the lower part of the ice cake ccmes in contact with the mineral soil lying below, the upward growth of the palsa begins to increase and it attains a dome-shaped form. Information concerning the time when each palsa began to be formed could be perhaps cbtained by studying the crystal structure of the palsa ice, in the same manner as the formation of glaciers has been investigated at Spitzbergen (Palosuo and Schytt, 1960) and the ice covering moraines at Kebnekajse (Ostrem, 1962). Although the formation of palsas has occurred and still occurs continuously, it is the belief of the writer that the continental sub-borealperiod was apparently very favourable for the development of palsas.
OTHER OBSERVATIONS ON PALSAS About 200 m east of the two closely-studiedpalsas on Kelottijänkä lies a group of lowish palsas. Their surface has numerous exposed patches, where the wind has eroded the superficial peat. In addition, various-sized fissures are common, and along the flanks of the palsas large pieces of peat have slid d o w n into the adjacent body of water,the pals lagg. Furthermore, on the tops of these palsas are m a n y depressions and pits of various sizes. These observations indicate that these palsas have already reached the peak
Investigationson palsas in Finnish Lapland
FIG.10. Palsa formation of anunusual type found in the Aaitsajoki palsa region.
1
of their development and are n o w in the process of rapidly disappearing. Finally, a few words will be said about observations made in the Aatsajoki palsa region located about 15 km to the north from Kelottijänkä. Figure 10 shows a palsa formation differing from the normal type. There is only a sparse cover ofvegetation on the surface of the palsa,but m a n y fissures are present. In the upper part of this formation is a 60 c m thick layer of peat which is unfrozen in the summer. Below this is permafrost, consisting of alternating layers of mineral soil (fine sand and silt) and ice. The top of this formation rises about 1.5 m above the surroundings. As far as is known, no previous reports have been made in the literature about similar formations in Finland. This can be regarded as a mineral soil palsa, which goes
under the n a m e of “pingo” in the literature.Pingos are found, among other places, in the arctic regions of Alaska, Greenland and the Soviet Union. In this connexion, reference can be made to the summarized compilation published by Maarleveld (1965)on pingos as well as palsa investigations in general. It is possible that the Aatsajoki pingo representsthe final state of a disappearing palsa, in which the mineral soil, originally lying at the bottom of the basin beneath the palsa, was raised in the same manner as was found on the basis of profile C of the large palsa at Kelottijänkä (Fig. 3). In Sweden and Norway, Lundqvist (1951 and 1953), Svensson (1962) and W r a m m e r (1965) have reported finding palsas with a permafrost mineral core.
Résumé Recherches sur les hydrolaccolithes (palses) de la Laponie Jinlandaise (M.Salmi)
L’auteur a étudié en Finlande,dans le nord-ouestde la Laponie, la structure des palses par rapport au sol et aux marais environnants. I1 a concentré ses observations sur deux palses voisines. L’une, au contour régulier, a un diamètre d’environ 15 mètres et un sommet aplati qui dépasse de 0,5 à 1 mètre le marais qui l’entoure. L’autre est oblongue et de forme sinueuse.
Longue de 85 mètres et large de 8 à 42 mètres, elle a la forme d’un bouclier. L e centre de cette palse est à environ 3,5 mètres au-dessus du marais. Les talus sont parfois abrupts,à cause de l’érosion, et la surface présente des crevasses de différentes dimensions. L’hiver, les sommets des palses sont couverts d’une mince couche de neige gui fond au début du printemps. Au voisinage des palses la couche de tourbe a de 2 à 2,5 mètres d’épaisseur, mais elle va en s’amincissant latéralement.
151
M. Salmi
Pendant l’été, la couche supérieure de la palse est formée de tourbe non gelée d’une épaisseur de 50 à 80 c m . Au-dessous, o n trouve un n o y a u de glace pérenn e mêlée de tourbe et aussi de limon et de sable. L a proportion de tourbe décroît à mesure qu’on descend, Les bords du n o y a u de glace sont souvent presque verticaux. D a n s la plus petite des d e u x palses, le noyau de glace a moins de 3 mètres d’épaisseur et repose entièrement sur de la tourbe n o n gelée, si bien que la palse flotte dans le marais. D a n s la plus grande, le noyau de glace mesure jusqu’a 6 mètres d’épaisseur et sa partie centrale repose sur un sol minéral. C e
n o y a u renferme des couches de limon et de sable de 1 à 2 mètres d’épaisseur. Latéralement, il est entouré de tourbe non gelée et d’eau. Ces deux palses se trouvent dans le bassin d’un ancien lac glaciaire. Les sédiments de limon et de sable provenant du lac glaciaire reposent sur une moraine. Un marais s’est d’abord formé dans le bassin, et il est évident que les palses sont beaucoup plus récentes. L a plus grande de celles-ci présente déjà des signes de dégénérescence, la plus petite, au contraire, est toujours active.
Discussion E. SCHENK. The observations of Dr. Salmi are really very important. The fundamentals are measurements and therefore facts. They provide evidence to prove that a palsa originates where the moore is deepest. While the process of water accumulation and ice formation has ceased in the surroundings, it is continuing in this deeper part. The ice mass will become thicker and thicker and therefore arise over the surroundings.This is the moment of the birth of a palsa. Another way s e e m s possible too. For example when the original deeper part of the moore is filled with m u d or clay: then the clay or m u d will supply the palsa with crystallizing water, so that cryoturbation takes place and frozen mineral soil forms the core of the palsa. Did you observe what happens when the vegetation cover of the palsa disappears? W h a t will be the consequence?
M. SALMI. If the vegetation cover of the palsa disappears, various erosive processes begin to wear off the superficial peat. As a consequence the frozen part of the palsa begins to melt. O n its surface pits are formed, which at times are filled with water. It is evident that these events mark the beginning of the disappearance of palsas. In m y investigations I did not especially study that problem. L. AARIO.It has been difficult to decide in which conditions the palsas have grown. In Petsamo the palsas I have investigated had begun their development after the arrival of the spruce, i.e. in under 2,000 years. M y pollen diagrams show the end of the growth of palsas occurred not very long before the present time. However, all the palsas in Finland are dead, i.e. their lichen covers do not form peat. During the last hundred years or so no great change in the climate has been observed. Thus the cause of the end of the palsa growth is still unknown. A. JAHN. M y question is about the terminology. Are there differences between the “palsas”, “pingos” and “bugors”? Dr. Salmi divided palsas into two groups: peat palsas and
J 52
palsas with mineral soil. As far as I know there is a third group: palsas with ice core. Do such palsas exist in the northern part of Finland?
M.SALMI. I have not found palsas with ice core in northern Finland. Further drilling is needed to obtain more information on the structure of palsas.
W.PRUITT. I a m happy to see that a greater understanding of palsas has been reached since these are of biological as well as cryopedological interest. They form an important part of the biological system of the Subarctic. Cariboufeed on the palsa lichens since these lichens are usually snowfree. In North America a genus of voles, previously believed to be rare, have been found to be limited markedly to palsas, where they are quite common. Several species of birds nest primarily in palsa trees, the only trees in an expanse of mire. Palsas elevate a m a s s of peat above the general watersoaked level. Here the peat dries and the organic meteria1 is m a d e available to the ecosystem where it supports a more varied vegetation than when water-soaked.The release of the previously immobilized energy is important for the biological community. Perhaps when we understand the genesis of palsas we could initiate their formation and thus manipulate one aspect of biological productivity in the Subarctic. .W. G. MATTOX. Have you made any temperature observations which shed light on your ideas of palsa development?
M.SALMI.During drilling, in August, I made some measurements of temperature in palsas. They were not made systematically and the results m a y not be quite exact. The measuring was disturbed by the freezing of the drill holes or by their becoming filled up with water or peat. M y results lay in the frozen palsa just about O0 C.
Investigations on palsas in Finnish Lapland
Bibliography AUER, 1927. Untersuchungen über die Waldgrenzen und Torfböden in Lappland. ( C o m m . Inst. Quaest. Forest. Finl. 12.) KEINONEN, L. 1961. Keskustelua routailmiöistä. Terra, vol. 73, no. 4. 1890. Pflanzenbiologische KIHLMAN, A. O. (KAIRAMO) Studien aus russisch Lappland. Acta Soc. F. FI. Fenn., vol. 6, no. 3. LUNDQVIST, G. 1951. E n palsmyr sydost o m Kebnekaise. Geol. Fören. Stockholm Förh.,vol. 73, no. 2. . 1953. Tillägg till palsfrågan. Geol. Fören. Stockholm Förh., vol. 75, no. 2. MAARLEVELD, G. C. 1965. Frost mounds. A summary of literature of the past decade. (Mededelingen van de Geologische Stichting, Nienwe Serie, N o 17.) MIKKOLA, E. 1941. Enontekiö. Suomenmaa, vol. IX, no. 2. Porvoo. OHLSON, B. 1964. Frostaktivität, Verwitterung und Boden bildung in den Fjeldgegenden von Enontekiö, FinnischLappland. Fennia, vol. 89, no. 3. PALOSUO, E.; SCHYTT, V. 1960. Till Nordostlandet med den
_-
/ Bibliographie svenska glaciologiska expeditionen (Summary: To Nordaustlandet with the Swedish Glaciological Expedition). Terra,vol. 72, no. 1. RUUHIJÄRVI, R. 1960. Ifber die regionale Einteilung der nordfinnischen Moore. Ann. Bot. Soc. Vanamo, vol. 31, no. 1. . 1962. Palsasoista ja niiden morfologiasta siitepölyanalyysin valossa. (Zusammenfassung: Über die Palsamoore und deren Morphologie im Lichte der Pollenanalyse). Terra,vol. 74, no. 2. SALMI, M. 1963. O n the subfossil Pediastrum algae and molluscs in the late-quaternary sediments of Finnish Lapland. Arch. Soc. Vanamo, vol. 18, no. 2. SVENSSON, H. 1962. Några iakttagelser från palsområden (Summary: Observations of palses). Norsk Geogr. Tidskr., vol. XVIII, no. 5-6. W R A M N E R , P. 1965. Fynd av palsar m e d mineraljordkärna i Sverige. Geol. Fören. Stockholm Förh., vol. 86, no. 4. OSTREM, G. 1962. Ice cored moraines in the Kebnekajse area. Bid. Peryglacjalny, vol. 11.
__
153
Permafrost and frost structures in the subarctic area E. Schenk
ON THE BOUNDARIES OF THE SUBARCTIC AREA Permafrost is frozen ground which does not melt through one or several summers. W h e n this occurs the yearly heat budget in the soil is negative. The area showing these conditions is the Arctic with permafrost and its structures. One might think that it would be easy +to describe the subarctic area in a similar w a y as a zone where the yearly heat budget of the ground is sometimes negative and sometimes positive. Under such conditions permafrost of former times (Pleistocene) will be preserved in the depth of the ground while permafrost of recent times is developing and collapsing. Furthermore this heat budget would be seen in the occurrence and the temperature of ground water at the depth of Auctuating ground temperatures. But until n o w w e have not had enough data to undertake an evaluation in this direction,to describe or to demarcate the subarctic area in this way. Therefore, the first assumption for the limitation of the subarctic area and for the classification of arctic and subarctic frost features still needs to be described. W e cannot say that the subarctic area begins here, where permafrost features occur sporadically, and that it ends there, where permafrcst is continuous. Even the boundary of continuous permafrost is drawn in a different w a y by several authors (Black, 1954; Lundqvist, 1962; Brown, 1969; Katz, 1930; Frenzel, 1960). The difficulty increases if we try to adopt the definitions which are used in geography and climatology. Finally the comprehensive paper by Troll on the frost structuresand frost climates of the Earth offers no significant characteristic of subarctic frost features. H e prefers the term “subnival zone” with patterned ground of different climatic conditions (Troll, 1944, p. 557).
The problem finally is complicated because of the occurrence of fossil arctic frost structures and of the active arctic pattern on high mountains in the subarctic area with its significantvegetation cover. Therefore,it is an open question as to whether there are any typical subarctic frost features at all.
RELATION BETWEEN THE SIZE OF FROST PATTERN AND THE CLIMATE Frost features, of course, can develop only if the freezing temperature can penetrate a soil with characteristic properties affording a high water content and sufficient water supply. T h e frost penetration depends on the time of freezing and on the quantity of water to be transformed into ice.The ratio offreezing timeto water content controls the differences in development of the types of patterned ground. The patterns are originated by short but frequent freezing periods as well as by long but single freezing periods during the year. Small patterns develop when the freezing soil layer is thin and well suppliedwith water. They also develop when the soil layer is thick but with a water supply restricted to its upper part (often by deep-freezing temperature). Large patterns occur only in deep freezing soils with high water content. Their formation also depends o n repeated freezing and thawing. Deep freezing, however, is mostly connected with seasonal freezing, but also with secular freezing periods. So w e understand that small patterns m a y occur in all climatic zones of the Earth: in the arctic and subarctic as well as in the alpine mountains of the tropic region and of course in the zones with a moderate climate. Furthermore w e understand that the distribution of larger patterns is restricted to the
155
E.Schenk
climatic zones with long frost periods, of the arctic and subarctic areas and of the high, mountainous regions,in so far as the high water content or supply is afforded by precipitation, melt water or accumulation through freezing and thawing itself. These climatic conditions replace the high groundwater level upon impermeable layers and permafrost. The huge patterns of frozen ground, such as tundra polygons, ice wedge nets, and pingos, however, are characteristic of areas with arctic climate. Their formation can occur only when the freezing layer is very thick and the freezing period very long. Accordingly the thawing period is very long too, although this process advances along the base and along the top surface of the frozen soil layer as well. Subarctic frost features, therefore, n o w occur on the surface, while arctic features are still dominant in the deeper ground since the end of the Pleistocene period.
ORIGIN AND DYNAMICS OF FROST STRUCTURES The formation of permafrost and frost structures is a process of increasing entropy and the frost structures which w e observe are the final product of systemfunctions as I described in an earlier paper (Schenk, 1955~).It is not possible to describe precisely the single factors which are enclosed in this polydisperse system of soil-water-gas including also vegetation, so w e cannot therefore calculate the frost effects exactly. Nevertheless it is possible to understand and to estimate the formation of frost structures in a w a y confirmed by facts. All frost structuresare developed by the dehydration of soil particles and the hydration of originating ice crystals. Osmotic and van der Waal forces are responsible for these processes and for the water movement in the freezing soil, which is caused by the originating and growing ice crystals. This process continues even under a pressure of about 2,000 kg/cm2 because the freezing-point decreases (Table l) and the concentration of the liquid increases with an increase in pressure. When this pressure is reached then the cryoturbation is stopped and another modification of ice develops (Bridgmann, 1912). The absorption energy of the ice crystal, that is the force of crystallization and of course of the ice veins and ice lenses,therefore, is the original force and the motor for the formation of frost structures (Schenk, 1955b). This formation takes time. The water of the capillaries and of the large pores must be consumed first,then the liquid water of the absorption film of the soil particles,and finally the water molecules of the vapour which surround the soil particles as well as very small ice crystals. Under the growing pressure caused by the volume expansion and the overload of freezing water the
156
TABLE 1. Characteristics of ice crystals controlled according to size (after Stackelberg,1964, p. 69)
Diameter of the small pores (cm) 3.10-4 10-8 3.10-3 Melting point of an ice crystalwith that diameter (o C) -0.4 -0.04 -0.01 -0.004 -0.001 Possible pressure (max.)in large pores (atm.) 4 0.4 0.13 0.04 0.013 Thickness of a soil layer which can be elevated by this pressure (cm) 2 O00 200 70 20 7
small ice crystals become unstable. This means that the small ice crystals will be consumed by the larger crystals (Ostwald,1960, p. 698 and 782;Everett,1961; Stackelberg,1964). This process controls the formation of ice aggregations in the frozen soil in contrast to ice segregation in the freezing soil. This formation can happen only under the conditions of a sufficientlylong frost period. It overwhelms the effects of high frost penetration speed, which w e observe for example in a quickly-frozen clay or fine sand as in the form of compact homogeneous frozen soil. N o w , in silt and clay and in organic materials the water content is higher than in coarse-grainedmatter. Furthermore the freezing temperature is decreased under the pressure of crystallization and vapour. It is obvious that under these conditionsthe formation of frost structures depends mainly on physico-chemical properties of the soil particles,because they determine the form of distribution of the water and its content and the size of the first ice crystals. Thus w e canunderstand the continuous growing of frozen structures. It can occur only in areas with very long frost periods. This condition exists only in the arctic and subarctic areas. Only in these areas do w e find huge polygons, pingos and palsas, while all other structures occur also in a wide variety in other climatic zones. But what is the cause of this variety? W e k n o w that the alternation of freezing and thawing (i.e. regelation) is a main condition for the development of patterned ground. But this alternation must be connected with the accumulation of water. In the arctic zone this is guaranteed by the impermeable permafrost in the subsoil and by the water absorbed by the soil which is often mixed with organic material. With greater distance from the arctic area and the nival boundary vegetation is an increasing factor and the appearance of the pattern changes. D u e to the vegetation the raw soil also has changed into a certain soil type and it seems that this is the reason why even thin soil layers are able to develop specific patterns as described and
Permafrost and frost structures in the subarctic area
clasdied into climatic groups by Troll (1944) and Frenzel (1960).
RELATION BETWEEN FROST FEATURES AND SOIL TYPE W e should, therefore, consider the soil types because their physico-chemical properties, which determine frost activity, are very different. But on this subject w e have not passed the first stage (see Kubiena, 1953; Tedrow and Drew, 1962). Soils are the immediate result of climatic conditions, vegetation cover,hydrological conditions,and original material. Therefore,w e can expect pedology to provide better characteristics of the frost features of the soils than w e n o w possess.Afirst step was made by Kubiena, w h o classified all patterned grounds into semi-terrestrial or groundwater soils and subhydric soils. In these main groups palses and string bogs are also to be included. If w e attempt to go into more detail w e see that certain types of pattern are restricted to the arctic area-with its raw soils of periglacial Frostschuttand other ones to the subarctic area, with increasing content of organic materials and peat formation when stretching from north to south and crossing the subarctic tundra and “sub-boreal” forest zones. As they disappear, stone nets, stone rings, stone stripes and polygons are replaced more and more by earth spots, earth hummocks and peat mounds of various forms and sizes. Grigoriew (1930) described their relation to the different climatic conditionsin Eurasia,Hopkins and Karlström (1955) their distribution in Alaska, Brown (1969) their occurrence in Canada, S p e h m a n n (1912),Sörensen (1935)Thoraninnson (1951)and others the pattern of Iceland. Frost activity destroys the plant roots because of the bending of the surface layer upon an ice accumulation (aggregation) growing in the subsoil. While the surface layer is expanded through this process and roots are being torn, the subsoil is shrinking due to loss of water which is carried away towards the freezing front. This gives us the convex surface form of frost pattern. This process advances into the depth so that freezing and thawing produce accumulation and heaving of earth in the centre of the pattern. Thus the earth often seems to be finally squeezed out. The spots of bare soil in the vegetationcover,sometimes surrounded by open cracks, exist in the subarctic and in the arctic zones (Schenk, 1955~).Single and densely associated hummocks or small palsas, with earth or peat (ash in Icelandic thufurs) in the core, are more frequent and become a familiar frost feature everywhere in the subarctic area. The surface of the bogs especially that of their higher parts in the strings, is clearly shown by the micro-relief.Big palsas are characteristic of thick peat layers and deep freezing so that thick, permanent ice
lenses are formed; sometimes even the subsoil of the peat cryoturbated upward. The surface of these high frost mounds also consists of small and large hummocks (called “pounikkos” in Finland) set closely side by side. They characterize the seasonalIy frostactive layer. According to my observations in Scandinavia, Iceland, Canada and Alaska these forms of frost pattern in the vegetation cover appear with optimal and maximal frequency in the subarctic area according to its climatic conditions. They afford high soil humidity, not by precipitation but by water absorbed by the organic material and by slowly melting ice cores below the vegetative cover of the frost mounds. There is no sharp delimitation of their occurrence in the south. In the north, however, the origin of string bogs is connected with the collapse of permafrost in the subarctic area and the boundary of originating string bogs runs parallel to the line of continuous permafrost (Schenk, 1964). The distribution of palsas furthermore shows the unstable climatic conditionswhich govern both developmentand collapse of permafrost. Big palsas are formed when frost penetration reaches a deeper part of the moore so that water accumulation and ice formation continue there while this is stopped in the surroundings.Palsas develop also from water accumulation and cryoturbation in clay material at the bottom of the moore. In northern Lapland they often are lined up along rivers (Ruuhijärvi, 1960) which afford long-lasting water supply during the freezing process. Palsas m a y originate in a sequence of hard,long winters and cold summers and m a y collapse during a period of mild winters and w a r m summers. The palsas collapse also if the protection cover of
TABLE 2.Frost pattern in the arctic and subarctic zones Arctic zone
Subarctic zone
R a w soil:
Soils with much organic material and with a vertical differentiation under a thick vegetation cover and its relics;
sometimes with raw humus material: tussock and hummocks; patterned ground in the frost active surface layer (stone nets,stone rings,stone and earth stripes,polygonal cracks); continuous permafrost; with giant polygons,ice wedge nets and pingos
formation of peat cover and peat bogs; hummocks, thufurs, pounikkos,earth spots; small and big frost bumps, palses; string bogs, with all combinations ofpalsforms (bumps) with permafrost cores (sporadic or discontinuous permafrost)
157
E. Schenk
their ice cores is damaged. Then the ice begins to melt quickly because of the improved heat conduction and earth often flows out. In contrast to the pingos which have the same origin as palsas there are indicators for unstable climatic conditions if w e refer this to the formation of permafrost at the annual temperature of -20 C to -60 C.
CONCLUSION Although permafrost has disappeared or is collapsingin the subarctic area the frost period is so long that ice
aggregation and permafrost occur in small and in large frost bumps (pals forms), also in string bogs with a typical frost pattern in the surface layer according to the soil formation through the vegetation cover and its specific water budget. Hummocks, thufurs, pounikkos, mounds, palsas, string bogs belong to the subarctic zone (Table 2). Thus w e have some typical subarctic frost patterns, but w e are not able yet to draw exactly the northern and southern boundaries of the subarctic frost-pattern zone. It m a y be that a more detailed knowledge of ill enable us to classify the arctic and subarctic soils w patterned ground of both the Arctic and the Subarctic.
Résumé Formation et structures du pergélisol
(E.Schenk)
Etant donné la différence existant entre la constante diélectrique de l’eau (82) et celle de la glace (2)’ le potentiel électrique des cristaux de glace est environ quarante fois plus élevé que celui de l’eau. Les molécules liquides sont donc attirées vers les cristaux de glace. Dès que le sol gèle, les molécules d’eau qui entourent les particules de sol sous la forme d’une pellicule d’eau d’absorption et d’eau osmotique sont fixées par les cristaux de glace. I1 se produit ainsi une différence de pression osmotique dans la pellicule d’eau, ce qui déclenche le mouvement de l’eau vers la zone de congélation. En conséquence, des couches et des lentilles de glace se forment et évoluent sous l’effet des qualités physico-chimiques du matériel pédologique. Par suite de la consommation de l’eau (humidité du sol), les couches du sol subissent un rétrécissement qui peut atteindre jusqu’à 20% de leur volume, et il se forme un réseau polygonal de fissures et de cas-
sures. Sur les parois de celles-ci,de nouveaux fronts de congélation commandent le mouvement de l’eau et la croissance des cristaux de glace givrés (pipkrake), qui portent souvent,sur leur sommet,des particules de sol arrachées aux parois des fissures. D e l’air est emprisonné entre les cristaux de glace en voie de croissance. C’est pourquoi la glace et la croissance des coins de glace sont caractérisées par des rangées de particules de sol et de bulles d’air. Les analyses chimiques,palinologiques et minéralogiques indiquent que cette glace provient de mouvements horizontaux de l’eau dans la couche subsuperficielle. I1 est donc exclu que de l’eau de fusion en provenance de la surface ait pénétré dans les fissures pour former les coins de glace par congélation rapide. Les coins de glace des sols réticulés, les polygones de toundra,les buttes (palses et pingos) et lestourbières réticulées (aapamoors) résultent donc du mouvement de l’eau et de la formation de glace au cours de phénomènes de congélation et de fusion. Ces phénomènes ont été montrés dans un film.
Discussion L. AARIO.The attempt of Dr. Schenk to give a general explanation for m a n y subarctic phenomena is very attractive. I believe that he has been right in m a n y cases. There are, however, some difficulties also. The rise of the strings and the “rimpis” are very much dependent on the annual freezing and melting of water, but it is in the first place a biological event, a differentiation in the growth of the peat in the strings and in the wet parts. I and m a n y students have never observed the deep cracks that should have been formed at the limit of the w e t and dry parts. Rimpis and strings grow in kermi-raised bogs side by side simulta-
158
neously, rimpi parts always at a slightly lower level (the different levels of the ash layer in the example of Dr.Schenk can be caused by this). The similarity between the microstructure in the laboratory experiments of freezing and the string-pattern of the aapa-fens is plausible, but roughly, a rather similar pattern is formed also in some calcareous and silicious hot springs, where the pattern has nothing to do with the freezing. The similarities are therefore not sufficient evidence to show the pure mechanical origin of the string pattern that should be caused by frost activity. The freezing mechanism alone can probably bring about m a n y
Permafrost and frost structures in the subarctic area
of the subarctic activities and plays a part in m a n y others, but the biological and hydrological factors mostly co-operate and are often the most important.
E. SCHENK.Deep, open cracks, I think, do not occur as a rule. I found open cracks between string and rimpi in a few places only: in the area of Jokkmok and in the Pelly river area north-east of Whitehorse. Here, and also in m a n y other places where cracks or slick surfaces were not observed, there was no doubt that the flat surfaces of the peat bogs were originally there when the ash layer was deposited. Their displacement occurred later as in the case along a tectonic fault. Often, between string and rimpi, a flexure or an overthrust m a y be found. Also the black, muddy layer and other layers geologically recognizable in the subsoil have been inclined upstream by a later mechanical but not a biological process. The growth of Sphagnum and other hydrophites developed after this, generally in a downstream direction because of the lower water level. The tectonic deformation of sections in a frozen peat layer is not the condition sine qua non. It m a y happen when permafrost and annual frost thaw faster from below than from the surface so that there is a solid in-between layer. If thawing reaches the bottom
Bibliography BLACK, R. F. 1954. Permafrost. A review. Bull. Geol. Soc. Amer., vol. 65, p. 839-856. BRIDGMANN, P.W . 1912. Verhalten des Wassers als Fliissigkeit und in fünf festen Formen unter Druck. 2. anorg. Chemie,vol. 77,p. 377-455. BUDEL, J. 1937. Eiszeitliche und rezente Verwitterung und Abtragung im ehemals nicht vereisten Teil Mitteleuropas. Peterm. Geogr. Mitt., Erg. H.229. . 1959. Die Klimazonen des Eiszeitalters. Eiszeitalter und Gegenwart,no. 1. BROWN, R. J. E. 1969. Permafrost as an econological factor in the Subarctic. Ecology of the subarctic regions. Proceedings of the Helsinki symposiumlEcologie des régions subarctiques. Actes du colloque d’Helsinki. Paris, Unesco. (Ecology and conservation/Ecologie et conservation, I.) EVERETT, D. H . 1961. The thermodynamics of frost damage to porous solids. Trans. Farad Soc., vol. 57, p. 1541-1551. FRENZEL, B. 1960. Die Vegetations- und Landschaftszonen Nord-Eurasienswährend der letzten Eiszeit und während der postglazialen Wärmezeit. I. Teil Allgemeine Grundlagen. Akad. Wiss. u. Lit.; Abhdl. Mathem.-Naturwiss. kl. Jahrgang 1959, Nr. 13. GRIGORIEW, A. A. 1930. Der ewige Frostboden und die diluviale Vereisung. In: (Der ewige Frostboden. Sammelwerk, hrsg. u. d. Akad. d. Wiss. USSR.),Leningrad. (In Russian.) HOPKINS, M . D.; K A R L S T R ~ MT. , N.V. 1955. Permafrost and ground water in Alaska. Washington, D.C.,United States Government Printing Office, p. 113-146. (Geological Survey Professional Paper 264-F.) KATZ, N. J. 1930. Zur Kenntnis der Moore Nordosteuropas. Beihefte zum Bot. Centralbl., vol. 16, part. II, p. 297-394.
--
of the frozen peat, m u d layers, peat layers with ice lenses, or other thawing layers, it gives rise to an underground movement of the drain water (accumulated by frost) which begins to move owing to any gradient of pressure or slope. This flow affects the surface layer which will settle down UP creep. The characteristic arc forms of the strings correspond to this fact. This can be derived also from calculations which are used in soil mechanics. The downstream, open arc forms result from higher speed in the underground whereas the upstream open arc forms result from higher speed on the surface. Therefore it is easy to recognize the primary effect of water movement and its relation to the arc forms of the strings in general. But there are, of course, biological effects in this ecosystem, which are favoured predominantly by frost activity. These differ wheD upstream or downstream of a string, they completely changed the original mechanic structure and in a way such that they are now reversed, hardly to be recognized in very old aapa moores and not at all in the raised bogs. Thus the biological factors and frost activity deformed the string bogs in accordance with the climatic and hydrological conditions in Finland, from north to south.
I Bibliographie KUBIENA, W. L. 1953. Bestimmungsbuch und Systematik der Böden Europas. Stuttgart, Enke-Verlag,392 p. LUNDQVIST, J. 1962. Patterned ground and related frost phenomena in Sweden. Suer. Geol. Unders. Avhdlg. Ser. C.,no. 583. OSTWALD;1960. In: Eggert (ed.), Lehrbuch der physikalischen Chemie. Stuttgart,Hirzel-Verlag,844 p. RUUHIJÄRVI, R. 1960. Ober die regionale Einteilung der nord-finnischenMoore. Ann. Bot. Soc. Vanamo, vol. 31, no. 1. SCHENK, E.1955a. Die periglazialen Strukturbodenbildungen als Folgen der Hydratationsvorgänge im Boden. Eiszeitalter und Gegenwart, no. 6, p. 170-184. . 1955 b. Die Mechanik der periglazialen Strukturböden. Abh. Hess. Landes.f. Bodenf., vol. 13. . 1964 Entwicklung und Zusammenbruch der Strukturen des Dauerfrostbodens. Report of the Sixth International Congress on Quaternary, Warsaw, 1961, vol. IV: Periglacial section. Lódz. S ~ R E N S E N ,Th. 1935. Bodenformen und Pflanzendecke in Nordostgrönland. Medd. o m Gronl.,vol. 93. SPETHMANN, H. 1912. Ober Bodenbewegungen auf Island. Zeitschr. Ges. f. Erdk. STACKELBERG,M.von. 1964. Die physikalische Deutung der Frostaufbrüche. Die Umschau in Wissenschaft und Technik, vol. 64, p. 68-71. TEDROW, J. C. F.; DREW, J. V. 1962. Arctic soil classification and patterned ground. Arctic, vol. 15, no. 2. TROLL, C. 1944. Strukturböden, Solifluktion und Frostklimate der Erde. Geol.Rdsch. vol. 34, no. 7/8. THORARINSSON,S. 1951. Notes on patterned ground in Iceland. Geograph. Ann. no. 3-4.
---
159
Frozen-groundmorphology of northeasternmost Norway Harald Svensson
INTRODUCTION From the point of view of Scandinavian geomorphology the northeasternmost part of Norway is of great interest,because this region is situated nearer to the great Euro-Asiatic permafrost belt than any other part of Scandinavia.For that reason there ought to be chances of finding transitional forms of permafrost, providing possibilities to study the processes of development and degeneration of frozen ground morphology. It could also be supposed that fossil forms m a y exist in this area, forms from earlier periods of colder climate, when the border of the continuous permafrost zone m a y have migrated westwards and touched northern Scandinavia. Because of the existence of traces of periglacial fossil structures in southern Scandinavia, it would also be of great interest to get a chance of comparison with the recent frost processes acting in this northerly district. These were the premises when w e started the survey of frozen ground morphology in this northern district in 1961 at the Department of Geography of the University of Lund. By means of aerial photographs a number of types of patterned ground and permafrost features were immediately found that initiated further work in the field. As there are plans of making a trip to northernmost Norway within the excursion to Finnish Lapland, I have chosen to give my contribution to the symposium the form of an introduction to the frozen-ground morphology of northeastern Norway and its problems.
PALSAS The most frequent feature of permafrost in the area is the palsa (Svensson, 1962~).The form of the palsas varies from singular, well-confinedindividual mounds
(Fig.1) to a complex of hillocks. The m a x i m u m height of a palsa observed in the district is 7 m.T h e lowest situated palsa bog is found at the 16 m level. The shortest distance to the sea-shoreis about 100 m. S o m e characteristics of the palsa morphology of the Varanger district are listed below. 1. M a n y palsas are surrounded by a small water body, the pals lag (Fig.2). It is thought to have been generated under the influence of the growing palsa in such a w a y that the surrounding bog surface is depressed under the weight of the palsa (cf. G. Lundqvist, 1951). 2.With the development of the pals lag the thermoerosional processes can grow more intense (cf. Fig. 2). 3.In the high palsas the peat cover is always dissected by vertical penetrating furrows, varying in width from 1-2 mm to a few centimetres. In summer the fissures are open to a depth of 40-60cm. Further down they are filled with an ice vein. T h e possible causes of fissuring are: (u) an interior breaking force generated at the growing of the palsa, (b) thermal contraction of the surface layers at low winter temperatures or rapid cooling of the frozen peat, (c) drying. It seems that the last possibility should be left out, as n e w fissures are observed in early spring, when the surface layers are still very wet. Probably the second possibility (thermal contraction) is the most valid. Anyhow, a wedging activity is responsible for the first widening of the fissures, caused by the increase in volume by freezing of water in the deeper parts of a fissure. 4. For the erosion of palsas the dissection of the surface is of great importance. Whole blocks of the insulating peat cover will burst or slide d o w n from the flanks of the palsas (cf. Fig.2). 5. Small,wind-erodedsurfaces without any vegetation sometimes occur o n the high palsas. 161
11
H.Svensson
FIG.1. Two palsas, 3.5 m and 5 m high, at the inner part of the Varangerfjord.
FIG.2. Palsas in a stage of erosion.
T H E SOIL STRUCTURE OF A P A L S A T h e palsas of the inner part of the Varangerfjord contain in general a core of frozen minerogenic soil below the peat mantle. F r o m the study of a vertical section to a depth of 2 min one of the highest palsas (Svensson, 1964a)the following characteristics can be quoted: 1.Because of the open fissures of the peat mantle thawing h a d been able to continue into the minero-
162
genic material to a depth of 60 c m below peat, i.e. 115 crn below the surface. 2. T h e minerogenic top soil below the peat consists of a layer of sand with s o m e gravel and sporadic stones. This layer is largely concentric with the surface of the palsa and surrounds the rninerogenic core proper, which is composed of finer fractions, mainly silt. 3. T h e permafrost surface is approximately conform-
Frozen-ground morphology of northeasternmost Norway
able to the palsa surface, gently sloping along the longitudinal axis of the palsa and with a steeper gradient near the flanks. The ice (or water) content (per cent of dry soil) was determined as 71.1 per cent and 73.3 per cent respectively for two pieces of the frozen core. The water content largely exceeds the capillary saturation of the soil. Water has been supplied from the soilbeneath (cf. Beskow, 1935; G. Lundqvist, 1951) on freezing, thus causing the growth of the palsa. The structure of the frozen soil shows the following characteristics. 1. The water content has crystallized in lamellae of pure ice which give the cutting a bluish colour. 2.The ice is not confined to horizontal or subhorizontal layers. On the contrary, vertical and subvertical lamellae frequently occur, giving the structure a more uniform appearance. 3. Biconvex ice lenses appear. They are connected with ice layers upwards and downwards by vertical or perpendicular ice laminae. 4.The frozen soil is divided into aggregateswith mainly plane surfaces. W h e n the separating ice laminae and the ice of the soil melts, the aggregates preserve this plane surface for some time. The occurrence of degenerated and collapsed palsas in an area can sometimes give the impression of a rapid amelioration of climate. As the palsas run through different stages, however, it is quite normal that degenerated palsas occur without having any climatic bearing. T o come to a conclusion about climatic changes from the palsa morphology it is necessary to study and compare conditions in different palsa areas. As is well known glacier variations are often used to prove climatic fluctuations.As to the questions of the palsas as indicators of climatic fluctuations it is worth noting that there are different factors of the climate (specially the winter climate) acting at the genesis of palsas on one hand, and glaciers on the other. Extremely low winter temperatures are, for instance, of great importance for the development of palsas but of no real effect for the glaciers. For a glacier snow is a condition of life,but for the genesis of palsas a snow cover is only a disadvantage.
THE THERMAL STRUCTURE OF A PALSA A study of the temperature regimen of a palsa in the Varanger district has been performed by two assistants from our institute, Lindqvist and Mattsson. By means of an isopleth diagram from their investigation (Fig.3) some facts can be summarized (Lindqvist and Mattsson, 1965, p. 43-46). 1. T h e palsa is dominated by temperatures below O0 C. The temperatures exceed O0 C only during
the summer and early fall, and then only in the uppermost peat layer. 2. Permafrost occurred with certainty from the depth of 1 m and downward. 3. The temperature extremes were greatest in the surface layer and were suppressed and delayed with the depth under the surface. The high summer temperatures are concentrated entirely in the surface layer of the palsa. 4.As a result of, first, a relatively slight heat surplus in the palsa surface (cool summer climate) and, second, a pronounced heat consumption in the upper surface of the frozen layers in the melting of the ice here,an effective barrier is formed against a heat wave penetrating into the palsa after the summer. 5. During the winter, on the other hand, the low surface temperatures result in a pronounced penetration of a cold wave into the palsa, after the upper peat layers have frozen. There no longer exists any barrier in the form of heat energy liberated in the freezing. 6.The low temperature conditions of winter form at a lower level a concerted winter cold wave. Its effect can still be traced at a depth of 4 m but here it is markedly reduced in strength and occurs first in the middle of the summer. 7.Immediately after the snow melting the temperature rises generally in the upper palsa layers, especially in the peat layer. Nevertheless freezing temperatures characterize all layers for another month or so. 8.The transition to positive temperatures begins in the palsa surface and continues only slowly downward as a result of high heat consumption in the ice melting. 9.Below a depth of 1 m the palsa is characterized by a very stable thermal structure and at 4 m the temperaturevaries during the year between the extreme -0.40 C and -1.20 C. 10.The annual mean temperature at 2 m has been calculated to -0.470 C, at 3 m to -0.760 C and at 4 m to -0.700 C. The relatively low annual mean temperatures of the two deepest measuring points are possibly affected by the weather conditions of earlier years. The thickness of the snow cover on the heath and on the palsa is also shown in the figure. F r o m the diagram it is seen that during the winter the palsa is characterized by only a slight snow cover, whereas the flat heath is distinguished by a heavier accumulation of snow (Lindqvist and Mattson, 1965). Before I leave the palsa morphology I would like to mention the existence in some plateau areas of low, flat hillocks and their degeneration stage, small, ring-ridged,circular or oval lakes (Fig.4). The diameters vary from 15-35m. The form occurs in flat depressions. The ground consists of clay and silt. No
163
H.Svensson
n
1
O
-1
-2
I_
,<--., , . --_-.. ., ,_
-3
-4
FIG3. The temperature conditions in a section of a pals shown in isopleths.(Measurements were made once a week.) At the top of the diagram the snow depth on the flat heath outside the pals and on the pals are represented by the unbroken and the broken line respectively.At the bottom of the diagram the temperature in the pals bog is shown at the depth of 1 m.
peat cover is present. This form of frost mound seems to be acquainted with the palsas and m a y be a transitional form to the pingo. T h e flat domes have probably originated at the growing of hydrolaccoliths in the oversaturated soil of the depressions. F r o m the high mountains in the Abisko district of Sweden, R a p p and Rudberg (1960) report a type of ring-ridged lakes of similar character but developed in coarser material. By means of aerial photographs some individual forms have been detected that are very near to the real pingo. As in the case of the ring-ridged lakes the forms shown on aerial photographs are delineated by a closed contour (Svensson, 1964b). These forms,however,stand out as true hummocks,the tops ofwhich are depressed, giving them a crater-likeshape. This pingo-like form or form group is found in high positions, not exactly in top situations but on slightly sloping ground in the upper part of mountain slopes. In Greenland Müller (1959)observed pingos of the open system type on slopes with an angle of up to 80. As stressed by Dylik (1963)the thermokarst forms discussed in the literature, and especially the pingos,
164
generally are recent, not fossil phenomena. This fact is a disadvantage for an effective comparison with clearly old or fossil variants as in the case of these north-Scandinavian forms. Via these pingo-like forms I have passed over to the fossil forms of permafrost that are observed in northeasternmcst Norway.
FOSSIL ICE-WEDGE POLYGONS The most frequent, but yet not specially widespread, form of fossil permafrost features is the ice-wedge polygons (Svensson, 1963). They can be found especially on raised deltas or marine terraces on the coast and on glacifluvial deposits in the interior (Fig.5). The first area in which this ground pattern was observed was at Bussesund at the outer part ofthe Varangerfjord (Svensson, 19623). The pattern is outlined in the ground by linearly extending furrows, 5-30 c m deep. The furrows generally contain peat and are then more or less overgrown by a dense vegetation which obtains some shelter in the shallow furrows.
Frozen-ground morphology of northeasternmost Norway
FIG.4.A ring-ridgedlake on Finnniarksvidd a. Oblique aerial phiOtC)graph from abouit 200 m height.
FIG.5. An outwash delta at Björnevatn (Porsangerfjord) with a network of polygons. Approximate scale 1 :10,000 (enlargement from the original scale 1 :20,000). Aerial photograph by Wideröes flyveselskap, Oslo.
165
H.Svensson
A t Bussesund t w o furrows of the polygon net are cut through in a gravel pit, whereby casts of icewedges corresponding to the furrows are exposed (Fig.6). T h e ground pattern of Bussesund can b e regarded as the type locality of the ice-wedge polygons of the coast district. I would, however, like to mention a tetragonal pattern of quite another character.In the aerial photographs from the 500 m high mountain Bugtkjölen a pattern stands out with the s a m e geometric characteristics as the other patterns, a uniform orthogonal
network. In the ground, however, it is quite different a n d has the appearance s h o w n by Figure 7, i.e. a block field of quartzitic sandstone, completely without vegetation (excepting lichen). It would certainly have been quite impossible to observe the pattern o n the ground, if w e h a d not k n o w n from the aerial photograph that it did exist. T h e lines of the pattern consist of indistinct furrows in the rough block surface. W e have tried to reach the solid rock base of the block field by picking up stones. Digging is of course impossible. A t the depth of 120 c m below the surface
FIG.6. A fossil ice wedge from the p olygon area of Bussesund (VarangerfjoNrd).
FIG. I. The patterned block field of Mt.Bugtjölen (the Varanger Peninsula).
166
Frozen-ground morphology of northeasternmost Norway
w e had not yet reached the rock, and working further d o w n in this w a y was impossible. Notable is the absence of fine-grained material in the block field. In my opinion the uniform network must be considered as a frozen ground pattern. In the development of the pattern by frost-wedgingprocesses the regular fracturesystems of the quartzite have however exerted a distinct influence as to the directions and the s y m m e try of the pattern. Ice wedges and ice-wedgenets in solid rocks are known from the investigations of Washburn (1950), Büdel (1960) and Rudberg (1963).
THE AGE OF THE FOSSIL ICE-WEDGEPOLYGONS The age of the patterns is an interesting problem in the investigations of the fossil ice-wedgepolygons of northernmost Norway. In the coastal district the polygons occur on raised deltas at the highest shoreline and on the nearest marine terraces below that level. The polygons are
thus younger than these shorelines which are of late glacial age. Another point of attack of the problem is the fact that the frost processes which initiated the formation of the ice wedges must be older than the organic material in the furrows. There are, however, great difficulties in obtaining representative samples for radiocarbon dating from the shallow peat layers of the polygon furrows. For the Bussesund locality radiocarbon dating has been performed for the lowest peat layer of two different furrows, giving the age of 4890 + 70 years B.P.and 4400 &90 years B.P.respectively. Probably the age of the polygons m a y be thus determined, by stating a m a x i m u m age from the shoreline chronology and a minimum age from the radiocarbon dating. More datings are, however, needed to prove this theory and to give possible conclusionsregarding the place of the polygons in the late glacial or postglacial time-scale and their climatic evidence for the time of genesis.
Résumé Morphologie des sols gelés dans ,!’extreme nord-est de la Norvège (H. Svensson) Les études ont consisté en interprétations de photographies, suivies de recherches sur le terrain. Les régions de pergélisol n’ont pu être encore délimitées en raison du manque de photographies aériennes pour certains districts. On ne trouve de pergélisol que dans des régions isolées,la plupart du temps marécageuses, mais parfois aussi sur les hautes montagnes. Des formes fossiles, contemporaines de périodes à climat plus rigoureux,se rencontrent surtout sur les terrasses marines et les deltas surélevés.
F o r m e s actives L a forme la plus typique est celle de la palse. Dans de nombreuses régions côtières de l’extrême nord de la Norvège, les palses contiennent un noyau de sol minérogénique (limon et argile) recouvert de tourbe. Les fondrières à palses occupent des dépressions où se sont accumulés des sédiments marins et fluvioglaciaires pendant la période de déglaciation. En raison de sa forte capillarité et de la grande quantité d’eau qu’il peut retenir, le sol à grain fin a joué un rôle dans la formation des palses et a fortement contribué à les développer. L a hauteur maximale des palses, au fjord de Varanger, est de 6,5 mètres. L a fondrière à palses la moins élevée est située à 16 mètres au-dessus du niveau actuel de la mer. Les palses élevées sont souvent fortement érodées.
L’érosion est accélérée par le fait que la couverture de tourbe qui les isole est souvent abondamment entamée par des sillons. Dans certaines régions,les palses sont déjà à un stade avancé de dégénérescence, qu’on peut attribuer à des causes climatiques. D’autre part, on constate aussi l’apparition de nouvelles palses. En mesurant la température à l’intérieur de certaines palses, on a découvert que la réserve de froid dans le noyau n’est pas particulièrement importante. A l’intérieur du pays (province de Finnmark) on trouve parfois des monticules gelés non recouverts d’une couche de tourbe isolante. I1 est probable qu’ils résultent du développement d’un hydrolaccolithe dans le sol à grain fin des dépressions à fond plat. On les remarque davantage lorsqu’ils en sont au stade de la dégénérescence, c’est-à-dire quand ils ont pris la forme de mares circulaires à bords relevés.
F o r m e s inactives Parmi les formes inactives, les plus communes sont les polygones à coins de glace. On trouve aussi des formes effondrées qui rappellent les pingos. Afin de découvrir l’âge des polygones à coins de glace fossiles on a appliqué la méthode du carbone 14 à la tourbe tirée d‘un interstice entre ces polygones. L’âge de cette tourbe était de 4 890 f 70 ans. On se prépare à effectuer les nouvelles datations nécessaires pour minimiser les erreurs inhérentes au choix de l’échantillon dans la mince couche de tourbe.
167
H.Svensaon
Discussion V. OKKO.T w o years ago Dr. Thorarinsson showed me a tundra polygon at Jokullheimar, on the western border of Vatnajokull, Iceland. The polygon occupies a spot that was deglaciated 35 years ago at the most. N o permafrost exists there.
H. SVENSSON. Last summer I had the opportunity to visit the same area of Iceland during an excursion, conducted by Dr. Thorarinsson. In m y opinion the polygons of
Jokullheimar are not ice-wedge polygons (permafrost phenomena) but polygons created by seasonal frost.
Bibliography BESKOW, G. 1935. Tjälbildning och tjällyftning m e d särskild hänsyn till vägar och järnvägar. Sver. Geol. Unders., Ser. C.,no. 375. BUDEL, J. 1960. Die Frostschutt-Zone Südost-Spitzbergens. Coll. Geogr., vol. 6. DYLIK, J. 1963. Traces of thermokarst in the Pleistocene sediments of Poland. Bull. Soc. Sci. Lettr. Lodz, vol. XIV, no. 2. HOPPE, G.;OLSOON-BLAKE,I. 1963. Palsmyrar och flygbilder. Ymer Arg., vol. 83. LINDQVIST,S.;MATTSSON,J. O.1965. Studies on the thermal structure of a pals. Lund Stud. Geogr., Ser. A. vol. 34, LUNDQVIST, G. 1951. En palsmur sydost om Kebnekaise. Geol. Fören. i Stockh. Förh. vol. 73, no. 2. LTJNDQVIST,J. 1962. Patterned ground and related frost phenomena in Sweden. Sver. Geol. Unders. Ser. C, no. 583. MULLER, F. 1959. Beobachtungen über Pingos Detailuntersuchungen in de kanadischen Arktis. Medd. om Grönl., vol. 153, no. 3. RAPP, A.;RUDBERG, S. 1960. Recent periglacial phenomena in Sweden. Biul. Peryglac., vol. 8.
168
A. RAPP.M a y I ask a question concerning s o m e of the pictures shown of the details of ground-ice structure in various cuts in mineral soil? If these structures were photographed close to the temporary frost table in summer it is possible that they will thaw later in the summer and consequently belong to the “active layer” and not to the permafrost?
H. SVENSSON.The pictures were taken in the middle of August 1964. Ice laminae and ice lenses of the same character as shown by the picture occur throughout the minerogenic core of the palsa. They can hardly be suspected to be a sort of annual ice layer.
/ Bibliographie __
. 1964. Studies on periglacial phenomena in Scandinavia 1960-63.Bid. Peryglac., vol. 14. ; GUSTAFSSON,K.;JOBS, P. 1962. Iskilar i Padjelanta? Ymer, 1962. RUDBERG, S. 1963. Morphological processes and slope development in Axel Heiberg Island, Northwest Territories Canada. Nachr. d. Akad. d. Wiss.in Gött. II. Mathem. Physikal. KI. Jahrg. 1963, no. 14. SVENSSON, H. 1962a. Några iakttagelser från palsområden. Norsk Geogr. Tidsskr., vol. 18. 1962b. Note on a type of patterned ground on the Varanger peninsula, Norway. Geogr. Ann., vol. 44. . 1963. Tundra polygons. Photographic interpretation and field studies in North-Norwegian polygon areas. Norges Geol. Unders. Arbok, 1962. . 1964a. Structural observations in the minerogenic core of a pals. SV. Geogr. Arsb. Arg.,vol. 40. . 19643. Traces of pingo-like frost mounds. L u n d Stud. Geogr., Ser. A, vol. 30. WASHBURN, A. L. 1950. Patterned ground. Rev. Canad. Géogr., vol. IV, no. 3-4.
-_
-
-_ -_
Remote sensing as an ecological tool Philip L.Johnson
ECOLOGICAL NEED FOR REMOTE SENSING The world problem of our time has been h o w to manipulate the world’s resources and environments to gain a better life; our successes are phenomenal in proportion to resource exploitation. The problem of coming generations is h o w to manipulate populations to conserve a cherished standard of living. It is very clear ill review the ledger of the world’s to those w h o w people and resources that a population explosion is upon us (for example see Mudd, 1964). Even at the present moment, over half the world’s population is underfed,unhealthy,poorly housed,uneducated and in general underprivileged. War, famine and pestilence are fixed in our education as Malthusian antidotes for over-population. But already the major h u m a n diseases can be controlled. Wars in recent history have never effectively stemmed the growth of populations (Borgstrom, 1965). It is sobering to note that more people were born in this world in 1965 (65 million) than died in all of the Second World W a r (50 million). Famine is more prevalent than w e realize and perhaps more closely related to economic and political unrest than w e can document. Ecology is one area of investigation certain to contribute to the future for this is the study of organisms versus their environment-the modern dilemma. Ecological thinking is already evident in m a n y areas of technology by one n a m e or another. M a n y ecologists feel that this discipline has an opportunity to bridge the gap between academia and the very real and extensive resource and environmental problems before us. Without belabouring an argument for environmental biology, I should like to propose four kinds of ecological inquiry that are susceptible to remotesensing techniques. Remote sensing is the non-contact
acquisition of information in any portion of the electromagnetic spectrum from aerial platforms. This information m a y apply to: (a) inventory and mapping of resources;(b) quantitizing the environment; (c) describing the flow of matter and energy in the ecosystem; and (d) evaluating change and alternative solutions for management of ecosystems. Aerial photography and remote-sensorimagery have the potential in ecology that the spectrophotometer has demonstrated in physiology. In fact, aerial photography m a y be thought of at the ecosystem level as the converse of the electron microscope in molecular biology; each depends upon the spectral reflectance, absorption, and transmission characteristics of the respective samples. The development of imaging technology for displaying signals from various portions of the electromagnetic spectrum including ultra-violet, visual, infra-red,and microwave bands provide n e w information about vegetation and its environment. INVENTORY AND MAPPING
Solution of m a n y resource problems depends on adequate assessment of physical and biological characteristics integrated over areas of a few square feet to thousands of square miles. Maps of these characteristics are logical forms of communication. This approach is well illustrated by advances in geological exploration including magnetometer surveys and in forest inventories. Important biological properties of ecosystems potentially measurable by remote-sensing techniques, singly or in multispectral combinations, include: leaf area, stem volume, species diversity, weight and chlorophyll content of vegetation; kind, density and biomass of larger animal populations; heat, water vapour and carbon dioxide fluxes of the Earth’s surfaces; water content of soils and vegetation; and depth and density of snow.
169
P.L.Johnson
QUANTITIZING T E E E N V I R O N M E N T
In addition to what is it and where, most resource problems require quantitative data. In some cases even estimates of magnitude applied to large areas are sufficient for meaningful planning. In more sophisticated treatments, remote sensing, particularly the narrow bandpass instruments, can generate accurate numbers. The recent development of a laser system as an airborne profilometer for studies of microtopography appears capable of portraying changes in elevation on the order of a few centimetres (Rempel and Parker, 1965). Radio ice-sounding techniques permit measurements of the ice/bed-rockinterface through ice thicknesses of 2,500 m (Rinker et al., 1966). The recent development of heat- and magnetic-sensing tools, sensitive at distances on the order of a degree, is a product of military necessity which is rapidly finding applications in the civilian economy. F L O W OF M A T T E R A N D E N E R G Y
Failure to understand the processes active in nature is frequently the cause of faulty resource planning. T o appreciate function as opposed to structure in ecosystems requires the study of processes and interaction between different organisms as well as between organisms and environment. Frequently, measures of metabolic activity are desired. These are the most difficult answers to obtain from remote sensors as well qs on the ground. Nevertheless, detection by airborne sensors of metabolic products or physical changes caused by active biological processes are important clues for inferences and conclusions about the direction and quantity of matter and energy flow in ecosystems. EVALUATING C H A N G E A N D ALTERNATIVE SOLUTIONS
Alteration of natural ecosystems is, of course,manifest in all resource problems. Without change-that is depletion, erosion,pollution,accrual, or epidemic-the problem is seldom recognized. This, perhaps, is the easiest type of information to procure by repetitive aerial surveillance and has been exploited with photography in the visual wavelengths. Until repetitive aerial photography is widely practised (and financed!), longterm and widespread change, m a n caused or natural, will be difficult to assess. Once trends of change or the consequences of our technology are evaluated, alternatives can be developed from the same data. This is the most promising application for photography from spacecraft.
170
PHYSICAL BASIS FOR REMOTE SENSING If w e are to utilize the informationcontent of airborne sensor systems, it is essential that w e appreciate the fundamental energy and matter relationships responsible for the images to be analysed. Various portions of the electromagnetic spectrum are exploited for information about the matter which emits or reflects energy quanta (Fig.1). There are three basic types of systems available for remote sensing from airborne or satellite platforms. All three are usually processed to present two-dimensional or pictorial displays: photography in the near visible spectrum 300 to 1,000 mp; optical-mechanical scanners in the infra-red wavelengths, 1 to 40 p; and passive microwave and radar for selected bands from 1 mm up to 1 m. Radar is an active system in the microwave frequencies in which the appropriate energy is generated in the aircraft and directed toward the ground. The radar return of reflected signal is captured by an antenna specific to that wavelength. T h e intensity of the energy returned is primarily a function of terrain aspect relative to beam direction and secondarily related to the dielectric properties of the reflecting material. The advantages of radar imaging are its independence of weather and diurnal conditions. Potential information is a function of the matter/ energy interactions peculiar to the sample. Energy absorption, emission, scattering and reflection by any particular kind of matter are selective with regard to wavelength,and are specific for that species of matter, depending upon its atomic and molecular structure (Colwell et al., 1963).The actual signal recorded differs basically in frequency, or intensity but it is also a resultant of (a) attenuation by the intervening atmosphere and (b) the fidelity of the electro-mechanical system employed. E N E R G Y A N D P L A N T RELATIONSHIPS
Most of the earth’s land surface is mantled by some kind of vegetation. Primarily, it is the signature of foliar surfaces of vegetation that is received by remote sensors. W h a t happens to solar energy incident upon leaves? In the visible and near infra-redwavelengths reflection or emission from the leaf’s cuticle and epidermis is relatively minor and not selective. In the red and blue ends of the visible spectrum 80 per cent or more of the energy is absorbed by chlorophyll, while perhaps 40 per cent of the green wavelengths are reflected. Energy from the near infra-redis little affected by chloroplasts, but is greatly affected by spongy mesophyll tissue. This tissue appears white when the chlorophyllous palisade tissue is removed or chlorophyll removed. Energy which penetrates to the mesophyll and is reflected from it is of greater intensity
Remote sensing as an ecological tool
P H O T O G R A P H Y
+
3000Å
3900Å
4900;
6000Å
6600Å
X-Rûï
7600Å
i0,OOOA
LONG W A V E LOW F R E Q U E N C Y
HIGH FREQUENCY AY
c
AC
AUDIO
UV
W A V EL EN GTH
TMOSPHERIC TRANSMISSION
GY-MATTER RACTIONS
DETECTORS
SENSORS
FIG.1. The electromagnetic spectrum is the basis for the development of remote sensors.(Modified from Colwell et al.,1963.) in the near infra-red than in the visible wavelengths (Colwell etal., 1963).Energy absorbedin one wavelength and emitted in another,for example fluorescence,m a y be important to both ultra-violet and near infra-red detectors; however, the quanta lost from plants as fluorescence is probably less than a few per cent of the available light energy. A change in plant vigour m a y often result in loss of turgor in foliar mesophyll tissue. As a result, a great loss in reflectance of near infra-red energy occurs long before any change can be detected in the visible wavelengths-that is, before any pigment transformation. An excellent discussion of the spectral properties of plants was published by Gates et al (1965).
T h e foregoing discussion suggests potential application of colour and infra-red emulsions to problems of disease detection, to effects of herbicides, to internal water stress, to radio-active radiation effects, and to the pigment structure of ecosystems. PROCESS OF I N F O R M A T I O N EXTRACTION
Certain limitations are imposed on the image analyst. First,he is restricted to his experiences or other groundbased information about (a) analogous areas and the academic discipline to which he is applying the analysis. No one is potentially in a better position to extract information about vegetation from aerial
(a)
171
P.L.Johnson
ORAIH
RESOLUTION
II m m 4
DISTAI CE
COWTRAST
DISTAI CE
ACUITY
\c L
C2 54 ~ ~
JJ DISTANCE
DISTA W C E
FIG.2. Optical density expressions of four film properties. photography than ecologists. Second, he must k n o w the scale of the imagery for both quantitative and qualitative analysis. Size, shape, and pattern are ill have different partially a function of scale and w meanings in satellite photography than in photomicrographs. Third, the interpreter must depend on observations of tone, texture, pattern and resolution. Tone or density variations m a y be measurable in gradations of a grey body or they m a y be expressed in terms of chroma and hue, tristimulus values, Munse11 colours, or other colour space notation in the ilm records case of non-grey bodies. Panchromatic f nearly 200 distinguishable shades of grey, whereas colour photography m a y record up to 20,000 separate combinations of chroma and hue (Fig.2). Photographic pattern or texture and spatial arrangements are frequently the most important clue to the identification, origin, or function of objects imaged. Resolution,the ratio of object size to grain size or noise
172
level, has improved tremendously with the increase in emulsion sensivity and the improvement in optical systems. A photographic pattern m a y be simple or complex depending on the number of variables. If w e can consider the potential information to form an organizational hierarchy (Fig.3), then the most general categories are immediately apparent (Johnson, 1966). A5 the number of variables increases, confidence in our inferences decreases and the amount of outside information required is greater. Photo-reading or species identification is the first step in inductive reasoning. Table 1 is an example of deductive criteria applied to the interpretation of vegetation.
Remote sensing as an ecological tool
PHOTO PATTERN INFERENCE LIMITATION ORGA NZZATZONAL HZERA RCHY
DEDUCTZVE REASONING
%CONFZDENCE
*
vegetation
1
‘s
$
* INDUCTIVE REASONING
TABLE1. Deductive approach to the interpretation of
WOUTSZDE ZNFOtMATZON
FIG.3. A concept of potential information in photographs and the limitations of interpretation with increasing pattern complexity.
SUBARCTIC VEGETATION PATTERNS In contrast to the abrupt vegetation gradient associated with the topography of temperate zone subalpine areas, the latitudinal vegetation gradient in the Subarctic is frequently extended over m a n y miles. Examination of this subarctic formation in North America is also difficult because of the limited road system. Aerial photography may greatly assist the separation of local environmental effects from regional vegetation gradients. In interior Alaska an extreme continental climate prevails in the northern tension zone of tree growth. A very complicated vegetation mosaic (Fig.4) is the result of a long history offire superimposedon alluvium deposited by ancient and recent river meanders. T h e present vegetation pattern is a mixture of forest, shrub, and herbaceous stands in various stages of plant succession. The stability of the pattern appears proportional to the interaction between: (a) the stability of the frost table and the resulting drainage condition; (b) the elapsed time since the last fire or flood disturbance and the intensity of the burn or duration of the flood; and (c) the age of the organic or mineral substrate which is usually a function of topographic position. The area of interest on Y u k o n Flats is roughly 60 x 100 miles, containing a few small villages accessible primarily by aeroplane; this .area has been proposed as a hydroelectric site. It would be very difficult to even describe the vegetation of such a large and remote area without the use of aerial photographs. At each of 43 stands, plots 10 x 50 m have been described,harvested and weighed (Johnson and Vogel, 1966). Height, crown diameter, and twig growth of the dominant woody species were measured (Table 2).
1. General pattern types: satellite photography and smallscale mosaics Discrimination: forest shrub grass
cultivated barren mixed 2. Pattern features:mosaics and small-scaleprints Discrimination: areal extent and distribution boundary conditions complexity 3. Pattern elements:stereo pairs A. Tone and texture: spectral class (optical density) arrangement complexity and uniformity B. Site characteristics: landscape, natural or cultivated soil and rock type drainage type slope and exposure C.Structuralcharacteristics: canopy configuration crown types (diversity) density height 4. Pattern components: large-scalephotography A. Local adjustments and interactions: causative factors and origin B. Composition:species diversity C. Dynamics: interactions, processes, successional trends
The results of this study will establish the relationship
of aerial photo-pattern elements to specific plant communitiesin a large subarctic area (see Fig.16). P H O T O G R A P H I C INTERPRETATION
Information interpreted from aerial photographs is frequently best presented in m a p form. Three types of vegetation maps were prepared from 1 :5,000scale infra-red photography to indicate the photographic pattern of some c o m m o n subarctic vegetation types in interior Alaska. The vegetation types m a y be identified with typical stand patterns without commitment to boundary or ecotones between types (Fig.5). Figure 6 illustrates a characteristic vegetation pattern along the Steese Highway representing foothill forests and their response to slope aspect. A third kind of display is a strip m a p based on a photographic mosaic. The first photo-basem a p (Fig.7) is a transect across Birch Creek to a high terrace with thermokarst lakes.
173
P. L. Johnson
T h e vegetation patterns are typical of tributary drainages throughout the area. T h e second photo-base m a p (Fig.8) extends from Circle City on the Y u k o n River along the Steese H i g h w a y a n d represents a transect from the present flood plain with cottonwood, Populus balsamifera, to a well-drained valley slope covered with white birch, Betula papyrifera.
In constructing these vegetation maps, thirteen vegetation units were recognized from photographic tone a n d texture associated with ground information. These units were often intermixed a n d could be subdivided or augmented; however, these types were easily recognized o n the photography and ground sampling did confirm their existence. T h e m a i n criteria for
FIG.4.Aerial mosaic demonstrating a photographicpattern of some common subartic vegetation.
174
Remote sensing as an ecological tool
TABLE2. Net primary production in selected ecosystems Ecosyatem
Dry glm'
Location
Standing
Picea glauca Picea mariana Betula papyrifera Populus iremeloides
Yukon Flats Yukon Flats Yukon Flats Yukon Flats
Arctic tundra Alpine tundra W e t Meadow Geum Meadow Cares Meadow
Barrow, Alaska Wyoming Wyoming New Hampshire
Picea abies Betula verruCosa
U.S.S.R., Sweden,
70 23
Pseudoisuga menseisii
United Kingdom Washington
33
Prairie Oak Savanna Oakwood Maize
Minnesota Minnesota Minnesota Minnesota
Mixed Dipterocarp TemperateEvergreen Tropical Deciduous Tropical Evergreen Evergreen Gallery
Thailand Thailand Ghana Congo Thailand
Wheat, world average Rice,world average Pine plantation Algae culture, optimum
World World United Kingdom Japan
Sweden, U.S.S.R.
84 108 45
48
5 099 2 949 4 822 3 865
Per
Source year
73 31 102 71 40-97
Bliss, 1966
22 923 5 680
339 251
Ovington,1962
9 290
267 93 526 819 946
55
30
delineation of these units was a combination of tonal contrast, texture, topographic position and field experience. Identification of most species was simplified by stereo study; in fact, crown shape, height, and topographic position can be confidently inferred only with stereo examination. Accumulation of organic matter is a significant ecosystem process in northern latitudes. The process of bydrarch succession is well k n o w n and often described. F r o m a sequence of aerial photographs over some boreal lakes near Northway, Alaska, it m a y be possihle to measure the rate of advance or retreat of the rooted aquatic vegetation. By planimetering the open water or measuring the width of the bog mat onphotographs taken every few years, w e can describe the amount of change since 1941.Based on fieldwork in June 1964, w e k n o w the structure and species composition of both the bog community and the aquatic ecosystem (Figs. 9 and 10). The results of this brief study indicate (a) the types of lakes or portions of basins in which bog succession
Pieper, 1963
112 202 176
90
50 18
Johnson and Vogel, 1966
5 110 18 100 26 160 12 290 29 520
Johnson (unpubl.) Bliss, 1966
Ovington, et al., 1963
Ovington, 1962 523 683
344 497 3 180 4 530
Odum, 1959
is rapid, static or regressing, (b) the nature of the ecosystem in each instance, and (c) the variations in lake depth, chemical composition, temperature profile, and other physical parameters apparently associated with various rates of carbon fixation. R A D A R INTERPRETATION
Imagery from a radar system flown 22 October 1956 across Yukon Flats (Fig.11) iliustrates the landform units of this region in a swath approximately 40 by 160 miles. This negative print (Fig.11) reproduced strong signal returns as dark tones, low signal returns as light tones. The most striking patterns are related to topography and drainage systems because the signal was sensitive to changes in slope with respect to the aircraft. More subtle tones can be associated with vegetation patterns such as alternating muskeg and forested stands on the river terraces (A). Drained lakes, marsh, and vegetation in cut-off meanders often provide a weak signal (B and Fig.12).
P.L.Johnson
FIG.5. A vegetation type map without boundaries.Typical stand patterns are identified: 1, Gravel bar; 2, Willow (Salk); 3, Cottonwood (Populus tacamahacca); 4,White Spruce (Picea glauca); 5,Black spruce (Picea mariana); 6,Muskeg (LedumBetula); 7, Birch-Spruce (Betula papyrifera P. glauca).
-
-
Particularly extensive burn patterns are displayed (C) north of Fort Yukon. A number of water bodies in Figure 11 reflected strong signals and others weaker signals. It is likely that in October ice had formed in the areas of higher signal return while open water areas were strong absorbers of radar energy. Imagery fromradar flownin August 1962 is comparablewith vertical panchromatic photography in Figure 12. T w o points are evident: high energy returns were obtained from filled meanders (arrows)although two lakes at the bottom of the panchromatic illustration with floating vegetation were not distinguished by this radar from other water bodies, and the panchromatic vegetation pattern associated with various forests and muskeg vegetation is not distinguishable on the radar imagery. The bridge and land form configuration are not as evident on the panchromatic print.
176
T H E R M A L INTERPRETATION
Radiation in the thermalinfra-redwavelengths emitted
by terrain is captured and displayed in a manner distinctly different than near infra-red radiation (700900 mp) that energizes infra-red film.Thermal imagery is obtained with a passive system utilizing an optical-mechanicaldevice in which a rotating mirror scans the terrain beneath an aircraft in a continuous strip transverse to the flight line. Radiation emitted by the terrain is reflected and focused on a photoelectric detector in a pre-selected spectral band. These detectors capitalize on atmospheric windows at 3.5-5.5 and 8-14p. The amplified signal is commonly recorded on tape and later converted to a photographic,thermal strip map. The tones displayed represent variations in the relative intensity of the emitted radiation as a function of temperaturesand emissivity of terrain surfaces. That is,lighter tones connote warm-
Remote sensing as an ecological tool
er temperatures, although only relative temperature differences can be inferred from these thermal displays. The thermal images (Figs. 13 and 14) are strikingly similar to the simultaneous panchromatic photography flown at a scale of 1 :10,000in the afternoon of 9 August 1962. Five predominant density tones were distinguished on the thermal imagery. The lightesttoned areas and therefore the warmest are (a) sand bars (Fig.13), (b) cleared areas (Fig.13), (e) bare ground or sparse vegetation of low stature (Figs. 13 and 15), and (d) differential warming of sun-facing
slopes (Figs. 14 and 15). Dark-toned areas,apparently of low temperature, are exemplified by (a) lakes and streams (Figs. 13 and 14), (b) dense foreststands,especially white spruce, Picea glauca, and (c) topographic or forest shadows. Intermediate tones correspond to relative temperatures between these extremes. Disturbance of the vegetation by man’s activities was concomitant with a different thermal environment. For example, at Beaver prominent thermal patterns were associated with the village and air strip clearings. In Figure 13 a cabin roof (arrow) presented an entirely
FIG.6. Stereo triplet and type m a p of vegetation along the Steese Highway. Pg
White Spruce
Pm Black Spruce B p White Birch
Bn Dwarf Birch L g Labrador Tea S Willow Pt Aspen
177 12
P.L.Johnson
178
Remote sensing as an ecological tool
-
f rozan Ground
L
+
o
E
+
I
.m
?IDistance (metres)
Fig. 9.Transect across successional vegetation zones encroachinginto Chara Lake,Northway, Alaska.
different tone in both the panchromatic and thermal illustrations apparently caused by a different roofing material. An aluminum roof has a low emissivity and would appear as a cold thermal signal. Subtle changes in slope m a y be enhanced by thermal signals. Four thermal areas are readily identifiable in contrast to three areas on the panchromatic photograph in Figure 15. The thermal boundary between area 1 and area 2 corresponded to a differentially cool slope of a remnant terrace. The higher apparent surface temperature in area 4 (Fig.15) represented the vacillations of microclimate following a recent fire. The sparsely-vegetatedsilt slopes (arrow A) were warmer than the aspen stand (arrow B) on a similar slope aspect. In Figure 14,local topography is enhanced by solar warming of sand dunes and redeposited alluvial sands while older meander patterns are imprinted in the willow stands. The sand ridges were too xeric to support. trees; although aspen and willow were comm o n on moister sands. In these thermal images taken in the afternoon,the
primary attenuation caused by vegetation was associated with its physiognomy rather than species composition. The forest canopy was discontinuous, and therefore the vegetative surface presented to the thermal scanner was a partially shaded surface. The resolution of the scanner was such that it does not distinguish between sun-warmed crowns and shaded gaps. The mosaic of warmer and cooler leaves was displayed as some intermediate grey tone. T h e noise level of this imagery is indicated by the regular pattern imprinted on the Y u k o n River (Fig.13). T h e vertical lines correspond to scanner sweeps, the horizontal lines represent imperfections in the phosphor coating of the imaging C-scope.These examples demonstrate that the s u m of information from multispectra1 displays m a y be greater than each sensor system considered separately. POTENTIAL SYNOPTIC D A T A F R O M SPACECRAFT
A
major contribution of the technology developed to
179 12*
P. L. Johneon
FIG.10. Aerial photographs of boreal lakes at Northway,Alaska,in 1949 (A)and 1962 (B)illustrating hydrachsuccessionover a 13-year interval.
FIG.11. Radar image (negative print) across Yukon Flats on 22 October 1956. The major landforms are delineated.Thevertical white bars are tape marks on the film and the horizontalline represents the flight line beneath the aircraft. Qai, flood-plain and low-levelalluvium; Qat, Quaternary alluvial fan and terrace deposits; Qi, loess mantle with thermokarst lakes; Qd, glacial drift.
180
Remote sensing as an ecological tool
TABLE3. Anticipated orbital resolution'
Spectrum
System
Area at 200 Temperature Ground nautical sensitivity resolution miles (kmP) (aK)
(4
Visual 15 cm FL camera Infra-red High resolution scanner Micro- Radar, K a band, wave 8 mm
766
6-30
IO.1
60
i0.5
500-1,000
1. Abstracted from Peaceful Uses of Earth Observation Spacecraft, Val. 1. A n n Arbor, University of Michigan, 1966. (Publication7219-2-Fl3.)
explore space will b e to provide information to the natural sciences from Earth-orbiting spacecraft. Already satellites are telemetering valuable synoptic coverage of weather patterns. O n c e a spacecraft is in orbit the cost of repetitive photography will be minimal. S o m e hand-held colour transparencies obtained in June 1965 o n the Gemini IV mission are indicative
FIG.12. Panchromatic mosaic and radar image (positive print) of Birch Creek and the Yukon River north of Circle. The flight line of the side-looking radar was parallel to the north margin of the illustration.
of the quality of present space photography. As syst e m s are improved a n d designed specifically for space platforms, improvements in quality can b e expected. S o m e anticipated parameters for remote sensors orbiting at 200 nautical miles are presented in Table 3.
SUMMARY R e m o t e sensing is defined as non-contact acquisition of information, usually from aeroplanes, in a n y portion of the spectrum. In context with the urgent need of a n ecological basis for resource management, it is sugges-
ted that remote sensors m a y solicit information for: (a)inventory and mapping of resources; (b) quantifying the environment; (c) describing the flow of matter a n d energy; and (d) evaluating changes in the ecosystem. In order to utilize the potential information from photography, optical-mechanical scanners, or microw a v e antennae it is necessary to understand the fundamental matter a n d energy relationships responsible for the images to be analysed. T h e interpreter of aerial photography is also dependent o n (a) his experience a n d training, (b) the size, shape and pattern of objecta, a n d (c) the tone or density variations in the processed
film.
181
P. L.Johnson
FIG.13. Panchromatic aerial photograph and thermal m a p of Beaver, Alaska, on
the Yukon River.
182
Remote sensing as an ecological tool
FIG.14. Panchromatic aerial photograph and thermal m a p north of Beaver, Alas,ka.
183
P.L. Johnson
Fig. 15. Panchromatic aerial photograph and thermal m a p south of Stevens Village, Alaska.
184
Remote sensing as an ecological tool
ELEVATION
ALPINE T ü N D R A
-Iy-
MIXED
P plouc ond Beiulo w p y r
F O REST‘
I
IifiJ
J FIG.16. Generalized relationship of vegetation types to slope in interior Alaska. In interior Alaska a n extreme continental climate prevails in the latitudinal tension zone of tree growth. T h e present vegetation mosaic is a mixture of forest, shrub and herbaceous stands in various stages of plant succession. Their stability appears proportional to: (a) the stability of the permafrost a n d the resulting drainage condition; (b) the elapsed time since the last fire or flood disturbance and the intensity of the burn or duration of the flood; (c) the age of the organic or mineral substrate which is usually a func-
tion of topographic position. Examples of Alaskan subarctic vegetation were selected from a recent study of structure a n d standing crop in the Y u k o n Flats region to reveal the utility of aerial photography, radar and thermal imagery. T h e potential of synoptic ecological information from Earth-orbiting platforms is suggested as a realistic m e a n s of extrapolating local studies to m a n a g e m e n t size units in resource planning.
Résumé L a détection à distance en écologie
(Philip L. Johnson)
L a détection à distance est définie c o m m e l’acquisition d’informations sans contact direct, généralement à partir d’un avion, dans une portion quelconque du spectre. C o m m e il est urgent de donner une base écologique à l’exploitation des ressources, o n suggère que les personnes qui pratiquent la détection à distance essaient d’obtenir des renseignements aus fins suivantes: a) inventaire et établissement d e cartes des ressources ; b) quantification du milieu; e) description du. flux de matière et d’énergie; d) évaluation des modifications de l’écosystème. P o u r utiliser les renseignements que permettent
d’obtenir les photographies, les appareils explorateurs optico-mécaniques o u les antennes 2 ondes courtes, il faut comprendre les relations fondamentales entre la matière et l’énergie, qui expliquent les images à analyser. L’interprétation des photographies aériennes est également fonction: a) de l’expérience et de la formation du technicien qui l’effectue; b) de la dimension, de la forme et de la disposition des objets; c) variations de ton o u de densité dans la pellicule développée. A u centre de l’Alaska, un climat continental extrême règne dans la zone de tension latitudinale où croissent les arbres. L a végétation se présente actuellement c o m m e un assemblage de forêts, d’arbus-
185
P. L. Johnson
tes et de plantes herbacées qui en sont à diverses étapes du cycle végétal. L a stabilité de ces végétaux paraît être proportionnelle : u) à la stabilité du pergélisol a u temps écoulé et à l’assèchement qui en résulte ; depuis le dernier incendie ou la dernière inondation, et à l’intensité de celui-là ou à la durée de celle-ci ; e) à l’âge du substratum organique ou minéral, qui est généralement fonction de la situation topographique. L’auteur donne quelques exemples de végétation subarctique de l’Alaska, pris d’une étude récente
a)
sui la structure culturale et les récoltes sur pied de la région de Y u k o n Flats, afin de montrer l’utilité des photographies aériennes, du radar et des images thermiques. Les informations écologiques synoptiques qu’on peut obtenir à partir de plates-formes placées sur orbite circumterrestre fourniraient, de l’avis de l’auteur, un m o y e n pratique de tirer, à partir d’études locales, d’utiles indications relatives à l’administration et à la planification des ressources.
Discussion F. E. ECHARDT. Vous avez démontré comment, dans u n avenir pas trop lointain, il sera possible, au moyen de différents types de capteurs installés dans des avions ou des satellites, d’obtenir des renseignements sur la structure et le fonctionnement des écosystèmes. J’aimerais savoir dans quelle mesure, à l’état actuel de développement de ces techniques, les phénomènes observés à partir d‘avions et de satellites sont effectivement en rapport avec la structure et le fonctionnement de ces écosystèmes. I1 est connu qu’il est souvent difficile, m ê m e en utilisant des émulsions spéciales, de distinguer, à l’aide de photographies aériennes, des espèces végétales différentes et, à plus forte raison, des écosystèmes.
P. L. JOHNSON. Certainly no one quarrels with the use of aerial photography for topographicmapping. There are m a n y examples of species identification, forest-type maps, and timber production that have used aerial photography
Bibliography BLISS, L. C. 1966. Plant productivity in alpine microenvironments on Mt. Washington, New Hampshire. Ecol. Monogr., vol. 36, p. 125-155. BORGSTROM, G. 1965. The hungry planet. New York, Macmillan. 487 p. COLWELL, R. N.;BREWER, w.;LANDIS, G.;LANGLEY, P.; MORGAN, J.; RINKERJ.; ROBINSON, J. M . SOREM, A. L. 1963. Basic matter and energy relationships involved in remote reconnaissance: Report of subcommittee I, Photo Interpretation Committee. Photogrammetric Engineering, vol. 39, p. 761-799. GATES, D.M.;KEEGAN H.J.; SCHLETER, J. C.; WEIDNER, V. R. 1965. Spectral properties of plants. Applied Optics, vol. 4,p. 11-20. JOHNSON, P. L. 1966. A consideration of metbodologv in photo interpretation.Proceedings of the Fourth Symposium on Remote Sensing of the Environment, p. 719-725.A n n Arbor, Michigan, University of Michigan, Institute of Science and Technology.
186
successfully.Further examples m a y be sought in the annual Proceedings of the Remote Sensing Symposium published by the University of Michigan. The purpose of this paper was to encourage ecologists to seek the answers to your question. Unfortunately, very few ecologists have been interested in this rapidly expanding technology. It is sufficient to emphasize that appropriate data must be obtained on the ground in conjunction with any evaluation of remote sensors. Since the patterns on aerial displays are a function of the interactions of energy with the ecosystem, these patterns must connote information. The biological significance of this information must be established for each example. T o approach the problem by asking if all taxa, societies or permutations known on the ground can be “seen” in a given photograph seems decidedly less profitable. As remotesensing technology yields greater quantitative data, statistical correlations will be more plausible; eventually the use of information theory will permit greater automation of the ’ interpretative process.
/ Bibliographie __Region, ; VOGEL, T. C. 1966. Vegetation of the Y u k o n Flats Alaska. Hanover, N.H.,Cold Regions Research and Engineering Laboratory. (Research Report 209 (in press).) MUDD, S. (ed.). 1964. The population crisis and the use of world resources.The Hague, W.Junk. 562 p. ODUM, E. P. 1959. Fundamentals of ecology, 2nd ed. Philadelphia, W.B. Saunders Co. OVINGTON, J. D. 1962. Quantitative ecology and the woodland ecosystem concept. Adv. in Ecol. Res., vol. 1, p. 103-192. --;HEIFKAMP, D.;LAWRENCE, D.B. 1963. Plant biomass and productivity of prairie, savanna, oakwood and maizefield ecosystems in central Minnesota. Ecology, vol. 44, p. 52-63. PIEPER, R. D. 1963. Production and chemical composition of arctic tundra vegetation and their relation to the lemming cycle. Ph.D.thesis, University of California, Berkeley, California, 95 p.
Remote sensing as an ecological tool
REMPEL, R. C.; PARKERA. K.,1965. A n information note on an airborne laser terrain profiler for micro-relief studies. Proceedings of the Third Symposium on Remote Sensing of the Environment, p. 321-337. A n a Arbor, Michigan, University of Michigan, Institute of Science and Technology. RINKER, J. N.;EVANS, S.; ROBIN, G. de Q. 1966. Radio
ice-sounding techniques. Proceedings of the Fourth Symposium on Remote Sensing of the Environment,p. 793-800. A n n Arbor, Michigan, University of Michigan, Institute of Science and Technology. UNIVERSITY OF MICHIGAN. 1966. Peaceful uses of earthobservation spacecraft. A n n Arbor, Michigan, Institute of Science and Technology. 57 p. (Report 7219-1-F(1).)
187
Soilsofthe subarctic regions J. C. F.Tedrow
Any discussion of soil development in the subarctic regions immediately poses the question of defining these regions. M a n y investigators1 have discussed the delimitation of the subarctic boundaries on the basis of various criteria, but even investigators employing the same general criterion find agreement difficult. Probably no two geographers plotting the subarctic regions would find that their boundaries coincided, nor would any two climatologists. In this report the Subarctic is defined primarily on the basis of vegetation, that is on those regions between the northern boundary of continuous forest and the line marking the northern extension of conifers. This criterion, though admittedly indecisive in m a n y situations,seems more realistic in a study of soil development than isotherms or other bases of delimitation. The definition, as used in this report, is based largely on the work of Polunin (1951) and, to some extent, that of Hustich (1953, 1960), w h o placed the arctic tree line at the polar boundary of spruce, larch, or pine, whichever occurs farthest north. Probably no criterion can be applied with rigidity and the one adopted in this report is no exception. At times it lacks definition and becomes indecisive. North of the tree line, for example, isolated stands of willow, alder, birch, poplar, and prostate forms of tree-likespecies are present. Another problem is that of subarctic boundaries in the land adjacent to the Bering Sea and those of Atlantic land masses; perhaps, here, some criterion in addition to vegetation should be considered. There are also problems in defining a tree line in the vicinity or large, high, land masses such as those in the Yukon Territory, Scandinavian Mountains, Northern Urals, East Siberian Highlands, and the Kamchatka Peninsula. In these sectors as well as in others the alpine tree line becomes confluent with the arctic tree line (Tedrow and Harries, 1960). Other factors such as marshiness,
shallow soils, length of time a landscape has been free of ice, together with the presence of perennially frozen ground, influence the position of the tree line. Griggs (1934),Sheludiakova (1938),Larson (1965)and Bryson et al. (1965), among others, have discussed the dynamics of the tree line, pointing out why it cannot be considered stationary over long periods of time. Scotter (1963) has discussed some of the effects of burning on the displacement of forest boundary and the edaphic changes induced by fires. Haugen (1965) reported that timber line in interior Alaska approximates the 100 C July isotherm and that the growing season temperatures probably remained relatively constant for at least the past five centuries, T h e subarctic regions, as considered in this report, are shown in Figure 2. The subarctic region of Alaska is based on the work of Cigafoos (1958) and that of Canada from the work of Porsild (1957). In Norway, Sweden,and Finland,the Subarctic occupies the region between the closed pine forests and that of the northern extension of pine or pine seedlings,largely as outlined by Hustich (1960).The delineation of the Subarctic in northern Eurasia is largely from information published in the Agricultural Atlas of the U.S.S.R. (1960)and boundaries,in substance,correspond to the shrub and hillock tundras, subtundra sparse forests, northern taiga forests and certain mountainous areas described by Sochava (1954).The work of Tikhomirov (1960) was also used in establishing the subarctic region of the U.S.S.R.Thin forests of tamarack, spruce, and birch, together with the dark conifer forests of the northern taiga, the mountain forests of the East Siberian Highlands, and forests of K a m chatka, are collectively designated as belonging to the subarctic region, but it would probably be more 1.
Hare (1951), Polunin (1951), Rosseau (1952), Hustich (1953), Sochava (1954), Ritchie (1959), Tikhomoriv (1960), Britton (1957), and others.
189 13
J. C. F. Tedrow
realistic to refer to certain sectors of these regions as subalpine rather than subarctic, especially in the southern portions. Beginning with the northern forest areas and moving northward, the continuous canopy thins out and appears as grove-like communities with smaller trees, broken canopy, and patches of tundra. Continuing further northward, the portion of land covered with tundra becomes greater at the expense of forest, until, still further northward, only groves of trees or single trees exist (Fig.1). Using earlier analogies, it might be assumed that the land colonized by tree species would have soils with podzolic characteristics, whereas that occupied by tundra vegetation would not. But this does not appear to be the case. Apparently there is no conclusive evidence of a rhythmic, qualitative, pedogenic change directly at the forest-tundra boundary, but there is fairly reliable eviden e of a group of slowly changing pedogenic processes occurring in a northsouth direction across the forest-tundra transition zone. Perhaps,if any one factor could be singled out as changing directly at the forest-tundra transition, it would be the quantity of organic matter present over the mineral soil. Well-drained forested sites usually, but not invariably,have more organic matter present on the surface than do the adjacent well-drainedsoils of the tundra. With the information n o w available it is improbable that soil characteristics as such could be used for the delineation of the subarctic regions. W h e n Dokuchaev compiled a soil m a p of northern Eurasia, there was no question that vegetation served as a prime criterion in the plotting of soil boundaries. That there is a qualitative pedogenic change at the northern tree line has been inferred but it has never been demonstrated.
Field investigations throughout the subarctic regions during the past twenty years have shown that the pedogenic problems are far more complex than pioneer investigators indicated.
ENVIRONMENTAL FACTORS PRECIPITATION
Precipitati0,n in the subarctic regions generally approximates to 10-20inches per year with increasing amounts near the oceans. In subarctic Alaska annual precipitation approximates to 25-40 inches in the western coastal regions, decreasing to some 15-20 inches in the interior basin. Throughout most of subarctic Canada from the Y u k o n Territory to the vicinity of Hudson B a y annual precipitation totals about 10 inches. East of Hudson Bay, precipitation increases from about 15 inches to 40 or more inches in the coastal sectors of Newfoundland. In the Eurasian Subarctic and in North America a similar rainfall pattern is present. The Subarctic of coastal Norway records about 40-70 inches of annual precipitation with a decrease to about 20 inches eastward through northern Sweden, northern Finland,and the Kola Peninsula. In subarctic U.S.S.R. precipitation decreases from about 20 inches in the western sector to approximately 10 inches in the east. There is a general increase in precipitation in the vicinity of the Kamchatka Peninsula,reaching as m u c h as 20-40inches per year in the south-easternportion. In general, throughout all of the subarctic region, including the coastal sectors, precipitation tends to decrease from south to north. The drier interiors of both continents reflect the low quantities of effective precipitation in the welldrained soils. In sectors where annual precipitation approximates 10 inches the soils do not show the podzolic character that they show in areas of greater precipitation. There is some evidence that soils throughout the Subarctic also derive a considerable amount of moisture from atmospheric condensation (Gorodkov,1939). TEMPERATU R E
In addition to vegetation, it appears that the mean July temperature values are also important in delineating the subarctic regions. Most of the subarctic region of Alaska and Canada falls between the 500 and 600 F July isotherms,as is also the case in Eurasia. PERENNIALLY FROZEN G R O U N D
FIG.1. View of the forest-tundratransition in north central Alaska. (Photograph by R. K. Haugen.)
190
Most but not all of the subarctic region lies within the region of perennially frozen ground. In the Alaskan sector, the region east of Hudson Bay, Scandinavia,
Soils of the subarctic regions
U.S.S.R.are not
all underlain
The first report of soils in the Alaskan Subarctic was
by continuous perennially frozen ground; in m a n y sectors there is probably little if any present (Tedrow
by Kellogg and Nygard (1951), w h o indicated that
and sectors of western
and Harries, 1960). In eastern Siberia perennially frozen ground extends well south of the subarctic region. In some sectors of the Subarctic, permafrost plays an important part in the soil system,in others it is a minimal factor; apparently the greatest effect of permafrost on the dynamics of the soil is to be found in eastern Asia. Q U A T E R N A R Y GEOLOGY Most of the subarctic regions are mantled with Quaternary-agedeposits. S o m e sectors of subarctic Alaska, however, were free of ice during the Quaternary. The Quaternary history of eastern Siberia is uncertain and some of the mountainous areas probably escaped glaciation (Obruchev, 1935; Spizharskii, 1937; Anikeen, 1959). Apart from all these localities, during the Pleistocene epoch the entire subarctic region was probably covered with ice. VEGETATION
Vegetation of the subarctic varies in character from place to place. Reduced to its greatest simplicity, the subarctic of North America is vegetated mainly by white spruce in the western portion and black spruce in the eastern sector. In Eurasia the western sector of the Subarctic is colonized mainly by Siberian spruce and Scotch pine, while the easternsectoris made up largely of Dahurian larch (Tedrow and Harries, 1960). The complexities of the vegetation in the maritime sectors of the continents and the vegetative changes with latitude are beyond the scope of this report.
SOILS OF SUBARCTIC ALASKA The Subarctic of Alaska can be defined only in general terms. M a n y of the subarcticregions merge with alpine conditions,and vegetative definitions become tenuous. The southern front of the Brooks Range serves as a fairly realistic demarcation line between the Subarctic and the tundra proper,but in detail it is more of a zone than a line. South ofthe Brooks Range the drier valleys and low hills tend to be covered with forest whereas high elevations and bogs are treeless. Discontinuous forests occur south of the Brooks Range to the vicinity of the Y u k o n and Tanana Rivers. South of Fairbanks most of the mineral soils are covered with forests; the bogs and the higher elevations possess a tundra aspect. The western portion of Alaska from Kotzebue Sound south to Bristol B a y and the Aleutian Islands lacks, for the most part, a well-defined forest cover, a condition probably highly influenced by the maritime climate.
Lithosols and Tundra occupy about 90 per cent of the mountainous landscape. The foothills immediately south of the Brooks Range are covered with Mountain tundra1 and Subarctic brown forest soils with Bog soils, Lithosols, and alluvial deposits (Fig.2, No. 1). Throughout the foothills region between the Brooks Range and the Yukon Flats, soils have yellow, grey, or brown-coloured sola with little mineral horizon differentiation (Fig.2, No. 2). Wilde and Krause (1960) recognized Melanized raw-humus (Subarctic brown forest) soils, Micro-podzols, and associated hydromorphic soils. The report by Wilde and Krause not only furnishes valuable information on the soils per se, but correlatesthe brown-colouredsoils with the Latent podzol, Crypto-podzol,and Incipient podzol of Soviet investigators. Wilde and Krause verify the lack of perceptible iron and aluminium translocation within the profile in spite of the presence of raw humus, and further report that this condition is unique to northern regions: “Because of this peculiarity,these soils occupy a unique, totally independent position in the world’s family of soils. Their pattern of development cannot and should not be to or, [sic] rather confused with that of any other genetic soil groups.” Krauseetal.(1959)show the effects of slope direction on soil properties in the interior of Alaska. On the northern exposures a poorly-drained Highmoor halfbog is present, while on the south-facingslopes the Subarctic brown forest soils are present. Half-bog soil is colonized by a more mature stand of white spruce than is the case with the better-drained sites. Drew and Shanks (1965)presented a report on the soils of the Firth River area that is of specialinterest, because the study transversesthe forest-tundraecotone in the vicinity of the Alaska-Canada border. In the predominantly dry environments, Alpine rendzina, Calcareous regosols, and a dry phase of Bog soil were identiïied, while in the wet environments Calcareous low-humic gley, Upland tundra, Calcareous meadow tundra, and various B o g soils were recognized. While local lithology together with drainage appears to have a high regulatory influence on the kind of soil present, the occurrence of spruce forest is also important because it imparts a definite mor surface layer, especially on carbonate-bearingdeposits. North of the tree line in the Brooks Range, soils with Podzol-like morphological features are present in the valleys (Brown and Tedrow, 1964), while in the uplands Arctic brown soils in various stages of development are widespread (Tedrow and Brown, 1962). The soil data of Kellogg and Nygard (1951)includes the western sector of Alaska and most soils are designated as Tundra, Bog, Subarctic brown forest, 1. In this paper only the firet word in the n a m e of the Great Soil Group is capitalized.
191
FIG.2. The subarcticregion is shown as the shaded area. Soil regions within the Subarctic follow:(1) Lithosols and Tundra soils of the northern mountains; (2) Tundra, Bog, Mountain tundra and Subarctic brown forest soils; (3) Mountain soils with Subarctic soils; (4) Mountain soils (undifferentiated); (5) Subarctic soils with peat; (6) Rock outcrops with peat and Subarctic soils; (7) Podzol soils with rock outcrops,Peat and Grey wooded soils; (ô) Peat with alluvial soils and Subarctic soils; (9) Peat with alluvial soils and Brown wooded soils; (10) Podzol soil with rock outcrops and Peat; (11) Mountain forest grassland soils, Lithosols, Podzol, Braun waldboden, Gleic and Bog soils; (12) Mountain taiga with marsh; (13) Gley podzolic, Mountain taiga with marsh, Podzolic marshy and other soils; (14) Complex of Podzolicmarshy and Peat-marshy soils; (15) Mountain tundra, *Mountain taiga, Gley podzolic soils; (16) Podzolic-marshy, Gley podzolic soils;(17) Gley podzolic with Mountain tundra soils; (18) Mountain tundra soils; (19) Mountain tundra, Mountain forest grassland, and Grassland soils of the subarctic meadows.
and related soils. Tedrow and Brown (unpublished) found the darker-colouredsola, compared to soils in the east,in both Arctic brown and Tundra soils in the Cape Sabine area of north-western Alaska. Rieger (unpublished) tentatively classified the well-drained soils of the west coast of Alaska as Podzols and Dark well-drainedsoils of uplands.
SOILS OF SUBARCTIC CANADA Soil investigations in subarctic Canada date back only for the last twenty years, but during this time some highly important informationhas been presented (Leahey, 1947, 1949, unpublished). Leahey (1947) reported the following soil properties from mixed forest cover of spruce, birch, alder, and willow near Fort Norman, Y u k o n Territory: Depth
2-0in. Semi-decomposed moss and leaves, p H 5.2. 0-4in. Grey-brown fine sandy clay loam, p H 6.5. 4-10in. Light grey-brown fine sandy clay loam, p H 7.8. 10-18in. Pale olive fine sandy clay loam with some yellow-brown mottling, p H 8.3. 18-30in. Pale olive fine sandy clay loam with yellowbrown mottling, p H 8.3. 30-39in. S a m e as 18-30in. depth, p H 8.4. Leahey (1947) stated that the upland soils of the Y u k o n can be divided into two broad groups on the presence or absence of permafrost, and describes the soils thus: “The upland soils vary from dark-brown grassland soils, which m a y be correlated with the Chestnut great soil groups, to Reddish-brown forested soils which probably can be correlated with the Brown podzolic great soil groups . . . their solums are rather thin, seldom exceeding 18 inches except on very sandy materials. Mineral soils ...show little profile development except that the upper inches of mineral material
.
J. C. F. Tedrow
...is
weathered to a reddish-brown color. . . . F e w of the Y u k o n soils s h o w a n y evidence of a podzolic A, horizon. ...T h e characteristic A, horizon of podzol soils w a s only observed in a f e w lighter textured soils.” T h e B r o w n w o o d e d soil n a m e has been retained in Canada, and its wide distribution has been substantiated by further investigations by D a y and Leahey
(1957). Grey w o o d e d soil has been recognized throughout m u c h of Canada from Ontario westward to British Columbia and northward to the approaches of the Beaufort Sea and A m u n d s e n Gulf (Stobbe, 1960). Grey w o o d e d soil has a dark-coloured, friable, mineralorganic appearance; it is slightly acid in reaction and is well-supplied with bases. Stobbe relates the m o r phology of Grey w o o d e d soil to the action of high biologicalactivity u p o n the carbonate-bearing material. In the younger profiles reducible iron values are higher in the upper horizon as compared to the lower ones, and with increased time there is a marked increase in translocation of iron. There is also a m a r k e d eluviation of clay from the upper horizon of the Grey w o o d e d soil. E v e n though p H values (5.8 in the A,) remain comparatively high, the podzolic process is evident in the Grey w o o d e d soil (Wright et al., 1959). Wright et al. further report the presence of quartz, illite, montmorillonite, chlorite, kaolinite, and mixedlayer minerals in the clay fraction of the Grey w o o d e d a n d B r o w n w o o d e d soils as well as the alluvial deposits in the North-West Territories. P a w l u k (1960)substantiated the podzolic character of northern Alberta soils even though the base saturation w a s between 53 a n d 90 per cent; severe weathering w a s manifest in the minerals in the surface horizons. D a y and Rice (1964) reported chemical, mineral, a n d morphological features along a sequence of soils of the subarctic ecotone of the Y u k o n Territory. There w a s n o outstanding regional morphological change across the ecotone, but soils under the heavier forest cover showed m o r e podzolic affinities than did those o n the tundra fringes. Accordingly, soils in the tundra-forest transition at Reindeer Depot were classed as Subarctic orthic regosol and Subarctic gleyed acid b r o w n wooded. A t Inuvik, which marks the tundra-forest transition, Subarctic b r o w n w o o d e d soil is recognized, while in the forest zone at N o r m a n Wells, a Subarctic minimal podzol is present. F r o m the standpoint of soils, the subarctic region from Great Slave L a k e extending eastward to H u d s o n B a y and northern Quebec has been explored very little. Published reports are largely confined to a few generalized m a p s (Ellis, 1938;A look at Canadian soils, 1960). Feustel, Dutilly, and Anderson (1939) reported soil analyses from this sector but n o pedogenic interpretation w a s m a d e of their data. Most of the Great Slave L a k e - H u d s o n B a y sector continues to b e designated simply as Subarctic soils with Bogs and Peat (Fig.2, No. 6). T h e H u d s o n B a y Lowlands from Churchill to
194
James B a y are mantled with extensive peat deposits (Fig.2, No. 8). Larsen (1965)and Bryson et al. (1965)in the Ennadai reporL a k e area, Keewatin district (610N.,1010 W.), ted a fossil Podzol overlain by a poorly-developed Arctic b r o w n or Subarctic b r o w n forest soil. Apparently at one time forests covered the area, but fires about 3,500 years ago and again 900 years ago (based o n Cl‘ dates of charcoal) destroyed the forests; n o w in this region tundra-like vegetation is present.
SOILS O F SUBARCTIC SCANDINAVIA Soil studies were conducted throughout Scandinavia
by a n u m b e r of investigators during the 1920-1940 period, and their w o r k extended to the northern extremities of the subarctic region. Since 1940 there have been f e w pedologic investigators in this northern sector, so w e have to rely o n the older reports. For the sake of greater uniformity of nomenclature, throughout northern Eurasia the soil boundaries of northern Scandinavia (Fig.2) are abstracted from the w o r k of Gerasimov (1954).Terminology in Figure 2 is not always identical to that used in the text of this report. S t r e m m e (1927) headed a commission which published a generalized soil m a p of Europe, and Björlykke (1927)published a pedologic m a p of Norway, as did T a m m (1927, 1932) of S w e d e n and Frosterus (Glinka, 1931) of Finland. S t r e m m e and Hollstein (Stremme, 1927) showed that, in the Subarctic of Scandinavia, Podzolized forest soil (moderately developed), Podzolized forest soil (strongly developed), R a w h u m u s soil, Peat, and shallow soils with Podzols and T u n d r a soil are present. This subarctic region of Scandinavia is of special interest to the pedologist because of the presence of a steep west-east precipitation gradient along a fairly well-defined isotherm. S o m e of the coastal fringes in the vicinity of Narvik (Norway), receive annually as m u c h as 60-70inches of precipitation and the interior mountains s o m e 2540 inches, whereas in Sodankyla (Finland), slightly south of the Subarctic, annual precipitation is approximately 20 inches. In northern N o r w a y brownish and yellow-coloured soils are interspersed with Podzol soils (Björlykke, 1927). In the western fringes of northern N o r w a y there are m a n y soils of a lithosolic character and also certain highly-weathered soils abundant in h u m u s . Podzols are comparatively rare but are nevertheless scattered throughout the entire region of northern N o r w a y -even beyond the extremities of the tree line. In the vaHeys Peat soils are present as well as wet mineral soils. In northern Sweden, T a m m (1927, 1932, 1950) reported the presence of B r o w n forest soill (Braun 1. Approximates to the Brown podzolic soil of the United States.
Soils of the subarctic regions
Rockland Heaths-scrub willow Subalpine birch forest Altitude
Pine-birch forest
in6000 hoti
BLOCK FIELD 4000
TUNDRA 2000
FOREST I Podzol 8 Bog)
I
.
.
I
-
.-
FIG.3. Diagrammatic sketch showing the relation of vegetation and soils with altitude,near Abisko (Sweden). Waldboden) with severalvarieties of Podzol and hydromorphic soils, including peats under forest conditions, as well as treeless areas. The relation of Braun Waldboden of northern Scandinavia to the Braunerde in Germany is not well understood, but of late some excellent discussions have been forthcoming (Gerasimov, 1959). In the northern fringes of the pine forests (66015’N., 21015’ E.) Podzol soils are present with an A2 horizon as m u c h as 6 inches thick; this condition is also present in the soils formed on acid glacial drift of the non-forested areas. In the valley south of Saltoluokta (Sweden), there are virtually no trees, but the sandy eskers, k a m e terraces, and outwash deposits with arctic-like vegetation are mantled almost completely with Podzol soils. In this same valley fossil profiles show well-developed Podzol with a white A, horizon underlain by a strong reddish-brownB horizon buried beneath a more recent glacial deposit. Above the tun-
FIG.4 A landscape in northern Sweden;only the drier sites are colonized by trees.
dra zone,in northern Scandinavia the block field zoneis a scene of accelerated mass wasting (Rapp, 1960; R a p p and Rudberg, 1960; Rudberg, 1962) with virtually no vascular plants and no evidence of a genetic soil profile. Probably the main reason why the dark yellow and brown-coloured soils, which are so c o m m o n throughout most subarctic regions, are usually not present in the higher elevation of this locality is primarily due to the high amount of rainfall and mass wasting of the superficial materials. There is a fairly well-established altitudinal limit of Podzol soils in Scandinavia. In the north-central sector it varies from about 3,500to 4,200 feet, but in Torne Lappmark it approximates an altitude of about 2,000 feet (Nordhagen, 1927; Lothe, 1950). Figure 3 shows, in diagrammatic form, the vertical distribution of soils near Abisko (Sweden), the upper limit of Podzol soils being about 4,000feet. Dah1 (1956) has demonstrated that the altitudinal trend of the Podzol soil zone in Norway does not follow the vegetative zones. Bog soils are c o m m o n throughout the Subarctic of Scandinavia, and the percentage of land occupied by these soils increases eastward into Finland (Fig.4). In northern Finland, Rindell (Krische, 1928) published a semi-pedologicm a p showing the distribution of sands, loams, and moors. Frosterus (Glinka, 1931) showed the Subarctic of Finland as being mantled with Ferruginous podzols and Humus podzols, with m a n y Bogs. Gerasimov (1954)indicates that the soils of northern Finland have podzolic affinities with extensive areas of Gley and Bog soils.
SOILS OF SUBARCTIC U.S.S.R. W e now turn to the problem of soil formation in the forest-tundraecotone of the U.S.S.R.-a vast expanse of land covering over half of all the subarctic regions of the earth. Not only is this region extensive but the literature is rich in scientific accomplishments, dating hack for more than a century (von Middendorf, 1864). Despite this voluminous literature on the northern U.S.S.R., few pedologic studies have been directed to the specific problem of soil genesis within the forest-tundra transition. The m a p of Dokuchaev, published in 1899, shows Tundra soils north of the Arctic Circle,while the areas immediately to the south are shown as Forest soils and Rocky forest areas (Gerasimov, 1956). Glinka depicted the northern region as being mantled with Tundra and Podzolic soils (Fig.5). In 1937 Prasolov extended the forest boundary further north than had his predecessors. More recently Gerasimov (1956) published a m a p showing the distribution of soils ofthe world,includingthe subarctic regions. Soils of subarctic U.S.S.R.are first discussed in this report from west to east (Fig.2, No. 12-19).While
195
J. C. F. Tedrow
.40
40
I
60
‘i I
eo
\ I
100
120
J 140
FIG.5. A n early soils map of northern Siberia showing the relationship of the Subarctic to various soils. Soilboundaries by K. D. Glinka (1927). The darker shading indicates the subarctic region.
soil names of Figure 2 are essentially those used by Gerasimov (1954), m u c h of the discussion is based on the work of Ivanova, ROZOV, Glinka and Polyntseva. T h e various delineations shown in Figure 2 do not necessarily signify regional changes in the qualitative processes but indicate more often local changes due to variation in drainage, lithology, and orogeny. If Quaternary history, lithology, age, relief, and related factors could be held constant, perhaps the entire subarctic region of the U.S.S.R.as well as those of North America could be shown with three to five delineations, but nature does not present conditions in such simplified terms. The work of Gorodkov (1939) shows a western and eastern zone of gleyey and Weakly podzolic soils and this writer finds in Gorodkov’s work an implication of greater podzolic affinities in the western zone. Other Soviet investigators suggest that the podzolic process weakens eastward in the northern boreal regions of the U.S.S.R.Gorodkov (1939) further singled out the weakening podzolic process east of the Yenesei River. T h e Kola-Karelia sector of Figure 2 (No.12) is perhaps the most intensively studied portion of the subarctic regions. Glinka (1931) describes the widespread marshes and Forest-podzolicsoils with thin but conspicuous horizons throughout the Kola Peninsula. A s altitudes increase, the thickness of the horizons decreases and there is a transition to tundra soils. Krasyuk (Glinka, 1931) describes Yellow-podzolic,l Carbonate, Humus, and Marshy soils in the KolaKarelia sector. Rozov (1962)reports the presence of Dwarf or Thin podzols, some with ferruginous-humus or humus-ferruginousilluvialhorizons,throughout the Kola sector. On higher ground Mountain tundra with podzolic or tundra affinities are present, whereas on
196
the lower ground sphagnum bogs are widespread. Polyntseva (1958)described in some detail podzolized soils, Podzolic moors, Bottomland sod soils, and various mountain soils in the Kola sector, and augmented the discussion with extensive analytical data. More recently Dobrovolski (1963)discussed the geochemical aspects of soils in the Kola area,including the chemistry of trace elements. The subarctic region from the White Sea to the vicinity of the Ob River (Fig.1, No. 13) is designated as one of Gley podzolic, Mountain taiga with Marsh, Podzolic marshy, and other minor soils. Throughout the Archangel-Komi-northernUral sector, Ivanova (1962) describes the ridges with Gley-podzolic,Tundra-gley podzolized, Tundra illuvial-humus,Tundramarshy, and Bog soils. East ofthe Timan Ridge conditions appear to be somewhat drier than to the west. Throughout the low plains, however, conditions are generally marshy with m a n y soils formed from binary deposits. Zaboyeva (1965) divided the podzolic soils into four subof the north-easternEuropean U.S.S.R. zones: Sod-podzolic (up to 600 N.),Typical podzol (up to 630 N.),Gley-podzolic (up to 650 N.),and Gleypodzolic and tundra bog soils up to approximately 670 N.).The Gley-podzolic soils of the northern region have tongue-likefeatures of humus present within the profile and there is evidence of a downward migration of iron. The subarctic region from the vicinity of the Ob River to an indefiniteline paralleling the Yenesei River (Fig.2, No. 14)consists of a low, flat plain within the West Siberian Lowland. The soils are predominantly marshy, with some showing podzolic-marshy characteristics. Ivanova (1962)presented testimony as to the presence of poorly-drainedsoils in this sector with such terms as Gley-turfyand Humus-turfy soils. In the southern portion of the sector there are, under larch stands, Superficially podzolized soils and Weakly podzolic gley soils. S o m e of the podzolized soils show illuvial-humuscharacteristics. The vast subarctic region as shownin Figure 2 (NO. 15) extends from the Yenesei River eastward across the Olenek, Lena, and Indigirka rivers, and northward to the vicinity of the Laptev Sea. This region consists of predominantly mountainous terrain, and prior to 1950 had not been studied in detail; recently, however, m a n y factual reports have been forthcoming (Karaveyeva et al., 1965;Ivanova, 1962;Filimonova, 1965).Rozov (1962) describes this area as a complex of Gley-freezing taiga soil, Superficially-gley soils, and marshy soils, with the higher elevations having extensive outcrops. The soils of the mountains are characterized as Mountain tundra and Mountain taiga. Ivanova (1962) recognized the Gley-frozen-taiga soils of northern Yakutia and described the morphological features of these soils,supportingher descriptions 1. Not io be confused with Yellow podzolic soils of the subtropics.
Soils of the subarctic regions
TABLE1. Classification scheme of subarctic soils by Rozov (1956) Containing Soil group
Class
Atmospheric moisture
Sporadic ground water
Constant ground water
Biogenic
2 Frozen taiga
Taiga (ferruginous)
Frozen bog
3 Forested taiga
Podzolic Forest grey soil
2 Frozen taiga
Frozen solod (degraded alkali soil)
Gleic pale yellow taiga Podzolic boggy soil and Forest grey gleic Frozen gleic solod
Halibiogenic
Soddy carbonated soil
3 Forested taiga
Lithobiogenic
3 Forested taiga
High bog
Soddy gleic carbonated
with detailed analytical data attesting to the lowleaching potential of the area. Throughout this sector (Fig.2, No. 15) frost action apparently play6 a more important role in the soil system than is the case further west. The Olenek River region (Fig. 2, No. 16) is mantled with Podzolic marshy and Gley podzolic soils, with a variety of H u m u s carbonate soils being present. The central portion of the Indigirka River region (Fig.2, No. 17)is delineated from the main East Siberian Highlandsbecause the former consistsofless mountainous terrain than the surrounding region. The soils consist mainly of Gley podzolic and Mountain tundra varieties. The East Siberian Highlands extending from the Tundra zone southward to the Kamchatka Peninsula (Fig.2, No. 18)have been studied very little. The soils are designated as Mountain tundra. Liverovski and Rubtsova (1959)recognized the following varieties of soil in the taiga zone of Prymeria (just south of Fig. 2,No. 18): Brown taiga,Brown marshy taiga, S w a m p frozen, and Taiga illuvial-humic. The Kamchatka Peninsula (Fig.2, No.19) not only has certain altitudinal soil zones, but certain sectors have a maritime climate that affects vegetation and soils. The eastern sector is designated as Turfy acid coarse humus forest soils (Ivanova,1962).T h e western slope of Kamchatka has Peaty podzolized soils in the uplands, while in the low sectors adjacent to the Sea of Okhotsk extensive peat deposits are present.
SOIL CLASSIFICATION IN THE SUBARCTIC REGIONS Since 1950 there have been a number of attempts at generalized soil classification in the northern regions, but none of these classifications have been adequately tested for adaptability on a circumpolar basis. Soil
Low boggy soil
classification should embrace those soil properties that reflect c o m m o n origin, synthesis, leaching, and transformation of compounds within a system. The larger taxonomic units (Great Soil Groups or equivalent) should reflect similar physical, chemical, and biotic conditions, and the soils should possess similar sets of horizons, including number and sequence, and related morphologic properties. The subgroupings should provide not only for low-orderregional changes in qualitative processes but also for minor changes in soil materials. It is important to list several soil classificationproiscusses posals for the northern area. Ivanova (1956a)d' the general philosophy of Soviet concepts on soil classification in the northern part of the U.S.S.R. Under the general heading of Podzolic soils three types1 are recognized: (u) Tundra podzolic soils, (b) Podzolic, and (e) Turfy podzolic soils and H u m u s podzolic (humus turf) soils. In a subsequent report Ivanova (1956b) proposed a generalized soil classification scheme for soils including the Subarctic (Table 1). Zavalishnin (1955) lists three basic soil varieties within the taiga-forest zone: (a) Podzolic, (b) Turfy, and (c) Marshy. Rozov (1956) presented a scheme for soil classification in the subarctic regions and the contiguous forest regions, as shown in Table 2. Apparently there has been no attempt to apply these proposed classification systems (Tables 1 and 2) outside of the U.S.S.R.Although the general philosophies of Prasolov, Ivanova, and Rozov on soil classification mark an important step forward,m a n y critical problems still await the attention of investigators. Except for the proposals of the Soviet investigators, 1. Soviet inveatigators generally list taxonomic units from the larger to the amaller unit 88: Type [Great Soil Group]. Subtype [Famjiy]. Species [Series] and Variety [Type] (brackets indicate approximate North American equivalents).
197 14
J. C.F. Tedrow
TABLE2. Classification of subarctic soils by Ivanova (1956b) Types of soil Groups of soil formation (subclasses of soils)
Automorphic
Automorpho-hydromorphic
Class II' Boreal-permafrost taiga soil 1. Permafrost-taigasoils Taiga ferruginous soils
-
Pale yellow taiga
Pale yellow gleyey soils
Permafrost solods
Permafrost gleyey solods
-
Marshy-permafrost soils
2. Permafrost-marshy
3. Permafrost solonets soils
Hydromorphic
Class III1Boreal-taigaand forest soils
1. Podzolic forest soils 2. Turfy-taigasoils
Podzolic soils Grey-forestsoils Turfy-taiga (including the turfy-carbonated)
3. Marshy soils
-
Podzolic marshy soils Grey-forestgleyey soils Turfy-gleyeysoils
-
Marshy soils
1. Global group of classee of boreal soil formation.
the over-allproblem of soil classilkationof the Subarctic has been studied very little from the standpoint of the larger taxonomic units. Kubiena (1953)published a guide for the identificationofthe soilsof Europe which lists major varieties of soils on the European continent. While his discussion includes little on the Subarctic per se, he does imply the presence of several varieties of Podzol, Brown earth, Ranker, Bog, and Gley soils in the Subarctic of northern Europe.
REGIONAL PROCESSES WITHIN THE SUBARCTIC Soils of the subarctic possess few if any properties unique to the region. Perhaps if a single pedogenic property could be singled out for the area between the forested and non-forestedareas it would be that there is a tendency for a thicker organic mat to form under the forested conditions.As minor as these changes m a y be within short distances, the existence of slowly changing north-south pedogenic gradients within the subarctic regions has been clearly demonstrated by m a n y investigators. Soils with similar properties, however, are commonly present within the northern forest zone as well as the subarctic and even the arctic zone. Perhaps it would be realistic to state that while m a n y investigators emphasize the north-southchanges across the forest-tundra transition, other equally important changes also occur across certain altitudinal zones and longitudinal belts of the Subarctic (Fig.6). Based o n climatic data, fragmental reports, and observations, the greatest podzolic potential of the subarctic region probably exists within the subarctic regions of northern Scandinavia and in possibly Labrador, Canada (Fig.6). In these two locations the
198
podzolic process appears to manifest itself in the northern fringes of the forests as well as in some portions of the tundra proper. That there is a slowly changing longitudinalprocess within the subarctic region of Siberia appears to be fairly well established (Prasolov, 1934; Gorodkov, 1939). The drier aspect of East Siberia as campared to the western sectors of Eurasia has been brought out by a number of investigators. Podzol soils are widespread within the western portion of the Siberian Subarctic,but in the easternsector Pcdzol soilsapparently are less commonly present. Frost processes appear to play a more important than in the role in the soil systems in eastern U.S.S.R. western sectors (Fig. 6),a conditionthat is brought out throughout the Soviet literature. The sod process has been given a prominent place in the Soviet literature and is recognized as a distinct process in the northern forest zone,the Subarctic,and in certain locations within the arctic regions (Muir, 1961; Vilenskii, 1957). Vilenskii shows that the sod process is the dominant process from slightly north of 600 N. from the Baltic Sea eastward beyond the Ob River (Fig.6); some varieties of sod soils are also recognized north af this latitude. Vilenskii further states that Podzols and Turf or Sod-podzolicare the The sod most widely distributed soils in the U.S.S.R. process has been recognized only by Soviet workers and apparently there has been no attempt by investigators outside the U.S.S.R. to recognize this sod process in their classification systems. There are two areas of the Subarctic in which the penecontiguousland to the south is saline in character. South-western Alberta, Canada, has numerous solonetzic sciils (Bowser, 1960),as does the Yakutsk sector in the vicinity of the central Lena River (Vilenskii,
Soils of the subarctic regions
1932; Yelovskaya, 1965) (Fig.6). Near the northern solonetzic areas (Fig.6)are soils with special properties. Peripheral to the solonetzic soils of Alberta and extending northward to an indefinite line in the Fort Simpson-GreatSlave Lake sector is a group designated by Canadian investigators as Grey wooded soils (“ALook at Canadian Soils”, 1960). Straw yellow soils (Ivanova, 1962) occupy the northern periphery of the saline soils in northern U.S.S.R. It would be important to study these two widely separated sectors (Canada and the U.S.S.R.) not only from the standpoint of a solonetzicpodzolic gradient, but also to learn if these two regions have certain genetic affinities. The soils of the subarctic region adjacent to the Bering Strait appear to have unusually dark-coloured sola. Soviet as well as American investigatorshave described their occurrence. Whether the dark colours are a result of the maritime climatic conditions or some other factor (both sides of the Bering Strait have extensive surficial volcanic deposits) remains largely unknown (Fig.6).
BOG SOILS Since this report has centred around mineral soils, there has been little discussion of Bog soils. Plant composition,nutrient supply, acidity, moisture levels, frozen ground,and related parameters affect the nature and composition of the bogs. Scandinavian investigators have played a leading role in studies of Bog soils, although a review of their findings is beyond the scope of this report. The extensive literature on bogs within the subarctic region includes studies specifically related to regional variations in bog characteristics.Dokurowsky (1938),Kats (1959)and Radforth (in press) discuss certain variation in bogs in sectors of the Subarctic, but it is somewhat premature to suggest any well-
defined quantitative or qualitative changes resulting solely from “regional” processes.
SOME RECOMMENDATIONS FOR FUTURE STUDY This report is intended to give a generalized picture of the soils within subarctic regions. W e must recognize, however, that vast portions of the Subarctic have not been explored by trained pedologists and that an effort should be m a d e to rectify this deficiency. One of the major problems that hinders the orderly development of soil science in the northern regions is that of classification,including nomenclature. While in most scientific disciplines, investigators agree and understand such terms as quartz, Picea, Gastropod, Ordovician, and cyclonic, this is not necessarily the case with soils. Most naturalists have at least a speaking acquaintance with terms such as loam, Bog, Podzol, and Tundra, but there is considerable confusion and ambiguity in the second, third, and fourth orders of classification. In fact, there is a certain amount of discord even in some of the first orders of classihation. The Russian system of soil classification had diffused into Germany shortly before the First World W a r and subsequently into other western European countries, the United States, and Canada. While retaining certain terms such as Tundra and Podzol, investigators of various nations introduced m a n y n e w soil terms for local or regional conditions. These n e w soils and soil terms have, for the most part, been poorly defined and remain uncertainties in the global picture. In addition to the need for continued fundamental soil studies in the unknown subarctic regions,consideration should be given to the ultimate use of a uniform standard for nomenclature and classification within the subarctic regions.
Résumé Les sols des régions subarctiques (J. C. F.Tedrow) L a région subarctique est formée essentiellement par les terres qui s’étendent entre les forêts boréales formant une voûte continue et la limite septentrionale des conifères. Elle se situe généralement entre les isothermes de juillet de 1OOC et 15,60C.Elle est recouverte en majeure partie d’une couche superficielle de sédiments quaternaires reposant la plupart du temps sur du pergélisol. Les précipitations annuelles varient généralement de 25 c m à 50 c m , mais atteignent dans
les zones côtières des valeurs comprises entre 75 c m et 1,75 m. Les premiers chercheurs avaient signalé un6 modification pédogénétique qualitative à la limite de la forêt mais cela n’a jamais été prouvé. Certaines zones bien drainées sont couvertes de sols podzoliques (les uns jeunes, les autres à horizons bien définis), tandis que beaucoup d’autres sont constituées de divers sols bruns, jaunes ou gris avec des horizons assez peu différenciés.Les sols à gley occupent de vastes étendues peut-être la moitié des régions subarctiques. Les sols marécageux couvrent également
-
199
FIG. 6. Schematic diagram of some changes in regional pedogenic processes within the Subarctic.
-
4Podzo I izati on decreases
-e Frost action increases ‘MHH~ Podzolic
process tends to weaken
Northern limit of extensive sod podzolic soils Subarctic regions with high precipitation Subarctic regions with relatively high precipitation and dark-coloured soils Solodized straw-coloured soils
Grq wooded soils
k-3
Solonetzic s o i ~ s
d’importantes superficies, surtout dans les parties basses. B o n nombre de sols minéraux ont des horizons A et B du type zonal mais présentent en profondeur certaines caractéristiques des sols à gley. Les chercheurs emploient, pour la plupart, la terminologie consacrée (podzols, sols à gley, sols marécageux) pour décrire les sols des régions circumpolaires, mais ils ont aussi adopté beaucoup de noms de sols nouveaux surtout pour tenir compte des conditions locales. Dans la zone subarctique, les phénomènes pédologiques vont en s’atténuant du sud au nord, mais il ne semble pas qu’une modification pédogénétique qualitative vraiment nette survienne exactement à la limite de la forêt. Dans les zones côtières, surtout en Scandinavie, la podzolisation est plus marquée que dans les zones intérieuresplus sèches. C’est,semble-t-il, sur les sols de la zone subarctique de la Sibérie orientale que le pergélisol exerce le plus d’influence.
J. C. F. Tedrow
Discussion E. SCHENK. I mentioned during the previous session that it would be useful and even necessary to classify the pattern
and structures of the freezing soils of the Arctic and Subarctic with the assistance of pedologists. I myself made the attempt to classify the frost structures in accordance with Kubiena’s classification. N o w we see that a very great step forward has been taken by Professor Tedrow w h o records nineteen soil groups. As far as I can see at the moment the occurrence of these soils agrees with the occurrence of vegetation types and types of frost features such as patterned ground, palsas, string bogs, etc., which are shown in the maps published in Frenzel’s book on the evolution of vegetation in Eurasia after the last glaciation. With regard to the ecology of the arctic and subarctic it would be best if a pedologist tried to compare soil-vegetation-frostpattern so that w e would get a real natural classification of frost structures. In this way it would be easier to understand their wide variation of types and to see the real problems, for their origin and mechanics are no longer a problem.
J. C. F. TEDROW.There appears to be little doubt that vegetation, patterned ground and pedology should be an integrated study. W e have gained a great deal of pedologic information from botanists and cryogeologists while in return pedologic contributions to the other two disciplines have been rather small. It must be remembered that in comparison with botany and cryogeology we have very few individuals working on the subject of northern pedology. S o m e years ago Professor J. V. Drew of the University of Nebraska and I collaborated on some studies dealing with the problem of patterned ground and pedology but this was more of a preliminary rather than a complete study. I believe that we should abandon the concept of using forms and configuration of patterned ground in a genetic sense in pedology. Terms such as polygonal soil, structure soil, etc., really have little or no meaning in soil processes. If, for example, w e say here is a polygonal soil, we m a y see a tundra profile, a bog, a polar desert or merely rubble without horizon differentiation. On the other hand if we tie together the two components: type of patterned ground and genetic soil type, w e then have a realistic approach to the problem. Dr. J. Grown used this approach in the Brooks Range with success and I elaborated on the subject at the symposium held in Abisko (Sweden) in 1960. There are sitiiations in which certain types of patterned ground are associated mainly with one soil variety-for example, string bogs indicate poorly-drained soils. O n the other hand certain polygonal forms occur on m a n y varieties of soil. Dr. Schenk‘s remarks certainly point out a glaring shortcoming in northern studies and I hope that his statements will serve as a catalyst for studying this long-neglected problem. J. MALAURIE. Vous avez signalé qu’il n’y avait pas, du point de vue pédologique, de différences fondamentales entre les régions arctiques et subarctiques, les phénomènes étant seulement plus lents au sud. Ce point de vue est important, car il permettrait de mieux définir nos problèmes. Je regrette vivement que, hier après-midi,la discussion ait été 202
quelque peu abrégée à cet égard. I1 s’agit pourtant d‘un problème majeur pour le géomorphologue, qui tourne autour du polygénisme des formes haut-arctiques dans les roches consolidées et qui n’est pas -l’idée sous-jacenteà la géomorphologie subarctique étant que les formes des secondes sont, c o m m e génétiquement, dérivées des premières -sans conséquences méthodologiques pour l’analyste du modelé du terrain subarctique. Les formes arctiques n’étant pas spécifiquement définies il est bien clair que l’expression “subarctique” doit être réduite à sa seule signification topologique, au moins pour le géomorphologue. Cette question mériterait peut-être d e provoquer la création d’une commission méthodologique, à l’issue de ce colloque. Cela dit, j’aimerais que vous nous précisiez si, en terre d’Inglefield (nord-ouest du Groenland), d’où vous dites revenir et où j’ai moi-même dressé une carte géomorphologique au 1:200 O00 en 1950-51,je souhaiterais savoir, dis-je si le plateau présente bien une opposition pédologique et végétale tranchée entre le Sud-Ouest (abords du fjord de Foulke) et la région de Qaqaitsut (glacier d’Humboldt). Cette distinction avait été suggérée par le regretté Thorild Wulff, explorateur suédois, en 1917, lors de la Seconde expédition de Thulé. Aucune recherche en ce sens n’a été entreprise depuis, à m a connaissance.
J. C. F. TEDROW. Unfortunately I did not have the opportunity to travel throughout Inglefield Land and have seen neither of the two locations mentioned. I located our field camp 5 miles south of Rensselaer Bay, about 1 mile west of the main river emptying into the head of the bay and our studies were confined to a radius of about 5 miles from our camp site. Therefore I will have to limit m y discussion to this location. The most c o m m o n soil in the area is polar desert. Polar desert soil is sparsely colonized by vascular plants-perhaps less than 1 per cent but genetic soil horizons have developed as is the case in soils of the more temperate climates. The organic component in the polar desert soil at times is at a low level but in other cases it is considerable, the latter contribution is probably by algae. Tundra soils are also present in Inglefield Land but they occupy only a small percentage of the landscape.While I a m retaining the term tundra, these soils cannot be considered as typical. Instead these tundra soils should be considered as a northern polar or para variety. Until such time as w e study the problem more fully and have an opportunity to conduct laboratory studies and evaluate the data I don’t believe that I can add a great deal more. I presume that our party examined at least 50 square miles of landscape just south of Rensselaer Bay and in this locality no peats were observed. C. O. i T ~ M a~y ~I ask . Dr. Tedrow to comment a little more on the effect of the parent material on soil processes in the Subarctic.M a n y of the soils in the Subarctic are young, and their particle size and minerological composition are very variable.
J. TEDROW. Time did not permit complete treatment of the subject of subarctic soils but I a m delighted that you raised the question on the effect of parent material on the
Soils of the subarctic regions
pedogenic processes. There is no question that parent materials play an extremely important role in soil formation, not only within the Subarctic but in other regions as well. The coarse-texturedmaterials have a high leaching potential and are usually base-deficient.Such coarse-texturedmaterials give rise to accelerated leaching and accordingly we commonly find soils with podzolic features present not only within subarctic but arctic regions as well.B. N.Gorodkov pointed out that the tundra regions of northern Siberia had podzol soils on m a n y sand deposits and I have made similar observations well north of the tree line in Alaska. I have also seen well-developed podzols in the northern treeless sectors of Norway and Sweden. With loamy or especially with clayey soils there is more
Bibliography Agricultural atlas of U.S.S.R. 1960. Moscow. 308 p. (In Russian.) A look at Canadian soils. Agricultural Institute Review (Canada),vol. 15, 1960, p. 9-60. ANIKEEN, N. P.(ed.). 1959. Stratigraphic tables and geologic maps ofthe Northeast U.S.S.R.Moscow, Ministry of Geology and Conservation, U.S.S.R.Academy of Sciences. B J ~ R L Y K K E ,IC. O. 1927. Soil types and soil profiles in Norway. Proc. First International Congress of Soil Science, vol. 4, c o m m . 5, p. 223-269. BOWSER, W.E. 1960. The soils of the prairies. In: A look at Canadian soils, Agricultural Institute Review (Canada), vol. 15, p. 24-26. BRITTON, M . E. 1957. Vegetation of the arctic tundra. Eighteenth Biology Colloquium,Corvallis,Oregon. p. 26-61. BROWN,J.; TEDROW, J. C. F. 1964. Soils of the Northern Brooks Range, Alaska, 4: Well-drained soils of the glaciated valleys. Soil Sci.,vol. 97, p. 187-195. BRYSON, R.A.; IRVING,W . N.;LARSEN, J. A. 1965. Radiocarbon and soil evidence of former forest in the southern Canadian tundra. Science, vol. 147, p. 46-48. DAHL, E. 1956. Mountain vegetation in South Norway and its relation to the environment. 373 p. (Skrifter utgitt av Det Norske Videnskaps-Akademi i Oslo I. Mat.-Naturv. Klasse. No. 3.) DAY, J. II.;LEAHEY, A. 1957. Reconnaissance soil survey of the Slave River Lowland in the Northwest Territories of Canada. Ottawa, Canadian Department of Agriculture, 44 p. -_ ; RICE, H . M. 1964. The characteristicsof some permafrost soils in the Mackenzie Valley, N.W.T.Arctic, vol. 17, p. 223-236. DOKUROWSKY, W.S. 1938. Die Moore Osteropas und Nordastiens. Handbuch der Moorkunde. Berlin, Borntraeger, 117 p. DOBROVOLSKI,V. V. 1963. Landscape-geochemicalfeatures of mountain tundras of Kola Peninsula. Pochvovedenie, no. 2, p. 25-32. DREW, J. V.;SHANKS, R. E. 1965. Landscape relationships of soils and vegetation in the forest-tundra ecotone, Upper Firth River Valley, Alaska-Canada. Ecological Monographs, vol. 35, p. 285-306.
of a tendency to have a brown-coloured soil similar to the Braunerde or Braun Waldboden and with the more clayey soils there is increased tendency towards poor drainage. I believe that your second sentence is more of a statement rather than a direct question. There have been m a n y ideas set forth on the problem of correlating soil development with absolute chronology and I believe that most of them are fairly straight forward. American investigators have shown that the older moraines in southern Alaska had better developed soils than the younger moraines. Dr.Jerry Brown of Hanover, New Hampshire, however, worked on m o r e recent glacial deposits of the Brooks Range in Alaska and established that very often the mineral composition w a s as important in soil development as was the age factor.
/ Bibliographie ELLIS,J. H.1938. The soils of Manitoba. Winnipeg, Man., Economic Survey Board, 112 p. FERNALD, A. T. 1964. Surficial geology of the central Kobuk River Valley, Northwestern Alaska. 31 p. (U.S. Geological Survey Bulletin 1181-K.) FEUSTEL, I. C.; DUTILLY, A.; ANDERSON, M. S. 1939. Properties of soils from North American arctic regions. Soil Sci.,vol. 48, p. 183-199. FILIMONOVA,L. G. 1965. Characteristics of the taiga soils of the Alden and T o m m o t regions of the Yakut A.S.S.R. Pochvovedenie,no. 3, p. 13-19. GERASIMOV,I. P. 1954. (Soil M a p 1 :4,000,000.)Dokuchaev Institute of Soils, U.S.S.R. Academy of Sciences. . 1956. Soil m a p of the world. Priroda,vol. 10, p. 5-13. . 1959. Brown forest soils. Pochvovedenie,no. 7,p. 69-80. GLINKA~ K.D.1931. Treatise on soil science. (Translated from the Russian and published by the Israel Program for Scientific Translations.)Washington, D.C., OTS.674 p. GORODKOV, B. N. 1939. Peculiarities of the arctic top soil. Izv. Gosud. Geogr. Obshch.,vol. 71, p. 1516-1532. GRIGGS,R.F. 1934. The edge of the forest in Alaska and the reasons for its position. Ecology, vol. 15, p. 80-96. HARE, F.IC. 1951. S o m e climatological problems of the arctic and sub-arctic. In: T. F. Malone (ed.), Compendium of Meteorology, p. 952-964.Boston, American Meteorological Society. HAUGEN, R. K. 1965. Climatic implications of timberline in interior Alaska. INQUA,7th Congress (Abstracts),p. 197. HUSTICH, I. 1953. The boreal limits of conifers. Arctic, vol. 6, p. 149-162. . 1960. Plant geographical regions. In: A. S o m m e (ed). Geography of Norden, p. 54-62. Oslo, Cappelens Forlag. IVANOVA, E. N. 1956a. Classification of soils of the northern part of European U.S.S.R. Pochvovedenie, no. 1, p. 70-88. . 1956b. A n attempt at a general soil classification. Pochvovedenie,no. 6, p. 82-102. . 1962. Various chapters in: Soil geographical zoning in the U.S.S.R. U.S.S.R.Academy of Sciences. (Translated by the Israel Program for Scientific Translations. New York, D. Davey & Co., 480 p.) KARAVEYEVA, N. A. et al. 1965. Peculiarities of soil
__-
__ __-
203
J. C. F. Tedrow
formation in the tundra-taiga frozen regions of Eastern Siberia and the far east. Pochvovedenie, no. 7, p. 26-36. KATS, N. Y.1959.Bogs and peats of North America. Pochvovedenie, no. 10,p. 44-53. KELLOGG, C. E.;NYGARD, I. J. 1951.Exploratory study of the principal soil groups of Alaska. U.S.Department of Agriculture. 138 p. (Agricultural Monographs, no. 7.) KOSHEL’KOV, S. P. 1961. Formation and subdivision of forest floor in southern taiga coniferous forests. Pochvovedenie,no. 10,p. 19-29. KRASYUK, A. A. 1925.As quoted by K. D. Glinka (1931), Treatise on soil science. KBAUSE, H.H.;RIEGER, S.; WILDE, S. A. 1959. Soils and forest growth on different aspects in the Tanana Watershed of Interior Alaska, Ecology,vol. 40 p. 492-495. KREIDA, N. A. 1958. Soils of the eastern European tundras. Pochvovedenie,no. 1,p. 62-67. KRISCHE, P. 1928. Die Verteilung der landwirtschaftlichen Hauptbodenarten in Finnland (Bodenkarten). Berlin, Parey. KUBIENA,W.L.1953.The soils ofEurope. London, T.Murby. 317 p. LARSEN, J. A. 1965. The vegetation of the Ennadai lake area, N.W.T.: Studies in subarctic and arctic bioclimatology. Ecological Monographs, vol. 35,p. 37-59. LEAHEY, A. 1947. Characteristics of soils adjacent to the Mackenzie River in the Northwest Territories of Canada. Proc. Soil Sci. Soc. Amer., vol. 12,p. 458-461. . 1949.Factors affecting the extent of areable lands and the nature of the soils in the Yukon Territory. Proc. Seventh Pacific Science Congress,vol. 6,p. 16-20. . Soils of the arctic regions of Canada. (Unpublished.) LIVEROVSKI,G. A.; RUBTSOVA, L. P. 1959.A classification scheme of soils of the plain territories of the Soviet Far East. Pochvovedenie, no. 4,p. 60-70. LOTHE, A. 1950. Alpin podsoleringsgrense i Vesterålen. Tidsskr.for det norske landbruk. Arg., vol. 57,p. 199-213. MIDDENDORF, A.VON. 1864.Sibirsche Reisen,vol. IV, part 1. Ubersicht uber die Natur Nord. und Ostsibiriens, Vierte Lieferung: Die Gewaschse Sibiriens, p. 525-783. St. Petersburg. MUIR, A. 1961.The podzol and podzolic soils. Adv. in Agro., no. 13,p. 1-56. NORDHAGEN, R.1927.Die Vegetation und Flora des Sylenegbietes. Eine pflanzensoziologische Monographie. 612 p. (Skrifter utgitt av Det norske videnskaps-akademii Oslo. I. Mat-naturv. klasse. No. I). OBRUCHEV, S.V. 1939. Past glaciation and Quaternary history of the Chukotsk region. Izvestia, p. 129-146. U.S.S.R. Academy of Sciences. PAWLUK, S. 1960. S o m e podzol soils of Alberta. J. Soil Sci.,vol. 40,p. 1-14. POLUNIN, N. 1951. The real arctic; suggestions for its delineation, subdivision and characterization. J. Ecol., vol. 39,p. 308-315. . 1955. Aspects of arctic botany. American Scientist, vol. 43,p. 307-322. POLYNTSEVA, O. A. 1958. Soils of the southwesternpartof the Kola Peninsula. Published by the Israel Program for Scientific Translations. Washington, D.C., 135 p. (U.S. Dept. of Commerce, OTS 61-11498.)
_-
_-
__
204
PORSILD, A. E. 1957. Atlas of Canada, plate 38, Ottawa, Department of Mines and Technical Surveys. PRASOLOV,A. E. 1934.Concerning the problem of the classification and nomenclature of soils. Trudy Pochvennogo instututs im. V. V. Dokuchu‘eva,vol. VIII. RADFORTH, N. W.1963.The origin and significance of oiganic terrain features. Proc. International Conference on Permafrost, Purdue University (United States). (In press.) RAPP, A. 1960. Literature on slope denudation in Finland, Iceland, Norway, Spitsbergen and Sweden. Zeitsch. fur Geornorph., vol. 1,p. 33-47. ; RUDBERG, S. 1960.Recent periglacial phenomena in Sweden. Biuletyn Peryglacjalny, vol. 8, p. 143-154. RIEGER, S. Report of investigations of soils near the west coast of Alaska. (Unpublished.) RITCHIE, J. C. 1959. The vegetation of Northern Manitoba. III. Studies in the subarctic. 56 p. (Arctic Institute of North America technical paper 3.) ROSSEAU, J. 1952. Les zones biologiques de la péninsule Québec-Labrador et 1’Hemiarctique. Cunad. J. Bot., vol. 30,p. 436-474. ROZOV, N.N.1956.Principals of elaborating a genetic classification of soils.Pochvovedenie,no. 6,p. 76-83. . 1962. Various chapters in Soil geographical zoning in the U.S.S.R.U.S.S.R. Academy of Sciences. (Translated by the Tsrael Program for Scientific Translations. New York, D.Davey &Co., 480 p.) RUDBERG, S. 1962.Geology and morphology of the “Fjells”. Biuletyn Peryglacjalny, vol. 11, p. 173-186. SCOTTER, G.W.1963.Effects of forest fires on soil properties in northern Saskatchewan. Forestry Chronicle, vol. 39, p. 412-421. SHELUDIAKOVA, V. A. 1938.The vegetation of the Indigirka river basin. Soviet Bot., no. 4-5,p. 42-79. SIGAFOOS, R. S. 1958. Vegetation of Northwestern North America, as an aid in interpretations of geologic data. p. 165-183.(U.S. Geological Survey Bulletin, 1061-E.). SOCHAVA, V. B. 1954.The geobotanical m a p of the U.S.S.R. Priroda,vol. 10,p. 36-42. SPIZMARSKII,T. N. 1937. Geological sketch of the LenaIndigirka district. Trudy Arktisch. Inst., vol. 87,p. 313-
__
__
366. STOBBE, P. C. 1960. The great soil groups of Canada. In: A look at Canadian soils, Agricultural Institute Review (Canada),vol. 15, p. 20-26. STREMME, H. 1927. General map of the soils of Europe. Danzig, International Society of Soil Science. TAMM, O. 1927. Die Klimatischen bodenregionen in Schweden. Proc. First International Congress of Soil Science, vol. 1, c o m m n . 1, p. 269-285. . 1932. D e r braun Waldboden in Schweden. Proc. Second International Congress of Soil Science, vol. 5, comm.5, p. 178-189. .1950.Northern coniferous forest soils. Oxford, Scrivener Press. 253 p. TARGUL’YAN, V. O. 1959.First stages of weathering and soil formation on igneous rocks in the tundra and taiga zones. Pochvovedenie,no. 11,p. 37-48. TEDROW, J. C. F. 1963. Arctic soils. Proc. International Conference on Permafrost, Purdue University (United States). (In press.)
__
Soils of the subarctic regions
-_. ,
B R O W N , J. 1962. Soils of the Northern Brooks Range, Alaska: Weakening of the soil-formingpotential at high arctic altitudes. Soil Sci., vol. 93, p. 254-261. _-. , .1965. Soils ofarctic Alaska. Presented at INQUA meeting, Boulder, Colorado. (Unpublished.) _-. , HARRIS,H . 1960. Tundra soil in relation to vegetation, permafrost, and glaciation. Oikos Acta Oecologica Scandinavica, vol. 11, p. 237-249. , * DREW, J. V.; HILL, D . E.; DOUGLAS, L. A. 1958. Major genetic soils of the Arctic Slope of Alaska. J. Soil sci., vol. 9, p. 33-45. TIKHOMIROV,B. A. 1960. Plantgeographical investigations of the tundra vegetation in the Soviet Union. Canad. J. Bot.,vol. 38, p. 815-832. VILENSKII, D.G. 1932. Brief history andprinciplesummary of investigators of saline and alkali soiIs in the U.S.S.R. Proc. Second International Congress of Soil Science, vol. 5, c o m m . 5, p. 309-326.
__
--
--. 1957. Soil science, Moscow. Translated by the Israel Program for Scientific Translations. Washington, D.C., OTS, 488 p. WILDE, S. A.; KRAUSE, H. H. 1960. Soil forest types of the Yukon and Tanana Valleys in subarctic Alaska. J. Soil Sci., vol. 11, p. 266-279. WRIGHT, J. R.;LEAHEY, A.; RICE, H . M . 1959. Chemical, morphological, and mineralogical characteristics of a chronosequence of soils on alluvial deposits in the Northwest Territories. Canad. J. Soil Sci., vol. 39, p. 32-43, Yelovskaya,L.G.1965. Saline soils of Yakutia.Pochvovedenie. no. 4, p. 28-33. ZABOYEVA, I. V. 1965. Gley-podzolic soils of the Northeast European U.S.S.R. Poehvovedenie, no. 7, p. 14-25. ZAVALISHNIN,A. A. 1955. Discussion on the present-day problem of the taxonomy and classification of soils. Pochvouedenie no. 4,p. 69-81.
205
Soil ecosystems in .subantarcticregions Francesco di Castri René Covarrubias Ernst Hajek
INTRODUCTION In this report data are given on the structure of terrestrial ecosystems of some subantarctic regions. The information will permit the discussion of aspects related to ecological regulation and function in these environments, leading to theoretical conclusions which apparently are also valid for the analogous subarctic zones. This research belongs to a series of investigations that the Working Group on Terrestrial Ecology of the Universidad de Chile has carried out in the last six years, covering an area from the humid and dry tropical regions of South America, mainly of Chile, Paraguay, Brazil and Argentina, up to the Antarctic continent. The main purpose is to study, under different climatic and geological conditions,the problem of biotic diversity, with its structural,functiona land biogeographical implications. Previous to a deeper analysis, the following commentaries must be made. 1. Not all biological elements that integrate these ecosystems have been considered,since at the present status of our knowledge it is impossibleto survey all species living in an environment, even so simplified as the antarctic one. Primarily, soil invertebrates have been studied (Arthropoda and Annelida) and in a minor degree some plants (mosses, lichens, algae and gramineae). Only fragmentary data were obtained on aquatic invertebrates of the soil (nematoda, rotifera, tardigrada). Populations of protozoa and bacteria were not studied. 2. Not only soil communities in a strict sense were analysed,but also related zoocenosis. M a n y observations,therefore,have been made on fauna of mosses, lichens, algae and organic deposits, such as faeces of marine birds and seals,marine algae on the shore, penguin and sea-gullnests, etc.
3.According to the considerations made in previous paragraphs,it is evident that in this paper w e do not always refer to the ecosystem in its exact meaning. In effect, autotrophic strata are sometimes absent, sometimes they have not been analysed. Also, w e have never studied the activity of microbial organisms (decomposers). 4.W e do not pretend to include here a complete revision of literature dealing with soil ecosystems of subarctic and subantarctic regions. W e will depend mainly on the results of our o w n research with a quantitative and synthetic approach based on principles of information theory. W e want to cite only some general papers in relation with Antarctica and particularly its fauna of arthropods and others with large bibliography (Covarrubias, 1966; Gary, 1962; Dalenius, 1965; Gressit and Weber, 1959; Gressit and Leech, 1961; Llano, 1962; Piyor, 1962).
STUDIED REGION The results presented in this paper are restricted only to the islands located south of the meridional point of the American continent, as well as the septentrional zone of the antarctic peninsula. Most of our observations refer to the South Shetland Islands. Results have been selected from these zones, not included in the antarctic circle, but almost every one south of the antarctic convergence, because here particularly is found an antarctic tundra,which physiognomically is more like its analogous arctic formations. Furthermore, the boundaries of the antarctic convergence for fixing a differentiationbetween antarctic and subantarctic regions is surely valid for aquatic fauna,but their meaning is problematic in relation to soil ecosystems. So, if a vegetational criterion for delimitation is
207 16*
F. di Castri;R. Covarrubias;E.Hajek
adopted, w e can affirm that the whole area analysed is found south of the tree line,but north-ofthe meridional limit of flowering plants, like Deschampsia and Colobanthus (Llano, 1962). Undoubtedly humid or dry tundras exist in the southern sector of austral America, but their presence is due often to edaphic (poor drainage) or altitudinal causes. A comparison between these soil communities close to Antarctica and some of them located in the magellanic archipelago will be included. D u e to the fact that the focusing of this report is not descriptive, it seems unnecessary to give detailed information on the studied area. Also, most of its ecological conditions will stand out in the following comparison of subantarctic and subarctic territories. S o m e geomorphologicaland biological data on the same zone can be found in the papers of Araya and Hervé (1964)and Follmann (1965). Climatic trends of the subantarctic regions close to South America m a y be deduced from Figure 2, in which the hythergraphs (mean monthly temperature in the ordinate, monthly precipitation in the abscissa) of some selected stations (Fig.1) are shown. Since temperature constitutes directly or indirectly the principal limiting factor in polar and subpolar zones, graphical representation of six ranges have been added, from mean temperatures below O0 C to those above 100 C; the progressive darkening of these figures correspondsto an increase of thermic unfavourability from north to south. The main conclusion from Figure 2 is the great climatic uniformity of these stations,thus revealing a clear oceanic effect. Also, the stations located in the
50‘
c
90”W
60. W
30.W
FIG. 1.Geographicallocation of selected subantarcticstations. (1) San Pedro;(2) Puerto Eden;(3) Puerto Consuelo;(4)San Isidro;(5) Navarino;(6) Diego Ramiree;(7) Chilean Station “Gabriel González Videla”; (8) Stanley, Falkland Islands; (9) Grivitken, South Georgia Island; (10) Laurie, South Orkneys. 208
Pacific (left side of the figure) show a greater favourability than the corresponding Atlantic ones (right side); apart from this, the zone of the drift ice reaches further north on the Atlantic than on the Pacific.
COMPARISON WITH SUBARCTIC AREAS Our purpose is not to compare by means of a general review the subantartic and subarctic regions, but to comment on the main differences and similarities only in aspects influencing the structure of subpolar soil communities. The most important difference is the unequal proportion of land and sea in both zones. Whilst large continental masses of North America, Asia and Europe belong to the subarctic area the subantarctic territory is very discontinuous with little extensions of emerged land. This is the basic factor from which a series of climatic and biogeographical consequences are derived. On the other side,the antarctic lands encircle a real continent, Antarctica, which in other geological ages had a fundamental role in the distribution of numerous groups of terrestrial invertebrates (palaeoantarctic lines). This is indubitable considering the present biogeographicalfindings,so the theory of intercontinental bridges or Wegener’s principles are accepted. The main climatic differences m a y be pointed out from a comparison of Figures 2 and 4. The climate is more equal in the subantarctic areas, even comparing them with the most maritime of the subarctic regions included between Greenland and Scandinavia (stations 4, 5,6 and 8 of Figures 3 and 4). Also in boreal zones, the location at the border of the coast of continents involves continental climatic trends, with increasing thermic amplitudes and concentration of rainfall in summer. This is seen for North America in stations 1, 2 and 3 and for European and Asian Russia in Stations 9, 10 and 11 (Fig.3 and 4). This continentalism, which is implied in the subarctic areas, is not unfavourable from a biological point of view. It is true that lower winter temperatures are reached than in austral regions,but this happens when the snow cover provides a good protection for soil organisms. On the other hand summer temperatures, higher in boreal zones and coincident with the rainy period, permit relatively favourable conditions for plants and animals of the arctic tundras. Furthermore, a simple observation of the stripes representing thermic favourability along the year (Fig.2 and 4) shows that at the same latitude,the arctic zones offer more favourablelife conditions than the antarctic ones. The extreme isothermic conditions of the subantarctic zones appear also in Figure 5, presenting daily thermic oscillations registered at Robert Island (South Shetlands) during the most favourable period of the
Soil ecosystems in subantarctic regions
FIG.2. Hythergraphs and mean temperatures of subantarctic stations corresponding to Figure 1.
-20
8
?
O
L
F
.20
-20
I
-20
I
5,
16L
I,!
-=-OLI 20
-20
,
O
JAf010JCYAMJ
,
200
,lo,
400
rnrn
-20
o
200
boo
mm 0 L
a
below
00
c
Ea between 00 C and 2.50C
Temperature in 0 C Precipitation in m m
i Ul between 2.5O C and 5.0° C i-
2
between 5.0° C and 7.5O C between 7.50 C and 10.00 C 0 above 1O.OoC
209
F. di Castri;R. Covarrubias;E.Hajek
90' 120"
60"
I
E O"
180'
W
120"
I
60'
90'
FIG.3. Geographical location of selected subarctic stations. (1) Barrow Point; (2) Frobisher; (3) Chesterfield Inlet; (4) Jan Mayen; (5) Akureyri; (6) Ivigtut; (7) Spitsbergen; (8) Inoy;(9) Archangel; (10) Anadir, Novo Mariinsky Post; (11) Okhotsk. summer. On some days,the difference between maxima and minima scarcely exceeds lo C. Regarding the hydric factor, austral zones have a good distribution of precipitation,which in some areas close to the American continent becomes limiting due to excess. Air humidity (relative and absolute) and cloudiness tend to be higher than in arctic zones. Undoubtedly hydric deficit exists for plants in both regions due to low temperatures impeding the utilization of water. Neverthelesa,it does not represent a real aridity from a climatic point of view. Subpolar regions are in general very ancient geologically and both have undergone strongly the effects of pleistocenic glaciations. Mountainous physiography of the subantarctic zone would have enabled the persistence of a greater amount of nunataks. After glaciations, the recolonization by terrestrial organisms of territories close to Antarctica has been made principally through transport by wind or water, without considering the possible survival of prepleistocenic forms. Contrasting with the Arctic, there did not exist the areal continuity necessary for the immigration of large terrestrial species from regions less influenced by ice. Concerning soils, solifluction is probably the major ecological factor; the abrupt topography of these subantarctic lands makes soils less stable and even less developed than the arctic ones. A fundamental fact for antarctic terrestrial ecosystems is the constant introduction of materials of marine origin through the excreta of birds which nest here; this leads to the formation of so-called ornithogenic soils. But the strongest differences are of biological nature.
210
Referring to vegetation, the austral area has not the extended contact zone between tundra and taiga which proposes one of the most interesting problems of the Arctic. T h e line marking the meridional growth limit of trees in the southern hemisphere would pass, hypothetically, through an area almost completely occupied by oceans. Also, the taigas are restricted to small areas of the Andes where a biological continentalism is found;only here is it possible to observe little extended boundaries between taiga and tundra. The dominant plant formation in the southernmostpart of the American continent is the magellanic cold woodland with oceanic climate and wet soils; from a floristic and faunistic point of view, it seems to be a degradation of Valdivian rain forests, due to cold. Antarctic tundra looks very poor when compared with the Arctic,having only three species of flowering plants (Llano, 1962) and these always very scarce. The antarctic terrestrial fauna is represented by a number of groups m u c h lower than in the arctic, not true land vertebrates,since all depend from marine ecosystems for their subsistence. On the other hand aquatic fauna close to Antarctica is relatively diversified. Seasonal and cyclic migrations characterizing m a n y elements of subarcticterrestrialecosystems have little significance here. Finally, the utilization of natural resources is very different, restricted for Antarctica only to those of marine origin. Different also are the conservation practices, due to absence of stable h u m a n populations in the austral pole. In spite of the differences shown here, there exist also great similarities between both subpolar zones. M a n y of their ecological problems therefore might be enclosed within the same theoretical frame. W e would like to point out now, schematically,the main c o m m o n features. 1. The duration and location of the favourable period, that is,the lapse in which thermic conditionsbecome less limiting. 2. The response of vegetation to climate, at least concerning tundra formations. 3. The pedogenesis. 4.The structure of terrestrial ecosystems, with little complexity, scarcely integrated and corresponding to a low level of maturity. 5.The type of evolution, being natural selection directed mainly by abiotic elements. 6. The presence of a few peculiar factors, whose biological effects are not well known, such as photoperiodicity, snow cover and permafrost.
TERRESTRIAL ECOSYSTEMS IN SUBANTARCTIC REGIONS The following considerations refer only to the region directly studied by us, but at least some general
Soil ecosystems in subantarctic regions
2El
FIG.4. Hythergraphs and mean temperatures of subarctic stations corresponding to Figure 3.
'C
.C
-
-20 O
ZOO
4OOmm
D
200
4OOmm
*C
-20
20
O
ZOO
4OOmm
O
200
40Omm
5
9
11
-20
'C
.C
2El .C
-20 O
-
$m below
Zm
ooc o'c
between
and 2.5-C
b e t w e e n 2.5.C
2m
zm
200
400mm
T e m p e r a t u r e in O C Precipilallonin m m
a n d 5.0'C
b e t w e e n 5.OoC a n d 7.5.C b e t w e a n 7.5.C a n d 10.0.C a b o v e 1O'C
conclusions might be projected to other subantarctic and even subarctic zones. W e have anticipated that analysed materials were principally the populations of aerobíontic arthropoda, generally k n o w n as soil mesofauna. T h e y deal mainly with mites and springtails. T h e major part of the samples were quantitative, being the extraction of animals m a d e by m e a n s of Berlese-Tullgren funnels. More methodological explanations can b e found in previous papers (Covarrubias, 1965a, 1966). Observations of present paper are based o n 150 samples, m o r e than 250,000 arthropods being counted and classified. Other large materials of the s a m e zone, which have been partially considered for preliminary
commentaries, are under study. M u c h of the data here presented, especially in Table 1 refer to a series of 68 quantitative samples with a total of 111,762 specimens. T h e criteria adopted for experimental design and interpretation of results are the s a m e explained in later publications, from which only t w o references will be given (di Castri et al., 1964; di Castri and Astudillo, 1365~). In the first of these papers, details are included o n the properties and use of indices based o n information theory, Shannon-Wiener and Brillouin's expressions, repeatedly cited in this report. M a n y of our theoretical discussions are coincident with the points of view of Margalef (1963).
211
F. di Castri; R. Covarrubias; E. Hajek
NOISE I I I I I
I F
27 29 31
Dec. 1965
2
4
6
e
io 12
-
IC 16
18
I
L
I
T
E
R I
4 + +
Days
Jan. 1966
FIG.5. Daily maxima and minima temperatures at Robert Island (South Shetlands).
FIG.6.Schematic representation of the main factors involved in ecological regulation of biotic diversity.
T o facilitate the exposition, w e shall group the subjects to be c o m m e n t e d o n in this section in three parts : ecological regdation, biocenology, and c o m parison with other ecosystems. In the first part, principles based o n cybernetics and mainly o n information theory will often be used, touching o n problems of structure, function, ecological succession and stability in these ecosystems. T h e classical aspects of frequency, dominance, abundance, etc., will be discussed in the second part, o n biocenology.
structural complexity and maturity of the system and in a certain sense also o n its stability (homeostasis). T h e structure, therefore, is not considered from a statical and descriptive point of view, but synthetically as a base of the system conduct. Figure 6presents a schematic view, a very elementary model, of the most important factors involved in the ecological regulation of the system. T h e principal variable is diversity; it is positively influenced by time, so that the complexity has a tendency to b e greater as the period of evolutive and successional history of the system increases. Alternatively, climatic or edaphic perturbations (environmental noises) interfere with the progressive increase of information. Also, as succession advances, “stabilizing filters” appear ( h u m u s layers, plant protective cover, etc.) which are interposed between the m a i n variable (diversity) and the extrinsic factors, diminishing or suppressing the transmission of variety and filter. If the interferences are very strong of the efficiency of the filter is poor, the system is controlled by extrinsic abiotic factors (climatic or edaphic); but if the filteris able to isolate the system of external influences in a high degree, then conditions of autocontrol arise; in this case intrinsic biotic factors (integration of species) b e c o m e dominant. Naturally, the problem is m u c h m o r e complex. In the mentioned papers, this has been discussed with a theoretical and mathematical approach o n the basis of examples provided by investigations carried out with the s a m e focusing from the tropics to Antarctica. Let us analyse from this point of view the situation in subantarctic ecosystems. T h e values of m e a n diversity, expressed as information in bits per individual, are given in Table 1 and Figures 7 a n d 8.Those appear-
ECOLOGICAL REGULATION
Points of view adopted here have been illustrated in previous papers and particularly in a recent revision (di Castri and Astudillo, 1965~). Nevertheless, it seems convenient to present t h e m in the shortest way, before discussing results of our subantarctic investigations. T h e structure of a n ecological system can b e measured by the n u m b e r of elements (species) integrating it a n d the frequency of t h e m (distribution of individuals within the species); this corresponds to a measure m u c h used in ecology, the biotic diversity. With higher diversity the possibility of interrelations between species increases and thus the probability of n e w channels of communications; as a result, a greater a m o u n t of information can be established in the system (Margalef, 1957). According to a structural approach, biotic diversity can be equated with information (negentropy or informative entropy); these terms will often be used in this paper with the s a m e meaning. A measure of information will give antecedents o n the degree of
212
Soil ecosystems in subantarctic regions
ing in the table and in Figure 7 were obtained using the formula of Brillouin; those of Figure 8 by means of the expression of Shannon-Wiener.These values are always very low for all communities when compared with ecosystems of other regions, exception made by cold and w a r m deserts. In this sense,subantarcticecosystems are characterized by a structural simplicity and a little degree of maturity. On the other hand, filtering mechanisms of this closed system DiversityZFilter also have low efficiency. The isolating effect of litter and mull does not exist here. Out of this, the strata of fruticose (like Usnea fasciata) or crustaceous lichens (e.g. Caloplaca cinericola) assure a very poor environmental protection for mixed populations of soil animals. A more favourable situation is found in the dense associations of mosses where diversity shows its maximal value. Another mechanism, capable of giving some independence from climatic conditions, is the possibility of soil fauna to reach deeper and more protected layers; neither can that be accomplished efficiently in these ecosystems,given the nature of soil, rocky or too wet and therefore little oxygenated. W e shall n o w discuss the determining causes of this state of little maturity and complexity in subantarctic ecosystems. These might be reduced fundamentally to: deficiency in the positive action (short time passed) or excess of negative feeding (strong environmental perturbations). Undoubtedly subantarctic ecological systems are historically recent,in any case after the last glaciations. In all of these regions rich forest formations existed, but they were completely destroyed due to palaeoclimatic changes. Nevertheless,there are in the world equally recent ecosystems (in alluvial plains, volcanic territories, etc.), which, notwithstanding,show greater structural maturity. On the other hand, when the possibility of recolonization from neighbouring continents is considered, it seems that long distances do not represent an impossible barrier for terrestrial invertebrates, due to transport by birds, sea currents and winds. The significance of the last factor is pointed out by Grescitt et al. (1961)in the Antarctic area trapping of airborne insects. If a smallnumber ofzoological groups and species is present here (see Table i), this is due not so m u c h to the geographical impossibility of n e w species arriving from far territories,but to the very selective environmental conditions,which permit only the survival of a few well-adapted or euryoic species. Synthetizing,the short historical age of the system has an influence on the low diversity,but it does not seem to be the main factor. In other words, it can be said simplistically that subantarctic ecosystems have a young character due not to the lack of time,but to the extreme abiotic interferences which have almost detained their development towards maturity. It would be more extensive to discuss subantarctic
abiotic factors, analysing them as a source of “noise” for terrestrial communities. W e are considering only a few of them. Environmental perturbations can be climatic or edaphic; on the other hand, considering the pattern of transmission according to the theory of communications, they can be classified as incessant or intermittent. L o w temperature is the main climatic perturbation. Its character is almost incessant in these isothermic regions; nevertheless, at the soil surface thermic ranges are larger, this being here favourable for soil organisms because higher temperatures are reached. The wind is an important source of constant interference, but principally of faunistic homogenization; facilitating the dispersion of arthropoda, the wind reduces total biotic diversity and makes possible the genetic flux between populations widely separated by sea. At the present state of our knowledge,it is impossible to interpret synthetically the biological effects of the light/darknessalternates and snow cover. Investigations carried out in the Italian Alps (di Castri, 1960; di Castri and Astudillo, 19653)demonstrate that snow would have a mechanical filtering action, protecting soil communities against thermic perturbations. The soil instability by solifluction is probably the most important edaphic interference; obviously, this process is aggravated by slopes. Other edaphic “noises” also of permanent action are the low penetrability of skeletal soils, permafrost and the high water-content in soil. More intermittent is the effect of soil humidity,which can pass from saturation states (in summer) to others of “physiological” aridity (in winter) following thermic changes. Accordingly from the previous discussions,w e shall synthetize some general facts. First, there is no doubt that ecosystems are controlled here by abiotic causes, climatic and edaphic ones. This means a deficient homeostasis, a very low integration of species,a characteristicpattern of evolution of them, a particular form of population growth and some functional projections which will be analysed later. Second,it is evident that the major part of perturbations are those which presuppose a transmission of incessant variety. Following the theory of communications, systems subjected to them come to have a certain “adaptation”, which implies the existence of some transmission mechanisms capable of passing over the noise. T o understand what this adaptation is at the level of communities,it is necessary to make a digression on population density. In relation with this, data are given in Table 1 and Figure 7. In contrast to what happens for biotic diversities, mean densities in subantarctic ecosystems are clearly higher than those observed for the same groups in any other terrestrial community. An extreme case is the density detected in a sample of terrestrial algae (PrasioZa crispa):
213
F. di Castri; R.Covarrubias; E.Hajek
116,832 arthropods of the mesofauna for 1,000 cc of material, almost exclusively oribatei (Covarrubias, 1966); it is by far a top value of density in comparison with any other soil biocenosis. Therefore, in terrestrial communities of this region a high replication of individuals of the same species is very often found. This repetition of elements of the same class, which can be defined as redundancy, represents the adaptativemechanism of the system for assuring the transmission of messages (biocenotic or genetic) in spite of the constant perturbation of environment. Certainly, this diminishes the possibility of increasing the total information. Apparently, this transmission mechanism is not economic, but the input of organic substances of marine origin by birds and seals assures to these extreme communities an important source of energy for this kind of regulation. Faunistic homogeneity which is demonstrated at sample level is also verified when a series of samples is considered. Thus another characteristic of these ecosystems is evidenced,their great affinity. For example, in an altitudinal transect in which ten samples were taken with values of diversity fluctuating from O to 2.21 bits per individual,the diversity of the universe was only 2.23 bits. This demonstrates that when new elements are added, little information is provided, due to their similarity. In tropical zones the opposite is observed. On the other hand, if absolute information of subantarctic ecosystems is low, the study of a sample or the finding of a species enables the researcher to predict with great probability the finding again of a sample of the same conditions, or individuals of the same species. This is due to redundancy, faunistical homogeneity and affinity between systems. In this restricted sense, the relative information for the researcher is high. This consideration obliges one to approach the problem of sampling in this region in a different form to that adopted for the opposite tropical ecosystems. In Figure 7 the different zoocenoses are characterized through mean values of diversity and density. There are two groups of communities: those of the left of the figure (from mosses to terrestrial algae) with a relatively high diversity and low density; those of the right (from faeces to penguin nests) in which generally an opposite situation exists. In the first group animal communities in close contact with plant associations are considered, thus integrating a real ecosystem with producers and consumers, although in the most simplified way. In the second, animal populations depend on more unstable substrata, such as deposits of dead organic substances (faeces, nests and marine algae on the shore). The intermediate position of terrestrial algae would be due to their temporaltransitoriness(Covarrubias,1965b). T h e faunistic diversity in a contact zone between associations oflichens and algae was higher than when fauna
214
within each association were separately analysed;
this is an edge effect (ecotone). Finally, the "jump" in the values of diversity,from zoocenosis ofthe second group to those of the first one, might be interpretated by the arising of a n e w type of control of biological nature. Without becoming dominant, this would be opposed to extrinsic abiotic controls, thus providing comparatively greater conditions of stability. T h e changes of diversity observed in Figure 7 could be due also to the fact that studied communities would represent different serial stages of an ecological succession. This states a very interesting problem: to k n o w if subantarctic environments, in spite of their extreme conditions, permit a certain successional progress. Figure 8 shows the results of another series of samples grouped according to a probable successional order; diversity was calculated by the ShannonWiener expression, which has already been used to mcasure the changes in informative entropy during
-
Density = A--A individuals/l OOGcc
= Diversity (bits/individual/cc)l O00 0 0
4 0 O00
4 I
I I I 30.000
I I
I I I I I I
I I
I I
20.000
I I I I l
I I
1o.ooc
O
FIG. 7. Diversity and density of mesofauna in some subantarctic environments.
Soil ecosystems in subantarcticregions
o1 , NESTS
1
FAECES
SOILS
-
TERRESTRIAL LICHENS M O S S E S ALQAE
FIG.8. Progression of mesofauna diversity in subantarctic tundras. succession (di Castri et al., 1964). This process has in the subantarctic region the same tendency evidenced for other ecosystems, i.e., the progressive increase of information,but everything develops at a lower level of diversity. With a certain statistical determinism, it could be stated that succession is analogous to a programme, during whose development the system acquires a memory permitting a progressively greater regulation, in the sense of a major possibility of decisions, against environmental factors. Certainly, this memory would be very poor in subantarctic ecosystems. Nests and excreta of marine birds would be pioneer stages of this succession, which therefore depends at the beginning on marine elements. The marked increase of diversity passing from faeces to soil (Fig.8) would be due to the intervention of a biological control because of overlaying plants; nevertheless, this cannot be considered a true relay, since abiotic control is still dominating. Referring to climax, w e think that it would be reached at the leveloflichens, which here form the superdominant group in number of species and vegetal mass (Follmann 1965). The highest faunistic diversity in mosses can be explained by post-climaxconditions (higher humidity compared with climax) favoured by edaphic causes (flat position, accumulation of water coming from the thaw, etc.). There exist two other ecological aspects which have been analysed by means of indices based on information theory: the vertical distribution of diversity in the soil profile and the altitudinal zonation (Covarrubias, 19656). Given little depth of these soils, only a layer at the level of lichens and mosses and another one a few centimetres under the ground were studied. Results are
not definitive, because the highest diversity is sometimes present at the surface, sometimes in the soil. This would lead one to think on the possibility of little vertical migrations of soil arthropods, but w e have not yet investigated the possible causes of this. T w o factors of contrasting effect must be considered: one, the advantage of organisms to locate themselves close to the surface where greater thermic oscillations are produced,mainly at well-exposedplaces, therefore reaching higher temperatures; two, a negative phototaxis of these arthropods. Nevertheless, this phototarris would be little accentuated, according to our recent investigations with the experimental application of darkening boxes. Also, results on altitudinal zonation of diversity are hard to explain; sometimes high values correspond to the top of mountains,sometimes to the foot,but never to the middle. The first case can be due to the effect of nunatak, since some peaks were excluded by the last glaciations and are even irregularly covered with snow in winter,thus conserving a largerfaunisticcontingent. On the other hand, maxima at the base of mountains are due probably to minor slopes providing greater stability to soil. Up to this point, we have analysed only problems of regulation at the organization level of communities. W e will n o w discuss some aspects of regulation at the population level. Unfortunately w e have little data about this; in addition, focusing deeper on this problem, it is necessary that the ecological approach be accompanied by a genetic one. In any case, as at the community level, regulation is very scarce for these populations. The fluctuations of density between different and also in the same environments are the greatest we have detected. They are probably related to the form of growth of these populations.In spite of not yet having fulfilled the mathematical study of population increment in the colonization of introduced substrata,it seems that it can be represented approximately by means of an exponential curve. This could have been expected in ecosystems dominated by abiotic controls, where population growth is almost independent of density. Terrestrial food chains are short and of little complexity, occurring almost exclusively as simple interspecific relations. A m o n g analysed groups, collembola, oribatei, acaridiae and diptera larvae seem to belong to the trophic level of primary consumers, dependent on plants and deposits of organic substances. Only gamasides and a part of prostigmata would be secondary consumers, apparently being the end of these chains. Also, in some mixed populations of soil arthropods, w e have evidenced the presence only of primary consumers (phytophagous or detritus eaters). Referring to the evoluiionary processes in these populations, it is necessary to point out that in those systems with little integration between species, with a prevalently intraspecific competition and a selection
215
F.‘di Castri;R. Covarrubias;E.Hajek
directed by abiotic factors, species show a tendency to opportunism and ecological versatility.In other words, they advance little from plasticity toward a progressive specialization, and so ecological niches remain relatively large. This is also made manifest in the fact that almost all of the studied species occupy very different habitats; a quantitative approach allows us to detect, nevertheless, a certain preference. W e have tried to obtain more data on the plasticity and colonizing power of these species by means of a series of investigations, still continuing, in which substrata not existing in natural conditions in these regions, such as faeces of domestic animals, mature soils of different kinds, litter and humus (mull) were introduced. Obviously, all were previously sterilized. In general, all these were immediately colonized, but an important difference was observed. The two most similar substrata to local subantarctic conditions (faeces and soils) maintained relatively homogeneous and constant densities. For the contrary, those most “strange” ones (litter and mull) were at the beginning attractive places, higher densities being reached in the mull than in the controls, but after seven weeks they were almost completely evacuated. This shows the plasticity and at the same time a little marked preference of these organisms. W e believe that other investigations being carried out on population dispersion, approached mathematically like a contraposition,almost a pulsation, of tendencies to aggregaiion and expansion,might supply a valuable informationfrom an ecologicaland genetical point of view. Little personal data can be added about functional problems in these ecosystems.Brief following considerations are based partially on preliminary observations, partially on theoretical postulates. It is evident that these ecosystems are not “economic”, since in each transference there is high dispersion of energy with sharp jumps of the energy flow (ecoforce). But this strong dispersion is balanced by the fact that the total number of transformations is very low, given the simplicity of the system and the short food chains. By this means, it might be presumed that the productivity/ biomass ratio reaches high values in these populations. Autosufficiency of these systems is poor, because m a n y of them are based on nitrogenic supplies coming from marine ecosystems. A last consideration is about the stability of these systems; out of a biological interest it is related with problems of exploitation and conservation of natural resources. Everything affirmed previously indicates that the processes of internal regulation (homeostasis) of these simple and immature systems are particularly deficient. Instability is clearly evidenced by severe fluctuations in the size of subantarctic populations, showing in the intensity of these changes an approximately inverse ratio with biotic diversity (information).
216
Nevertheless,w e have seen that here exists a great plasticity at the level of the systems and of the species integrating them. Species are opportunist, easily dispersed and with a rapid population growth rate. Considering the type of natural selection to which they are subjected, w e might presume that their adaptation to abiotic factors must be high, not so to the changes of a biocenotic nature. Consequently, in subantarctic zones it is necessary to avoid the introduction of species coming from neighbouring continents, which here might be free of any competition. BIOCENOLOGY
Data for discussing some quantitative relations between groups of mesofauna are given in Table l. All sampleshave been obtained during antarctic summers. The analysis of results could be approached according to three points of view: by habitat, biocenotic parameter and animal group. Nevertheless, the table being easy to interpret, only some general considerations will be presented. Generally,some of the terms appearing in the table are almost exclusively used at the level of species and ill be useful to define not of groups as is done here. It w the sense given to them in this paper. W e mean by abundance, the percentage of individuals belonging to an animal group in relation with the total of specimens collected. The presence is the percentage of samples where individuals of a given animal group are found. The dominance is the percentage of samples in which the analysed group has reached the highest density. First commentary is about the little number of groups in these collections. Out of those listed in the table, some specimens of tarsonemini, araneida, psocoptera and enchytraeidae have been obtained, obviously not considering soil aquatic animals (protozoa, rotifera, tardigrada and nematoda), which are always very abundant. These groups of mites and springtails represent for mesofauna the “essential frame” of soil communities, since they are present in almost every natural and man-influenced ecosystem of the world. W e have previously discussed the low amount of species and the high number of individuals per sample. W e would like to point out that the values of mean density are of only a relative interest, because it is practically impossibleto obtain stable means of density for populations subjected to so rough fluctuations. Also the number of plant species is very low. Out of this,no clear relation between floristic and faunistic diversity exists. F r o m a physiognomic point of view, the abundance of prostigmata and entomobryomorpha confers to these communities a “xeric” aspect,that is,something like that of communities living in arid zones of south America.
Soil ecosystems in subantarctic regions
TABLE1. Quantitative description of subantarctic mesofauna b y habitat
Mesofauna
Mean density (no. per 1,000
Density range
4
N o of species (mean per sample)
Range no. of species
Abundance
(%)
Presence (Yo)
Dominance
(%)
Mean diversity (bits/ individual/ cc) 1,000
Diversity range
Mosses Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae
45.03 245.06 36.58
0-666.67 0-1085 0-412
-
-
62.21 4 836.84 4.80
0-240 4-93647.50 0-105.70
1.36 3.36 0.68
0-5 0-6 0-2
-
-
1.00 1.91 0.05
0-3 1-3 0-1
0.41 3.88 0.42
63.60 95.50 59.10
22.70
-
-
-
-
0.95 94.18 0.17
72.70 100.00 4.50
4.50 72.70
100.01
-
99.90
0.27 58.65 28.27 0.27 0.61 11.93
52.90 94.10 70.60 5.90 47.10 76.50
-
- c
5 230.52
Total
156-95257.60
8.36
2-14
0.88 2.69 1.19 0.13 0.50 1.44
0-4 0-5 0-3 0-2 0-2 0-4
7.50
0.2-22.3
4.80
0-19.7
Lichens
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae
6.36 1454.47 732.18 4.92 18.16 313.02
0-44 0-7852 0-8512 0-78.81 0-216 0-2548
-
-
-,
Total
-
2 529.11
-
- L
I -
-
--
-
0-16368
6.83
0-17
100.00
4-508 0-16288 0-116764
1.75 1.75 1.50
1-3 0-3 0-3
1ou.00 66.70 66.70
-
-
0.40 19.81 71.92
1:25 1.25
1-2 1-2
1.59 0.28
100.00 100.00
I
-
50.00 25.00 -
25.00 -
100.00
Terrestrial algae
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae
145.55
6 246.03 29 896.00
-
-
583.18 98.70
44-1468 16-212
-
~.
Total
36 969.46
868-116382
-
7.50
-
-
-
6-11
100.00
-.
-
-
33.33 33.33
-
33.33
-
-
3.60
0.7-6.1
6.50
1.2-10.4
5.10
0-13.5
Gramineae
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae Total
-
-
2.66
-
-
-
0-2
0.32
33.30
-
-
0-2 2-2
0.49 99.19
33.30 100.00
-
-
0.67
-
-
4.00 1323.25
0-12 700-1577.78
0.67 2.00
-
1329.91
0-8
-
-
-
-
-
-.
-
-
700-1577.78
3.34
2-6
100.00
-
100.00
0-36 0-1445 0-2736 0-8 0-310 0-5545
0.85 1.00 0.31 0.08 0.92 1.46
0-3 0-3 0-2 o-1 0-3 0-3
0.68 10.12 23.25 0.07 4.09 61.79
41.70 41.70 16.70 8.30 66.70 83.30
9.10
Soils
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae
7.87 114.77 211.37 0.62 54.37 966.01
-
-
-
Total
l 355.01
-
I
-
-
-
-
9.10 81.80
-
_ .
0-7305
4.62
0-11
100.00
100.00
217 16
F. di Castri; R.Covarrubias; E. Hajek
TABLE1 (continued) M e a n density (na. per 1,000
Mesofeuna
4
Density range
No of species (mean per sample)
Range no. of species
Abundance
Presence
Dominance
(Yo)
(Yo)
0.02
50.00
25.00
-
-
-
(yo)
Mean diversity (bits/ individuali cc) 1.000
Diversity range
Faeces
2.00
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae
-
166.00
-
0-1
0.50
-
-
75.00
0-456
0-1
1.54
-
-
0.75
2.00 10 597.00 1.00
0-4 0-33024 0-4
0.25 1.00 0.25
0-1 1-1 0-1
0.02 98.41 0.01
50.00 75.00 25.00
75.00
8-33484
2.75
2-4
100.00
-
100.00
0.90
0.4-2
0-20 0-7616 0-92 0-8
0.67
0-2
0.26
-
-
33.33
-
1.00
0-3
98.45
33.33
0.33 0.33
0-1 o-1
1.19 0.10
100.00
0.27
0-0.80
10 768.00
Total
0-4
-.
-
-
-
-
_ _ -
Nests 6.67
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae
-
2 538.67
-
30.67 2.67
-
Total
2 578.68
-
0-7736
-
2.33
It will b e observed in the table that all these habitats, with or without plant cover, have been colonized by terrestrial mesofauna. In 94.11 per cent of the samples, animals of these groups have been found; all the rest (samples of penguin nests) are not azoic, since they always contained populations of nematoda. In addition, m a n y h a n d y collections of mesofauna living under stones, enable us to affirm that this environment is also inhabited in most of the situations. W e have already discussed the marked faunistic homogeneity and therefore the great biocenotic affinity. Nevertheless, a quantitative approach to the problem proves s o m e preferences. T h e clearest e x a m ple is the proportional distribution of groups for mosses and lichens. In the first habitat entomobryomorpha are dominant, showing the m a x i m a values of abundance and presence. For lichens, a m o r e xeric environment, all these values are favourable for prostigmata. On the other hand, prostigmata are often absent from scarcely-integrated communities with l o w diversity, such as fauna from faeces and nests, which s e e m to be pioneer stages, whilst they are dominating in the climax associations of lichens. A greater successional maturity could b e assigned to this group, which includes species belonging to m o r e advanced trophic levels. In general, entomobryomorpha and prostigmata are the t w o most important groups in soil a n d related environments, as it can b e seen from Table 2 on the total of a series of samples. T h e y are followed by
-
33.33 33.33
~
0-7
-
.
-
,-,
-
100.00
__.
-
oribatei, which in s o m e favourable situations (terrestrial algae) reach the highest absolute densities (116,7&l individuals per 1,000 cc of material); poduromorpha, apparently the most hygric group a n d with a certain edaphic preference; a n d gamasides, predatory mites whose densities are always low, but less fluctuating. A m o n g mites, acaridiae and tarsonemini are also present, but their frequencies are by far lower. COMPARISON WITH OTHER ECOSYSTEMS
I
218
W e should like to note here briefly the position and structural similarities of subantarctic ecosystems in relation to the analogous ones already studied in other regions with the s a m e methods a n d approaches. T h e first comparison must b e m a d e with communities of the extreme south of Chile, since this area is s o m e times regarded as subantarctic. T h e studied zones are located mainly o n the M u ñ o z G a m e r o Peninsula and also near P u n t a Arenas and o n Navarino Island; the dominant plant formation is the evergreen cold rain forest. According to our investigations, great similarity is demonstrated between these zoocenoses and those of the Valdivian rain forests. On the contrary, little structural and faunistical ties exist with antarctic ecosystems analysed in this paper. Figure 9 shows a sequence of increasing diversity for different communities of arthropods in the magellanic region. E v e n the lowest values are higher than those recorded
Soil ecosystems in subantarctic regione
TABLE2. Quantitative description of subantarctic mesofauna-total of a series of samples Mean density (no. per 1,000
Mesofauna
4
Density range
No of species (mean per sample)
0-666.67 0-16288 0-116764 0-78.81 0-1468 0-93647.50 0-105.70
1.04 2.10 0.82 0.04 0.78 1.56 0.03
Range of species
PO.
Abundance
Presence
(%)
Wo)
Mean Dominance diversity (bits/ (%) individuail CO) 1,000
Diversity range
~
Gamasides Prostigmata Oribatei Acaridiae Poduromorpha Entomobryomorpha Diptera larvae Total
28.31 812.76 2 122.29 1.28 99.87 2 571.16 1.61 5 637.28
__
0-5 0-6 0-3 0-2 0-3 0-4 0.1
57.35 69.12 55.88 2.94 61.76 86.76 2.94
for subantarctic communities in Figure 8. Also, the distribution of individuals within the species is c o m pletely different. Finally, climax conditions for these terrestrial communities are reached in magellanic forests corresponding with litter, substratum obviously absent south of the timber line. For these reasons, magellanic environments have been excluded from the general analysis m a d e in the first part of this report. Before passing over to a comparative discussion with ecosystems of farther territories,it is convenient to examine Figure 10. Here curves of successional advancement are represented in three very different regions: palar, temperate and humid tropical. Naturally, it deals with schematic curves, which nevertheless are based o n effective results of investigations carried out with the purpose of observing analogies and dif-
2.94 20.59 10.29 -
4.41 55.88 -
-,
- 8
6.37
0-116832
0.37 14.75 32.61 0.03 1.56 50.61 0.07
0-17
-
100.00
94.11
-
ferences in parallel series of ecological successions. T h e elements appearing in this figure (diversity, filter, time and noise) have been already theoretically explained in relation to Figure 6.Noise is expressed by m e a n s of the n u m b e r of climatic unfavourable months, according to approaches discussed in previous papers (di Castri, 1964; di Castri and Hajek, 1964). T h e three curves have the s a m e tendency, but they are displaced in different ranges of diversity, in accordance with the major or minar control by abiotic conditions. Polar communities m a y reach a lower degree of diversity and maturity as compared with m o r e favourable environments. Thus, m a n y of the structural and functional characteristics of polar communities depend o n their l o w degree of maturity, or in other words, o n the youth type of communities.
Fi I ter
>
I
5
0Iropical -.I c I n + . -
5 2 c 2.
.-
-
In
0 0 L'ICHENS
SOILS
MOSSES
pioncer siage climax
LI'TTER
FIG.9. Progression of mesofauna diversity in magellanic cold forests.
FIG.10. Theoretical curves of successional advancement in polar, temperate and tropical regions.
219
F. di Castri;R. Covarrubias;E.Hajek
This fact has been also recognized by Fischer (1960), focusing the problem of latitudinal variations of organic diversity on the basis of the number of species belonging to different animal groups. For this reason w e could presume that analogies with subantarctic communities should be found also in other ecosystems,on condition that neither would they have reached a high degree of maturity. Youth character can correspond to the three possible conditions which are outlined now. During a series of researches w e have tested in the field these theoretical possibilities; similarities are not only referred to diversity, but also to instability of populations and biocenosis, to the dominating influence of abiotic factors and in general to the major part of the aspects already discussed under the heading of ecological regulation. First condition. That the simplicity and structural immaturity be maintained by too strong environmental interferences. It is the case better fittedto subantarctic communities. W e have verified similar situations for Andean tundras and moorlands and for fields under periodical flooding. With some differences, the structure of desert zoocenosis corresponds to this type. Second condition. That the immaturity be a transitory step, which can be surpassed with time. This occurs in pioneer stages of all successions.Particularly, early stages of hydroseres in w a r m deserts have shown close analogies with the structure of subantarctic systems. Third condition. That the immaturity be a regressive phenomenon, produced by intervention of m a n cultivating lands, pasturing livestock and felling timbers. The constant decrease of information in the m a n influenced systems (state of disclimax) was touched in several papers, whose conclusions have been recently synthetized (di Castri, 1965). In this sense, subantarctic ecosystems are alike in m a n y aspects to cultivated fields subjected to irrigation and periodic agricultural rotations. It is interesting to emphasize that m a n y conclusions coming from research on antarctic ecosystems (naturally simple) can often be projected to cultivated lands (artificially simplified). Synthetizing, these antarctic terrestrial environments seem to constitute the less complex natural example of an ecosystem at the climax stage. Given also its geographical isolation and the very little influence of m a n in the past and present, they represent zones of an exceptional scientific meaning for verifying in the field basic principles of ecology.
SUMMARY S o m e results of quantitative research are presented on structural aspects of terrestrial zoocenosis in antarctic regions, mainly the South Shetland Islands and the septentrionalpart of the Antarctic Peninsula.
220
This region is compared from an ecological point of view with analogous subarctic territories. Studied animals belong in general to the aerobiontic mesofauna of soil and related environments, such as mosses, lichens,nests and faeces of marine birds,algae, etc. This fauna was extracted from the samples by means of Berlese-Tullgrenfunnels. In total 150 samples were analysed, more than 250,000 arthropods being counted and classified. Biotic diversity was the main subject considered, with approaches based on information theory. For measuring diversity, as information in bits per individual, Shannon-Wiener’sand Brillouin’s formulae were used. Results are discussed under three headings : ecological regulation, biocenology, and comparison with other ecosystems. The following are the main conclusions. 1. Diversity is very low and density very high, when antarctic zoocenoses are compared with those of South American ecosystems already studied with the same focusing and similar methodology. This situation,interpreted as a condition of redundancy, would represent the adaptation of the system against the constant environmental interferences of climatic or edaphic nature. 2.Density of each animal group shows strong fluctuations, being the intensity of them varying approximately inversely to the diversity of the zoocenoses. These fluctuations demonstrate the existence of unstable and scarcely integrated systems, subjected to abiotic controls. 3. The diversity is higher for zoocenoses associated to an autotrophic stratum (lichens and mosses) in comparison with those independentof plants. 4.In antarctic environments, ecological successions with a progressive increase of diversity have been verified,but in these processes only very low levels of maturity and complexity are reached. 5. Most of the species have large ecologicalversatility, invading and colonizing almost every natural and introduced environment. Nevertheless, at a quantitative scale a preference can be detected. 6. Great faunistic a5nity between the different samples and places exists. This is due to the plasticity of species and the easy transport, mainly by wind.
7.Dominant groups of the antarctic mesofauna are entomobryomorpha (collembola) and Prostigmata (acarina). 8. The physiognomic aspect of these zoocenoses is of a “xeric” type. 9. Structural similarities with ecosystems of the extreme south of South America are poor. 10. On the contrary, great structural analogies exist with some pioneer stages and with cultivated fields (disclimax), already studied in the South American continent.
Soil ecosystems in subantarctic regions
Résumé $eos,ystèrnes du sol dans les régions subantarctiques (Francesco di Castri,et al.)
L’auteur présente certains résultats de recherches quantitatives sur les aspects structuraux des zoocénoses terrestres dans les régions antarctiques,notamment les Shetland du Sud et la partie septentrionale de la péninsule antarctique. I1 compare du point de vue écologique cette région à des territoires subarctiques analogues. Les animaux étudiés appartiennent pour la plupart à la mésofaune aérobiotique du sol et des milieux qui s’y rattachent,c o m m e les mousses,les lichens,les nids et les déjections d’oiseaux marins, les algues, etc. Cette faune a été extraite des échantillons au moyen d’entonnoirs de Berlese-Tullgren.Au total, 150 échantillons ont été analysés,et plus de 250 O00 arthropodes ont été dénombrés et classés. Les recherches ont porté principalement sur la diversité biotique, étudiée au moyen de méthodes fondées sur la théorie de l’information. Pour mesurer cette diversité, en binons d’information par individu, . on a utilisé les formules de Shannon-Wiener et de Brillouin. Les résultats sont étudiés sous trois rubriques :régulation écologique, biocénologie, et comparaison avec d’autres écosystèmes. Les principales conclusions sont les suivantes: 1. L a diversité est très faible et la densité très élevée quand on compare les zoocénoses de l’Antarctique à celle des écosystèmes sud-américainsdéjà étudiés du m ê m e point de vue et avec des méthodes analogues. Cette situation,interprétée c o m m e un état de surabondance, représenterait l’adaptation du système malgré les interférences continuelles, climatiques ou édaphiques, du milieu.
2. L a densité de chaque groupe d’animaux varie fxtement ; elle est à peu près en raison inverse de la diversité des zoocénoses.Ces variations attestent l’existence de systèmes instables et à peine intégrés, subissant des contraintes abiotiques.
3. L a diversité est plus grande pour les zoocénoses associées à une strate autotrophe (lichens et mousses) que pour celles qui sont indépendantes des plantes.
4.On a constaté l’existence,dans les milieux antarctiques, de successions écologiques progressivement plus diversifiées ; mais, au cours de ces processus, les niveaux de maturité et de complexité atteints restent très faibles. 5. L a plupart des espèces font preuve d’une grande souplesse écologique, car elles envahissent et colonisent presque tous les milieux naturels, ou artificiellement introduits. Néanmoins, à l’échelle quantitative, on peut déceler des préférences. 6. I1 existe une grande affinité faunistique entre les différents échantillons et lieux. Ce fait est dû à la plasticité des espèces et à leur transport facile, surtout par le vent. 7. Les groupes qui prédominent dans Ia mésofaune antarctique sont les entomobryomorphes (ccllemboles) et des acariens. 8.L’aspect extérieur de ces zoocénoses est du type “xérique ”. 9.Les ressemblancesstructurales avec les écosystèmes de l’extrémité sud de l’Amérique du Sud sont faibles. 10.Au contraire, il y a d’importantes analogies structurales avec certaines phases pionnières et avec les espaces cultivés (disclimax) déja étudiés dans le continent sud-américain.
Discussion F. E. ECKARDT. J’ai suivi avec le plus vif intérêt votre excellent exposé. Vous avez su, d’une manière très claire, mettre en évidence l’importance que présente pour l’écologiste la connaissance non seulement du fonctionnement mais aussi de la structure des divers systèmes écologiques. Au cours de votre exposé, vous avez employé le terme “ soil ecosystem ” pour désigner des ensembles fonctionnels comprenant les organismes vivants du sol et le milieu édaphique avec lesquels ces organismes échangent de l’énergie et de masse. Lors du Premier Colloque international sur les écosystèmes,qui s’est tenu à Copenhague au cours de l’été 1965,sous les auspices de l’Unesco,le terme “écosystème” était utilisé par presque tous les participants pour désigner
une subdivision de la biosphère caractérisée par un certain degré d’invalidité structurale et fonctionnelle et possédant non seulement une strate hétérotrophe,mais aussi une strate autotrophe permettant la conversion de l’énergierayonnante provenant du soleil en énergie chimique potentielle.Dans le langage de la cybernétique1 il s’agissait, par conséquent, de système ouvert du point de vue de l’énergie (l’énergie d‘origine solaire,après dégradation,est réémise vers l’espace) 1. Cf. Remarques préliminaires concernant la structure et le fonctionnement des écosystèmes et l’organisation d u Colloque de Copenhague, Actes du Colloque, de Copenhague. Paria, Unesco. (Recherches sur lei resBowces naturelles,
XV.)
221
F. di Castri; R. Covarrubias; E.Hajek
et de l’information,mais fermé en ce qui concerne le contrôle (par exemple contrôle des populations). Pensez-vous qu’il serait possible de concilier ces deux m a nières d’employer le terme “ écosystème”?
F. DI CASTRI.J e suis fondamentalement d‘accord sur la définition d’écosystème que M . Eckardt a donnée, surtout dans le sens que la présence d’une strate autotrophe est nécessaire pour pouvoir se référer à un véritable écosystème. Dans notre travail il y a plusieurs indications à ce sujet, que je n’ai pas pu discuter dans m o n exposé résumé, faute de temps. Sans doute, l’expression “écosystème du sol”, employée conventionnellement, n’est pas très précise. Si, parfois, on la mentionne dans le sens d’écosystème terrestre, en considérant aussi les mousses, les lichens, les graminées, etc., d’autres fois aucun organisme producteur n’est compris dans la biocénose. J’ai touché ce problème dans plusieurs publications antérieures, en définissant c o m m e “ subsystème édaphique ou
Bibliography ARAYA, R.;HERVÉ, F. 1964.Estructuras en playas actuales y antiguas, islas Greenwich y Robert, South Shetland. Com. Esc. Geología Uniu. Chile, no. 6,p. 1-5. CASTRIF. DI. 1960.Prime osservazioni sulla fauna del suolo di una regicme delle Prealpi Venete (Monte Spitz, Recoaro). Atti Ist. Ven. Sc. Lett.Arti,vol. 118,p. 475-493. . 1964. Interpretación bioclimática de las biocoras de Chile de acuerdo a su periodo de actividad biológica. Bol. Prod. anim. (Chile),vol. 2,no. 2,p. 173-186. . 1965.Cmsideraciones sobre el estado de disclimax en las zoocanosis edáficas. In: Progresos en biología del suelo, p. 333-34,l. (Monografías no. 1. Unesco, Montevideo.) __ , HAJEH, E.R.,1964.Introducción a la bioclimatología de Chile.Bol. Prod. anim. (Chile),Serie A, no. 1.(Monografías sobre ecologia y biogeografía de Chile.) ; SAIZ, F. 1964. Aplicación de la teoría de la información al estudio de las biocenosis muscícolas. Bol. Prod. anim. (Chile),vol. 2,no. 2,p. 153-171. ; ASTUDILLO, V. 1965a.Revisión crítica de las aplicaciones de la teoría de la información en zoología del suelo In: Progresos en biología del suelo, p. 313-331.(Monografías no. 1. Unesco, Montevideo.) __ -. . 1965b.Análisis de algunas causas abióticas de variación en la densidad de la fauna del suelo. In: Progresus en biologia del Suelo, p. 371-377.(Monografías no. 1. Unesco, Montevideo.) COVARRUBIAS,R. 1965a. Densidad y diversidad biótica de los Invertebrados terrestres en la Antártica. III Congreso Latinoamericano de Zoología, Santiago. . 1965b. Estructura de las zoocenosisterrestres antárticas. In: Progresos en biología del suelo,p. 343-357.(Monografías no, 1, Unesco, Montevideo.)
-_
-_
__
__
__
-_
222
endogé” cet ensemble fonctionnel et structural. Seulement en union avec le “snbsystème épigé” on arrive à un écosvsteme dans l’acception classique du mot, avec des conditions d’autosuffisance et d’autocontrôle (autocommande). Entre ces deux subsystèmes ouverts il y a u n flux constant d’énergie et d‘information, mais pas dans les mêmes proportions. E n effet, le subsystème endogé reçoit surtout de l’énergie, tandis que le flux d’information est dirigé en plus grande partie vers l’épigé. Le problème se complique dans les milieux subantarctiques, étant donné que plusieurs biocénoses édaphiques ne tirent pas l’énergie d’une strate autotrophe superposée,mais d‘apports nitrogénés d‘origine marine (excréments et nids d’oiseaux marins). D’autre part, je suis convaincu que le concept “fermé” est plutôt relatif vis-à-visles systèmes écologiques naturels, m ê m e en ce qui concerne l’information. E n général, il y a toujours des échanges d’information entre des systèmes (ou subsystèmes) avec différents degrés d e maturité et de complexité.
/ Bibliographie __ . 1966. Observaciones cuantitativas sobre
los inverte brados terrestres antárticos y pre-antárticos. Publ. Inst. Antártico Chileno,no. 9,p. 1-60. CRARY, A. P. 1962. The Antarctic. Scienti$c American, vol. 207, no. 3,p. 2-15. DALENIUS, P. 1965.The acarology of the Antarctic regions. In: P. van Oye and J. van Mieghem (eds.), Biogeography and ecology in Antarctica, p. 414-430. (Monographiae biologicae, 15.) FISCHER,A. G.1960.Latitudinal variations in organic diversity. Evolution,vol. 14,p. 64-81. FOLLMANN, G. 1965. U n a asociación ni’-ófila de líquenes epipétricos de la Antártica occidental con Ramalina terebrata Tayl et Hook. como especie caracterizante. Publ. Inst. Antártico Chileno, no. 4,p. 1-18. GRESSITT,J. L.;WEBER, N. A. 1959.Bibliographic introduction to Antarctic and subantarctic entomology. Paci$c Insects, vol. 1, no. 4, p. 441-480. ; LEECH, R. E. 1961. Insect habitats in Antarctica. Polar Record, vol. 10,no. 68,p. 501-504. -_. --; LEECH, T. S.; SEDLACEK, J.; WISE, K. A. J. 1961. Trapping of air-borne insects in the Antarctic area (part. 2). Pacijic Insects,vol. 3,no. 4,p. 559-562. LLANO, G. A. 1962. The terrestrial life of the Antarctic. Scientijic American, vol. 207,no. 3,p. 212-230. MARGALEF, R. 1957.L a teoría de la información en ecologia. M e m . real Acad. Ciencias Artes,Barcelona, vol. 32,no. 13, p. 373-449. . 1963. On certain unifying principles in ecology. The American Naturalist,vol. 97,no. 897,p. 357-374. PRYOR, M.E.1962.S o m e environmental features of Hallett Station, Antarctica, with special reference to soil arthropods. PaciJic Insects,vol. 4,no. 3,p. 681-728.
__
__
Soil temperatures in river bottom stands in interior Alaska L. A. Viereck
INTRODUCTION In a region where frozen ground underlies a large percentage of the land surface, knowledge of the interaction of permafrost and vegetation is necessary to an understanding of forest ecology. Most of the flood-plainsof interior Alaskan rivers are underlain by permafrost of varying thickness; frozen ground is lacking, however, beneath present and recent river channels and in freshly-depositedalluvium. As a river meander advances, permafrost forms concurrently with the development of vegetation on the n e w deposits on the slip-off side of the river (Benninghoff, 1952; Drury, 1956; Péwé, 1957). The formation and rise of the permafrost table is at least partly caused by vegetation. Where vegetation has been Iemoved from flood-plain areas with permafrost near the surface, the depth to perennially frozen ground has increased to as m u c h as 14 m (Péwé, 1957). This study attempts to show differences in soil temperatures resulting from plant succession on the river flood-plains in interior Alaska. Four stands on different ages of river alluvium were studied.These are a feltleafwillow (Salix alaxensis Cov.) stand,a 50-yearold balsam poplar (Populus balsamiferu L.)stand, a 120-year-oldwhite spruce (Picea glauca (Moench) Voss) stand, and a 220-year-old white spruce/black spruce (Picea muriana (Mill).B.S.P.)stand. These four ecosystems represent a plant succession sequence on the floodplain of the Chena River.
THE REGION CLIMATE
The study area is located adjacent to the Chena River about 50 k m north-eastof Fairbanks at an elevation
of 200-220m (Fig.1). The climate is continental with cold dry winters and w a r m summers. Yearly precipitation averages 287 mm with a m a x i m u m in late summer (August with 56 mm) and a minimum in late winter (April with 6 mm).Annual snow-fallaverages 151 c m with an average m a x i m u m accumulation of 74.4cm. Mean annual temperature is -3.40 C with a mean monthly temperature of -23.90 C in January and 15.50 C in July. Table 1 presents average monthly temperatures, precipitation, and snow-fall for Fairbanks and the deviation from normal for the period of this study (1 October 1964 to 30 September 1965). The period of study was characterizedby an unusually cold winter (Fig.2). The averageDecembertemperaturewas 10.5OC below the normal of -22.00 C. January and February were also subnormal.March, in contrast,was 8.60C above the normal monthly average. The remaining months of the year were within 3.00 C of normal and the year averaged approximatelylo C below the normal of -3.40 C. The weather pattern was characterized by rapid fluctuations over a wide range of temperatures, the m a x i m u m for the period being, 29.50 C on 8 June and the lowest -500 C on 27 December, a range of
800 C. VEGETATION
Regional vegetation consists of stands of white spruce on well-drained soils and south-facing slopes, and black spruce, larch (Larix laricina (DuRoi) K.Koch), muskeg, and bogs on poorly drained or permanently frozen soils, and on north-facing slopes. Fire has been an important environmental factor; successional stages of aspen (Populus tremuloides Michx.) and paper birch (Betula papyrifera Marsh.) are predominant throughout the area. Plant succession along braided and meandering
223
L. A. Viereck
FIG.1. M a p of Alaska with location of study area on the Chena River.
--
500 Kilometres
TABLE1. Average temperature, precipitation, and snow-fall for Fairbanks and deviations from normal for the period 1 October 1964 to 30 September 1965
Month
Average temperature
Deviation
(“C)
Average precipita- Deviation tion (mm)
Average 6nOW-
Deviation
fall (cm)
Oct. Nov. Dec. Jan. Feb. March April May June July Aug. Sept.
-3.2 -15.6 -22.0 -23.9 -19.4 -12.8 -1.4 $8.4 +14.7 +15.4 +12.4 +6.4
-0.8 -10.5 -3.8 -8.4 +8.6 $0.8 -2.4 -1.8 +0.2 -2.8 +2.6
21.6 15.2 13.7 22.6 13.2 10.2 6.4 18.0 35.3 46.7 55.9 27.9
-8.1 +6.6 -5.1 -20.8 -5.1 -3.3 +5.6 -14.5 -5.8 -11.4 -18.3 +25.6
21.8 20.8 22.4 29.2 26.2 18.0 7.9 2.3
-10.2 +25.1 $13.2 -25.1 -15.2 -16.0 +4.1
T T T
-
1.5
-1.5
Year
-3.4
-1.3
286.8
-54.6
151.1
-32.2
224
+l.l
-1.5
-
streams in interior Alaska is diagrammatically shown in Figure 3. Colonization of freshly formed alluvium is by light seeded herbs and shrubs, primarily S a h species. These shrubs grow rapidly and are flooded almost every spring,but they are able to send out adventitious roots and continue to grow. Within 10 to 15 years balsam poplars, established at the same time as the willows, overtop the shrubs and begin to shade them. By the time balsam poplars are 50 years old the shrub layer of willows has been replaced by Rosa aciculuris Lindl. and Viburnum edule, (Michx.) Raf., and there is a thin herbaceous cover on the forest floor, primarily of Equisetum species. At this stage in succession seedlings of white spruce become established. The spruce grow slowly at first but by the time they are 50 to 100 years of age they have grown taller than the balsam poplar, which begin to die and disappear from the stand. Stands of white spruce m a y persist for 200 to 300 years along the lower terraces of the river, especially where there is occasional flooding. As the white spruce stands mature a thick m o s s layer
r
MAR
I
APR
I
MAY
I
JUNE
I
JULY
I
AUG
I
SEP
I
OCT
I
NOV
I
DEC
I
JAN
1
FEB
'I
'F
30
80
20 60
IO 40
O 20
-10
-2o
O
WILLOW
I
20
-40 -30
.40
-50
'OG
' II I
BUCK SPRUCE
I
I I
WHITE SPRUCE
i I I
BALSAM POPLAR
/BARE; ! SILT I
RIVER
I
BLACK I SPRUCE i
l
!
FIG.3. Diagrammatic cross-sectionof typical distribution of vegetation and permafrost across a meander of a river in interior Alaska. 225 17
L.A. Viereck
of Hylocomium splendens (Hedw.) B. and S. a n d Rhytidiadelphus triquetrus (Hedw.) Warnst. forms on the forest floor. These mosses develop a n organic layer that m a y b e c o m e 20-30 c m thick. Soil temperatures are lower beneath this thick insulating layer and permafrost can sometimes be found in the older white spruce stands. On the older terraces, permafrost is prevalent and close to the surface. Slow-growing black spruce stands become the dominant vegetation. Farther from the river black spruce stands are often replaced by bogs w h e n s o m e disturbance of the organic layer causes shallow thawing of permafrost. Thawing creates a wet depression suitable for the establishment of bog species. This process has been described by both Drury (1956) a n d Benninghoff (1952). O n the higher terraces, often Pleistocene in age, permafrost with large ice masses is prevalent. T h e most widespread vegetation types are black spruce and open bog, except where fire has resulted in successional stands of birch, aspen, and white spruce.
THE STANDS Four undisturbed stands were selected to represent typical stages of succession o n the flood-plain of the Chena River. These stands represent surfaces of 15 to 20 years, 50 years, 150 years, and 250 years. WILLOW STAND
This stand consists of areas of open gravel with scattered young balsam poplar surrounded by dense shrub clumps of Salix alaxensis. A f e w forbs and grasses are scattered throughout the stand. T h e moss, Rhacomitrium canescens (Hedw.) Brid. forms a loose l o w m a t between the willow clumps. Scattered throughout are young saplings of white spruce, most of which are less than 1 m in height. Density of the tree species are: balsam poplar, 3,850 per hectare; white spruce, 5,360 per hectare. T h e soil is coarse river alluvium with a 1-2c m layer of silt o n the surface. Litter exists only under dense shrubs. Parent material varies little with depth, the a m o u n t of sand in samples greater than 2 m m being from 92 to 98 per cent while clay content varies from 0.2 to 4.2 per cent. BALSAM POPLAR S T A N D
T h e second stand in the successional series is a 47year-old balsam poplar stand. Trees have a density of 980 per hectare; dominant trees are 24-27 m in height and have a n average diameter of 21.3 c m . Beneath the balsam poplar canopy is a 1-1.5 m high shrub layer of Rosa acicularis and Viburnum edule. Scattered throughout the layer with a density of
226
180 trees per hectare are seedlings and saplings of white spruce. There is a rather dense herbaceous layer primarily of Equisetum aruense L. and E.pratense Ehrh. There is no well-developed soil profile within the stand. Parent material consists of coarse river gravel overlain by 1 m of alluvial silt and sand. T h e upper 8 c m is composed of alternating layers of partially decayed leaf litter a n d river silt deposited during recent periods of light flooding. A 1-2c m leaf litter lies loosely o n the surface of the silt. Tree roots are found throughout the upper 65 c m but are concentrated from 10 to 20 c m below the surface. Analysis of the silt l o a m in the upper 50 c m shows it to be 35 per cent sand, 65 per cent silt, with less than 1 per cent clay. W H I T E SPRUCE S T A N D
T h e third stage sampled in forest succession is that of a 120-year-old white spruce stand. M a n y old decaying balsam poplar persist a m o n g the spruce. Density of spruce is 810 trees per hectare with a n average diameter of 24.4 c m and a height of 27-30m for dominants. Beneath the forest canopy are scattered clumps of Alnus crispa (Ait.) Pursh a n d a loose shrub layer of Rosa acicularis and Viburnum edule. A thick, nearly continuous, m a t of Hylocomium splendens and Rhytidiadelphus triquetrus covers the forest floor. A n u m b e r of herbs, the most important being c o m m o n horsetail, are interspersed in the moss mat. Parent material consists of coarse river gravel overlain by 45 c m of finer river gravel a n d 40 c m of silt loam. A 4 c m organic layer is overlain by 5 c m of undecayed mosses. T h e upper layer of silt averages 40 per cent sand, 55 per cent silt, and 5 per cent clay. T h e lower layers of gravel average 90 per cent sand, 8 per cent silt, and 2 per cent clay. Roots are scattered d o w n to 85 c m but the m a x i m u m root zone is from 5 to 20 cm. W H I T E SPRUCEIBLACK S P R U C E S T A N D
In the stand of mixed black and white spruce, the older, larger white spruce are 220 years old and average 27.7 c m in diameter a n d 24 m in height. Between the older trees are 140-year-old black and white spruce and a n occasional paper birch. Total tree density within the stands is 1,450 trees per hectare. Beneath the evergreen canopy is a scattered layer of Alnus crispa, Rosa acicularis, Viburnum edule, and Ribes triste Pall. A dense m a t of feather mosses covers the forest floor. Conspicuous during the s u m m e r is a nearly continuous layer of three species of Equisetum. Parent material in this stand consists of silt and sand to a depth of a least 150 c m overlaying coarse river gravel. A b o v e the silt layer is a 10 c m organic stratum overlain by 17 c m of partially decayed and green mosses. T h e base of this moss layer is considered the zero level for placing soil thermistors. T h e roots are primarily in the 5-10 c m layer of the organic zone.
Soil temperatures in river bottom stands in interior Alaska
TABLE2. Depth of silty layer,depth of moss and leaf litter, and texture of the upper 20 c m of soil in four successional stands on the Chena River
Stand
Depth of silty laver
Depth of moss nnd litter
White spruce
1-2 65 40
O 2 5
White spruce/ black spruce
150
17
Willow Poplar
Camposition o
"'lt
94 3 35 65 40 55 10
80
Clay
3 1 5 10
The silty lower layers are composed of 10-15 per cent sand,75-80per cent silt and 10 per cent clay. Table 2 summarizes some characteristics of the soil profile for the four stands.
METHODS Colman fibre-glass temperature-moisture units were placed at depths of 5, 10, 20, 50, 100, and 150 c m and were read weekly with an alternating current o h m meter. Air temperatures were taken with standard m a x i m u m and minimum thermometers and a thermograph in a bird-house type of shelter (Frazer, 1961). S n o w depths were measured in three snow stakes placed so as to determine average deposition within each stand. Colman and Hendrix (1949) pointed out the usefulness of the fibre-glass units for determining the point at which soil moisture is frozen. A sudden and significant rise in resistance of the Colman moisture units, usually accompanied by a rapid decrease in temperature below O0 C,indicates that the moisture in the soil has frozen. Three indications of thawing were used. Return of soil moisture readings approximately to pre-frozen levels indicated that some thawing had taken place even though the soils often remained hard and at O0 C for a month after this level was reached. Depth measurement with a steel thermistor probe was considered the most reliable method to determine the upper surface of the frozen layer. In the coarse gravel layers where the probe could not penetrate,a sudden increase in temperature above the 00 C mark was used as an indication of complete thaw.
RESULTS A N D DISCUSSION Significant differences in annual soil temperature patterns occur among the four stands. Rates of freezing and thawing are given for each of the four stands in Figure 4. Freezing occurred in the surface Iayers of the soil approximately when average weekly tempera-
tures remained below O0 C. Freezing at depth was most rapid in the willow stand,slower under balsam poplar and white spruce, and slowest in the white spruce/ black spruce stand. Freezing in the willow stand was completed to the 150 c m depth in two months whereas the poplar and white spruce stands required three months. In the white spruce/black spruce stand soil moisture units did not indicate hard freezing at the 50 c m level until 16 February, four months after freezing began at the surface. In Figure 4 two lines are used to indicate the time of soil thawing. The first indicates when soil moisture was approximatelythat of the previous fall.T h e second line indicates the time at which the soil could be penetrated by the soil probe. In the willow stand, the second line indicates the time at which temperatures rose significantly above 00 C. Thawing was progressively later in each of the successional stages. Significant warming occurred in soils in the willow stand when air temperature maximums reached O0 C,but while there was still 25 c m of snow on the ground. Soil temperatures remained at O0 C until the average daily temperature went above O0 C. Thawing was completed throughout the soil profile by 18 May. Thawing was not completed in the balsam poplar and white spruce stand until 30 June and 10 July respectively. In the white spruce/black spruce stand m a x i m u m thawing to 50 c m occurred in September. The relationship of frozen to thawed ground was difficult to interpret in the white spruce/black spruce stand. A frozen layer from 40to 80 c m was encountered in August when units were first buried in the ground. Beneath this the ground was thawed and at a temperature of lo C. During the entire year temperatures at 100 and 150 c m remained between 1.750 C and 0.50 C and moisture units never indicated complete freezing.The upper surface of the frozenlayerretreated to 55 c m in September but refreezing took place again in October so that at the last date of probing, 9 November, the surface of the frozen layer was again at 45 cm. Evidently a layer from 55 to 80 c m remained frozen during the entire year and seasonal frost did not penetrate to the 100 c m level. Soil temperatures at all depths throughout the year can best be shown by annual isotherms (Fig.5). In these graphs the O0 C to 20 C represents a period that can either be thawed or frozen. Temperature fluctuations are greatest and most rapid in the willow stand. Temperatures at 5 c m range from 260 C to -80 C. Major changes in above-ground temperatures are quickly transmitted d o w n through the gravels. During the spring the isotherms are nearly vertical and close together, indicating rapid thawing and warming. At 150 c m thereis a 160 C annual temperature range. In the balsam poplar stand temperatures at 5 c m cooled to -100 C in the winter but warmed to only
227
L.A. Viereck
cm. O
' U N J U L A U G SEP
10 20
50
WILLOW 100
150
O IO
t
i
;
l
I
:
l l
1 l
I t
20
50
I
BALSAM
POPLAR
100
I50
O 10 20
50
100
150
O IO 20 50
I
WHITE S P R U C E / BLACK S P R U C E
frozen 150.
HttL FIG.5. Annual soil isotherms in the four stands.
228
Soil temperatures in river bottom stands in interior Maska
100 C in the summer, due probably to the shading of the ground surface by the balsam poplar and to the insulation of the leaf litter. The spring isotherms are more widely separated and descend at a more oblique angle than in the willow stand. Fluctuations near the surface were not as pronounced as in the willow stand. In the white spruce stand the temperature range at the 5 c m level was similar to that of the balsam poplar stands, -10 0 C to 100 C, but fluctuations were less pronounced and the individual isotherms did not dip as steeply or as deeply. Distance between isotherms is great, especially from O0 C to 20 C. The spruce stand was considerably slower to w a r m in the spring than the balsam poplar and willow stands. In the white spruce/black spruce stand soil temperatures remained relatively constant for the entire year. The seasonal range at the 5 c m level was from 100 C in early August to -50 C in mid-December. Isotherm lines run conspicuously in a horizontal rather than a vertical direction. At 150 c m temperatures fluctuated less than 10 C during the entire year. The effects of the time lag in warming of the soil and the lower degree of warming of the soils in the spruce stands are most important during the growing period. In the two spruce stands m a x i m u m root devel-
MAR
1
APR
I
MAY
I
JUN
I
JUL
I
AUG
opment is at approximately the 10 c m soil depth. Figure 6 shows soil temperatures at this depth. During the growing period, approximately 20 M a y to 20 July, soil temperatures at 10 c m averaged 17.80 C in the willow stand, 8.80 C in the balsam poplar stand,1.60C in the white spruce stand, and 0.60 C in the white sprucelblack spruce stand. Differences in the soil temperatures at all depths during the summer can easily be seen in Figure 7 which shows vertical representation of the average soil temperatures between 4 M a y and 31 August. Average soil temperatures during this period are highest in the willow stand, and cooler in each of the later successional stages. In the white spruce/black spruce stand the average temperature for the period approached O0 C from 10 cm to the 150 c m depth. Only the upper 10 c m shows any conspicuouswarming during this period. Air temperatures did not differ significantly among the four stands. Average weekly temperatures for the year in the willow stand and white spruce/blackspruce stands are graphed in Figure 8. As has been shown previously for an area of similar latitude and climate (Percin, 1960), densely timbered stands have a somewhat lower air temperature in summer than open
I
SEP
[ OCT I
NOV
I
DEC
I
JAN
I
FEE
OC OF
20
15
IO
WILLOW BALSAM POPLAR WHITE SPRUCE WHITE SPRüCE/BLACK SPRUCE frozen
-
60
- 50 . 40
-5
-I
o
FIG.6.Soil temperatures at 10 c m depth in the four stands. 229
O 5 IO
FIG.7. Average soil temperature profile for the period 4 May to 31 August.
20 stands. In winter, when differences in the amounts of incoming solar radiation are slight between the two stands, differences in air temperatures are negligible. Even in summer, differences in air temperatures are slight and cannot account for any great differences in soil temperatures: the average air temperature from 1 M a y to 31 August was 10.00C forthe willow stand and 9.60C for the white spruce/hlack spruce stand. Winter accumulation of snow varied considerably in the study area. S n o w depths for the willow and white spruce/black spruce stands are shown in Figure 9. Interior Alaska is a relatively windless region during the cold winter months and consequently within the spruce forests much snow is caught and retained by the crowns and never reaches the ground. S n o w depths are greatest in the willow and in the balsam poplar stands,and least in the two spruce stands. Even in late spring,snow persisted in the willow stand as late as it did in the spruce stands except for one late storm in M a y when the snow fell on warmed soil in the willow stand and on frozen moss in the spruce stands. S n o w
50
! ' I
I
!I
100
-WILLOW - -BALSAM POPLAR WHITE SPRUCE
I 1;
II I>!'
A
O #AR
ir-
WHITE SPRUCE/ BLACK SPRUCE 4
2
I
6 A?R
IO 12 14 16
8
1
MAY
1
JUNE
I
150 18 2 1 O c JUL
1
AUG
I
SEP
OC
I
oc1
I
NOY
1
DEC
I
JAN
1
FEE
OF
eo 20 60 10 40
O 20 -10
-20
-1-1
WILLOW WHITE SPRUCE/BLACK SPRUCE
O
-30
20
-40
40
-50
60
FIG.8.Average weekly temperaturesforthewillow and white spruce/blacksprucestandsfrom 1 October1964to 30 September1965.
230
Soil temperatures in river bottom stands in interior Alaska
FIG.9. Snow depth for the willow and white spruce/black spruce stands.
cm MAR 55..
1
APR
I
MAY
I
JUNE
I
JUL
1
AUG
1
SEP
1
OCT
I
NOV
1
DEC
I JAN I
I
I
FEB
0 WILLOW WHITE SPRUCE/BLACK SPRUCE
I
depths were at a m a x i m u m in late December and decreased throughout the rest of the winter due to packing and sublimation.
CONCLUSIONS
,
Differences in the soil temperature régime in the four river bottom stands can b e related to t w o factors: texture of parent material and thickness of the insulating organic layer. Older stands o n the flood-plain s h o w thick deposits of fine alluvial silt deposited during flooding. Frost penetration and thawing are slower in finer deposits than in coarser deposits because finer deposits hold m o r e water and thus m o r e latent heat of fusion is contained within the soil (Geiger, 1965). Permafrost is m o r e likely to occur closer to the surface in fine than in coarse soils. Changes in plant cover effect changes in soil temperatures. T h e insulating effect of the accumulated moss or organic layer is the most significant difference a m o n g stands. B r o w n and Johnston (1964)report that the thermal conductivity of dry peat is 0.0017cal/cm/oC/sec whereas that of saturated frozen peat has a conductivity of 0.0056 cal/cm/0C/sec. T h e thick organic layer a n d moss layer in the spruce stands in this study should have a thermal conductivity close to that of peat. During the w a r m periods in the s u m m e r the organic layer dries, resulting in l o w thermal conductivity, thus preventing heat penetration into the soil. In the autumn, the organic material becomes saturated by a u t u m n rains. W h e n frozen in winter it acts as a better thermal conductor than w h e n dry in s u m m e r . T h e thick organic layer thus allows m o r e heat transfer in winter from ground to atmosphere than in s u m m e r from atmosphere to ground. Soils with a thick organic layer are colder during s u m m e r than those with a thin or with no organic layer. An insulating moss layer helps
l
I
l
l
to maintain a permafrost layer in m a n y soils in the boreal region. T h e successional development of vegetation, especially the development of a thick moss mat, produces colder soils a n d eventually permafrost. T h e frozen soil prevents water percolation and the resulting wetter soils inhibit tree growth while favouring further moss development. Ultimately fast-growing timber stands along the rivers give w a y to slow-growing black spruce a n d bogs, the t w o most prevalent vegetation types o n permafrost soils.
SUMMARY 1. T h e soil temperature régimes in the four stages of river bottom succession are significantly different.
2. Soil froze earliest and deepest in early successional stages a n d later and less deep in later successional stages, T h e insulating effects of a m o s s layer and the presence of finer river alluvium probably account for slower freezing in the spruce stands. 3. M o r e s n o w accumulated in earlier successional stages, though freezing w a s faster and soil temperatures colder in these stands. 4. Large differencesin time of thawing occurred between the stands. In the earliest successional stage thawing w a s completed by the end of M a y , whereas in the oldest stand, thawing did not begin until the end of M a y and w a s never completed, a continuous frozen layer being present between 55 and 80 c m . 5. Fluctuations in soil temperatures are most rapid and greatest in the willow stage and are less rapid and of less magnitude in the later successional stages. 6.Soil temperatures during the growing season are warmest in the earlier stages of succession and colder in the later stages.
231
L. A. Viereck
Résumé L a température du sol dans les peuplements forestiers du fond des vallées à l’intérieur de l’Alaska (L. A. Viereck)
On trouvera dans cette étude des comparaisons entre la température d u sol à quatre stades différents et l’évolution du peuplement forestier dans la vallée de la Chena, près de Fairbanks, Alaska. On a enregistré la température du sol à des profondeurs de 5, 10,20, 50,100 et 150 c m dans un peuplement de saules ( S a h alaxensis Cov.), sur un banc d e gravier récemment mis à découvert, un peuplement de peupliers baumiers (populus balsamijera L.)de cinquante ans, un peuplem e n t d’épicéas (Picea glauca (Moench) Voss) vieux de cent vingt ans, et un peuplement mixte de deux variétés d’épicéas - Picea glauca et Picea mariana (Mill.) B. S. P. de deux cent vingt ans. On a constaté qu’au cours des premiers stades de l’évolution, le sol gèle plus vite et plus profondément et atteint des tempéra-
-
tures plus basses qu’aux stades suivants. L a plus grande différence de température d u sol entre les peuplements a été observée pendant le dégel et pendant la période de croissance des plantes. D a n s le peuplement de saules le dégel était terminé à la fin de mai, tandis que dans le peuplement mixte d’épicéas, il n’a pas c o m m e n c é avant la fin de m a i et n’a jamais été total, u n e couche gelée continue s’étant maintenue à une profondeur de 40 à 80 c m . Pendant la saison de croissance, les températures à 10 cm de profondeur ont atteint les m a x i m u m s suivants : saule, 22%; peuplier baumier, 14% ; Picea glauca, 9oC ; Picea mariana, 2%. L’action isolante d’une épaisse couche de mousse et de dépôt d’alluvions fluviales microgrenues explique sans doute le refroidissement plus lent et le dégel plus tardif dans les peuplements d’épicéas, par rapport a u x peuplements de saules et de peupliers baumiers.
Discussion J. MALAURIE. Avez-vous observé une stratification sur le plan de l’humidité de la surface du sol à la table du permafrost ? E n terre de Hall, il a été noté qu’il est une zone intermédiaire peu humide constituant du point de vue de la diffusion thermique c o m m e un “ coussin protecteur” (cf. Davies Greeland Symposium, Copenhague). D’autre part, et cela est un peu en dehors de votre c o m m u nication, avez-vous mesuré les temps et température lors de la surfusion de l’eau vive des rivières de Fairbanks (abord du Campus) ?
L. A. VIERECK.I have only one set of soil moisture units in place above a permafrost layer. The soil above the frozen layer remains wet during the summer because the water cannot penetrate through the frozen layer and because evaporation from the moss layer is low. I have attempted to observe if there is a migration of moisture to the freezing layer in the autumn but have not recorded any dehydration of the surrounding layers. Perhaps the “thermal cushion” that you mention occurs under very different circumstances from that which I have studied in interior Alaska. No, I have made no measurements of the temperature of the water in Alaskan rivers. I believe that Dr. Benson of the Geographical Institute of the University of Alaska has
232
recorded super cooling of river waters during the period of freezing.
C. O. TAMM.Nobody will question your statement that a thick m o s s layer is a better insulating agent when dry than when in a wet condition. However, I have a suspicion that speaking of thermal conductivity as the decisive factor m a y he an oversimplification. W-hile summer heating m a y be mainly a question of incident radiation and conductivity of the soil material, it seems at least possible that cooling by mass flow of cold air near the ground and into soil cavities m a y play some role, until the ground receives a thick snowcover. I should welcome a strict physical analysis of the heat transport in soils beneath different canopies (with their characteristic patterns of snow accumulation).
L. A. VIERECK. I agree with you that there is undoubtedly more involved in the effects of the moss and organic laver in the transfer of heat to 2nd from the soil than its thermal conductivity. The cooling effects of the evaporation of moisture from the moss surfaces as a result of transpiration is one of these. In years of late snow-fall,movement of cold air near the ground m a y be important but in interior Alaska, where snow usually remains on the ground from October to late April or M a y and where winds are very light,its effects m a y not be very great.
Soil temperatures in river bottom stands in interior Alaska
Bibliography BENNINGHOFF, W. S. 1952. Interaction of vegetation and soil frost phenomena. Arctic, vol. 5, p. 34-44. BROWN, R. J. E.;JOHNSTON, G. H . 1964. Permafrost and related engineering problems. Endeavour, vol. 23, p. 66-72. COLMAN,E. A.;HENDRIX, T. M. 1949. The fiberglas electrical soil-moistureinstrument. Soil Science, vol. 67, p. 425-438. DRURY, W. H. 1956. Bog flats and physiographic processes in the upper Kuskokwim region, Alaska. 130p. (Contr. to the Gray Herbarium, No. 178.) FRASER, J. W. 1961. A simple instrument shelter for use in
Bibliographie Forest Ecology studies. 10 p. (Canada Dept. of Forestry,
Tech. Note no. 113.) GEIGER, R. 1965. The climate near the ground. Cambridge, Mass., Harvard University Press. 611 p. PERCIN, F. DE. 1960. Microclimatology of a subarctic spruce forest and a clearing at Big Delta, Alaska. 162 p. (Quartermaster Res. and Eng. C o m m a n d Tech. Rpt. E. P. 130.) PÉwÉ, T. 1957. Permafrost and its effect on life in the North. Arctic Biology, p. 12-25.Oregon State College.
233
On the study of the ecology of subarctic vegetation I. Hustich
It is not easy to decide h o w to present such a broad and general subject as the ecology of subarctic vegetation. If I were a more modest m a n I should talk today only about some details, which I have studied during recent years; a more concrete observation, regardless of h o w small, always seems more useful than attempts to generalize things which perhaps are already vaguely defined. However, as a basis for discussion-and that is the main reason why w e are meeting here-I will try to present subjectively some features concerning the study of subarctic ecology. The first thing is to define the words “ecology” and “subarctic”. Regardless of the fact that so m u c h research has been done in ecology, it is still necessary to explain what it is: “Plant ecology” is a general term to cover the study of the plant/environment relationship, and in particular the importance of this relationship for the development and production of species and vegetation units with regard to habitat and regional distribution. I have, as you notice, slightly modiíied a well-known and generally accepted definition of the concept ecology by adding the word “production”, which is significant for modern ecological research. The word production is not here restricted to the applied science aspect; I have used it to give added importance to a significant quantitative tool, even if the word production in this respect should need some clarification itself. M u c h is thus crammed into this science. However, the ecological approach,regardless of “pure” or “applied”, w ill be more important in the future than at the present time. When, as seems to happen quickly just now, the relationship of m a n to his environment changes from being an interesting scientific question to an urgent need of knowledge for a mankind having to live on an overcrowded globe, ecology (or better
the ecological sciences) will certainly be more and more in the foreground. By this I do not want to advance a n e w “environmentalism” or a “determinism” of the rigid kind which is still to be found among some geographers. I have simply stressed the necessity to know the ecological mechanism not only between plant growth and an environment more or less untouched by man-as is still the case in large parts of the Subarctic-but also between plant growth and an environment which has been greatly modified or perhaps entirely changed by man. W e admit, if w e are honest, that phytogeographers have generally considered an environment, which has been greatly modified by man, as a “secondary” object of study compared with the natural or virgin environment. And what is the Subarctic? I must confess that it is rather interesting to take part in a symposium in which no one seems to k n o w exactly what the subject (i.e. Subarctic) really is and where everyone seems to have his o w n definition. The following is a definition which is very m u c h my o w n and which unfortunately is not commonly accepted: the Subarctic in a truly phytogeographic sense is the forest tundra, it is the narrow (or indeed very broad in some places of Siberia and Canada) “ecotone” beween the polar tree line and the boreal forest region proper. Lately,and particularly during the preparation of this symposium, I have noticed that m a n y scientists like to push the southern limit of the Subarctic more to the south,i.e., to the vaguely defined line where the closed or the continuous boreal forest begins; I refer to Professor Mikola’s paper on subarctic forests presented at this symposium. I also greatly appreciated Professor Blüthgen’s careful compilation of the relevant literature regarding this complicated question. I propose that w e push this need of clarification forward by bringing together (it could be done by, for instance, the Arctic
235
‘
I. Hustich
Instituteor Unesco) a few scientistsfrom differentparts of the Subarctic to work out a definition of the concept “subarctic”, a definition upon which later w e could all agree. As the definition in itself indicates the problems of ecology are in principle the same everywhere on the globe. But in each phytogeographical region, such as in the Subarctic,special factors operating in the environment come into focus.How can w e otherwise define these regions? (It seems that w e always need a classification and that our workis mostlylimitedto altering or even demolishing classifications made by others.) Regardless of h o w w e defineit the subarctic has-so it seems for me-been neglected compared with the quantity of biological research which has been done, for instance, in the Arctic. One reason, psychological, is that the Arctic is generally a more beautiful and interesting area than the Subarctic; also from the botanical point of view, particularly phytocenologically, it seems more clear-cut and, therefore, easier to investigate. Until recently,some parts of the Subarctic,northern Scandinavia, Finland and parts of the European U.S.S.R., have been more investigated than extensive parts of North America and Siberia, where w e can still find some of the largest biologically and geographically unknown “middle north” areas in the world. F r o m an ecological point of view, however, this regionally uneven localization of research is not of such a great importance.The subarcticbelt is in general very similar from east to west. A field biologist w h o has been working, let us say, in northern Europe and Canada, can observe almost at once this great similarity the two subarctic areas which are so far apart. This is not only true of the isolated mountains (cf. for instance, Hustich, 1962) or the shores, it is also true of the main forest types in the northernmost forests. It is thus by no means a mere chance that Cajander’s well-known forest-typesystem, based on the ground vegetation pattern as a key to the productivity of the forest, was easiest to apply in the subarctic parts of North America and Eurasia (cf. Kalela, 1961; Kujala, 1945; Hustich, 1949, etc.). T h e same is also true for other ecosystem theories which have been worked out in the northern regions. Ecological research in one part of the Subarctic can thus be applied quite easily to other parts of the same region, because of the circumpolar uniformity. It can also be done because m a n y species occur over almost the entire Subarctic.Other speciesofimportance in the vegetation are either vicarious, substitution or corresponding taxa: opinions vary considerably as regards the true taxonomical identities of m a n y of these taxa. Field botany should be more concerned with pointing out which species are the c o m m o n ones (and pay great attention to their ecology, because these are the species important in the vegetation and the landscape) and less interested in the often more or
236
less “philatelistical” job of listing the rare plants. However, according to the taxonomists, no single tree species c f the forest tundra (or of the Subarctic) seems to occur both in Eurasia and in North America. This is worth mentioning even if the reason for this might be accidental, due perhaps to the deep-felt respect the taxonomists seem to have for the Bering Strait. (Compare also the numerous ideas on biological and geological links between North America and Europe expressed during a symposium in Iceland in 1962, see Löve and Löve (1963).) F r o m an ecological point of view: which are the significant criteria of subarctic? This depends on h o w w e limit the Subarctic, of course. But pending a definition of the subarctic region which could be more commonly accepted, the following characteristic features must be mentioned: (u) the short growth season; (b) the great annual variations in growth, caused by the annual variations in temperature rather than precipitation; and (c) the duration of the snow cover. These featuresalso mark-but to a more pronounced degree-the arctic and the boreal region; nevertheless, w e can hardly study the ecology of the productivity of the Subarctic without going deep into the three above-mentioned factors. There seems to be no direct correlation between permafrost and other subarctic features. Also, I have not mentioned above the light climate. However, this factor varies, so m u c h within the subarctic region that it cannot be singled out as a typical subarctic feature. In the Subarctic we have “arctic”1ight conditions in northern Norway, at Mackenzie (Alaska), near Taimyr and in other parts of Siberia, where the subarctic region runs close to latitude 710 N. (i.e. up to the polar tree line). South of Hudson B a y and in southern Kamtchatka the Subarctic,on the contrary, reaches south to 500 N.with a light-climateof its own. Nevertheless, there are several interesting studies related to the response of growth and assimilation to the light-climatein differentlatitudes (cf Mooney and Billings, 1961) which are of importance for the understanding of subarctic ecology. A m u c h more typical ecological problem of the Subarctic is the snow cover and its significance for plant and animal life. It is remarkable h o w m u c h plant ecologists have neglected the study of the importance of the snow cover and winter conditions in general; both the zoologists and the foresters have done more in this field. (As a comment to the excellent paper on the snow cover by Mr. W. Pruitt, I should like to point out the difference in approach to this problem between plant and zoo-ecologists.A plant ecologist certainly needs a “time coefficient” as an addition to the snow index presented by Mr.Pruitt. W e must also realize that the varying height of the snow cover does not necessarily imply that the temperature on the ground itself, i.e. in and on the vegetation, varies in the same degree.) The ecological importance of ihe
O n the stiidy of the ecology of subarctic vegetation
snow cover is certainly a subject which deserves more attention than has so far been given to it; such research will have some useful practical applications. It should also add a n e w dimensionto a specific and often neglected sector of plant ecology,i.e. plant phenology. In a subarctic environment this problem cannot properly be understood if w e do not literally dig into the snow and study the winter conditions in general. The Subarctic should form a transition between the boreal forest region and the Arctic. This means that the entire polar tree and forest line as well as the forest tundra problem must be a central object of study also from a purely ecological and pragmatical point of view. Fortunately, m u c h has been done already in this field of research by Renvall, Eide, Heikinheimo, Andreev and others (see literature quoted by Mikola, 1969; Sirén, 1961; Tikhomirov, 1956; Hustich, 1949). Here I will only illustrate the main pattern of the change from the closed forest to the treeless tundra. This northern transition line, or better, transition belt, between the forest and its outposts is one of the main phytogeographical ecotones, comparable in ecological importance to the transition belt between the forest and the desert. The forest tundra (Ljesotundra, Waldtundra), forms in its ecological and “microtopagraphic” details an interesting mosaic pattern. Norin (1961)has pointed out that the forest tundra is more than a simple transition belt; it is a biocenological entity in itself, an opinion which is more or less the same as I have advanced in an earlier paper. The short growth season towards the north and the great influence of temperature as an external growth regulator, are, as mentioned above, two typical features which should be taken into account in every ecological study in the Subarctic. In this respect w e must also remember that the climatic control is of primary importance and the edaphic control of secondary importance for the distribution pattern and the ecology of a species. F r o m the southern margin of the Subarctic to the polar tree line there is an increased “climatic hazard coefficient”, a concept which I introduced twenty-fiveyears ago and which certainly could be more used as a tool in ecologically-directedproductivity research. The climatic hazard coefficient (i.e. the size of the variation coefficientin an annual growth series) clearly increases towards the north. The variations in growth of trees in the forest tundra and in the Subarctic in general correlate particularly well with the annual variations of the July temperature because the growth processes are, in fact, more or less concentrated in this month. It is, however, a rough measurement only, but very intricate multiple regression analyses (see, for instance, Sirén, 1961) also show more or less the same thing. Of course, w e still k n o w too little about the real mechanism (i.e. the physiology of the response of growth to climate. W h a t w e k n o w so far is rather superficial; however, Dr. R. Sarvas’
paper presented at this symposium, introduces some interesting n e w aspects of this problem. The flowering and fructification processes, and also the vegetative growth (thus, including the production of dry matter) of trees and perennials, vary annually in the Subarctic to such an extent that all investigations relating to the productivity of a subarctic area must be carried out for several years if such productivity research is to have any meaning at all. M a n y sins have been and will be committed in this respect. Aspects relating not only to the annually but also to the periodically varying growth (such as the climatic fluctuation in the thirties) must be included in ecological research, particularly in the Subarctic (cf. Erkamo, 1956; Tikhomirov,1956). The need for ecologically correct productivity research has been “in the air” for m a n y years. Now the International Biological Programme has taken up this problem in all its width; it is also a sound idea to let the “Biological Year” cover a period of at least five years (cf. Worthington, 1965, and my comment above). The subarctic region is in general believed to be poor in productivity per areal unit. According to a recent Russian report (Bazilevitch and Rodin (1964) quoted by Ovington, 1966) the production of dry matter in metric tons per year is in the Arctic about 1 ton, in the Subarctic about 5 tons, but in certain more southerly regions over 30-50 tons; these productivity observations are,however, still based on a small material only. Compare in this respect the proceedings of a “productivity symposium” (Lieth, 1962) and for an excellent detail study on the productivity of a c o m m o n subarctic forest moss species, see T a m m (1953). M a n y years ago I illustrated how, in some years, the favourable temperature in combination with the light-climatein northern Finland can allow a production of organic substance as high or higher than in southern Finland,per areal unit. Theoretically, it is in some ways easier to carry out productivity research in the Subarctic than in other regions, except the Arctic. The growth capacity of, for instance, an open one-storied northern forest can easily be measured and its potential more or less forecast. Incidentally, the ecology of lakes and rivers in the Subarctic is a sector worth still more attention even though m u c h attention has recently been focused on such research. The study of the productivity of the lakes-a truly ecological problem of great practical significance-is neverthelessin its infancy in large parts of the Subarctic. The population of the Subarctic, defined earlier according to the “broader” alternative, is probably 8-10million. The area, the “middle north”, to use a recently-coined expression, is thus very thinly populated. This is one reason why a noticeable,and in the long run dangerous, carelessness with nature is the mark of man’s efforts in large parts of the Subarctic.
237
I. Hustich
One is seldom keen to save what one thinks one has more than enough of. W e have started late with the ecological research of m a n y significant land-use problems in the Subarctic. The influence of changes in groundwater level,caused by the construction of large artificial Iakes, is one such problem where lack of co-operation between ecological research and accelerated practical application has been evident. There is also m u c h ecological work still to be done, for instance on the effects ofthe drainage of peat lands. The peat lands of the Subarctic have, however, in contrast to lakes and rivers, been the object of intensivephytogeographic research in northern Europe and the Soviet Union. It is an important tribute to the research activity of earlier phytogeographers that the necessary basic study of morphology and stratigraphy of peat lands started m a n y decades ago; this,. as so m u c h research earlier and later, started with no particular thought being given to its practical usefulness. Today, the theoretical, ecological study of bogs and fens and research as regards their practical use go hand in hand; such co-operative efforts have also started regarding reafforestation in the north. Phytosociology (in a rather orthodox form) was a very popular science in Scandinavia some decades ago. At the beginning the theoretical approach to phytosociology markedly neglected the ecology of the species and the “plant societies”.The reason for the earlier popularity of phytosociology was that in the 1920s and 1930s several university professors just happened to be interested in pure phytosociology. I mention this detail only because w e should not forget such h u m a n factors w h e n discussing the various directions which a branch of science takes at different times in different countries. Today w e can say that ecologicallyill based productivity research is rather modern and w advance, as I hope, rapidly. It must, however, be stressed that w e still need phytosociology or phytocenology as an important primary tool also for an ecologically-directed inventory of the vegetation. Itis time n o w to end the unnecessary friction between the different “sociological” schools and to adopt some general ecosystem classification,at least within a large region such as the Subarctic. Such a classification of subarctic vegetation, if generally adopted, would greatly aid fundamental knowledge of
’
the different regions and give a better basis for produc-
tivity comparisons. During a symposium on forest types and ecosystems, held in Montreal in 1959, I was somewhat concerned with the rather inflexible individual and national attempts-also observed later -to maintain so m a n y different theories and schools in this field. Recently Ovington (1966) said that “comparative measurement in natural and man-managed areas in m a n y parts of the world should greatly improve our understanding of what determines the ‘productivity’ of plants and animals”. I believe that the Subarctic one of the few great reserves of land,fresh air and water which are still untapped on this planet, is a natural area for an increased international research effort and one of the efforts of the final session should be to strongly emphasize this need. I believe that plant ecology is the basic and natural science for such a task. W e need,however, much more co-operation to avoid unnecessary duplication of work; w e need the exchange of students and exchange of ideas on a greater scale than hitherto. The fact that the ecology ofthe subarcticvegetationis in its main features so similar in different parts of the subarctic zone makes these exchanges easier. But ecology, particularly when directed towards productivity research, is a more concrete and experimental science than phytocenology,a fact which gives us greater hope of achieving international co-operationin this field. I a m well aware that 1 have neglected m a n y important aspects-our problem today is so big that it is impossible to cover all sides of the question. A few simple questions for discussion emerge from what I have stated above. 1. W h a t is the subarctic region? 2.Which are the main characteristics of this region? 3. H o w can w e accelerate the study of the ecology of this region? 4.iHow can w e rationalize our terminology so that at least the primary concepts are defined similarly in Eurasia and North America? 5. H o w can we, for instance, standardize ecological equipment (i.e. using more or less the same instrument for the same purpose) and reach a better co-operation between scientists all over the Suharctic?
Résumé Sur l’étude
de l’écologie de la végétation subarctique
(I. Hustich) ’
L’auteur expose que les facteurs mésologiques particuliers dont dépend la végétation dans la région
238
subarctique, outre bien entendu les facteurs généraux qui agissent partout, sont : a) la brièveté de la saison de croissance ; b) les grandes variations annuelles de croissance dues aux variations annuelles de température (plutôt qu’aux précipitations) ; c) l’enneigement. En
O n the study of the ecology of subarctic vegetation
raison du caractère circumpolaire de la région subarctique (surtout si l’on n’entend par zone subarctique que la région de la toundra forestière) les recherches écologiques effectuées dans la partie eurasienne de cette zone peuvent être utilisées dans la partie nordaméricaine, et vice versa. Cette étude souligne en outre l’importance de la limite de la forêt, qui constitue l’une des principales “frontières” phyto-géographiques. Le coefficient d’aléas climatiques, qui augmente vers le nord, est une
caractéristique dont les écologistes qui travaillent dans la région subarctique doivent se préoccuper plus qu’ils n e l’ont fait jusqu’ici. L’auteur signale aussi le rôle que joue le P r o g r a m m e biologique international dans l’avancement des études écologiques relatives à la région subarctique, et la large nécessité d’une plus grande coopération internationale en matière de classification phyto-écologique et de recherches sur la productivité.
Discussion F. E. ECKARDT. A u début de votre conférence, vous avez proposé une nouvelle définition du terme “écologie” en faisant intervenir le concept de “productivité ”. Je regrette u n peu cette proposition et cela pour deux raisons. D’tvbord je pense que l’on ne devrait pas augmenter le nombre de définitions de ce terme, qui est déjà utilisé dans plusieurs acceptions différentes. Ensuite, parce que le concept “productivité” est, lui-même,très difficile à définir. Personnellement, je crois qu’il serait raisonnable d’employer le t e r m e <‘ ecologie” ou bien dans l’acception ancienne désignant une science qui a pour objectif l’étude de l’ensemble des rapports entre les organismes vivants et le milieu ambiant, ou bien dans l’acception plus moderne adoptée lors du premier colloque international sur les écosystèmes, tenu à Copenhague en 1965. Cette acception du terme était la suivante : l’écologie est la science qui étudie la structure et le fonctionnement de la biosphère ainsi que de ces constituants, les écosystèmes. I
I. HUSTICH. M y definition of the concept “ecology” was only intended to point out the “productivity” aspect, which I believe is an important part of an ecological investigation at a time when “land use” is not only a catchword but a necessity. Ecology is such a wide concept that its definition must be vague, as is illustrated by the definition made in Copenhagen in 1965 and referred to by Dr. Eckardt. W . PRUITT. I would like to second heartily Professor Hustich‘s comments. They bring to mind what I consider m a y be another importance of the Subarctic. As this planet gets more and more crowded, ecology will become more and more the centrai pillar of h u m a n survival. Thus there will be increased need for dissemination of ecological concepts. In the Subarctic various ecological principles are displayed with simplicity and directness, because the subarctic ecosystems are relatively simple. I look upon the Subarctic as a great teaching laboratory, where ecological principles can be studied by people from the temperate and tropical zones where the ecosystems are too complex for easy understanding. M a n y inland universities today require their doctoral candidates to spend time at a marine biological station; I visualize the time when temperate and tropical universities will require their candidates to spend time at a subarctic biological station.
F. E. ECKARDT. Le professeur Hustich a mis l’accent très pertinemment sur la nécessité d’intensifier la recherche écologique dans les régions arctiques et subarctiques. J e pense qu’il convient d‘ajouter à cet égard que de telles recherches devraient être réalisées de préférence par des équipes composées de micrométéorologistes, écophysiologistes, édaphologues et phytosociologues, et que ces équipes devraient travailler en rapport étroit avec des spécialistes du traitement de l’information. O n ne peut pas ne pas regretter le gaspillage de matière grise auquel aboutissent souvent les méthodes de recherche actuellement utilisées. E. HULTEN. You said that field botanists should pay more attention to the c o m m o n species and not to the rare species which have merely “philatelic” interest. I was shocked, as in m y opinion the rare isolated species provide a key-one of the very few objective keys in fact-to the older history of the flora.Taken together from different areas the occurrence of these rare plants presents a consequent picture and helps us to interpret the changes that have taken place in the flora and vegetation between the present time and earlier periods. I hope that ecological field workers also will pay increasing attention to the rare plants they come across. Ecological papers are mostly lamentably destitute of such information.
I. HUSTICH. I merely stressed the importance of knowing in detail the ecology and physiology of the c o m m o n plants, which dominate the vegetation cover. In m a n y subarctic areas only 20-30 species are quantitatively-from the point of view of production, for example-really important. I m a d e the comment, I must confess, mainly to shock you a little (Hulten: You succeeded!), and by irritating you to a comment, to stress m y point that w e should not-as seems to happen too often-neglect the trivial elements in the flora (which usually, as w e know, are not well represented in the botanical museums, for instance). L. AARIO. W e should pay more attention to the ancient timber lines, I don’t mean so m u c h in this case the timber line in the post-glacial period-that is important in other respects too-but the more real historical timber line, in order to eliminate the h u m a n interference. In the Alps of the Tyrol the timber line has gone down about 300 m since mediaeval times and there is nothing to be seen in the landscape
239
I. Husticb
of this former “pre-human” timber line. In Fisherman Peninsula on the coast of the Arctic Ocean the Norwegian fisherm e n have destroyed the birch woods. In Finland, Sweden and Norway the Lapps have damaged the wood line, too, but we do not know h o w much. Thus the further timber line must be stated, before w e are able to declare the right climatological and other ecological data that determine the location of the timber line.
E. SCHENK. You mentioned three environmental factors of subarctic ecology: the short growth season, the great annual variations in growth caused by annual temperature, and the snow cover. A fourth factor-permafrost-is not mentioned, although there are large areas of permafrost. Permafrost is important for the temperature of the soil, for the moisture content of the soil, for the groundwater table in the soil and deeper ground, for the water-level in rivers
and lakes, etc. I ask, therefore, w h y you did not mention permafrost as a significant factor of subarctic ecology ?
I. HUSTICH. The permafrost is, of course, of importance and I mentioned it in passing. However, the correlation between the permafrost and the phytogeographical belts is not very clear and we need-precisely because of the reasons outlined by Dr. Schenk-much more research into this matter.In old feather moss forests, on water divides in southern Labrador, ice under the moss cover is still found in July. In certain years this ice might not melt at all. This means that the southern border of permafrost is not clear and we could use the expression “semi-permafrost” for the pattern in such areas. In passing, I would mention that the concept permafrost is in need of a clear definition, I refer to Dr. Brown’s paper,
Bibliography / Bibliographie ANDREEV, V. N. 1947. Recherches géobotaniques du nord de Petchora. Leningrad. Sovetsk. Botan., vol. XV, no. 4. BAZILEVITCH. N. I.; RODIN, L. E. 1964. Types of biological cycles of ash elements and nitrogen turnover for the main natural zones of northern hemisphere. Genesis, classi5cation and geography of soils in the U.S.S.R. Moscow, U.S.S.R. Academy of Sciences. ERKAMO, V. 1956. Untersuchungen über die Pflanzenbiologischen und einige andere Folgerescheinnngender neuzeitlichen Klimaschwankung in Finland. Helsinki, Ann. Bot. Soc. Vanamo, vol. 28, no. 3. HARE, F.K . 1959. A photoreconnaissance survey of LabradorUngava. (GeographicalBranch, Ottawa, Memoir 6.) HUSTICH, I. 1949. O n the forest geography of the Labrador peninsula. Helsinki, Acta Geographica, vol. 10, no. 2 . 1962. A comparison of the floras on subarctic mountains in Labrador and in Finnish Lapland. Acta Geographica, vol. 17, no. 2. KALELA, A. 1961. Waldevegetationszonen Finnlands und ihre klimatischen parallelltypen. Helsinki, Arch. Soc. Vanamo,vol. 16, Suppl. K ~ C H L E R ,A. W. 1966. Analyzing the physiognomy and structure of vegetation. Ann. Assoc. Amer. Geogr., vol. 56., no. 1. KUJALA, V. 1945. Waldvegetationsuntersuchungen in Kanada m i t besonderer Berucksichtigung der Anbaumöglichkeiten Kanadisher Holzarten. (Helsinki, Aead. Scient. Fenn., ser. A, vol. IV. LIETH, H.(ed.). 1962. Die Stoffproduktion der Pflanzendecke. Stuttgart, G. Fischer Verlag. LOVE, A.; LOVE, D. 1963. North Atlantic biota and their history. Oxford, Pergamon Press.
_-
240
MIKOLA, P. 1969. Forests and forestry in subarctic regions. Ecology of the subarctic regions. Proceedings of the Helsinki symposiurnlEcologie des régions subarctiques. Actes du colloque d’Helsinki. Paris, Unesco. (Ecology and conservation/Ecologie et conservation, I.) MOONEY, H.A.;BILLINGS,W.D. 1961. Comparative physiological ecology of arctic and alpine populations of Oxyria digyna. New York. Ecol. Monogr., 31. NORIN, B. N. 1961. W h a t is the “forest-tundra”? Botan. Journal, vol. X L V I . Moscow, U.S.S.R.Academy of Sciences. OVINGTON, J. D. 1966. Measuring the fruits of the land. New Scientist,London, 3 March. RITCHIE,J. C. 1956. The vegetation of northern Manitoba, I. Canad. J. Bot., vol. 34. SIREN, G. 1961. Skogsgränstallen som indikator för klimatfluktuationer i Norra Fennoskandien under historisk tid. C o m m . Inst. For. Fenn.,vol. 54, no. 2. STÅLFELT, M. G. 1960. Vcäztekologi. Stockholm, Scandinavian University Books. Symposium on forest types and forest ecosystems during the ninth botanical congress in Montreal, 1959. Silva Fennica, vol. 105. Helsinki, 1960. TAMM, C. O. 1953. Growth, yield and nutrition in carpets of a forest moss (Hylocomium splendens). Stockholm, Medd. Stat. Skogsforskn. Inst., vol. 43. TIKHOMIROV, B. A. 1956. O n the preservations of forests on their northern limit. The vegetation of the Far North of the U.S.S.R., I. Moscow, U.S.S.R. Academy of Sciences. WORTHINGTON, E. B. 1965. The International Biological Programme. Nature, (Lond.), vol. 208.
A critical limit of primary production for the survival of arctic alpine plants in the northern Pennines of England D. J. Bellamy and W.M. Tickle
The
area of the northern Pennines of England k n o w n as Upper Teesdale has long been famous for its rich and peculiar flora (Backhouse, 1843; Valentine, 1965), which includes a number of pre-alpine, alpine, arctic alpine and subarctic plants (Pigott,1956). The majority of these can be found in s o m e abundance between the 300 m and 600 m contours which, in an area at approximately latitude 54040’N.,is indeed remarkable. Manley (1942)describes the climate at 840 m on the western bosndary of Upper Teesdale in the following words: “. . w e therefore form a conception of an excessively windy and pervasively wet autumn, a very variable and stormy winter with long spells of snow-cover,high humidity and extremely bitter wind, alternating with brief periods of rain and thaw. April has a m e a n temperature little above freezing point and sunny days in M a y are offset by cold polar air, while the short and cloudy summer is not quite w a r m enough to support the growth of trees. Throughout the year indeed the summits are frequently covered in cloud.” He labels the climate of the high western ridge as subarctic but emphasizes the marginal nature of this upland climate by saying that “a relatively slight increase in the frequency of w a r m anticyclonic summer weather would allow a rise in m e a n temperature almost sufficient to permit the growth of trees”. Yet over 300 m lower o n both north- and southfacing slopes, ecosystems harbouring arctic alpine plants are widespread in conditions which might therefore be regarded as marginal for their survival. Figure 1 shows a climatic diagram for Upper Teesdale and for true arctic and alpine stations. The proportion of English gardens situated below the 300 m contour and the abundance of arctic alpines in our seedman’s catalogues should be proof enough that these planta will grow at even lower altitudes and in dryer situations, with m u c h longer growing periods than the arctic wastes of Upper Teesdale.
.
W h a t then is the factor which restricts the range of these plants in natural situations?
The importance of competition, or at least the lack of it, to the survival of arctic alpine plants has been discussed widely; of direct relevance are the papers of C o o m b e and White (1951) and Pigott (1956). There is, however, little or n o empirical data available concerning this important factor. T h e work to be outlined w a s not primarily directed at obtaining the necessary data but s o m e of the results obtained are of direct relevance, and are therefore presented here.
T H E STUDY The w o r k w a s basically a study of the floristic m a k e - u p and productivity of the “wetland ecosystems” of U p p e r Teesdale in relation to their mineral supply (Bellamy, 1965). Wetland ecosystems in the limited sense of this work are defined as ecosystems which have a water-table, above, at, or very near the substrate surface, the substrate remaining saturated throughout the year. These m a y be subdivided into t w o main groups. 1. T h e Blanket Mire-ecosystems occupying water partings which are covered with a blanket of peat (which varies in thickness from a few centimetres to 4 m), their water and mineral supply being the rain falling directly on them (Bellamy and Holland,
1966). 2. Flushed ecosystems-these m a y be subdivided into three types: (u) ecosystems which occupy the drainage axes of the blanket mire and are affected by run-off from peat-covered catchments; (b) ecosystems which are supplied with spring-water derived from the acid rocks of the area (mainly quartz dolerite) and from glacial drift formed from 241
D.J. Bellamv and W.M.Tickle
Reschen (1,494m)
Moorhouse (556rn)
(131
5,10
1870
FIG.1. Climatic diagrams for Upper Teesdale (Moorhouse meteorological station at 556 m) and for true arctic and alpine stations.
10. -- 30. 1
-- 21. 51.
3.t
quartz dolerite; and (c) ecosystems which are
SOIL STUDIES
supplied with spring-waterderived from the extensive deposits of carboniferous limestone in the area and from glacial drift and boulder clay formed from the limestones.
THE VEGETATION A representative series of the wetland ecosystems was mapped and its vegetation described using standard phytosociological techniques (Braun Blanquet, 1964), combined with point quadratting. Differential analysis of the floristic data using the formula given by Kulczynski (1949), indicates the existence of a cline of wetland associations which fall into seven groups on the grounds of similarity of species content. The direction of variation of this cline appears to be related to the increase in the effect of flowing bicarbonate-richground water passing from the blanket mire ecosystems to those developed in spring-watersderived from the limestone. Figure 2 gives the Kulczynskidiagram,which shows the results of the differential analysis of the reduced matrix of sites used in the production studies. 242
Soil samples were collected from each association described. T h e samples were dried and aliquots extracted with 2N a m m o n i u m acetate (J’efferiesand Willis, 1964), and the extracts analysed for calcium and magnesium (atomic absorption spectrophotometer) and sodium and potassium (flame photometer). Total carbonates were determined by shaking aliquots of dried soil with standard acid followed by back titration using bromothymol blue as indicator. The p H was determined on fresh soil samples made liquid by the addition of distilled water.
PRODUCTION STUDIES Sixteen wetland complexes covering the seven floristic types were fenced in early 1965 against grazing by sheep and cattle. These ranged in altitude from a site at 440 m to one at 640 m. In order to minimize damage to the rare communities,increment cropping was carried out on only two sites. For this purpose the enclosures at either end of the altitudinal range were sited around communities which are widespread in the area, and which contain n o rare plants. These
A critical limit of primary production for the survival of arctic alpine plants
FIG. 2. Kulczynski diagrams for the reduced matrix of sites used in the production studies. Explanation of symbols
@ Ericion tetraiicis
0
Molinion
A
Cardanino-Montion
0 Caricion iasiocarìae 0 Caricioncanescenti-fuscae
0
Eriophorion latifolii
O unclassified moss community
Symbols denoting floristic groups
Shoat prodLICliO” g/mf ‘Y r
43 1 415 310 383 268 449 461 579 272 323 344 266 295 272 193 422 548 414 99 140 136 150 118 164 108 53 150 142
were sampled at monthly intervals by paired, clipped quadrats,the area of each quadrat being half a square metre. As soon as the standing crop showed maximization in the highest enclosure, cropping of the other enclosed communities was begun, gradually working downwards. All the crops were treated in exactly the same way. They were first sorted into Monocotyledons, Dicotyledons and Bryophytes, and then each component was sorted into current and previous years’ growth. After sorting, the components were dried to constant weight at 1000 C. F r o m the results a mean figure for the net annual aerial production per square metre was calculated. Throughout the discussion this will be referred to simply as “shoot production”.
DISCUSSION M a n y of the Teesdale communities which contain arctic alpine species are very open, their total vegetation cover being less than 30 per cent. As it seems ill be greatest reasonable to infer that competition w in communities with m a x i m u m cover value, only those results for communities with a cover value of 100 per cent will be discussed here. Figure 3 shows the values for shoot production
plotted: (a) against the combined content of calcium and total carbonate per 100 g dry weight of soil (as a rough edaphic scaler), and (b) against altitude (as a rough climatic scaler). Manley (1945)states that for the northern Pennines “a slight increase in elevation is accompanied by a remarkably large decrease in the length of the growing season (a shortening of 10 days for 80 m”. It m a y readily be seen that there is no correlation between shoot production and either of these factors. However, one interesting fact emerges. All those communities which contain arctic alpine species (shown on the figures in solid symbols), have an annual shoot production of less than 150 g. Comparable figures for the shoot production of a whole range of true alpine tundra ecosystems are given in Table 1. T h e figures are all taken from Bliss (1966)and 75 per cent of them have a shoot production of less than 150 g. It is also interesting to note that the highest shoot-production figure given by Bliss is from a community dominated by woody perennials, (tundra heath). This is certainly ecologically similar to the blanket mire ecosystems of Upper Teesdale which are dominated by Calluna vulgaris and which have a shoot production of 272 g. O n e other point of similarity between the Upper
243
D.3. Bellamy and W.M. Tickle
Teesdale ecosystems containing arctic alpines a n d the true tundra ecosystems, is the calorific value of the plant material (Table 2). Bliss c o m m e n t s o n the high calorific values of tundra
plants linking t h e m to the storage of fats a n d surmising o n the importance of energy-rich food stores in ecosystems with a short growing season.
TABLE1. Shoot production in grammes dry weight per square metre
Bliss (1966)
Upper Teesdale
Grass spring Brown m o s s spring Grass spring Grass spring Polytrichum heath Brown moss spring Tall rush mire Brown moss spring Tall rush mire Grass spring Grass spring Tall rush mirc Brown moss spring
579 548 461 449 431 422 415 414 383 344 323
310 Heath
Sriowbank Sexijraga-Antennmia Artemesia dry meadow Calamogrostis wet meadow Sedge meadow .
283
200 193 192 180 176
Carex dry meadow
150
Sibbaldia-Agrostis
141
Moist swales Heatb-rush meadow
295 272 272 272 268 266 193
164 150 150 142 140 136
Heather moor Grass spring Brown moss spring Tall rush mire Brown moss spring Brown moss spring
Sedge-rush sward Cushion community Sedge-rush sward Cushion community Carex turf Carex turf
24-125 124 118
Sedge-rush heath Carex-Deschampsia wet meadow Geum-sedge, south slope Carex-Deschampsia, moist meadow Arenaria-Eriogonurn,dry meadow Heath-rush fellfield Turf sites Diap nsia Carex-Geum turf G e u m - S a h meadow
108 99
244
Cushion community Carex turf
13-102 78 74 11-68 67
6-60 58 53
Geum-sedge,north slope Carex-Deschampsia Cushion plants Carex-Sib baldia-Erigeron
Sedge-rush sward
112 112 112
45 40 28 21
Cushion community
A critical limit of primary production for the survival of arctic alpine plants
11 ,OOf
=
:
500
o
o 0
0
For explanation of symbols see Figure 2.
O0
O
o
,-300-
o
E
O
k 8 200‘Z
:150-m
FIG.3. Shoot production per squaremetre plotted against (a) the combined content of calcium and total carbonate per 100 g dry weight of soil and (b) altitude.
oo
o
o
. I
-c
-
~100-~
o r rn
TABLE2. Comparative figures for Upper Teesdale and Mount Washington (Bliss, 1962) C&/g
Evergreen shrubs Herbs Mosses
represent early seral stages which are characterized
by low production, which are held at this stage by grazing pressure, a n d in which the arctic alpine species can survive and in fact, m u s t have survived since the late glacial.
ash free dry weight
Mt. Washington
Upper Teesdale
5 098 4 601 4 410
5 239 4 710 4 419
It would appear that the Teesdale ecosystems have m u c h in c o m m o n with true alpine tundra ecosystems, and it would therefore Seem reasonable to speculate that the critical figure of 150 g dry weight shoot production per square metre is of m o r e than local significance in the survival of the plants in question. In true arctic and alpine areas, harshness of climate, and especially the short growing season, m u s t limit the production level of the ecosystems. Also, instability of the substrate due to frost action and wind, rain and rime erosion could easily account for the maintenance of the ecosystems in a n open “skeletal” form. In U p p e r Teesdale the action of herbivores, mainly sheep, could well account for the latter effect (Pigott, 1965), a n d the dramatic effects brought about by fencing. A t the present state of knowledge all w e can d o is surmise that the special Teesdale communities
APPENDIX T h e following is a list of the arctic alpine species occurring in the communities referred to in this paper. Care% capillaris L. Juncus alpino articulatus Chaix. KObresia simpliciuscula (Wahlenb.) Mackenzie. Saxifraga aizoides L. Thalictrum alpinum L. Tojieldia pusilla (Michx.) Pera. Catascopium nigritum Brid. G y m n o s t o m u m recurvirostrurn H e d w .
ACKNOWLEDGEMENTS This w o r k w a s carried out while one of us (W. Tickle) held a n NERC research studentship. W e would like to thank the R a b y and Strathmore Estates for permision to w o r k o n their lands, a n d the Nature Conservancy for permission to w o r k in the U p p e r Teesdale National Nature Reserve.
245
D.J. Bellamv and W.M.Tickle
Résumé U n seuil critique de la production primaire pour la survie de plantes alpines arctiques dans les Pennines septentrionales (Angleterre) (D.J.Bellamy et W.M. Tickle).
L e róle de la concurrence c o m m e facteur de régulation de la répartition des espèces botaniques qui caractérisent les peuplements arctiques, alpins arctiques et alpins a été souvent étudié. D a n s la région des Pennines septentrionales (Angleterre) connue sous le n o m d’Upper Teesdale (vallée supérieure de la Tees), il existe entre les courbes de niveau de 300 à 600 mètres, une série d’a écosystèmes de mouilles» (bourbes et flaches), dont certains contiennent des plantes alpines arctiques. Un grand n o m b r e d’entre elles apparaissent en peuplement squelettique là où le couvert végétal total ne dépasse pas 30% et où la concurrence doit être très limitée. Cependant toute une partie de la flore alpine arctique
de cette région peut se trouver dans des peuplements où le couvert végétal atteint 100%. L a mesure de la production primaire aérienne nette d’un ensemble de ces peuplements indique que le seuil critique est de 150 g r a m m e s de poids sec par mètre carré et par an. On n’a trouvé aucun cas d’espèce alpine arctique croissant dans des écosystèmes où la production aérienne annuelle nette est supérieure à ce chiffre. On a constaté un seul cas d’écosystème où la productivité est inférieure à 150 g r a m m e s et où il n’existe pas a u moins une espèce alpine arctique. Les auteurs examinent les facteurs qui peuvent maintenir la production à un bas niveau. U n e p o m p a raison avec les chiffres concernant la production d’autres écosystèmes où il existe, à l’état de reliques, des espèces alpines arctiques et à la production des écosystèmes de toundra proprement dite montre que le seuil critique obtenu dans la vallée supérieure de la Tees présente plus qu’un intérêt local.
Bibliography Bibliographie BACKHOUSE, J. ; BACKHOUSE, J. 1843. A n account of a visit to Teesdale in the summer of 1843. Phytologist, vol. 1, p. 892-5. BELLAMY, D.J. 1965. Studies on the productivity of mire ecosystems. (Bericht 9 Int. Kong. fur Univ. Moor und
Torf.)
-_ ; HOLLAND, P. 1966. Determination of the net aerial production of Calluna vulgaris (L.). Hull in northern England. Oikos, vol. 17, no. 2. BLISS, L.C. 1962. Caloric and lipide content in alpine shrubs. Ecology, vol. 43, p. 753-757. . 1966. A comparison of plant development in microenvironments of arctic and alpine tundras. Ecol. Monog., vol. 26, p. 303-337. BRAUN BLANQUET, J. 1964. Pflanzensoziologie. Wien, Springer Verlag, 865 p. COOMBE, D.E.; WHITE, F. 1951. Notes on calcicolous communities and peat formation in Norwegian Lapland. J. ECO^., vol. 39, p. 33-62.
__
246
JEFFERIES, R.L.; WILLIS, A. J. 1964. Studies on the calcicole-calcifuge habit. 1. J. Ecol., vol. 52, p. 121-138. KULCZYNSKI, S. 1949. Peat bogs of Polesie. M e m . Acad. Sci. Cracovie,ser. B, p. 1-356. MANLEY, G.1942. Meteorological observations on D u n Fell, a mountain station in northern England. Quart. J. R. Met.SOC.,vol. 68, p. 151-162. . 1945. The effective rate of altitudinal change in temperate Atlantic climates. Geogr. Rev., vol. 35, p. 408417. PIGOTT, C. D. 1956. The vegetation of Upper Teesdale in the North Pennines. J. Ecol., vol. 44, p. 545-586. VALENTINE, D.H . 1965. The natural history of Upper Teesdale, p. 70. Newcastle upon Tyne. WALTER, H. 1963. Climatic diagrams as a means to comprehend the various climatic types for ecological and agricultural purposes. In: The water relations nf plants, p. 3-9.Oxford, Blackwell.
_-
Plant ecology and succession in some nunataks in the Vatnajökull glacier in South-eastIceland E.Einarsson
INTRODUCTION Since about 1890 the glaciers of Iceland have, as a whole, been retreating considerably; they have thinned and shrunk especially during the period 1931-60which shows a considerable rise of air temperature in Iceland as compared with the period of 19.0130 (Thórarinsson, 1943 ; Eythórsson, 1963). Vatnajökull, the biggest glacier in Europe, which covers a little more than one-twelfth of Iceland, has been retreating like the other glaciers. Breidamerkurjökull, being Vatnajokull’s biggest southern outlet, has thus retreated 1,254 m during the period 1931-60(Eythórsson, 1963) and at the same time become somewhat thinner. D u e to this almost continuous retreating of the glaciers, numerous areas in Iceland earlier covered by ice have become free of ice during the last fifty years, and for some of these n e w nunataks their accurate or approximate “age” is known. T h e aim of the investigations reported in this paper was to try to find out h o w two n e w nunataks in Breidamerkurjökull were colonized by plants and to follow the succession of the vegetation, partly by comparing the vegetation of these nunataks to the vegetation of an old nunatak area.
THE NUNATAKS In Breidamerkurjökull,about 18 km from the margin of the glacier,there is a nunatak area named Esjufjöll (Fig.1). It consists of four steep mountain ridges, their highest peak being 1,640 m , mostly built up of palagonite tuff and breccia together with some basalt, which have probably been ranging up above the surrounding ice through out the post-glacial time (Einarsson, Th., personal cofimunication). The Esjuf-
jöll nunataks have been k n o w n for a long time (Björnsson, F., 1951) and are mentioned in old travelogues (Olafsen and Povelsen, 1772) but there has been some confusion as to the n a m e of the area and Esjufjöll has sometimes been confused with another nunatak area named Mávabyggdir, which is also situated in Breidamerkurjökull but about 5 km farther west. The flora and vegetation of Esjufjöll were, however, almost unknown until late July 1950 when four brothers from the farm Kvísker, which is the nearest farm to Breidamerkurjökull, made ail expedition to these nunataks. The Björnsson brothers from Kvísker, all of them being exceptionally keen and clever amateurs on natural history, found a fairly rich flora and luxuriant vegetation in some sheltered places in the Esjufjöll nunataks (Björnsson,H., 1951). Just after 1940 the Kvísker brothers observed that a small area in the middle of Breidamerkurjökull, situated about 13 km from the edge of the glacier (Fig.i), had become free of ice. In their opinion the ice most likely retreated from this area in the late 1930s (Björnsson, S., 1958), and in 1957 and 1958 they made expeditions to the area which they named Kárasker and found thirty-three species of vascular plants growing scattered in the area (Björnsson, H., 1958). Kárasker, which is composed of basalt and rhyolite (Einarsson, Th., personal communication) and almost completely covered by basalt-rhyolite moraines (Fig.2), is a mountain slope, sloping from 50 to 200 to the east-north-east.In early September 1965 it was about 500 m broad and 1,200 m long, the altitude being from 580 to 760 m. During an expedition which the present author made to Esjufjöll and Kárasker in late July 1961, accompanied by Mr. Hálfdán Björnsson from Kvísker, it was observed that a n e w small area, situated about 1 km west of Kárasker but at approximatelythe same distance from the edge of Breidamerkurjökull, was
247
E.Einarsson
FIG.1. Map of Breidamerkurjökull.
just becoming free of ice, the gravelly surface of the area still being so wet and soft that it was impossible to step on it without sinking to the ankle. This area which the present author has named Braedrasker, i.e. word the sker of the brothers (from Kvísker)-the “sker” meaning a relatively small flat gravelly or rocky area surrounded by ice or water-is a mountain slope (Fig.3) mostly composed of basalt and sloping from 100 to 200 to the south-east.In 1961 the area was completely without vegetation. In early September 1965 it was about 100 m broad and 150 m long, the altitude being from 680 to 740 m.
248
THE VEGETATION OF BRAEDRASKER In August 1963 the present author,guided by Mr. Hálfdán Björnsson, investigated Braedrasker. It had then been colonized by a single species of vascular plants, Trisetum spicatum, and three species of mosses Pohlia wahlenbergii, Rhacomitrium canescens and Ceratodon purpureus. Only one very small specimen of Trisetum
and some very few specimens of the mosses were found in the gravel between the rocks and boulders in the oldest part of the nunatak. During an expedition to Braedrasker in September 1965, where the present author was accompanied by the bryologist Bergthor J’ohannsson and guided by Mr. Hálfdán Björnsson, fifteen species of vascular plants were found scattered in the gravel in the oldest
Plant ecology and succession in s o m e nunataks in the Vatnajökull glacier
FIG.2. Khasker, covered by moraines. The southern part
of Breidamerkurjökull in the background.
Early August 1963.
FIG.3. Braedrasker at a distance of about 200 m. Early September 1965.
249
E. Einarsson
TABLE1. Vegetation discovered in Braedrasker on seven plots, each of approximately 1 square metre. For cover estimation the Domin scale was used. Taxa
Plot number
Discovered J
Arabis alpina 1965 Cerastium alpinum 1965 Cerastium cerastoides 1965 Epilobium lactijorum 1965 Luzula spicata 1965 Mìnuartia rubella 1965 Oxyria digyna 1965 Poa alpina vivip. et non vivip. 1965 Poa jexuosa 1965 Poa glauca 1965 Sagina intermedia 1965 Sagina procumbens 1965 Saxifraga caespitosa 1965 Sedum annuum 1965 Trisetum spicatum 1963 Ceratodon purpureus 1963 Philonotis tomentella 1965 Pohlia wahlenbergii 1963 Rhacomitrium canescens 1963
2
3
4
5
6
+
7
+ +
++
+
+ + ++ + +
++
+
+
central part of the nunatak together with eight or nine species of mosses. No lichens or fungi were observed. Vegetation analyses were carried out in seven plots, which were clearly marked for future studies of plant succession,the results of which are found in Table 1, which includes all the vascular plants and the most prominent moss species. The vegetation cover in all the plots was less than 1 per cent. The most c o m m o n vascular plant was Pou alpina vivip. and, as a whole, grasses were the most prominent plants in Braedrasker which, however, because of the very scattered plants almost hidden between the boulders gives one, at the first look,the impression of being without any vegetation. S o m e of the plant species, present in the area, were not found in any of the seven plots specially marked for future studies of the succession. The most prominent moss species was Philonotis tomentella.
THE VEGETATION OF KARASKER Hálfdán Björnsson (1958) describes the vegetation of of Kárasker as being fairly luxuriant in some sheltered depressions and rivulet beds and dominated by grasses and other vascular plants. On the more exposed gravel flats between the depressions,however, Björnsson reports the vegetation to be very sparse and scattered but mostly dominated by the same species. Thirty-three species of vascular plants were
250
found in Kárasker (Table 2), the most c o m m o n ones were Cerastium cerastoides and Pou alpina. Three species of mosses, not identified, are reported but the mosses were found in few places and were m u c h less prominent than the vascular plants. No lichens of fungi were observed. In a small rivulet,however, some filamentous algae were observed. During the 1961 expedition five additional species of vascular plants and five species of mosses were found (Table 2). No lichens or fungi were observed. T h e vegetation was still dominated by grasses but mosses seemed to be a little more prominent than before. By far the most c o m m o n species were Pou alpina and Cerastium cerastoides but Arabis alpina, Deschampsia alpina, Oxyria digyna, Pou jlexuosa and Trisetum spicatum were also common. In 1963 three additional species of vascular plants were observed in Kárasker. (Table 2.) T h e mosses were obviously more prominent than in 1961. One fungus, Russula alpina, was observed. On flat rocks in the central part of the nunatak some thalli of a crustaceous lichen were found together with Rhaocomitrium canescens (Fig.4)the biggest of them being 5-8mm in diameter but without any apothecia and as a whole so immature that it was not possible to identify them, although they all seemed to belong to the same species. S o m e analyaes of the vegetation in the most sheltered places were made, and the results are found in Table 2, all the species of vascular plants being included but only the most prominent moss species. Pou alpina was the most c o m m o n plant in the nunatak and was found in all the quadrats where analyses were made and Cerastium cerastoides, Arabis alpinu, Deschampsia alpina and P o a jlexuosa were found in most of the quadrats. On the more exposed gravel flats, the vegetation was m u c h more sparse and scattered, but also dominated by Pou alpina, especially in the central and the southern part of the nunatak, other c o m m o n species being Pou jZexuosa, Saxifraga caespitosa, Saxifraga oppositfolia, Trisetum spicutum
and Rhacomitsium canescens. On few gravel flats in the northern part even Arabis alpina was dominating (Fig.5). Most of the plant specimens in Kárasker were observed to be very healthy looking and seemed to be doing well (Fig.6). S o m e soil samples were collected for analysis (Table 4). In 1965 two additional species of vascular plants were observed in Kárasker,(Table 2) and some species of mosses. Vegetation analyses were also made but as the quadrats where vegetation analyses were carried out in 1963 had not been marked it turned out to be impossible to find them again with the single exception of quadrat number 1. Analyses were therefore made in n e w quadrats, which were, however, clearly marked to m a k e possible future studies of plant succession, and therefore the results are somewhat different from the results of 1963. Quadrat number II is the same as quadrat number 1 and here it can be
TABLE2. Vegetation in s o m e of the
m o s t sheltered places of Kárasker (quadrats of 1 square metre). F o r cover estimation the D o m i n scale w a s used. Figures in parentheses in each quadrat refer to the percentage of vegetation cover
Taxa
Ara bis alpina Armeria maritima Cardaminopis petraea Cerastium alpinum Cerastium arcticum Cerastium cerastoides Deschampsia alpina Draba norvegica Epilobium anogallidifol. Epilobium lacti$orum Festuca rubra var. mutica Festuca vivipara Gallum normanii Gnaphalium Supinum Luzula spicata Minuartia rubella Oxyria digyna Phleum commutatum Poa alpina Poa jlexuosa Poa glauca Ranunculus glacialis Rhodiola rosea Sagina intermedia Sagina procumbens Sagina saginoides Salix herbacea Salix callicarpaea Saxifraga caespitosa Saxifraga cernua Saxqraga nivalis Saxijraga oppositifolia Saxifraga rivularis Sedum annuum Sedum villosum Silena acaulis Silene martitima , Taraxacum sp. Thymus arcticus Trisetum spicatum Veronica alpina Veronica fruticans Viscaria alpina Brnchythecium rivulare Bryum afine Ceratodon purpureus Hygrohypnum sp. Philonotis tomentella Pohlia annotina Pohlia proligera Pohlia wahlenbergii Polytrichum alpinum Polytrichum juniperinum Rhacomitrium canescens
1958 1958 1958 1958 1961 1958 1958 1958 1958 1958 1958 1958 1958 1961 1958 1961 1958 1958 1958 1958 1958 1958 1958 1958 1963 1965 1961 1963 1958 1958 1958 1958 1958 1965 1958 1958 1963 1958 1958 1958 1958 1958 1961 1965 1961 1965 1965 1961 1961 1961 1961 1961 1961 1961
+
+ +
+ 1
3
+
5 1
2
+
1
1
+ + +
1
+
5
+
1 4 3 4
1 + + 1
+
4 2
8
3
+
2
1 2 4
3 2
7
1
+
+
2 1
2
+
+ +
4
2 4
+
3
4 2
1
4
2 2 3
1
2
2
1
2
i
+ 6
3
1
+
+
2
3
1
1
+ +
+
3
+
1
t
+ '
1
+
1
2
2
2
1
3
+
1
1
1
1
+
+ t 1
8
1
4
3
2
7
2 4
3 2
Mosses
Russula alpina
2
3
4
3
2
1
2
+
1 3
1
+
1963
1. Taxa were discovered in 1958 by the Kvfskerbrothers;in 1961, 1963 and 1965 by the present author. 2. Quadrats II and III were in the same depression;IV,V and VI in another, and IX,X and XI in a third.
25 1
FIG.4. From Kárasker. A dike in the middle of the nunatak with small clusters of Rhacomitrium canescens and an unidentified lichen. Early August 1963.
FIG.5. From Kárasker.Arabis alpina scattered among the boulders and completely dominating the sparse vegetation. Some grasses (Poa alpina, Poa flexuosa and Trisetum spicatum) can also be seen on the picture. Early August 1963.
252
FIG.6.From Kárasker.Oxyria digyna;the height of this vigorous specimen is about 40 cm. Early August 1963.
seen that s o m e changes have occurred. Cerastium cerastoides and Epilo bium lactiflorum were observed in the quadrat in 1965 but not in 1963. A s a whole the vegetation of the sheltered depressions has not changed m u c h although mosses are still becoming m o r e prominent. T h e s a m e species of vascular plants are the dominating ones, i.e. P o a alpina and other grasses, but Cerastium cerastoides and Arabis alpina w-ere not as common-at least not relatively-as before. On the gravel flats the vegetation w a s s o m e w h a t m o r e prominent than before the dominating species being Poa alpina, Deschampsia alpina, P o a jleauosa, P o a glauca, Oxyria digyna, Saxifraga caespitosa, Trisetum spicatum and Phleum commutatum. N o w h e r e on the gravel flats could a quadrat of 1 square metre with vegetation cover m o r e than 25-35per cent be observed, but in s o m e of the depressions the cover w a s almost 100 per cent in certain places. All the plant taxa found in Braedrasker a n d Kárasker are also found in Esjufjöll and are c o m m o n in South-east Iceland.
T H E VEGETATION O F ESJUFJÖLL
As expected the flora of the Esjufjoll area is m u c h richer than the flora of Kárasker and Braedrasker. T h e Kvísker brothers (Björnsson, H.,1951) report seventy-nine species of vascular plants from Esjufjöll. In 1952 one additional species w a s found (Sv. Sch.
Thorsteinsson, personal communication) a n d during the 1961 expedition to the Esjufjöll nunataks sixteen m o r e species of vascular plants were found there by the present author (Table 3). T h u s ninety-six species of vascular plants, three taxa of Hieracia and one taxon of T a r a x a c u m included, are n o w k n o w n from the Esjufjöll nunataks, which is a surprisingly high n u m b e r compared with the Icelandic flora as a whole which contains only about 440 vascular plants, Hieracia a n d Taraxaca excluded, a n d a n u m b e r quite comparable with a n y other Icelandic mountain area, not surrounded by ice, of comparable size a n d elevation. As also might b e expected the vegetation of Esjufjöll is not as uniform as in Kárasker, not to mention Braedrasker. In most parts of Esjufjöll, i.e. o n exposed gravel flats with sparse and thin soil, the vegetation is physiognomically alike the gravel-flat vegetation in Kárasker but mostly dominated by other species such as Cardaminopsis petraea, Silene acaulis a n d Silene maritima, although the species from Kárasker are also found here. In this aspect, therefore, the gravel-flat vegetation of Esjufjöll resembles the gravelflat vegetation of other Icelandic mountains m o r e than that of Kárasker. It is therefore obvious that the p r o m inent grasses of the Kárasker gravel-flat vegetation are pioneer species o n moraines which d o not play the s a m e important role in the established vegetation. There is a small bog in Esjufjöll completely dominated by the mosses Philonotis fontana, Philonotis
253
E.Einarsson
TABLE 3. Vascular plants in Esjufjöll Tana
Alchemilla alpina Alchemilla glomerulans Alchemilla vestita Arabis alpina Archangelica o$cinalis Arenaria norvegica Armeria maritima Bartsia alpina Botrychium lunaria Calamagrostis neglecta Cardaminopsis petraea Carex atrata Carex bigelowii Carex capillaris ssp. pors. Carex lachenalii Carex maritima Carex rujìna Cerastium alpinum Cerastium arcticum Cerastium cerastoides Chamaenerion latifolium Cystopteris fragilis Deschampsia alpina Draba norvegica Empetrum hermaphroditum Epilo bium anagallidifolium Epilo bium hornemanni Epilobium lactijorum Equisetum arvense Equisetum variegatum Erigeron boreale Erigeron unijorus
Discovered'
1952 1950 1961 1950* 1950 1961 1950 1950 1950 1961 1950 1961 1950 1961 1950 1961 1950 1950* 1961* 1950 1950* 1950 1950 1950* 1950 1950 1950 1950 1950 1950 1950 1950
Discovered'
Taxa
Euphrasia sp. Festuca rubra var. mutica. Festuca vivipara Galium normanii Gentiana nivalis Gentianella aurea Gnaphalium Supinum Harimanella hypnoides Hieracium alpinum Hieracium microdon Hieracium percome Juncus zrijìdus Juncus triglumis Juniperus communis Kobresia myosuroides Loiseleuria procumbens Luzula arcuata Luzula spicata Minuartia bi$ora Minuartia rubella Oxyria digyna Parnassia palustris Phleum commutatum Pinguicula vulgaris Poa alpina Poa jlexuosa Poa glauca Polygonum viviparum Potentilla crantzìi Ranunculus acris Ranunculus glacialis Ranunculus pygmaeus
Tana
1961 Rhodiola rosea 1950 Rumex acetosa 1950* Sagina intermedia 1950 Sagina procumbens 1950 Sagina saginoides 1961 Salix callicarpaea 1950 Salk herbacea 1950* Salix lanata 1950 Saxifraga caespitosa Saxifraga cernua 1961 Saxifraga hypnoides 1961 Saxifraga nivalis 1950 Sazifraga oppositifolia 1961 Saxifraga rioularis 1950 Saxifraga stellaris 1950 Saxifraga tenuis 1950 1961* Sedum acre 1950* Sedum annuum Sedum villosum 1950 Sibbaldia procumbens 1961' 1950* Silene acaulis Silene maritima 1950 Taraxacum sp. 1950 Thalictrum alpinum 1950 1950* 'Thymus arcticus 1950* Tqfieldiapusilla 1950* Trisetum spicatum 1950* Vaccinium uliginosum 1950 Veronica alpina 1950 Veronica fruticans 1950* Viscaria alpina Woodsia ilvensis 1961
Discovered'
1950 1950 1950 1961 1950 1950 1950* 1950 1950* 1950 1950 1950* 1950* 1950 1950 1950* 1950* 1950 1950 1950 1950* 1950 1958 1950 1950 1950 1950 1950 1950* 1950 1950 1950
1. Taxa discovered in 1950 were by the Kvísker brothers, Li 1952 by Thorsteinsson,and in 1961 by the present author. * Signifies those taxa discovered above 1,150 m o n the peak Steinthorsfell,
tomentella and Drepanocladus exannulatus but in the continuous moss carpet vascular plants like Saxifraga rivularis, Saxifraga stellaris, Cerastium cerastoides, Deschampsia alpina, Phleum commutatum, Pou alpina, Festuca rubra var. mutica, Calamagrostis neglecta, Carex lachenalii, Carex ruJina and some other species
are found. In sheltered places in the Esjufjöll nunataks,mainly on the South-western slopes of the mountain ridge named Skálabjörg which have a very favourable exposure, a very luxuriant vegetation is found which has almost made the nunataks famous in Iceland. Salix callicarpea shrubs with Rhodiola rosea, Bartsia alpina, Ranunculus acris (40-50c m high), Alchemilla glomerulans, Chamaenerion latifolium, Taraxacum sp., Hieracium sp., P o a alpina, Polygonum viviparum, Angelica archangelica (90-100c m high) and some other species form this interesting vegetation which also has some mosses, the most c o m m o n being Rhacomitrium canescens and Drepanocladus uncinatus. This
254
vegetation is m u c h m o r e luxuriant than the depression vegetation in Kárasker is ever likely to be because of the more favourable exposure and it is dominated by dicotyledones and not grasses. This vegetation is also as luxuriant as the most luxuriant comparable vegetation at this altitude in other Icelandic mountains. But it has one good character, Chamaenerion latifolium which is here, the only area in the whole of Iceland, to the best of the author's knowledge, growing abundantly in vegetation quite different from the gravelly river-beds and banks which are its usual habitat in Iceland. In Esjufjöll Chamaenerion thus is found in various vegetation types, just as in Greenland. In Table 3 the species of vascular plants found above the altitude of 1,150 m on the peak Steinthórsfell in Skálabjörg are characterized by a *, there being twenty-four,a higher number than found above that altitudein any other single Icelandicmountain. A m o n g these species are Chamaenericn latifolium, S e d u m acre
Plant ecology and successionin some nunataks in the Vatnajökull glacier
and Veronica fruticans which have here their highest k n o w n locality in Iceland.
SOIL CONDITIONS Five soil samples were collected in Kárasker during the 1963 expedition. As mentioned before Kárasker is almost entirely covered by moraines and in most parts of the nunatak the soil is therefore very sparse, mostly coarse soil with some clay found between boulders. The soil is thickest in the depressionswhere there has been a considerable vegetation cover for years and the vegetation cover is gradually becoming continuous in some places. All five soil (Table 4) samples were collected just under the vegetation cover in the quadrats which have the same numbers (Table 2) as the samples. The author is aware of the fact that the small number of soil samples and the few soil factors analysed cannot give any satisfactory knowledge of the soil conditions in the nunatak. They might, however, give some information ahout this moraine soil. The p H was measured by soil-water mixture test and the carbon by wet oxydation. The p H in four out of five samples is nearly the same, that of the remaining sample, number VII, being a little lower, which might be due to higher clay fraction in the depression where the sample was collected or to the fact that the moiaines in Kárasker are not homogeneous as they are of a basalt-rhyolite mixture. The carbon, i.e., the organic matter, content of the samples is very small as might have been expected, one of them, however, number IV,containing nearly three times as m u c h carbon as the sample with the lowest carbon content. This differencemight be due to more plant roots being mixed with the soil in sample IV than in the other samples.
TABLE4.Analyses of five soil samples collected in Kárasker during 1963. Sample no.'
PH
II IV
6.4 6.5 5.3 6.1 6.4
VI1 VI11
X
Carbon
(%) 0.36 0.60 0.28
0.24 0.22
1. T h e sample number relates to the quadrat number given in Table 2.
CONCLUSION The plant colonization of areas in Breidamerkurjökull which have recently become free of ice seems to start within two years after the melting of the ice. This seems to be about a year later than observed at Skaftafell, South-east Iceland (Persson, 1964) and in northern Sweden (Stork, 1963). The nunataks of Breidamerkurjökull, however, are isolated by the surrounding ice from areas with vegetation, whereas the areas in Skaftafell and in northern Sweden are moraines in front of retreating glaciers. The distance from Kárasker to the mountains east of Breidamerkurjökull is 9 k m and the distance to the mountains west of the glacier is 8 km. The distance between Kárasker and Esjufjöll is about 5 k m , the distance to the nunatak Mávabyggdir, where some few species of vascular plants have been found, is 3.5 km. As mentioned before the distance between Kárasker and Braedrasker is only about 1 km. Therefore there seem to be m a n y opportunitiesfor plant dispersalfrom these areas to the recent nunataks. The present author is of the opinion that the diaspores are probably windborne or blown along the surface of the ice to the new nunataks. Even whole plant specimens, mostly grasses, with roots and leaves have been found lying on the surface of the ice of Breidamerkurjökull far from the nunataks and the edges of the glacier. Grass species seem to be the pioneer plants in such areas accompanied by some few species of mosses. Lichens and fungi seem to invade at a later stage. Exposure to wind, snow cover and radiation of the sun seems to influence the establishment of vegetation cover very much. In the established vegetation the pioneer vascular plants do not play any important role.
ACKNOWLEDGEMENT Appreciative acknowledgement is made to the NATO Research Grants Programme, these investigations being a part of a more extensive research programme supported by NATO Research Grant No. 129, to Mr. Bergthór Jóhannsson,Museum of Natural History, Reykjavík, for identification of mosses, to Mr. Kjeld Hansen, Botanical Institute, Royal Agricultural and Veterinary College, Copenhagen, for investigating lichens samples, to Dr.Bjarni Helgason, Agricultural Research Institute, Reykjavík, for chemical soil analysis and last but not least to Mr. HálfdánBjornsson from Kvísker for excellent guidance to the nunataks and skilful assistance during the field-work.
255
E.Einarsson
Résumé Phyto-écologie et évolution du tapis végétal dans certains nunataks du glacier de Vatnajökull, dans le sud-est de l’Islande (E.Einarsson) Les végétations respectives de trois régions de nunataks d’âge différent ont été étudiées et comparées. Les nunataks les plus anciens et les plus importants sont ceux de l’Esjufjö11, dont l’altitude v a n e entre 700 et 1600 mètres, et qui émergent probablement des glaces qui les environnent depuis la fin de la période glaciaire. Leur fiore se compose de 92 espèces de plantes vasculaires (on en compte 440 espèces dans toute l’Islande) et d’une végétation de prairies herbeuses et de formations buissonnantes à saules, d’une surprenante luxuriance, a u x endroits où l’exposition et la couverture neigeuse favorisent sa croissance. L a seconde région, celle de Kárasker, se réduit à u n simple versant de montagne pierreux, d’une altitude de 550 à 760 mètres, et incliné de 10 à20 degrés vers l’est, qui s’est dégagé de la glace il y a une trentaine d’années. S a flore comprend 42 espèces de plantes
vasculaires dont 5 ont colonisé le nunatak depuis 1963. L a végétation se compose surtout de graminées poussant çà et là dans les pierres. Toutefois, dans certaines dépressions, qui sont probablement d’anciens lits de cours d’eau, la végétation est plus serrée, et là où l’exposition est favorable elle peut recouvrir presque entièrement le sol. L e troisième nunatak, qui est récent et n’a pas encore de n o m , est également un versant montagneux pierreux, d’une altitude de 650 à 700 mètres, exposé a u sud-est. Il a émergé des glaces a u début de l’été 1961 et était dépourvu de toute végétation à la fin de juillet de la m ê m e année. En août 1963, o n y a découvert une espèce de graminée et trois espèces de mousses; en septembre 1965 o n a pu y dénombrer 20 espèces de plantes vasculaires dispersées parmi les pierres et quelques nouvelles espèces de mousses. Plusieurs parcelles-témoins ont été délimitées dans ce nouveau nunatak ainsi que dans celui de Kárasker ; elles permettront de suivre l’évolution du tapis végétal.
Discussion E. SCHENE. In Iceland I saw rising palsas and collapsing palsas. You showed a picture of a bog with moss. Do you also have string bogs in Iceland?
E.EINARSSON.No, so far as I know there are no string bogs in Iceland.
V. OKKO.Is there any thermal activity in the nunataks you have described?
E.EINARSSON.No,none at all. H. LUTHER.Y o u stated that the diaspores probably were wind-borne or blown along the ice surface. Have you made
any observations regarding birds as potential diaspore carriers? E. EINARSSON. No, I have not made any such observations. In the Esjufjöll nunataks four bird species have been found nesting, namely ptarmigan, snow bunting wheatear, and arctic stua, and some more species have been seen there. The new nunataks are probably also occasionally visited by birds which might carry diaspores with them. But I do not think they do so. The fact that I have even found whole plant specimens with roots and leaves lying on the ice of Breidamerkurjökull far from “land” has helped m e to form the opinion that the plants are dispersed to these new nunataks mostly along the surface of the ice.
Bibliography / Bibliographie F. 1951. Esjufjöll og Mávabyggdir. Náttúrufraedingurinn,vol. 21, p. 99-108. B J ~ R N S S O N ,H. 1951. Gródur og diralíf i Esjufjöllum Náttúrufiaedingurinn, vol. 21, p. 109-112. . 1958. Gródur og d+alíf í Káraskeri. Jókull, vol. 8, p. 19-20. BJ~RNSSON, S. 1958. Könnunarferd i Kárasker. Jókull, vol. 8, p. 15-17. EYTHÓRSSON, J. 1963. Variation of Iceland glaciers 19311960. J6kull, vol. 3, p. 31-33. BJäRNSSON,
--
256
OLAFSEN, E. ; POVELSEN, B. 1772. Reise igennem Island, etc. Soroe. THÓRARINSSON, S. 1943. Oscillations of the Iceland glaciers in the last 250 years. Vatnajökull, chap. XI, p. 1-54. (Geografiska Annaler, XXV.) PERSSON, A. 1964. The vegetation at the margin of the receding glacier Skaftafellsjökull, South Eastern Iceland. Botaniska Notiser, vol. 117, p. 323-354. STORK, A. 1963. Plant immigration in front of retreating glaciers with examples from the Kebnekajse area, Northern Sweden, p. 1-22.(Geogra6ska Annaler, vol. XLV (1963).)
Écologie des plus importantes espèces de la faune subarctique A.N. Formozov
LES CONDITIONS D ’ E X I S T E N C E DES ANIMAUX DANS LES ReGIONS SUBARCTIQUES Les principaux traits de la nature de la zone des toundras déterminant les conditions de vie des anim a u x sont très connus : ce sont, avant tout, u n hiver long et dur, un soleil médiocre, une couche de neige très épaisse empêchant de se procurer la nourriture. Ce sont ensuite un été court et froid, une congélation d u sol en hiver, son humectation au printemps et en automne ne permettant pas de faire de bons terriers et de les bien utiliser. Un paysage ouvert, dépourvu d’abris sûrs prédomine dans les toundras. Ce n’est que dans la zone du sud des toundras et des toundras coupées de forêts qu’on peut s’abriter sous les buissons et les arbres contre la pluie et le vent. En hiver, les conglomérations de glaces, les amoncellements de neige, les saillies de roches peuvent servir d’asile. L a productivité de la végétation se composant presque uniquement de plantes pérenelles,de lichens et de mousses est assez basse. Ces défauts sont en quelque mesure compensés par une haute qualité nutritive de certaines espèces riches en vitamines C et A , et conservant pendant tout l’hiver leur verdure. Les espèces largement répandues dans les toundras sont : Rubus
v.
charnuernorus, vaccinium myrtillus, diginosuni, V. vitis idaea, Oxycoccus spp., Arctostaphylos uva ursi, Arctous alpina, Empetrum nigrum. U n e partie de
ces plantes rapportent toujours une récolte qui assure aux animaux la nourriture nécessaire. Certaines plantes fruitières se conservent très bien en hiver et, l’année suivante,peuvent être utilisées par les oiseaux avant la nouvelle récolte. C’est un fourrage unive’rsel qui est utilisé par les mammifères de tout genre (campagnols,isatis, hetmines, rennes, élans) et par les oiseaux de plusieurs ordres (lagopèdes,oies, canards, goélands, labbes, courlis, faucons,buses pattues).
Les vents permanents soumant à grande vitesse jouent également un rôle considérable dans la vie des animaux. En hiver, c’est à eux qu’on doit la rudesse du climat (l’indice en unités conventionnelles est calculé d’après la formule de Bodman). Dans les régions subarctiques d’Eurasie,l’hiver est surtout rigoureux dans la Nouvelle-Zemble,à Yamal, à Taïmyr et non pas dans les toundras de Yakoutie, situées à proximité du pôle de froid. Lors des chasseneige, m ê m e les espèces très adaptées au climat, telles que renards polaires, lièvres variables, rennes, se cachent et suspendent toute activité. Certaines particularités de leur métabolisme et la capacité d’accumuler de grandes réserves énergétiques sont propres à certaines espèces et races géographiques des animaux hivernant dans les toundras (harfangs des neiges, rennes, renards polaires, lièvres variables, d’après Schwartz). Ces réserves sont indispensables lorsque les animaux restent sans nourriture par suite de conditions climatiques très dures. Les vents d’été, également permanents et frais parce qu’ils soufflent du côté de la mer, entravent, à leur tour, l’existence des animaux. Par exemple, les faucons pèlerins essayent de protéger leurs petits et la femelle en faisant leurs nids sur des versants à l’abri du vent. Mais les représentants très caractéristiques des toundras, tels que harfangs des neiges et buses pattues, sont plus résistants que les faucons et nichent dans des endroits découverts, exposés au vent. Les lagopèdes muets se nourrissent en hiver sur des versants et des élévations dépourvus de neige. A la recherche de nourriture, des vols de lagopèdes émigrent des montagnes vers les vallées et des vallées vers les montagnes. Parfois, ils émigrent vers le nord, sur le littoral, et se nourrissent sous les sabots des rennes, déterrant la nourriture sous la neige. L’insuffisance de la lumière diurne est également une condition défavorable pour l’existence de plusieurs espèces 257
19
A.N.Formozov
d’animaux. C o m m e on le sait, le mergule nain et le Cepphus grylle mandtii hivernent dans les secteurs de l’océan Arctique, libres de glaces. On pourrait croire que ces animaux éprouvent beaucoup de difficultés lorsqu’ils cherchent leur nourriture eous l’eau, car la clarté est toujours insuffisante en hiver. Cependant, il est évident que la quantité de nourriture accessible est beaucoup plus importante que l’intensité d’éclairage. Un été court,la lumière diurne pendant plusieurs jours constituent des conditions particulières pour l’existence des animaux. Les espèces qui sont actives dans la journée utilisent au m a x i m u m toute la lumière du jour. Le développement des petits est plus rapide que dans les régions de latitude moyenne. Un éclairage prolongé permet d’avoir une “journée de travail plus longue”, d’obtenir plus de nourriture et de nourrir mieux les petits. Tous les oiseaux (Passeres,espèce de moineau) ont des interruptions dans leur activité, égales à 4 ou 5 heures et qui tombent sur la partie la plus froide de la journée,c’est-à-direde 23 heures à 4 heures du matin. L a période d’activité est égale à 19 ou 20 heures, tandis qu’au sud de la zone forestière, pour les oiseaux des mêmes espèces cette période est égale à 17 heures. Grâce à cet état de choses, le développement des petits oiseaux dans les régions subarctiques se fait à un rythme accéléré et ils abandonnent leurs nids plus tôt que cela ne se fait dans les régions situées plus au sud. Par exemple, les jeunes faucons,à Yamal, abandonnent leur nid dix jours plus tôt que les mêmes oiseaux en Europe centrale. Les petits oiseaux (Passeres) nourrissent leurs petits pendant 10 à 12 jours contre 14 o u 16 chez les mêmes oiseaux aux latitudes moyennes. D’après Kouznetzov,l’entomofaunearctique compte plus de 3 O28 espèces. Les oiseaux (Passeres) ont, en été, une nourriture abondante,grâce à l’existence d’un grand nombre d’insectes dans les toundras (Chironomidae, Tipulidae, Culicidae et autres). Les larves des insectes et les imagines jouent un rôle considérable dans l’affouragementdes courlis,qui sont extrêmement nombreux dans la toundra. Par exemple,dans la région d’Andyr (nord-estde la Sibérie) il y a plus de 30 espèces de courlis et la densité de leurs populations est parfois très grande. C o m m e on peut le voir d’après les recherches de Tchernov de 1961, 1964 et 1965, les liaisons trophiques des courlis embrassent la faune des invertébrés nichant dans la couche active du sol et parfois dans les bassins d’eau douce et dans la mer. L a proximité de la congélation est à la base du fait que Eisenia nordenskioldi et les larves du sol se trouvent dans la couche superficielle d’où elles sont faciles à dénicher. D’après les données de Tchernov, les larves et les imagines de tipulidae et surtout Tipula carinifrons, Prionocera lapponica, P. serrirostris, etc. joient un rôle important dans la nourriture des courlis. L e nombre moyen de ces larves atteint 50 et par endroits dépasse 200 par mètre carré. Les courlis nichant dans les toundras marécageuses s7en nour-
258
rissent. Les larves de tipulidae sont abondantes au début de l’été,mais vers la fin de juin elles disparaissent et c’est le tour des imagos. Les grandes femelles des espèces septentrionales ne volent pas, elles rampent sur la surface du sol. A la mi-juillet,elles disparaissent à leur tour et bientôt les courlis s’envolent vers de lointains pays d’hivernage. Les brusques contrastes entre l’été et l’hiver sont à la base des grandes différences entre l’aspect d’été et l’aspect d’hiver de la faune des toundras. Dans les régions où nichent 60 ou ‘i0 espèces d’oiseaux en hiver, il n’en reste qu’une G U deux (lagopèdes muets et, rarement, harfangs des neiges). Mais elles ne sont plus sédentaires et mènent une vie nomade. L a migration des rennes, des renards polaires et des loups a lieu deux fois par an. De lointaines migrations saisonnières présentent un caractère adaptatif propre à la plupart des oiseaux et des mammifères des régions subarctiques. Ce ne sont que les lemmings, consommateurs massifs des céréales, des Eriophorurns, des lichens et des mousses, qui émigrent à courtes distances et, en hiver, se déplacent vers les vallons où il y a des amoncellements de neige qui les protègent contre les intempéries. Les lemmings se reproduisent en été et en hiver et, certaines années, ils remplissent les toundras. Les variations dans le nombre des lemmings suivent des cycles de 4 ans. Le caractère des oscillations reste invariable au cows de plusieurs décennies, ce qui est très bien observé pour L e m m u s lemmus et certains autres microtinae du nord de la Scandinavie. Certains auteurs estiment que c’est à la structure simple des écosystèmes subarctiques qu’on doit la régularité des cycles chez quelques espèces dominantes (lemmings, renards polaires, harfangs des neiges : cycle de 4 ans ; les cycles de 10 ans Tont propres aux lagopèdes muets, aux lagopèdes des saules et aux lièvres variables). Cette idée n’est juste qu’en partie. Par exemple, les cycles de 4 ans sont caractéristiques pour les populations de Microtus arvalis, dans la zone moyenne de la partie européenne de l’URSS où la structure des écosystèmes est beaucoup plus compliquée que dans la toundra. Mais nous n’apons pas la possibilité de discuter en détail ce problème complexe. Faisons remarquer qu’il faut coneidérer les lemmings c o m m e les plus importants animaux subarctiques parce qu’ils exercent une grande influence sur la formation de la couche végétale et du microrelief et sur la vie de plusieurs oiseaux et mammifères subarctiques. Les lagopèdes des saules et les lagapèdes muets sont les représentants caractéristiques de l’ornitofaune du Nord. Des guillemots de Brünnich, des guillemots à miroir blanc et d’autres peuplant en été les “bird cliffs” des côtes boréales forment des colonies de nidification vraiment grandioses. Près de 2 millions de couples d’Uria lomvia nichent dans la NouvelleZemble et ils sont presque aussi nombreux au Groenland où, seulement à la “gigantic rookery cf Cape
lhologie des plus importantes espèces de la faune subarctique
Shackleton”, il y a près d’un million de couples. D e s centaines de petites colonies comprenant des dizaines et des centaines de milliers d’oiseaux de 5 à 8 espèces sont dispersées sur les côtes rocheuses. I1 y a des pays où l’on utilise les œufs, la chair et la peau de ces oiseaux. Telle est l’une des possibilités -peu variées --d’utiliser d’immenses ressoiirces en petits poissons et en invertébrésde mer dont se nourrissent les oiseaux, mais qui ne présentent pas beaucoup d’intéret économique. Les troupes d’animaux de mer, des cétacés et des pinnipèdes, des poissons de mer et de rivière, voilà les ressources biologiques des régions subarctiques qui méritent l’attention des participants à ce colloque. L’auteur du présent exposé se trouvait devant une táiche difficile : ou bien donner un bref aperçu sur l’écologie des plus importants animaux subarctiques, ou bien analyser quelques espèces caractéristiques.En choisissant la deuxième méthode, l’auteur a peut-être commis une erreur ;il espère cependant que les données soumises à votre attention pourront servir de base à une large et intéressante discussion.
LES RENARDS POLAIRES (ALOPEX LAGOPUS) Lc renard polaire est l’animal le plus caractéristique
des régions subarctiques. Son image pourrait servir d’emblème à cette zone géographique. L e renard polaire est largement répandu dans les toundras de l’Amérique du Nord, de l’Europe, dans les îles de l’océan Arctique ainsi qu’en Islande, dans les îles de Komandor, de Pribylov, dans les îles Aléoutiennes et autres. L a frontière sud de son aire de répartition pendant la période de reproduction et durant sa vie sédentaire se situe sur les continents, près des forêts, et, dans les toundras montagneuses,elle s’étend encore plus loin au sud. L a frontière nord de l’aire de répartition au cours de la période de vie sédentaire est située dans les régions arctiques sur les côtes du Groenland, dans l’île du nord de la Nouvelle-Zemble.En hiver, lorsque les renards mènent une vie nomade, on peut les vcir sur les glaces du bassin polaire (870 47‘ de latitude nord) à 1 025 km de tout continent (Routilevski, Ouspenski, 1957). Sur une vaste aire de dispersion dans la région polaire, il existe plusieurs races géographiques de renards polaires dont une partie est considérée par certains zoologues c o m m e espèces indépendantes (Alopex 1. beringensis, A. 1. pribilofensis, A. 1. halensis). Les animaux qui prédominent sont ceux dont le pelage d’été se change en fourrure d’hiver à longs poils, blanche, touffue et épaisse (le renard blanc). C’est la phase blanche du renard polaire ordinaire, qui diffère de la phase de couleur du renard bleu. L e renard blanc niche dans les toundras continentales. En Sibérie,sur deux mille ou trois mille renards blancs, il n’y a plus qu’un ou deux renards bleus.
Sur la majeure partie du nord du Canada, les renards de la chasse. Cependant, bleus ne représentent que dans plusieurs îles qui ne sont pas trop éloignées des continents, les renards bleus se rencontrent souvent et prédominent parfois. Par exemple, dans l’île de Kildin, non loin de la presqu’île de Kola, en 1927 il y avait près de 30 ou 50% de renards bleus. Dans les deux îles de K o m a n dor, le renard bleu prédomine nettement. Dans l’île de Medny, avant les travaux de sélection, le renard blanc représentait 20% ; aujourd’hui, il y a 100°/o de renards bleus. Dans l’île de Béring,parmi une masse de renards bleus, certaines années on peut rencontrer quelques renardsblancs. I1 est possible qu’ils atteignent cette île sur des glaces arrivant du nord en hiver. F.W.Braestrup,en 1941, étudia les statistiques de la chasse aux bêtes à fourrure dans l’ouest du Groenland, de 1793 à 1939,et utilisa d’autres données. D’après ces données, dans la population de la côte ouest du Groenland,il y a plus de 50% de renards bleus. L’auteur propose de les considérer c o m m e des populations géographiques spécifiques intitulées les “renards côtiers” (coast foxes) et les “renards de lemmings” (lemming foxes), c’est-à-dire renards des toundras ou renards continentaux. Ils ont des traits caractéristiques dans leur mode de vie et dans le nombre de représentantsde diverses phases de couleur. En réalité, il existe beaucoup plus de populations géographiques d’isatis qui diffèrent les unes des autres par les traits principaux de leur biologie et de leur structure génétique. L e trait c o m m u n propre aux renards côtiers et à ceux des petites îles consiste en liaisons trophiques avec les animaux de mer. L a plupart du temps, ces isatis se nourrissent de crustacés, d’oursins, de poissons, de cadavres d’oiseaux, de phoques, de cétacés ou d’autres animaux rejetés par la mer. En été, ils détruisent les colonies d’oiseaux de mer en atteignant des roches à pic où se trouvent leurs nids ou bien ils cherchent de petits oiseaux qui se sont écrasés au sol, au pied des roches inabordables. L a proie des renards polaires est alors abondante et ils enfouissent une partie des œufs dans la tourbe, près de la couche supérieure de congélation. Dans un tel garde-manger, les œufs se conservent parfois pendant plusieurs années (observations personnelles de l’auteur sur la côte de Mourman). Là où la mer a une grande productivité biologique et ne gèle pas pendant toute l’année, les renards côtiers sont toujours pourvus de nourriture et, par conséquent, sont relativement sédentaires. Leurs migrations, alors, ne sont pas lointaines, et ils n’entrent pas en contact avec les “renards de lemmings” (par exemple, les renards d’Islande). Leur nombre ne varie pas d’une facon brusque et régulière,ce qui est propre aux populations de renards polaires de toundra qui habitent les continents et les grandes îles où nichent les lemmings (Nouvelle-Zemble). Une partie des renards polaires
ly0
259
A. N. Formosov
habitant le littoral ouest du Groenland sortent1parfois sur les glaces et, lors des dérives, se trouvent e m portés par le courant du nord vers le territoire de Baffin Island. Selon Ch. Elton, les grands changements dans le pourcentage de renards bleus enregistrés à certains postes de la Hudson’s B a y Company n’ont lieu qu’aux endroits où arrivent dans la zone des renards bleus les animaux venant du Groenland ouest. L e renard polaire des toundras est un animal omnivore comme,d’ailleurs, tous les représentants de cette espèce. I1 mange volontiers des fruits (Empetrum nigrum, Vaccinium), des insectes,des poissons,des œufs, des oiseaux,petits et grands, des mammifères. I1 attaque parfois de jeunes phoques et des faons. Pendant les mois d’hiver difficiles, le renard polaire parcourt de grandes distances et suit les traces d’autres carnassiers. I1 mange les restes de phoques chassés par les ours blancs, de la neige imprégnée de sang, des restes de peau de renne; il déterre des campagnols enfouis sous la neige par l’hermine, etc. Cependant, la nourriture principale des renards polaires, ce sont des lemmings, dont les plus précieux sont : L e m m u s sibiricus, Dicrostonyr torquatus et Microtus. C o m m e on le sait, la couche de neige des toundras est très dure et solide à cause des vents. Dans les vallons et sur les versants où hivernent les lemmings,la hauteur de la couche de neige dépasse souvent un mètre. Cela rend la chasse des renards polaires aux petits animaux extrêmement difficile. Ils les cherchent et les trouvent grâce à leur flair et leur ouïe. Ils perçoivent le bruit fait par les lemmings lorsque ces derniers grignotent des feuilles sèches ou creusent des passages sous la neige. On trouve souvent des morceaux de feuilles dans les dents d’un lemming dévoré par le renard, ce qui nous fait supposer que le carnassier l’a attrapé au moment où il mangeait. Souvent, le renard polaire pénètre à travers l’épaisseur de neige vers les nids de lemmings. Cette méthode de chasse,lorsque les rongeurs ne sont pas nombreux, ne fournit qu’une proie médiocre, et les renards polaires souffrent de la faim. Lors des années de disette, les estomacs des renards polaires capturés sont vides ou remplis de charogne. Par exemple, d’après les données de L. Tsetsevinski, au cours d’une année où les lemmings manquaient dans la presqu’île d’Yamal, 68,1% des estomacs de renards n’étaient remplis que de charogne, tandis que, quand la chasse aux lemmings était satisfaisante,il n’y avait que 15% de renards polaires qui se nourrissaient de charogne. Dans la zone sud des toundras d’Yamal, les oiseaux peuvent parfois remplacer les rongeurs et servir de nourriture aux renards polaires. AU cours de l’hiver 1942/43,on trouva des oiseaux dans les estomacs de 36,20/,de renards polaires et il est à remarquer que, dans 30% des cas, c’étaient les lagopèdes des saules. Cependant, dans les toundras du Nord, en hiver, il y a très peu d’oiseaux et le renard polaire ne se nourrit que des restes d’oiseaux morts en automne
260
et ce n’est que rarement qu’il réussit à chasser des lagopèdes muets. U n e lacune importante de plusieurs études consacrées à l’écologie des “renards de lemmings” est qu’on a étudié le carnassier sans étudier parallèlement les rongeurs prédominant dans la toundra. Dounaeva et Osmolovskaia surent éviter ce défaut, grâce à quoi elles purent obtenir quelques données importantes dans ce domaine. L a laparatomie des renards polaires capturés en hiver démontre que les jeunes L e m m u s sibiricus constituent la nourriture du carnassier pendant toute la période froide, de novembre jusqu’à avril. On a trouvé souvent dans l’estomac des renards polaires des nichées de lemmings comprenant de 2 à 7 animaux, que les renards auraient pris dans le terrier. Plus rarement, des nichées d’oiseaux tels que Dicrostonyx torquatus furent enregistrées au cours des mois de mars et d’avril. Tsetsevinski trouva pendant l’hiver 1932/33,dans l’estomac d’un isatis, une nichée de Microtus gregalis et, au mois de janvier, une femelle de campagnol à 15 e m bryons. On a donc établi que le campagnol pouvait se reproduire dans les régions subarctiques sous la neige, au cours des plus durs mois d’hiver. C’est là un facteur qui fournit de nouvelles données précieuses pour l’étude des ressources fourragères de la toundra. Dounaieva et Osmolovskaia établirent que l’intensité de reproduction des lemmings sous la neige en hiver et au printemps variait d’une année à l’autre et était une condition importante de l’augmentation du nombre de ces animaux au cours de l’été suivant. L a plus importante période de reproduction des renards polaires se situe en hiver. Le cycle sexuel commence en janvier (période préparatoire), la copulation a lieu au mois de février et mars, la gestation tombe à la fin de l’hiver et se termine à mi-mai. L’apparition des petits coïncide avec le début du printemps. Afin que son organisme soit à la norme, le renard polaire a besoin, au cours de tout l’hiver, d’une nourriture abondante et de bonne qualité. Seuls les lemmings peuvent constituer une telle nourriture. Malgré la diversité apparente de nourriture en été, l’importance des rongeurs dans la vie des nichées de renards polaires est plus grande qu’elle ne l’est dans la vie des animaux adultes en hiver. Les petits des renards polaires (il y en a 15 ou 16, souvent 18, parfois 20 ou 22) ne sont bien nourris que lorsqu’il y a beaucoup de rongeurs dans la toundra. L e lemming sibérien sort pour se nourrir toutes les deux heures. Les abris qu’il utilise en cas de danger ne sont pas sûrs parce qu’ils ne sont pas profonds à cause des eaux souterraines et de la congélation. L e renard attrape facilement ces animaux et les transporte dans les terriers pour nourrir ses petits qui se développent alors bien et, en automne, la toundra se remplit de renards polaires. Par contre, au cours d’un hiver pauvre en rongeurs, les renards polaires ne se reproduisent pas et le nombre de terriers occupés se réduit considérablement (3 ou 15% contre 60 ou 70%
kcologie des plus importantes espèces de la faune subarctique
au cours des années favorables). Dans une partie des terriers,les femelles n’ont pas de petits et les nichées dans d’autres terriers ne comptent pas plus que 2 à 6 bêtes. Si,en été, le nombre des rongeurs se réduit pour une cause ou une autre, il arrive que les renards polaires abandonnent les terriers en y laissant leurs nichées qui périssent ensuite. L a méthode de calcul du nombre des terriers habitables et du volume moyen d’une nichée permet d’obtenir des données sûres pour évaluer le nombre des renards polaires à la prochaine saison de chasse. Cette méthode est souvent employée dans les toundras de l’URSS. I1 en existe cependant une autre : les chasseurs apportent souvent des renards polaires dans les postes, où leur peau est mieux traitée par des spécialistes. Cela permet d’obtenir de très bonnes fourrures et les carcasses des renards peuvent être soumises à un examen biologique. L’analyse du contenu de l’estomac des renards d’hiver est une méthode objective qui permet de connaître l’état de la population de rongeurs et en m ê m e temps de se faire une idée de la nourriture des carnassiers durant la période qui précède celle de la reproduction. Dounaieva et Osmolovskaia, qui avaient élaboré cette méthode, proposèrent de calculer le nombre des jeunes lemmings dans les estomacs des renards polaires capturés en hiver et démontrèrent qu’il était possible de prévoir le nombre des isatis pour la prochaine saison de chasse. Les cycles d’oscillations, de trois ou quatre ans, pour le nombre de lemmings entraînent simultanément des changements semblables dans les populations des renards polaires et des renards des toundras d’Eurasie et d’Amérique. L a monographie bien connue que Ch. Eton a publiée en 1942 contient des données intéressantes à ce sujet. Son auteur démontre par de multiples exemples qu’après une période où le nombre des renards polaires atteignait un maximum, il était de règle qu’une période de “crash” s’ensuive, où leur nombre se trouve extrêmement réduit. Aujourd’hui, il y a tolites possibilités d’étudier plus en détail comment une période optimale est suivie d’une période de réduction des populations de renards polaires. Les jeunes bêtes y composent un pourcentage considérable. Elles ne peuvent pas se fortifier au cours de cette période. On peut voir certaines bêtes ii la fourrure peu développée. Ces animaux, qui sont encore jeunes, se trouvent dans des conditions beaucoup plus difficiles que les bêtes adultes. Les renards polaires des toundras reprennent la vie nomade bientôt après la période de reproduction,ce qui arrive, en règle générale, aux mois d’août et de septembre. L a diminution du nombre de rongeurs, qui est d’ailleurs courante après une reproduction massive, la migration des oiseaux et d’autres facteurs réduisent les ressources fourragères de la toundra et obligent les renards polaires à se déplacer à la recherche de nourriture. L a migration commence et, par conséquent, les
contacts à l’intérieur de la m ê m e espèce se renforcent, la concurrence s’aggrave, ce qui facilite la contamination des bêtes. Lorsque la densité de population des renards polaires est grande, les migrations locales se transforment en migrations lointaines. L e marquage montre que les bêtes d’une m ê m e nichée peuvent se disperser à de grandes distances et dans des diiections différentes. Selon ces données, les mâles sont plus mobiles que les femelles. Certains renards polaires se rendent vers le nord et s’éloignent sur les glaces de la mer. U n e grande partie des renards polaires, au début de l’hiver, se trouvent déjà dans la toundra du Sud, dans la toundra coupée de forêts et dans la zone septentrionalede la taïga. Ils y pénètrent en suivantles rives des fleuves, n’osant pas traverser les grands bassins1. D’autres renards polaires se déplacent derrière les troupeaux de rennes. L e choix de la direction des migrations dépend de celle des vents. Des jeunes renards marqués sur la presqu’île d’Yamal, 55% s’éloignèrent à l’ouest, dans la direction opposée aux vents d’octobre. U n e partie des bêtes migratrices se dispersent dans la sous-zone septentrionale de la taïga où, en hiver, il y a beaucoup de lagopèdes des saules, de lièvres, de rongeurs. Cependant, les années de “maximums”, beaucoup de renards polaires pénètrent au fond de la taïga et certains d’entre eux s’éloignent à des centaines de kilomètres au sud de la toundra. On connaît des cas de capture des renards polaires aux environs de Leningrad, dans la haute Petchora, aux enlirons de Tobolsk, de Ienisseisk, dans les paye en amont des rivières Vilioui et Yana. Les renards polaires émigrent volontiers dans les vallées de grands fleuves où la couche de neige est moins épaisse et moins solide, et, dans l’est de la Sibérie, ils choisissent parfois pour leur voie de migration les crêtes des montagnes. Les peuples habitant la zone septentrionale de la taïga en Sibérie occidentale estiment que si, en automne, il y a une grande migration de harfangs des neiges vers le sud,il y aura des renards polaires. Ce jugement est tout à fait fondé, parce que ces hiboux,tout c o m m e les renards polaires, ne se reproduisent que lorsqu’il y a beaucoup de lemmings. Le manque de nourriture rend la migration de ces deux genres de carnassiers nécessaire. En Amérique, o n connaît des arrivées massives de harfangs des neiges qui se répètent à des intervalles réguliers de 3 ou 4 ans. Les hiboux y sont faciles à remarquer parce qu’ils viennent hiverner dans les régions du Canada et des lhats-Unis dont la population est très dense. A. O. Gross fit en 1927, 1931 et 1935 une analyse intéressante de ces migrations. Ch. Elton utilisa ces données et montra que les barfangs des neiges arrivaient en masse dans les années
1. Les renards polaires franchissent facilement des fleuves de 400 à 100 mètres. de largeur. mais ils périssent souvent loraqu’ils entreprennent le passage de fleuves plus larges.
261
A. N. Formozov
de m a x i m u m des lemmings, dans le nord du continent. Finn Salomonsen, en 1951, décrivit le m ê m e phénomène au Groenland. Les peuples du Nord connaissent depuis longtemps les migrations des renards polaires. Dans le pays en aval de l’Obi, on connaît très bien les périodes de l’arrivée des renards polaires d’Yamal ou des renards de Taïmyr. Les années de m a x i m u m des lemmings sibériens et par conséquent des renards polaires n’arrivent pas simultanément sur tout le territoire de la toundra. Par exemple,les variations dans le nombre de L e m m u s lemmus dans le nord de la presqu’île de Kola ne sont pas synchronisées avec celles du nombre de L e m m u s sibiricus des toundras situées entre la mer Blanche et l’Oural. Parfois, dans la toundra Bolchézemelskaia, il y a peu de renards polaires mais, au milieu de l’hiver en Sibérie occidentale, arrive une masse de renards polaires dont les fourrures sont d’une meilleure qualité que celles des renards polaires locaux. Cela représente un gain pour les chasseurs des pays en aval de la rivière Petchoura et de la région d’Arkhangelsk. L’ouvrage publié par Sdobnikov en 1940 confirme l’existence de ces migrations de renards polaires à travers les toundras. Au cours dee années 1937-1939, sur la presqu’île d’Yamal, on marqua avec des punaises piquées à l’oreille 172 renards polaires qu’on avait atrrapés près de terriers à l’âge de deux mois environ. Les données de cet ouvrage montrent qu’il y avait 31 retours des bêtes marquées. D e l’endroit du marquage,4 renards s’éloignèrent au nord,5 au,nord-ouest, 3 à l’ouest, 12 au sud-ouest,2 au sud, 1 au nord-est et 4 renards polaires furent capturés dans la région du marquage. Ces derniers cas concernaient les femelles. En m ê m e temps, Sdobnikov cite les données de son correspondant selon lesquelles on peut constater que, dans la région de l’embouchure du fleuve Ienissei,les mâles sontlespremiersà quitterleslieuxdereproduction. Ils forment une “marche riveraine” en amont du grand fleuve tandis que les femelles restent en grande partie dans les régions de leurs terriers, où elles attendent l’arrivée des reniles, après quoi elles se mettent ii suivre les troupeaux de rennes en longeant les terres entre les fleuves. Les captures les plus intéressantes furent les suivantes :no 215, mâle, marqué le 17 août 1937 aux environs de la ferme Labarovaia, capturé en mars 1940 dans l’île du nord de la NouvelleZemble (la distance en droite ligne est de 1 O00 km) ; no 127,femelle,marquée le 18 août 1937 dans la région de la rivière Paderate, capturée le 5 décembre 1937 sur la rivière Snopa dans la toundra de T y m a n (la distance en ligne droite est de 900 km) ; no 1412, mâle et no 1422,femelle,marqués le 27 juin 1937 dans la région du Nouveau Port, sur la côte est d’Yamal, tous deux capturés en décembre 1937 dans le delta de la Petchora,à une distance de 900 km.L a plus grande vitesse de déplacement par jour est donc de 14 k m . Elle fut établie suivant le déplacement d’un renard qui se trouva 6 la distance de 900 k m , 65 jours après
262
le marquage. I1 est évident que la vitesse réelle de déplacement dut être beaucoup plus grande. L a faculté d’accommodation et la résistance de ces animaux, leur aptitude à tolérer longtemps la faim, permettent aux renards polaires de parcourir de grandes distances dans dee conditions très dures. I1 faudrait augmenter le marquage de ces bêtes sur tout le territoire de leur aire de dispersion. Les oscillations cycliques du nombre des renards polaires et les migrations massives rendent très difficiles la planification et l’organisation de la chasse d’une part et, d’autre part, deviennent parfois la cause d’une perte inutile de plusieurs bêtee. On connaît depuis longtemps l’épizootie parmi les renards polaires, les renards roux, les loups et les chiens de trait. Les Russes appellent cette maladie la rage des toundras.Au Canada, on l’appelle sledge-dog disease et, au Groenland, on la n o m m e crazy disease. Au cours de ces dernières années, le Dr R. Kantorovitch étudia cette maladie dans les toundras de l’URSS. I1 se trouva que l’agent morbifique de cette maladie est une des variantes biologiques de celui de la rage classique, largement répandue dans la zone des toundras. Les traits particuliers de cette infection sont : période d’incubation très courte, paralysie des muscles du bassin et des extrémités. Les animaux malades ne manifestent pas de symptômes de rage furieuse,mais leur comportement s’écarte de la norme. L e virus en cause aurait été, par rapport à l’homme, la variante affaiblie du virus rabique. Après une vaste enquête et un examen sérologique des personnes se trouvant en contact avec des renards et des bêtes contaminées lors du traitement des peaux de renards polaires, on établit que ces personnes n9étaient pas atteintes et que, dans leur sang, il n’y avait pas d’anticorps neutralisant le virus. (Ence qui concerne la rage classique, les cas d’hydrophobie sont souvent enregistrés chez les personnes qui écorchent les renards et les chiens.) L’absence de cas semblables dans le Nord au cours de dizaines d’années nous permet de croire que la pathogénie de la variante arctique du virus s’est affaiblie pour l’homme. Lors d’un examen virusologique de 4 612 animaux (renards,polaires, renards, loups, hermines, gloutons, campagnols, harfangs des neiges), les virus de rage furent décelés chez 30% des renards polaires, chez 9% des renards roux et l’on ne trouva pas de virus chez les autres mammifères et chez les oiseaux. Les études effectuées dans les toundras, aux environs des montagnes de l’Oural, firent apparaître que les foyers d’infection les plus intenses se situaient toujours autour des aires de dispersion des renards polaires et le long de leurs voies de migration. Durant les années de maximums, au début de l’épizootie,le rapport des jeunes bêtes aux adultes fut de 11 à 1 ; parmi les jeunes, il y avait 58% de bêtes contaminées et 11% parmi les adultes. A la fin de l’épizootie, les mêmes indices furent de 6 à 1, soit 32% et 7%.
Ecologie des plus importantes espèces de la faune subarctique
Tout en tenant compte de certaines données suivant lesquelles on peut supposer que les virus de rage se conservent dans les populations de lemmings, de campagnols et de musaraignes, le Dr Kantarovitch effectua en diverses périodes de cycle épizootique l’examen virusologiqne de 2 835 bêtes et l’examen sérologique de 172 rongeurs. Dans la totalité des cas le résultat fut négatif. L a supposition qu’il existe un rapport étroit entre l’épizootie chez les chiens et l’épizootie chez les renards polaires fut confirmée. De 95 chiens de toundra soumis à l’examen, on réussit à dégager les anticorps neutralisant le virus de la rage chez 10% des bêtes. Cependant, dans le sang de 150 rennes, ces anticorps étaient absents. Donc, on peut dire que les populations de renards polaires, les plus importantes bêtes à fourrure des régions subarctiques et arctiques, constituaient la source principale du virus de la rage. L a mortalité parmi les renards polaires lors des périodes d’épizootie est très grande ; par exemple, au mois de mai et de juin 1933, dans la région de Se-Yakha, les Nénéans trouvèrent près de 130 bêtes mortes. L a plupart de ces bêtes portaient leur fourrure d’hiver et présentaient un engraissement satisfaisant. Dans la toundra Bolchezemelskaia, après les épizooties, on trouvait parfois deux cadavres par kilomètre carré. L’épizootie de la rage parmi les chiens de trait entraîne souvent la mort de ces bêtes à la période de chasse la plus intense et lors des transports des marchandises au nord du Canada, sur les côtes du Groenland et au nord-est de la Sibérie. L a vaccination antirabique des chiens dans les lieux oìi apparaissent souvent les renards polaires furieux, effectuée par Kantarovitch, fut très efficace. On peut recommander d’employer cette méthode sur une large échelle dans la région du Nord. Beaucoup de chasseurs et d’éleveurs de rennes dans les toundras de l’URSS nourrissent les jeunes renards, près de leurs terriers, de poisson, de chair de phoque, etc. L a mortalité parmi ces nichées est nettement réduite ; les bêtes restent longtemps dans la région de leurs terriers et sont plus faciles à chasser. On essaya également, à l’aide d’un affouragement massif, de retenir les bêtes dans un endroit donné afin de pouvoir procéder à la chasse immédiatement après la m u e d’automne. U n e partie considérable des renards polaires migrateurs viennent dans les régions assez peuplées, mais souvent ces bêtes n’ont pas encore leur fourrure d’hiver. En Sibérie, on pratique parfois la capture de ces bêtes et on les élève ensuite dans des cages jusqu’à ce que leur fourrure soit satisfaisante. Des poissons bon marché, des déchets provenant de l’élevage des rennes rendent ce système assez rentable. Il faut améliorer l’exploitation des populations de renards polaires des toundras. Cette exploitation sera plus efficace si l’étude de l’écologie de cette espece est plus approfondie.
LES RENNES (BARREN GROUND CARIBOU) Les rennes de la toundra ont leur aire de répartition dans la zone subarctique et arctique. Les rennes de la toundra diffèrent de ceux des forêts par certains traits adaptatifs de leur constitution et par certaines particularités de leur écologie. Les sabots des rennes de toundra sont plus larges,la fourrure plus épaisse et plus longue et la crinière est plus développée que celle des rennes de forêt. En hiver, le pelage est très clair, presque blanc. Chez certaines races géographiques (par exemple, Rangifer turandus pearyi), la fourrure reste blanche, m ê m e en été. I1 y a un ou deux siècles, le nombre des rennes peuplant les toundras de l’Ancien et du Nouveau Monde était énorme. Les migrations saisonnières de leurs troupeaux présentaient un tableau grandiose, difficile à décrire. En automne, aux endroits des passages de rivière, les troupeaux de rennes remplissaient à perte de vue les toundras et l’on aurait dit que c’était une forêt de bois (voir, par exemple, Vrangel, 1841, Voyage wers les côtes de Sibérie ... fuit en 1820-1824).L’ancienne densité des rennes prouve que la végétation de la toundra,maigre à première vue, possède une assez grande productivitébiologique,car, grâce à cette végétation, de grandes populations ont existé, non seulement de rennes, mais d’autres anim a u x comme, par exemple, de lemmings, d’oies, de lagopèdes. L a chair, la graisse, les peaux, la fourrure, les tendons, le sang et les entrailles des rennes furent, pendant des siècles, des produits importants dans l’économie de plusieurs peuples septentrionaux ne vivant que de la chasse, de la pêche et de l’élevage des rennes. I1 est à noter un trait particulier de cet animal nordique :sa capacité de concentrer dans l’organisme pendant un été court beaucoup de réserves nutritives,y compris celles de vitamine C.Les peuples du Nord et certains peuples de Sibérie qui utilisent en hiver la chair et le sang des rennes n’ont jamais connu le scorbut; de combien de tragédies, liées à cette maladie, l’histoire de l’exploration des régions subarctiques n’est-ellepas marquée ? Les populations de rennes de toundra ne sont aujourd’hui que les vestiges de celles qui existaient autrefois. Cependant, de nos jours, les troupeaux d’ongulés restent une source importante de nourriture et de fourrure pour les habitants de la plupart des régions du Nord. Ainsi, par exemple, A. W.Banfield affirme en 1954 que dans les Northwest Territories,ainsi que dans la zone nordique de trois grandes provinces du Canada, sur un territoire de 600 O00 m2, les rennes de toundra constituent la part essentielle de l’économie de près de 20 O00 Canadiens. Sur la plus grande partie du territoire des Northwest Territories, la vie des h o m m e s ne serait pas possible sans les rennes -dit-il en conclusion dans son compte rendu. Vers 1900, au Canada, d’après les données de ce savant, il y avait près de 1 750 O00 rennes de toundra. Dans les années
263
A.N. Formosov
1948-1950, les recensements spéciaux effectués à l’aide des avions établirent qa’il y avait près de 670 O00 rennes. Donc, au cours d’un demi-siècle,le nombre de ces animaux au Canada se réduisit de 62:/,. Outre la chasse intense faite à ces animaux, la cause de cette réduction est due à la disparition, à la suite d’incendies, d’une grande partie des pâturages de lichens (pour le renouvellement des pâturages brûlés, il faut plusieurs années). Dans la presqu’île d’Alaska, en 1921, d’après les données de Olaus J. Murie, il y avait près d’un million de rennes de toundra,y compris un certain nombre de rennes de forêt. Au cours des années ultérieures, la quantité de rennes diminua considérablement. Dans les régions subarctiqves d’Eurasie, le sort des rennes fut encore plus terrible. D e grands troupeaux de rennes domestiqués occupèrent les meilleurs pâturages de toundra et de forêt, refoulèrent les rennes sauvages vers les côtes de l’océan Arctique où la vdgétation est plus pauvre et l’hiver plus rude et barrèrent les voies de migration. En m ê m e temps, l’utilisationdes rennes de trait facilitait la chasse aux rennes sauvages et le transport des proies. I1 y a près de deux siècles, il était habituel de voir les rennes peupler les zones subarctiques de Norvège, les toundras alpines et la zone forestière de toundra. Vers le début de notre siècle, dans la région septentrionale de Norvège, les rennes furent pratiquement exterminés. A partir de 1930, la chasse aux rennes de forêt fut régularisée et elle ne se pratique à présent qu’avec des permis spéciaux. Dans les toundras de la presqu’île de Kola, les rennes ont disparu depuis longtemps,mais on peut les voir encore dans les régions forestières et montagneuses où ils sont aujourd’hui protégés. Vers la fin du X I X ~siècle, les rennes disparurent dans les toundras situées entre la mer Blanche et l’Oural, où l’élevage des rennes est très développé. Un petit nombre de rennes restent encore dans l’île du sud de la Nouvelle-Zemble.En Sibérie occidentale, les rennes de toundra se rencontrent encore au nord de la presqu’île d’Yamal et de Gydansky, dans l’île du sud de la terre du Nord, ainsi que dans quelques petites îles situées le long du littoral du continent. Les rennes sont surtout nombreux dans la région du nord de la presqu’île de Taïmyr, où, en 1961, lors d’un recensement spécial effectué par avion, on a établi qu’il y avait de 100 O00 à 110 O00 têtes et dans le delta du fleuve Lena et entre les fleuves Y a u a et Indiguirka, il y en avait près de 50 O00 ou 55 000. Dans les îles de Nouvelle-Sibérie(Kotelny,Fadeevski), il y a près de 5 O00 rennes. A l’extrême orient de la Sibérie, sur la terre de Tchoukotka et dans le bassin d’Anadyr, les rennes sont rares aujourd’hui, bien qu’au X I X ~siècle ils fussent très nombreux dans ces endroits. L e renne est un animal qui vit dans des pays à climat dur. I1 hiverne dans des régions où les froids atteignent -50 ou -600 C. En été, le renne préfère les endroits où il faut moins chaud et où il y a des vents
264
frais. C’est pourquoi les rennes ont l’habitude de se déplacer vers les côtes des mers froides. L e renne supporte mal la chaleur et souffre beaucoup des piqûres des diptères, qui sont surtout actifs pendant les périodes chaudes et sans vent. Les périodes chaudes sont rares dans les régions subarctiques mais, lorsque cela arrive, les rennes pâturent mal et, pendant un été court, ils n’ont pas le temps de rétablir leurs forces après l’épuisement de l’hiver et de faire des réserves énergétiques indispensables à la reproduction et à l’hivernage. Au cours de telles années, on constate chez les rennes une mortalité plus élevée et chez la nouvelle progéniture une très faible résistance. Dans les toundras d’Eurasie,les rennes se nourrissent de diverses espèces de lichens (plus de 50 : Cladonia, Cetraria, Alectoria, etc.) et des dizaines d’espèces de plantes supérieures. Les lichens sont riches en hydrates de carbone, ils sont faciles à digérer et à absorber, mais ils sont pauvres en albumine et, notamment, en phosphore et en potassium. Les éleveurs de rennes savent qu’un long affouragement par des lichens se termine normalement par des troubles du métabolisme. Dans la nature, m ê m e en hiver,les rennes cherchent à manger non seulement des lichens,mais aussi diverses plantes vertes. Ils déterrent des feuilles de Carex, des céréales. Cependant, au printemps, les rennes de toundra souffrent de la nourriture déficiente en albumine et en minéraux. Après ‘la disparition des neiges, les rennes mangent parfois les œufs des oiseaux, chassent des lemmings, se nourrissent des os éventés et des bois de rennes. Après l’apparition de la première végétation,les rennes deviennent très difficiles:ils ne mangent que les parties des plantes les plus tendres et les plus succulentes, riches en vitamines et,par conséquent,ils se déplacent sans cesse à la recherche de nouveaux pâturages. Au cours des 2 ou 3 mois d’été,les rennes sont capables de rétablir leurs forces et de se constituer une épaisse couche de graisse sous-cutanée.I1 est intéressant de remarquer que les accumulations de graisse sont surtout considérables chez les rennes habitant les régions où l’hiver est ties long et très rude. Ils utilisent leurs réserves énergétiques pendant cette période froide. Malgré la pauvreté nutritive des lichens,les rennes en mangent toute l’année et ces plantes prédominent dans le contenu de l’estomac de la plupart des animaux. I1 est à croire que c’est à cette particularité des liaisons trophiques que les rennes doivent leur existence dans les régions subarctiques et arctiques, où les lichens jouent un rôle important dans la végétation. I1 n’y a que l’Ovibos moschatus qui sache mieux utiliser les lichens que le renne.et qui puisse vivre dans les régions insupportables pour le barren ground carihou. Les rennes sont des animaux grégaires qui se nourrissent de plantes naines dont ils ne mangent qu’une petite partie. Un déplacement incessant et une vie
Écologie des plus importantes espèces de la faune subarctique
nomade sont les conséquences logiques d’une telle utilisation des ressources fourragères. En effet, les rennes de toundra se trouvent toujours en mouvement. L a vitesse de cette migration change souvent, car elle dépend de l’importance des troupeaux, de la saison, du temps et des pâturages. D e u x fois par an, ces déplacements à la recherche de pâtures se transforment en migrations saisonnières massives. Lors de ces migrations,les populations de rennes comprennent des milliers et des dizaines de milliers de têtes qui s’en vont par groupes à la finde l’été vers le sud, vers les lieux d’hivernage. Au printemps, ils se mettent en route pour le retour. Les migrations s’étendent,dans les toundras d’Eurasie, sur un territoire de 200 à 750 km dans les îles du Nord, sur 300 ou 500 k m et, au nord-est du Canada, sur 850 k m . C’est lors des migrations d’automne,lorsqueles troupeaux de rennes, bien nourris et pleins de force, approchent de la frontière septentrionale des forêts et des lieux habités, que la chasse devient plus intense. Les rennes choisissent les routes les plus faciles et évitent les grands obstacles. Voilà pourquoi les voies de migration dans chaque région sont presque toujours les mêmes et sont sillonnées de sentiers battus. Cependant,on peut constater parfois certains changements souvent dus à l’épuisement des pâturages se trouvant le long des voies. On a également obscrvé le déplacement des aires d’hivernage.I1 arrive qu’une partie des animaux hivernent dans les endroits où ils ont passé l’été, quittant ainsi les troupeaux. Dans certaines régions, l’endroit abandonné par une population se trouve occupé par une autre qui vient de toundras plus nordiques pour passer l’hiver. Cela arrive,par exemple, au nord de la Yakoutie, où les rennes des îles de Nouvelle-Sibérieviennent dans la toundra continentale pour la période d’hivernage. Selon A.W. Ranfield, les premières gelées détruisant les buissons et les herbes à la fin de l’été dans la toundra stimulent le déplacement des rennes. En se déplaqant vers le sud, le sud-est ou le sud-ouest,les rennes arrivent dans des toundras coupées de forêts. Les rennes restent au sud de la toundra si le temps est assez doux et, au contraire, ils s’abritent dans les forêts si les gelées sont précoces. Protégés par les forêts, les rennes se trouvent à l’abri des vents glacés et trouvent des pâturages plus riches en lichens et en buissons. L a couche de neige de la zone forestière, plus meuble que celle de la toundra, est facile à creuser. Dans les lieux d’hivernage, les rennes de toundra côtoient parfois ceuy des forêts. Cependant, la différence dans les délais de copulation empêche le croisement entre les animaux de ces populations. L a période de copulation chez les rennes de toundra a lieu beaucoup plus tard, et leurs petits n’apparaissent que lorsqu’il n’y a plus de neige et que les périodes de froid sont passées. Par exemple, les rennes de Taïmyr ont leur période de copulation au mois de novembre et les femelles mettent bas dans la seconde moitié de juin.
Suivant certaines données recueillies en Sibérie, ce sont les femelles gravides qui se déplacent les premières vers le sud, parce qu’elles sont pressées d’atteindre, avant le dégel des rivières et des lacs, les lieux éloignés où elles pourront mettre bas. Ces lieux sont évidemment stables pour chaque troupeau. Plus tard, ce sont les mâles et les jeunes bêtes qui commencent à se déplacer vers le sud et, enfin, ce sont les vieux rennes et les rennes les plus épuisés qui se mettent en migration. Souvent, les insectes diptères obligent les rennes à accélérer leur migration. Le désir de se déplacer vers le nord en été est, de toute évidence, un désir inné. I1 est très manifeste chez les rennes domestiques qui pâturent dans la nature. En été, sur les terrains clôturés, ils se rangent près de la paroi du nord et, lors des conduites,il est extrêmement difficile de tournerlestroupeauxdu côtédu sudl. Beaucoup de données et articles sont consacrés à l’écologie des rennes de toundra, mais le nombre de leur population est grand et le climat et la nature de leurs aires de répartition sont très différents. Certaines races géographiques et populations sont déjh exterminées, et d’autres se trouvent en voie de disparition. Voilà pourquoi il faut se hâter de les étudier et d’organiser leur protection. Au cours des deux dernières décennies, un nouveau problème s’est posé, portant sur la particularité de l’écologie du renne. Les lichens sont des plantes qui accumulent vite les particules radio-actives se trouvant dans les retombées des essais d’armes nucléaires. Dans la chair des rennes se nourrissant toute l’année de lichens,le strontium 90 et le cesium 137 s’accumulent extrêmement vite, plus vite que chez les autres mammifères. L e niveau de contamination des rennes (d’après l’analyse des squelettes) atteint 100 ou 200 unités de strontium 90,c’est-à-direque ce niveau dépasse considérablementcelui qui peut être considéré c o m m e “inoffensif” pour l’homme et pour les autres mammifères. Ainsi, l’utilisation des rennes dans la nourriture devient dangereuse pour les générations humaines à venir. Pour les mêmes raisons, l’acclimatation du bœuf musqué, un autre consommateur de lichens, dans la région subarctique, ne peut plus donner les résultats qu’on en attendait il y a 40 ou 50 ans. Ce n’est que dans le cas où tous les Etats renonceraient à l’utilisation des armes nucléaires et à tout essai de ces armes que l’atmosphère terrestre, les eaux, le sol et toute la biosphère pourraient se purifier de substances radio-actives extrêmement dangereuses pour l’humanité.
LE LOUP L’aire de dispersion des loups couvre presque toute la zone subarctique.Cette espèce est répartie d’une façon i.
Observation personnelle de l’auteur.
265
A. N. Formozov
égale et représentée par quelques sous-espèces. Les loups nichent dans les toundras continentales et dans la plupart des îles de l’océan Arctique qui ne sont pas éloignées du continent. L e grand loup blanc des toundras d’Alaska et du Canada (Canis lupus tundrarum) se rencontre parfois jusqu’à 830 de latitude nord, sur la côte du Groenland. L a période de copulation chez les loups de toundra d’Eurasie commence 3 nu 4 mois plus tard que chez les représentants de la m ê m e espèce dans la zone tempérée. Les louves mettent bas dans les terriers des renards polaires, qui sont alors élargis et appropriés aux nouveaux maîtres. En été, les louves sont bien pourvues de nourriture dans la toundra, grâce aux multiples oiseaux qui y viennent nidifier. Les loups chassent des lemmings, des faons, de jeunes lièvres et des oies adultes lors de la mue. I1 arrive parfois que les loups mangent aussi des renards polaires et des renards; ils poursuivent toujours des rennes et des bœufs musqués dans la toundra de 17Amériqiredu Nord. D e s migrations massives sont caractéristiques pour les loups des deux continents. Ces migrations saisonnières se font en m ê m e temps que celles des rennes et des rennes domestiqués. En automne, les loups abandonnent les îles arctiques et viennent sur le continent immédiatement après la congélation des golfes. Les loups détruisent constamment les troupeaux de rennes domestiques ; 54 ou 67% des pertes sont dues aux loups. Parmi les victimes des carnassiers, les faons figurent pour 40%, les femelles pour 30% et les mâles pour 2%. L e métier des bergers est rendu très difficile par la présence permanente des loups autour des troupeaux.Pendant les longues nuits et par mauvais temps, lorsque la visibilité est faible, il faut toujours se tenir aux aguets, noter les moindres signes d’alarme, faire le tour des troupeaux fréquemment, examiner les traces, tirer des coups de feu. U n e attaque imprévue provoque, dans les troupeaux, une panique, dont les carnassiers cherchent à profiter. D e s groupes de rennes se dispersent sur un large territoire et il devient très difficile de les réunir. I1arrive alors que certains rennes sejoignent aux rennes sauvages et disparaissent définitivement. En hiver, on profite parfois, et non sans succès, de certaines habitudes particulières des loups afin de prévenir leurs attaques. Par exemple, à la tombée de la nuit,les bergers tâchent de concentrer le troupeau et de le faire passer de façon qu’il ne laisse qu’un seul sentier derrière lui (les loups cherchent à s’approcher d u troupeau en profitant des traces de rennes). Sur le sentier, on laisse en vue de petits drapeaux ou des vêtements pendus. Parfois, on secoue simplement les arbres pour en laisser tomber la neige. Ainsi, les arbres sombres, les drapeaux, les vêtements, qui se profilent nettement sur la neige blanche, font peur aux loups et ces derniers n’osent pas s’approcher du troupeau. Bien que les loups de toundra fournissent à la
266
pelleterie de belles fourrures, les dommages qu’ils causent aux troupeaux justifient toutes les mesures prises pour les exterminer. Sur ce plan, on obtient de bons résultats lorsqu’en tire sur des loups d’un avion, les jours QÙ la visibilité est excellente. Cependant, cette méthode est onéreuse, et ne peut être utilisée dans la toundra coupée de forêts, où s’abritent bon nombre de loups des régions subarctiques. Durant les périodes d’épizootie de rage parmi les renards polaires, la m ê m e maladie est observée chez les loups dans les toundras du Canada et au nord de l’Eurasie. L’étude détaillée de l’écologie des loups de toundra est souhaitable afin de pouvoir mieux contrôler leur nombre.
L’EIDER
ORD INA1RE
Somateria mollissima est un des plus précieux oiseaux
des mers subarctiques. L a vaste région de nidification
de cet oiseau se trouve sur le littoral boréal d’Amérique et du Groenland,y comprisles îles environnantes, sur le littoral septentrional de l’Eurasie, à partir de l’Islande jusqu’à la Nouvelle-Zemble,sur le littoral et les îles des mers de Béring et d’Okhotsk. L a plus riche zone de nidification se trouve toujours dans les régions subarctiques. L’eider à duvet est parfois sédentaireet se retire en hiver sur la mer aux endroits OU elle ne gèle jamais. En d’autres régions,cet oiseau est migrateur et ses migrations saisonnières sont souvent lointaines (ainsi pour l’eider du Spitzberg, l’eider du nord-est de la Sibérie et d’autres). On a décrit beaucoup de sous-espèces de cet oiseau, dont quatre se distinguent nettement par leur apparence extérieure. Cependant, toutes ces sous-espèces sont semblables du point de vue de leur importance économique et de leur écologie et nous n’avons pas besoin de les caractériser toutes. L’eider à duvet est un oiseau côtier marin. até c o m m e hiver il évite le large et se nourrit dans une bande littorale dont la largeur ne dépasse pas 2 à 5 km. I1 préfère les golfes, les baies, les détroits entre les îles et les continents,où 1’011 peut toujours trouver des endroits bien protégés contre le ressac et où le benthos est assez riche. L’eider consomme presque exclusivement une nourriture animale. En plongeant à une profondeur de 2 à 5 mètres, il se procure divers mollusques (Mytilus edulis, Litorina, Buccinus, Telh a ) , des crustacés, des oursins et de petits poissons (Cottidae). Au début de l’été, les eiders mangent, en petites quantités,des insectes,des baies, et des plantes fraîches. Les côtes rocheuses découpées, peu escarpées et basses, sont surtout favorables aux eiders pour un asile d’été. Protégées par des pierres et des buissons de saules nains, les femelles y trouvent des lieux sûrs pour leurs nids, à l’abri du vent et des rapaces. I1 est également nécessaire que les rivages soient en pente douce afin de permettre aux eiders d’amener leurs
Ecologie des plus importantes espèces de la faune subarctique
petits vers la mer. Les eiders ne reviendront dans leurs nids qu’au printemps suivant.L a couvéepasse sa vie sur l’eaudans la régionlittorale.Pour sereposer,les femelles montent avec leurs petits sur des pierres à surface plate situées loin des rivages. Les côtes à profil découpé permettent de trouver toujours de bons endroits pour le repos et l’affouragement. L e littoral sablonneux, en pente douce, des régions subarctiques, n’est pas propice i la vie de cet oiseau. L e ressac, le déplacement incessant des glaces dévastent la faune des basses eaux. Les eiders ne trouvent ici ni nourriture, ni abri. On peut donc penser que, vu cet état de choses, les eiders ne nidifient pas dans l’île de Kolgouev, ni sur les vastes étendues des côtes boréales de Sibérie. Les destructeurs habituels des nids de ces oiseaux dans la toundra sont les renards polaires et les renards. Aussi les eiders ont-ils acquis un trait adaptatif qui est la capacité de trouver, pour la nidification,des îles et des îlots isolés du continent et inaccessibles aux deux espèces de renards. Voilà pourquoi, dans des îlots rocheux et éloignés,on trouve parfois des colonies d’eiders très nombreuses. A l’abri des carnassiers, leurs nids restent exposés aux rapaces. Li où l’élevage des eiders est bien organisé, on extermine les corbeaux et l’on régularise le nombre de goélands en enlevant systématiquement des œufs de leurs nids. (Ces œufs ainsi prélevés sont utilisés c o m m e nourriture OU conservés jusqu’à l’hiver, afin d’être ensuite employés c o m m e appât ou c o m m e nourriture pour les renards polaires.) I1 est à remarquer que les eiders sont répartis sur les côtes de façon inégale parce que les endroits de nidification doivent satisfaire à tous les besoins de ces oiseaux. Dans des lieux favorables se forment des colonies d’eiders qui comptent des centaines et des milliers de couples. L a densité de nidification est alors très grande. Par exemple, dans l’île de Krestovatik (Nouvelle-Zemble), dont la superficie atteint à peu près 300 m2, on trouve chaque année plus de 400 nids. D’habitude, la plupart des nids se situent le long de la lignc côtiPre assez étroite, dépassant légèrement quelques dizaines de mètres. I1 faut ajouter que, dans chaque région, les eiders utilisent tous les avantages du microrelief et occupent des abris caractéristiques bien connus par les habitants de la région. Ces facteurs facilitent la recherche des nids des eiders et le ramassage de leur duvet précieux. L a quantité moyenne de duvet par nid est de 17 à 20 grammes dans la péninsule de Kola, de 14,5 à 16,5 grammes en Islande. L e duvet du nid des eiders est extrêmement mou, léger, élastique, et possède une très basse condsctibilité de la chaleur. Ses qualités sont adaptées à la défense des œufs contre le froid,ce qui s’avère important parce que les œufs sont pondus sur un sol gelé légèrement recouvert d’une litière dans les conditions d’un climat frais et venteux. L e duvet plumé possède des qualités marchandes
tout autres et coûte moins cher que le duvet de nid. Il garnit le dessous de la poitrine et de l’abdomen de la femelle, qui commence à se plumer après la ponte du quatrième œuf et arrive vite à faire une sorte de calice en duvet où la ponte est cachée. Somateria spectabilis, une espèce voisine de l’eider à duvet, a également un duvet de nid de très bonne qualité, mais cet oiseau n’aura évidemment jamais un rôle aussi important que Somateria mollissima. Ses nids sont difficiles à trouver et le ramassage du duvet de S. spectabilis ne peut se pratiquer sur une grande échelle et ne se fait que par hasard. Certains traits particuliers, m ê m e insignifiants, de ces oiseaux (nidification isolée ou coloniale) déterminent leur rôle dans l’économie de l’homme. Ainsi,par exemple, l’eider ordinaire, en m ê m e temps qu’il tient à nicher en colonie, cherche à trouver une protection chez des animaux plus forts afin de défendre ses nids contre les rapaces. I1 est cependant à remarquer que le m ê m e trait se retrouve dans le comportement de plusieurs oiseaux des régions subarctiques et m ê m e dans celui des oiseaux des steppes. Les naturalistes ont remarqué depuis longtemps que, dans la toundra, les nids de diverses espèces de canards et d’oies sont situés au voisinage de ceux des Falco peregrinus OU des gerfauts. Ces rapaces défendent énergiquement leurs lieux de nidification contre tous les animaux capables de détruire les nids. Donc, le rapace, tout en défendant son nid, protège ceux de ses voisins et ne cherche pas à les attaquer. Un phénomène semblable est observé lors de la nidification de l’eider ordinaire aux environs des villages de pêcheurs où il n’y a ni chiens,ni chats et les h o m m e s reçoiventl’oiseau c o m m e le bienvenu et lui assurent un endroit c o m m o d e pour la nidification. Reconnaissant de cette attention, l’oiseau laisse dans le nid un petit amas d’un superbe duvet. L a plupart des femelles retournent dans le m ê m e nid. On profite depuis longtemps de cette habitude des eiders en les attirant dans des abris artificiels construits de pierre ou de tourbe. On arrive à créer de grandes colonies d’eiders dans lesquelles les oiseaux deviennent très confiants (Islande,Norvège, Nouvelle-Zemble,Estonie). Lorsque les lieux d’hivernage étaient bien choisis et protégés, on a observé une densité de 13 O00 nids par hectare (Isakov, 1956). L’aire de répartition de l’eider ordinaire est très étendue. L’écologie de plusieurs populations géographiques de l’espèce n’a pas encore été étudiée, ce qui laisse à penser que les méthodes de protection élaborées en Norvège et en Islande ne sont pas toujours satisfaisantes.En Islande, dès 1821, on promulgua un décret interdisant de tuer l’eider. Aujourd’hui, on l’appelle souvent la poule aux œufs d’or. En 1805, sur toutes les côtes d’Islande, il y avait 116 secteurs où nichaient les eiders et l’on ramassa 1072,75 kg de duvet. Vers 1914, le nombre des colonies de nidification était de 258 et le duvet ramassé
267
A. N. Formozov
atteignit 3 886 kg. Au cours de la période 1914-1929, la quantité annuelle moyenne de duvet fut de 3 848 kg maximum, 4 355 en 1916, et minimum, 3 238 kg en 1918. L e nombre des nids d’eider au cours de ces années était à peu près de 230 O00 à 250 000,dont 10 O00 se trouvaient sur la côte nord-est. Malgré la protection organisée des oiseaux, le nombre des eiders change suivant les années. L e “blocus de glaces” qui, certaines années, se produit sur les côtes, un temps froid et pluvieux en été s’avèrent défavorables,m ê m e pour cet oiseau habitué à un climat dur. En d’autres endroits on note des cas d’épizootie et d’invasion de vers intestinaux, qui causent une grande mortalité parmi les petits lorsqu’ils vivent sur la mer (mer Baltique, mer Blanche, en Finlande et en Estonie). Les maladies de ce précieux oiseau n’ont pas encore été étudiées. I1 est extrêmement souhaitable de connaître l’expérience de plusieurs siècles en Islande,en Norvège et dans d’autres pays, dans le domaine de la protection des eiderset de l’utilisationrationnelle de leurs colonies.
-
LES OIES ET LES CYGNES
*
Parmi les oiseaux consommateurs de masse verte brune,les oies tiennent la première place dans la faune des toundras. Dans les régions subarctiques,le nombre des espèces est beaucoup plus élevé que dans la zone tempérée et leur nombre reste toujours très grand malgré l’exploitation peu économique et de longue date de ces populations d’oiseaux. On peut dire sans risque d’erreur que les régions subarctiques sont une grandiose pépinière d’oies qui fournit du gibier aux pays des zones tempérée,subtropicaleet tropicale des hémisphères est et ouest. Dans la toundra d’Eurasie,il y a neuf espèces d’oies et deux espèces de cygnes. Dans le nord de l’Amérique, il existe huit espèces d’oies et une espèce de cygnes. U n e partie des oies, par exemple l’oie rieuse et le bernache cravant, occupent de vastes aires polaires lors de la période de nidification, d’autres nichent sur des terrains relativement étroits de la zone subarctique. Dans le nord de l’Eurasie, les oies rieuses et les oies des moissons sont surtout nombreuses. Les vols d’oies des moissons sont parfois si grands qu’ils ressemblent de loin à des nuages sombres s’élevant à l’horizon. Dans les régions subarctiques d’Amérique, les oies sont moins nombreuses, car le pays est en grande partie montagneux, et ces oiseaux préfèrent les toundras marécageuses,plates, à multiples bassins d’eau caractéristiqves du nord de l’Eurasie. Les oies sont monogames. Certaines espèces, tout c o m m e les cygnes, forment des couples pour toute la vie. Les oies arrivent dans les régions subarctiques assez tôt après les cygnes, lorsque la toundra est encore couverte de neige et que les terrains dégelés et les premières mares de neige fondue n’apparaissent
268
que par endroits. Les couples se mettent immédiatement à faire des nids, le plus souvent sur les pentes des collines. Toute la période de nidification est très bien appropriée à un été court. C’est un trait adaptatif important dans les conditions subarctiques. L a durée de la couvée et du développement des petits est assez longue chez les oies comme, d’ailleurs, chez tous les grands oiseaux. C’est pourquoi tout retard dans la couvée peut la compromettre ensuite, lorsque les froids et les orages surviennent avant que les petits soient forts (de tels cas sont enregistrés chez certaines espèces d’oies, de cygnes, de canards et de plongeons). L’auteur du présent exposé a vu, au sud du Kazakhstan, lors des migrations des oies des moissons, en octobre, quelques jeunes oiseaux abandonnés dont les rémiges primaires n’étaient pas assez développées pour leur permettre de suivre les adultes chassés de la toundra par le mauvais temps. Au cours d’un printemps froid, la plupart des oies des régions subarctiques ne nidifient pas et d’autres ont des couvées très réduites. Les années de nonnidification (non-breeding) ne sont pas rares en Eurasie et en Amérique. Les premiers temps après leur arrivée, les oies se nourrissent de feuilles vertes de céréales et d’Empetrum nigrum qu’elles trouvent dans des vallons. En cas de retour prolongé des froids et de la neige, les oies se trouvent privées de nourriture et meurent en grand nombre. Tel fut le cas dans la région d’Anadyr, au printemps1887,où beaucoup d’oiseauxfurentperdus à cause des froids.On pouvait attraper à la main les oies affaiblies. Un cas semblable fut observé en 1884. Outre les oies adultes, dans chaque région de nidification, il y a des volées de jeunes oiseaux de deux ou trois ans des mêmes espèces. Leur période de m u e a lieu plus tôt que celle des oiseaux adultes et souvent ils s’envolent à cette époque. On a établi que, dans la toundra maritime de la partie européenne de l’URSS, beaucoup d’oies rieuses et d’oies de moissons viennent pour la période de m u e sur la Nouvelle-Zemble.Les oies des moissons nichent dans la toundra coupée de forêts et dans la taïga en Sibérie orientale (sous-espèce Anser fabalis sibiricus) ; elles passent leur période de m u e dans la toundra en aval du fleuve Léna. Des déplacements semblables vers le nord sont aussi propres aux canards. Les canards de Miquelon nichent en masse dans la région d’Anadyr, sur la presqu’île de Tchoukotka, mais ils ne nidifient pas dans l’île de Vrangel. Cependant, des milliers de mâles de Clangula hyernalis se réunissent chaque année près de cette île pour la période de mue. L’attachement aux endroits de m u e chezles oies et chezles canardsde Miquelon est aussi grand que leur attachement au lieu de nidification. Des traits semblables sont enregistrés chez les oies en ce qui concerne leurs aires d’hivernage. I1 est évident qu’une volée d’oies, de caiiards ou d’autres oiseaux n’est pas une agglomération de hasard, mais c’est un groupement stable. C’est le facteur dont il
&cologie des plus importantes espèces de la faune subarctique
faut tenir compte en vue d’une meilleure exploitation et protection des espèces précieuses d’anseriformes. Au printemps, à mesure que la neige fond, les oies se nourrissent de feuilles vertes et d’Eriphorurn angustifoliurn des vallons. L a végétation de certains secteurs autour des marais se trouve pratiquement détruite par les oies et ce sont des mousses qui la remplacent durant les saisons ultérieures. Ainsi, dans les endroits de la toundra où les oies se trouvent en très grande quantité, les broussailles de plantes fourragères sont très exploitées,ce qui ne se produit pas dans les steppes de Kazakhstan, m ê m e aux environs de lacs très peuplés d’dnser anserl. Les exemples suivants permettent de se faire une idée de la densité de la population et du nombre des oies dans certaines régions subarctiques d’Eurasie. A. Kretchmar, ayant inspecté le bassin de la rivière Piassina, citait en 1965 les chiffres suivants : dans la région de la sous-zone du sud de la toundra, sur 100 hectares peuplés d’anseriformes,il y a 3 à 5 couples d’oies des moissons et d’oies rieuses. Dans les endroits propices,c o m m e les rivages des lacs,les îles,le nombre des oies est encore plus élevé. On rencontre beaucoup plus souvent l’oie des moissons et l’oie rieuse que le bernache à cou roux, dont la densité de peuplement est indissolublement liée à la présence de faucons nichant sur les rivages à pic des fleuves et des lacs de toundra. D e 2 à 6 couples de bernaches nichent d’habitude à côté de chaque nid de faucons. D e grandes volées d’oies se réunissent pour les périodes de m u e autour des bassins d’eau de la toundra. Pourtant,les bernaches à cou roux ne forment que des volées peu nombreuses (pres de 100 oiseaux par volée). En 1961, sur la rive gauche d’un affluent de la rivière Piassina, on compta près de 15 O00 oies des moissons et oies rieuses sur un terrain de 350 km. N. Pougatchouk, ayant effectué une série d’observations sur le littoral ouest de la presqu’île d’Yamal, au cours des années 1958-1959, obtint les chiffres suivants :sur 1 km2de surface des bassins d’eau, il y avait 66 oies des moissons, 60 eiders à tête grise, 60 canards de Miquelon, 27 cygnes de Bewick, 22 canards pilets, 15 oies naines, 4 canards milouinans, 2 sarcelles d’hiver,1 macreuse brune,1 canard souchet. Dans certains endroits du territoire inspecté par Pougatchouk, les conditions étaient très favorables à la nidification et, dans la toilndra absolument déserte, la densité de nidification des petits cygnes était de 25 couples par kilomètre carré. L e bernache cravant, espèce cyclopolaire, niche sur les côtes arctiques et dans les €les de l’océan du Nord. Branta bernicla bernicla, une sous-espèce,niche en été dans la zone maritime des îles de NouvelleZemble et de Kolgouev, ainsi que des presqu’îles d’Yamal et de Taïmyr. Les oiseaux de cette population passent l’hiver dans les secteurs de mer limités situés près du littoral d’Allemagne, des Pays-Bas,de France, des îles Britanniques et d’autres. Au siècle passé,
dans ces lieux d’hivernage “les cris d’innombrables volées couvraient entièrement le bruit de la mer” et “les essaims d’oiseaux ombrageaient la lumière”. C’est ainsi que N a u m a n n décrivait les troupes de ces oiseaux. Quelques décennies passèrent, et le nombre des oies de cette population se réduisit. F. Salomonsen estime que cette population ne se compose à présent que de 16 500 oiseaux. Les ornithologues d’Estonie ne comptaient, ces derniers temps, que 5 O00 à 10 O00 bernaches cravants. L a réduction du nombre des oies dans les lieux d’hivernage en Europe occidentale cause depuis longtemps de l’inquiétude aux organismes internationaux pour la protection des oiseaux. Certains auteurs croient que la disparition de l’herbe de mer (Zostera marina), nourriture principale des bernaches en hiver, amène la diminution de cette population. I1 est à noter que c’est là en effet la cause essentielle. I1 est incontestable que la chasse mal réglementée sur le territoire de toute une série de pays où hivernent ces oiseaux joue un rôle grandement négatif. L a petite oie des neiges (Chen coerulescens cqerulescens) fut très répandue aux X V I I ~ et X V I I I ~ siècles dans les toundras maritimes de Sibérie orientale. Elle nichait par grandes colonies et les nids de cet oiseau étaient disposés à proximité les uns des autres. Les œufs ramassés dans les lieux de nidification de l’oie des neiges constituaient une nourriture abondante et sûre pour la population indigène. I1 est naturel que ces lieux aient attiré l’attention des chasseurs. Les oies qui ont perdu leur couvée ne couvent plus la m ê m e année. Et les populations qui perdent leurs couvées d’année en année sont condamnées à la disparition. C’est le cas des oies des neiges dont les colonies étaient couramment exploitées par la population aborigène. L a chasse aux oies, lors de la période de mue, a également contribué à la disparition de cette espèce. siècle, les oies des neiges sont Vers le début du X I X ~ devenues extrêmement rares dans les toundras continentales de Sibérie, mais elles sont encore bien conservées dans l’île de Vrangel, qui est restée déserte jusqu’aux années vingt de notre siècle. C’est justement dans les années vingt que quelques familles de chasseurs vinrent habiter cette île, et la colonie des oies des neiges qui nichait au centre de l’île se mit à disparaître rapidement. Aujourd’hui, ces oies se trouvent protégées,ainsi que les ours blancs (d’après les récentes données, de 150 à 200 ourses viennent mettre has dans l’île de Vrangel). U n e recherche spéciale, effectuée en 1960, montra qu’il y avait dans l’île une colonie principale d’oies des neiges et quelques colonies moins nombreuses. Au centre de la grande colonie, sur une aire de 500 à 600hectares,la densité de nidification est très grande : 90 à 100 nids par hectare. L e nombre total des oies de cette colonie était d’environ 130 O00 couples. 1.
Observation personnelle de l’auteur.
269
A. N. Formozov
D’autres colonies moins nombreuses ne comptaient pas plus de 20 O00 ou 30 O00 couples. Les résultats d’un recensement des nids furent complétés par le dénombrement des oies adultes qui s’étaient établies dans les lieux de leur concentration après la période de nidification. On enregistra au total,par observation aérienne, 400 O00 ou 500 O00 oies des neiges adultes. Au cours des années 1961-1963,dans l’île de Vrangel, on bagua 4 522 oies. Vers 1965, le centre responsable de la mise des bagues aux oiseaux de l’URSS reçut 452 documents témoignant de la capture des oiseaux portant des bagues soviétiques et 81 oies attrapées en URSS portaient des bagues américaines. Ces oies étaient attrapées dans les lieux d’hivernage de nos oies des neiges. On établit que, dans l’île de Vrangel, les oies passent près de 3 mois et, à la fin d’aout, elles commencent la migration dans la direction du golfe de Béring. Dans la seconde moitié de septembre,tous8 les retours de bagues sont enregistrés sur la voie de migration passant par le Pacijic migration route, et la plus grande partie des bagues vient du centre de l’$tat de Californie, c’est-à-diredes lieux d’hivernage. I1 a été constaté que le pourcentage d’oies des neiges dans l’île de Vrangel est très élevé dans la chasse américaine (près de 10% du nombre des oiseaux bagués). Donc, sur le continent de l’Amérique du Nord, on avait chassé des oiseaux qui avaient niché en URSS et se trouvaient chez nous sous protection. Cela peut empêcher l’accroissement de la population qui a déjà commencé à exercer une influence sui la répartition des oies des neiges au nord-estde la Sibérie. L e cygne de Bewick est un oiseau propre à la toundra d’Eurasie. I1 niche à partir du nord-estde la presqu’île de Kola à l’estjusqu’au delta de Kolyma. Par endroits, son aire de répartition s’étend aux régions arctiques et, en Sibérie, elle entre dans la zone marécageuse de la taïga. Cet oiseau niche dans les toundras herbacées couvertes de mousse, de lacs et de rivières. I1 niche là où la neige fond vite et, plus rarement, au bord des lacs. Son nid apparaît c o m m e un amas plat de mousse, de feuilles sèches de Carex, de plumes et de duvet de la femelle. L e diamètre du nid est de près d’un mètre et sa hauteur est de 50 à 60 cm. Lorsque la neige fond, le nid se trouve à la surface et les cygnes, en arrivant sur les lieux, se mettent à le réparer. L a couvée commence à la fin de mai. Les cygnes tirent donc meilleur profit que les oies du court été pour leur reproduction. Ces oiseaux se nourrissent de verdure, de rhizomes d’hydrophite côtière et, parfois, de plantes terrestres, de petits poissons et d’invertébrés. L a chasse au cygne est interdite en URSS et les Cygnus bewickii qui hivernent sur la Caspienne et sur la mer d’Aral sont protégés. L’aire de nidification d’un cygne plus grand,n o m m é Cygnus Cygnus, occupe la zone de taïga, des steppes coupées de forêts,et s’étend aux régions subarctiques. Un petit nombre de Cygnus Cygnus islandicus nichent en Islande et leur aspect typique est signalé au nord
270
de la péninsule scandinave, dans les toundras européennes de l’URSS, à l’est de la Sibérie,dans la région d’Andyr et sur la presqu’île de Tchoukotski. Dans la région septentrionale de leur aire de dispersion, ces beaux oiseaux sont mieux préservés que dans la région du Sud, où plusieurs bassins d’eau sont transformés par l’activité des h o m m e s et s’avèrent inutilisables pour la nidification. L e Cygnus Cygnus est plus étroitement lié au milieu aquatique que le Cygnus bewickii. I1 fait ses nids dans de petites Iles ou dans les broussailles d’hydrophite, aux endroits peu profonds des grands lacs. Ces grands et forts oiseaux défendent énergiquement leurs nids et leurs petits contre les rapaces et les carnassiers. Ils n’ont pratiquement pas d’ennemi naturel. Cependant, les braconniers détruisent les nids des cygnes et emportent leurs petits. Le cygne siffleur (Cygnus colombianus) niche sur les côtes arctiques et dans les îles de l’Amérique du Nord, à l’ouest du golfe de Goudson. Cette espèce septentrionale est mieux conservée sur le continent du Nouveau Monde que le cygne trompette nichant en Amérique, dans la zone boréale, et qui est pratiquement exterminé. L’importance esthétique et culturelle des cygnes est incontestable et universellement reconnue. Voilà pourquoi il faut prendre toutes les mesures possibles pour protéger les populations de la toundra dans les endroits d’hivernage et lors des migrations. Malheureusement, des volées de cygnes venues de l’URSS, sont chassées avec des filets sur les lieux d’hivernage, par exemple au nord de l’Iran.
LES CANARDS L a plupart des canards des régions subarctiques, d’après leurs liaisons trophiques, sont des zoophages. I1 existe une loi de réduction du nombre des espèces et des exemplaires des phytophages du sud vers le nord. Dans la zone des steppes d’Eurasie, à Kazakhstan, 570/, des canards sont herbivores et 50% dans les steppes coupées de forêts,tandis que, dans la zone septentrionale de toundra, l O O ~ , des canards sont zoophages. Les mêmes lois sont observées dans la faune d’Amérique du Nord. Ce n’est que sur les bassins d’eau de la toundra coupée de forêts et de la toundra du Sud, qu’on signale des espèces telles que le canard pilet, le canard siffleur,la sarcelle élégante,espèces qui se nourrissent de plantes. Ils ne nichent pas plus au nord. L e fait est que la végétation aquatique des régions subarctiques est trop peu développée à cause d’un dégel tardif et d’une congélation précoce des bassins d’eau, dont la température est très basse. Ce n’est que dans des lacs peu profonds qu’il y a des broussailles de 3 ou 4 espèces de plantes seulement, dont Potamogeton alpinus et Potamogeton natans servent de nourriture aux canards. A titre d’exemple, mentionnons que, sur des lacs de steppe riches en
ficologie des plus importantes espèces de la faune subarctique
canards,la masse totale des plantes immergées est de 58 tonnes par hectare. Riches en plantes fourragères sont les bassins d’eau de la région centrale des prairies de l’Amérique du Nord. Dans cette région,il y a beaucoup de canards. En m ê m e temps,la basse température de l’eau contribue à une pénétration plus rapide de l’oxygène de l’air’ dans les bassins d’eau des régions subarctiques. Cela est favorable au développement et à la reproduction des crustacés microscopiques (Copepoda, Cladocera), des larves (Chironontidae, Culicidae, Trichoptera), etc. Tous ces animaux constituent la nourriture principale des canards et de certains poissons. Les petits et les femelles des canards de Miquelon, des canards milouinans, des eiders à tête grise, etc., se nourrissent, dans les toundras, de larves (Chironomidae, Tipulidae) et de mollusques. Les larves représentent une nourriture facile à digérer. L’existence de ces espèces de canards (3e groupe d’après le système proposé par Koutcherouk en 1948) est étroitement liée à la présence d’un milieu aquatique, où ils trouvent leur nourriture et un abri contre les rapaces. Les macreuses noires et brunes et les harles ne savent pas marcher sur la terre et sont liés aux bassins d’eau (4e groupe). Selon certaines observations effectuées sur la presqu’île d’Yamal, les espèces du 3e groupe commencent à couver lorsque les lacs de toundra dégèlent, à la fin de juin, c’està-dire vingt jours plus tard w e les oies (ler groupe). Les espèces du 4 e groupe nichent sur la presqu’île d’Yamal, sur les rivières,les ruisseaux,et commencent à couver au début de juillet, immédiatement après la baisse des eaux. Le 5 e groupe est représenté par les plongeons, qui ne sont pas du tout capables de se déplacer sur la terre. Ils couvent sur les rivages bas, tout près de l’eau, à mi-juillet,lorsque le niveau de l’eau est stabilisé. Les canards des espèces énumérées vivent avec leurs petits sur la surface des lacs presque entièrement dépourvus de broussailles et, par conséquent, ils n’ont pas d’abri naturel. Lorsqu’ils sont attaqués par des rapaces, les petits se mettent à plonger et la femelle, tâchant de les couvrir, se jette contre l’ennemi. Mais la victoire n’est pas toujours de son côté. Les grands goélands poursuivent les nichées, attrapentles plus faibleslorsqu’ilssont fatigués d’avoir plongé. On a observé, par exemple,la perte de nichées entières de canards de Miquelon et une diminution de 30% du nombre des jeunes oiseaux dans les couvées de canards milouinans au cours des quinze premiers jours de leur vie (Yamal). Les couvées grandies se déplacent des lacs vers les rivières, où il y a plus de nourriture,plus de mollusques,et descendent ensuite vers la mer. Les canards célibataires (femelles et mâles) se réunissent aussi dans ces endroits après la période de mue. Toutes les étapes suivantes de la vie saisonnière des canards jusqu’a leur retour dans les lieux de nidification sont liées à la mer. Grâce à cette vie maritime, ils ont moins à souffrir des chasseurs. Cependant,les canards
de mer sont souvent victimes de la pollution des eaux territoriales (oil pollution) et meurent par milliers dans les secteurs à navigation intense. Sur le plan de la protection des oiseaux, y compris les oiseaux subarctiques,les choses ont changé après 1936 sur le continent de l’Amérique du Nord, lorsque le Mexique se joignit à l’accord entre les Etats-Unis et le Canada en 1916 et lorsque, aux fitats-Unis,on réglementa les normes et les délais de chasse en fonction du nombre et de l’état des populations d’oiseaux. Des mesures identiques devraient être prises le plus vite passible dans les pays du Vieux Monde. A cette fin, Issakov rassembla en 1965 toutes les données indispensablessur l’état des populations géographiques des oies et des canards en U R S S , et les soumit au 7 e Congrès de l’Union internationale des biologistes du gibier. Les participants au congrès accueillirent favorablement cette intervention. I1 est à noter qu’en URSS la chasse aux canards et aux oies est permise durant trois mois et demi, tandis que, dans les pays d’Europe occidentale, où notre gibier passe l’hiver, la saison de chasse dure de cinq à sept mois. Aux Pays-Bas, on chasse chaque année près d’un million de canards et d’oies; au Danemark, o n en chasse 500 O00 ; sur le littoral méridional de la Caspienne, en Iran, on chasse chaque hiver près de 1200 O00 canards venus du Nord, etc. Tout cela nous montre que la protection, la reproduction et l’exploitation des ressources du gibier aquatique en Eurasie du Nord ne peuvent être réalisées que sur la base d’une coopération internationale étroite. Afin de connaître le rôle de divers pays et des diverses régions de l’URSS dans la reproduction et l’exploitation des canards et des oies, il est important d’établir les traits particuliers de la répartition saisonnière de leurs populations géographiques. Chacune de ces populations a ses aires de nidification,ses voies de migration et ses lieux d’hivernage.
CONCLUSION L’étude du monde animal des régions subarctiques attire depuis longtemps l’attention des chercheurs. C’est un monde original et pittoresque, dont les représentantspossèdent d’excellentes capacitésd’adaptation aux dures conditions d’existence. L’expérience concernant l’utilisation de ce monde animal date de plusieurs siècles, mais cette utilisation, faute de suivre des recommandations scientifiques ou à tout le moins, des règlements convenables, a provoqué la réduction ou l’extermination de certaines espèces précieuses. C‘est un fait bien connu qu’en très peu de temps on a exterminé presque totalement, dans les eaux septentrionales,les baleines bleues et les baleines du Groenland. Parmi les espèces exterminées se trouvent le grand pingouin et le Camptorhynchus labradorifis ou canard
271
A. N. Formosov
du Labrador. L e nombre des ours blancs s’est réduit considérablement et ces animaux deviennent aujourd’hui très rares dans les régions subarctiques; ils sont un peu plus nombreux dans les régions arctiques. Les aires de dispersion de beaucoup d’autres espèces précieuses ont diminué notablement. Les animaux des régions subarctiques se trouvent en état de lutte permanente pour leur existence. En outre, beaucoup d’entre eux sont très confiants et, par conséquent, faciles à exterminer si l’on néglige les normes rationnelles d’exploitation. AU COWS de ces dernières décennies, l’activité des associations scientifiques, les mesures législatives des gouvernements et les accords internationaux ont beaucoup contribué à la régularisation de l’exploitation des ressources du monde animal en général et des régions subarctiques en particulier. Cependant, il reste encore beaucoup à faire. A m o n avis, il est incontestable que la coordination des recherches sur l’écologie des animaux subarctiques effectuées par des savants de divers pays doit servir davantage cette cause. L’expérience montre que des recherches de plusieurs
272
années, effectuées selon un programme unifié, en des endroits soigneusement choisis, donnent le m a x i m u m d’avantages dans l’accumulation de données écologiques complètes et intéressantes. C’est pourquoi il est recommandé d’augmenter le nombre des parcs nationaux. I1 serait utile également de compléter les données obtenues à l’aide de questionnaires spéciaux envoyés aux chasseurs professionnels. Ce ne serait pas qne erreur de dire que les recherches au moyen d’expéditionssont passées de mode. Cependant, sur les cartes des régions subarctiques il existe bon nombre d’endroits où les écologues n’ont jamais mis le pied, et il faut tenir compte que, dans les régions où l’on a déjà effectué certaines recherches, les expéditions fournissent encore des données précieuses et des faits intéressants. U n e combinaison adroite de diverses méthodes de recherches, l’utilisation de nouvelles techniques, la coopération internationale, dont un exemple encourageant nous est donné par les recherches sur la nature de l’Antarctique, doivent accélérer et approfondir l’étude du monde animal des régions subarctiques.
Écologie des plus importantes espèces de la faune subarctique
Summary Ecology of the major species of subarctic fauna
(A.N. Formozov) The animal world of the subarctic regions is rather poor from the point of view of the number of species, but it is varied and its biomass is rich. Most of the endemic species and the tundra races of very c o m m o n species adapt themselves excellently to existence in the Northern regions. Marine, river and lake fish, marine mammals of the Pinnipedian and Cetacean orders, (land mammals and lands birds, all these are the precious biological resources of the subarctic regions,and they have been exploited for several centuries. Unfortunately, owing to uncontrolled exploitation, some valuable species are n o w almost completely exterminated and others are beginning to disappear. Subarctic animals have to struggle unremittingly for their existence. At the same time, m a n y of them are very trusting and consequently easy to hunt, if reasonable hunting standards are not observed. T h e author of this brief account analyses the ecology of a series of land mammals and birds, most of his data being collected in the subarctic zone of the U.S.S.R.Since there is very little heat, the strong winds which blow unceasingly in summer and winter play an important part in the life of the land fauna of that zone. F e w birds survive a long and extremely severe winter, with little daylight and a very thick blanket of snow. In the tundra regions, where some sixty or seventy species nest, only one or two stay through the winter (Lagopus mutus, Nictea scandiaca). In these same regions, however, rodents ( L e m m u s ,
multiply under the protection of a layer of hard snow, feeding on green planta preserved by the cold. As the summer is very short, such large birds as swans,geese,gerfalcons,snowy owls (harfangs) begin to nest very early, long before the snow melts. The continual daylight which lasts for several days in summer is propitious for various species. The high nutritive value of the vegetation, the abundance of certain insects (Tipulidae, Chironomidae, Trichoptera and others), whose larvae and chrysalises are easy to unearth, and the richness of the benthos and zooplankton are factors conducive to the development of the subarctic animal world. The average number per square metre of the larvae Tipula carinifrons, Prionocera lapponica and others, which are the principal nourishment of the curlews, is 50,and in some places even 200. That is one reason why curlews nest in the tundra (for example, over thirty species nest in the region of Andyr). In the years when the spring is cold and late, a “non-nidification”of geese and ducks has been noted. W h e n there are few lemmings, the large rodent-eating birds do not nest and the number of polar foxes suddenly drops. The Passeres lose their young when the weather is cold and wet. The reproduction rate in the,subarctic regions is very unstable, mortality sometimes being very high (Alopex lagopus, Lagopus lagopus, Aythya marila, etc.), and at the time of the great seasonal migrations characteristic of the subarctic regions,there is a considerable loss of life. Very great care is needed in exploiting the resources of the subarctic fauna. Dicrostony, Microtus)
273 20
The ecology and population dynamics of the barren-groundcaribou in Canada A. G.Loughrey and J. P.Kelsall
The barren-ground caribou, Rangifer tarandus groenlandicus (Linnaeus) of northern Canada is the only large ungulate adapted to life in both arctic tundra and subarctic forest. Before the coming of the white m a n it was a life-sustainingnecessity for the northern Indians and inland Eskimoes. Even today the gregarious and migratory barren-ground caribou is still an important source of food and clothing to several thousand Indians and Eskimoes w h o follow a hunting and trapping economy. The ecology and population dynamics of the barrenground caribou found on the central, northern mainland of Canada have been studied since 1948 by biologists of the Canadian Wildlife Service, Department of Indian Affairs and Northern Development. The Canadian animal is of medium size in comparison with other races of caribou and reindeer in North America and Eurasia. Males average 225 pounds (102 kilograms) and females 150 pounds (68kilograms) in weight. In north-central Canada barren-ground caribou occupy a range of 750,000 square miles (1,943,000 square kilometres). Approximately half of the range is tundra and the other half boreal forest or taiga. It is bounded on the west by the Mackenzie River and on the east by Hudson Bay. The Arctic Ocean forms the northern boundary, while the southern limit includes the northern portions of the three prairie provinces. The range extends through 15 degrees of latitude,from 550 N.to 700 N.and through a longitudinal range from 820 W.to 1300 W. F r o m the Mackenzie River delta in the north-west to the south-eastern range extremity near God’s Lake, Manitoba, is a distance of 1,500 miles (2,414 kilometres). (Kelsall,in press.) Almost all of the tundra and m u c h of the taiga caribou range is underlain by Precambrian granitic bed-rock. It is a country characterized by low relief
and rolling plains. The western and south-eastern parts of the winter range are underlain by more recent Palaeozoic sedimentary rocks. Annual precipitation is light, not exceeding 14 inches anywhere on the range. However, perennially frozen ground, surface bed-rock, and poorly developed drainage contribute to the accumulation of large amounts of surface water. This surface water varies from temporary melt-water tundra pools to great river and lake systems. Surface waters constitute 30 per cent of the area of the caribou range. Soil formation and plant growth are slow. In this paper I shall sketch the annual cycle, movements, food habits, reproductive physiology, and population dynamics of the barren-ground caribou. These findings are based largely on studies by Kelsall (in press), Scotter (1964), Pruitt (1959), M c E w a n (1963) and Loughrey (1952) carried out between 1950 and 1962. In m a n y cases results are preliminary and more intensive research both in the field and laboratory is required. The barren-groundcaribou of the central mainland are both migratory and nomadic. Between the taiga winter range and tundra calving grounds they make long, uninterrupted and directionally orientated movements twice each year. These migratory periods total three to four months of the year and m a y cover as much as 400 miles. During the remainder of the year the caribou are nomadic, constantly on the move, and often change direction unpredictably in response to local environmentalfactors. Caribou are seldom evenly distributed on the taiga winter range. Although certain areas appear to be favoured and support wintering animals each year, almost any area of suitable winter range m a y be occupied by animals in any year. However, caribou prefer specific calving grounds. Year after year they return to certain locations on the central northern
275
A.G. Loughrey and J. P. Kelsall
tundra. The calving areas have c o m m o n characteristics in that they are high, dry, and possess a more rugged relief than the surrounding country. In the central tundra four major calving grounds are k n o w n -the Coppermine Mountains, the coastal range of hills on either side of Bathurst Inlet, the elevated plateaux north and south of Beverly Lake, and the rugged granite hills of central Keewatin south of Chesterfield Inlet. In addition, isolated bands of cows m a y calve almost anywhere on hilly tundra country. The spring migration normally begins in midMarch with a gradual shift of animals, principally pregnant cows and their calves of the previous year, from southern forested ranges toward tree line. Depending on snow conditions the animals m a y remain near tree line from mid-April to mid-May. Bands of 20-200animals leave the tree line and head for the distant calving grounds 200-300 miles away. Typically, the bands travel in single file following the course of least resistance across chains of frozen, snow-covered lakes, eskers, or glacial ridges. W h e n conditions are favourable the animals keep moving throughout the long daylight hours, covering thirty or more miles per day. Deep, soft, or wet snow, open rivers and thin lake-ice impede travel. The leading bands of cows reach the calving area by the first week in June while bulls and non-breeding animals m a y still be leaving the forest hundreds of miles to the south. Parturition starts by the first of June and ends in late June, but most (70 per cent) of the calves are born between 9 and ~15June. At that time from 10,000 to 50,000 cows are assembled in each of the four major calving grounds. Caribou calves are precocious and when only a few hours old are able to follow the cows. W h e n parturition is completed bands of cows with their calves leave the higher ground for lower and greener pastures. During July and August the animals often disperse widely, sex and age groups intermingle, and nomadic behaviour is typical. Concentrations of up to 100,000 animals m a y build up at water crossings, only to disperse again over the tundra pastures. In late summer the animals drift southward toward tree line, often penetrating well into the forest. They then reverse direction and swing back out on to the tundra prior to the October rut. This late-summer movement is haphazard and lacks the m o m e n t u m and drive of the spring movement. Neither the triggering mechanism nor the purpose of this movement is understood. The rut takes place on the tundra and begins in early October. After the rut and usually coinciding with increasing frosts and snowfall the animals exhibit a strong migrational drive and move rapidly and directly into the taiga. Caribou occupy their winter range by December, and during the winter months they move through it in search of feeding areas. Densities of wintering animals vary from less
276
than one, to more than ten, per square mile. W h e n range and snow conditions are favourable,herds of up to 50,000 m a y remain more or less stationary for several months, individual animals in the area moving only in response to daily requirements. Their winter movements appear to be in response to snow conditions, areas covered by deep and packed snow being avoided. The foods and feeding behaviour of caribou and reindeer appear to be similar throughout their holarctic ranges. The s u m m e r or tundra range in northern Canada is apparently more than adequate to support the present caribou population. However, the forests of the taiga or winter range have been so badly damaged by fire as to limit a major increase in caribou numbers. On summer tundra ranges caribou move constantly. They graze, choosing a mouthful here and one there, stop feeding, and walk or trot a few steps before taking another mouthful. On the calving grounds the pregnant females and those with young calves frequent the higher and drier ground. Their diet is more restricted to lichens than is that of the males and nonbreeding animals, which seek out lower and greener pastures. These animals feed on the new growth of green grasses and sedges. The phenological progress of preferred number foods is as follows: green shoots of cotton-grass (Eriophorum spp.), mid-June; other sedges and grasses, and n e w leaves of glandular birch (Betula glandulosa), 1-10 July;then ground willows (Salix spp.) from 10 July. Throughout the summer, lichens of the genera Cladonia, Cetraria, and Stereocaulon are eaten as are the leaves of m a n y woody perennials such as Arctostaphylos, Vaccinium, L e d u m , Empetrum, and Rhododendron. Fungi are highly favoured and are actively sought (Loughrey, 1957). The winter feeding behaviour of caribou is adapted to obtaining food under snow. At that time most caribou are within the taiga. W h e n snow depth is not great caribou feed on exposed vegetation, pawing with their forefeet to expose additional vegetation. As snow depths increase a characteristic feeding behaviour is followed, in which the caribou, using their forelegs alternately,p a w crater-likepits through the snow. The animal feeds on the exposed vegetation at the bottom of the cone-shaped pit, then moves on a few paces and. digs another. These feeding craters seldom overlap. A large herd of animals moves slowly through the taiga, the individuals digging craters, feeding, and moving out on to the frozen lakes to rest and ruminate.Smaller herds tend to stay in a good area for several weeks, until the snow is extensively packed with feeding craters and trails (Kelsall,1957). The bio-energetics of obtaining adequate nutrition in a severe climate appear to impose certain thresholds on caribou with respect to snow depth, density, and hardness. As yet no detailed investigations of this
Ecology and population dynamics of barren-groundcaribou in Canada
phenomenon have been made. However,it appearsthat thresholds vary with size and nutritional level of the animal. This m a y explain why adult males travel farther into the taiga during the winter, since they are not only able to travel through deep snow more easily but also are able to satisfy their food requirements more readily. Studies by Pruitt indicate that ill obtain food through 60 c m of light adult caribou w snow (Pruitt, 1959). Crusted snow with a density of ill prevent them from reaching ground vegeta0.9 w tion, while a density of 0.2 to 0.4generally restricts animals to feeding in snow depths of 20 to 50 cm. The presence of granular snow in the under layer also affects feeding behaviour. A thick layer of granular snow causes the walls of the feeding crater to collapse at the base. In the taiga, snow depth, density, hardness, and granulation increase during winter. W h e n critical levels or thresholds are reached in an area, caribou either move on in search of more favourable areas or feed on arboreal lichens (Usnea spp. and Alectoria spp.). They m a y even move prematurely on to the tundra (Kelsall, 1957). Caribou generally feed only once or twice over a section of range each winter. In cold weather their crater digging and movements increase the hardness and density of the snow until thresholds are reached and the animals move on. Pruitt’s studies indicate that this threshold is reached when the number of craters exceeds 1,000 per hectare (Pruitt,1959). The above cycle of feeding activity promotes a rotation of winter pastures and ensures that an area of winter range is not overgrazed, an important behavioural pattern in view of the animals’ dependence on slow-growinglichens for food. Terrestriallichens form the most important component of the winter food of caribou, amounting to approximately 50 per cent of their diet (Scotter,1964). Fructicose lichens of the genus Cladonia are the most important. Other genera include Cetraria, Stereocaulon, and Peltigera. The arboreal lichens Usnea and Alectoria are taken occasionally. A large variety of woody plants, grasses, and sedges are also eaten in winter. These include plants of the genera Vaccinium, Empetrum, L e d u m , Arctostaphylos, Betula, Salix, Picea, Pinus and Alnus. Grasses eaten include the genera Deschampsia, Festuca, Calamogrostis, and Pou and sedge genera Carex and Eriophorum. Horsetail (Equisetum spp.) is a highly preferred
winter food. The stems of these plants often protrude above the frozen surface of shallow lakes. Caribou p a w through snow to reach them, often completely cleaning out a bed of Equisetum. This is in marked contrast to the usual more random feeding behaviour (Loughrey, 1952). Studies of the effects of forest fires on caribou winter range have been carried out by G.W.Scotter of the Canadian Wildlife Service. H e documented
the effect of forest fires on reducing both quantity and quality of forage in forests of various age classes (Scotter, 1964). Mature black spruce forest from 76 to 120 years of are produced an average of 940 pounds of air-dryforage per acre,while forests in the 1-10year class produced 230 pounds. The açcummulated production of high-value lichens varied from 246 to 1 pound per acre for mature and recently burned forests. Based on annual growth rates of 4.1 and 4.9 mm for Cladonia alpestris and C. rangifirina, Scotter estimated that nearly a century is required after a fire before these lichens reach former abundance. H e demonstrated that the rate of fire destruction of winter range in the last fifteen years has increased 3.1 times over the 1840-84period. Although forests have always been subject to destruction by fire, chiefly caused by lightning,increases in the rate of fire destruction can be attributed to increased h u m a n activity resulting from mining and settlement. As a result of our aerial surveys and field and laboratory studies of caribou over the past eighteen years much has been learned of the population dynamics and reproductive behaviour and physiology of barren-ground caribou. Yet m u c h remains to be confirmed and studied in detail. Working on a migratory species in .a remote arctic and subarctic environment has presented physical difficulties which have impeded progress. The following paragraphs outline the results of our findings based on investigations begun under Banfield in 1948 (Banfield, 1954). Early estimates of the numbers of barren-ground caribou contained m u c h speculation and romance and ranged as high as Seton’s 30 million animals (Seton, 1912). Kelsall has given an estimate for the population, before the advent of the white man, based on the size of the winter range, and its average sustainable carrying capacity. H e calculated the size of the productive winter range at 479,000 square miles (64per cent of the surface area). Based on a sustained carrying capacity of four to six caribou per square mile he estimated a primitive number of between 1.9 and 2.9 million animals (Kelsall,in press). Field studies in the period between 1920 and 1940, notably by Anderson, Hoare, and Clarke indicated that caribou populations of that time had decreased in number (Anderson, 1924; Clark, 1940; Hoare, 1926). Light, ski-equipped aircraft adapted to northern flying conditions became available in the late 1940s and made possible the first extensive aerial coverage of the herds. In 1948 and 1949 Banfield attempted the first systematic aerial survey of the entire mainland range. H e flew 32,000 miles of transects and recorded 359,000 animals. H e estimated a total population of 668,000 animals (Banfield, 1954). During the next five years biologists of the service carried out intensive aerial surveys of specific herds, and survey techniques were tested and refined. In 1955 another major survey was carried out. During
277
A.G. Loughrey and J. P. Kelsall
April and M a y the authors flew 44,000 miles of transects for a 10 per cent sample of the range, and recorded 39,000 animals (Kelsall and Loughrey, 1955). The total population was estimated at 279,000 animals plus or minus 20 per cent. Those results clearly indicated a considerable decrease in numbers (58 per cent) in the six years since the previous survey. At the same time that aerial surveys were flown, data on the size of annual calf crops were gathered for various herds. During the period from 1950 to 1956, winter and early spring calf crops ranged from 7 to 17 per cent of total animals or only 20 to 50 per cent of the potential annual increment. In 1957 an intensive two-year study of caribou ecology and population dynamics was initiated. A herd of 100,000 animals occupying the central part of the range was selected for study, and teams of biologists and technicians followed these animals through two complete annual cycles. By using light aircraft and camping gear the teams were able to study the animals on the ground during all seasons. Specialists in range studies, aerial photograph interpretation, diseases, physiology, and behaviour also took part in the study, which yielded m u c h new information. I shall summarize our results concerning male and female reproductive physiology; herd structure; parturition and productivity; mortality; and the rut. Laboratory examination of reproductive organs of both sexes and various age classes and field observations of the study herd indicatedthe followingpoints. 1. Males become sexually mature at 17 months, but do not become successful breeders until 4-5 years of age (McEwan, 1963). 2. Females become sexually mature at 17 months and remain fertile throughout their life span (up to 16 years) (McEwan, 1963). 3. Only 50 per cent of 2- and 3-year-old females become pregnant (McEwan, 1963). 4.Between 80 and 85 per cent of 4-year-oldand older females produce calves (McEwan, 1963). 5.The percentage of barren mature females is very small (less than 5 per cent) in normal rutting conditions (McEwan, 1963). 6. F e w cases of abortion of term foetuses were recorded. From observations of the animals on the calving ground, during two successive springs, w e obtained the following information on herd structure,parturition and productivity. 1. Sexual segregation is greatest just before and during the calving period,but there is no clear-cut sexual segregation during the annual cycle. Small numbers of mature bulls m a y be found with cows at almost any time of the year. 2. The sex ratio af adults varies from 40 to 60 males per 100 females. 3. Mature females, prior to calving, constitute 40 to 50 per cent of the herd.
4.Calving occurs throughout the month of June with 70 per cent of all calves born during a six-day period centred around 12 June. 5. No occurrence of twinning was recorded. 6. The sex ratio of new-born calves was 106 males per 100 females. 7. Potential calf crop (productivity) was 35-39 per cent of the herd.
8.Highest observed productivity in July was 27 per cent.
9.Barren cows and those that have lost calves tend to join the predominantly male bands during early summer. 10. Calves stay with their mothers during their first winter and tend to accompany the cows to the calving area. Female yearlings tend to remain associated with the maternity and nursery bands, but male yearlings leave and form small juvenile bachelor bands or join adult bulls and nonbreeding animals in the lower pastures. Banfield (1955) constructed a preliminary life-table for caribou indicating an average life expectation from birth of 4.1 years. In our study w e attempted to measure mortality and identify causes. W e found that: 1. Observed normal mortality of calves during their first year ranged from 50 to 70 per cent; in exceptional years mortality was even higher. 2. Greatest calf mortality took place during the first week of life and resulted from cumulative stress (heat loss) due to a combination oflow temperatures, high winds, and precipitation (Hart et al., 1961). 3. Water crossings (lakes or rivers) are believed to be an important cause of mortality during the first two months of life. 4.Conditions of deep, dense and hard snow on the winter range result in mortality, particularly to animals in their first and second winter. 5. The higher rate of mortality for males as indicated by the adult sex ratio, is believed to result from a combination of selective hunting pressure on adult males and a high mortality rate for males during the first two years. It is suggested that young males have a higher metabolic rate and are more subject to mortality from stress induced by insect harassment, water crossing, and the search for food in winter. 6. During the period from 1950 to 1955 observed average annual increment was 14.5 per cent, but h u m a n utilization often exceeded 20 per cent. 7. Extrapolations of natality and mortality for the study herd supplementedby aerial surveys indicated that the mainland caribou population underwent a further decline of 28 per cent between 1955 and 1962. The latest estimate (1962) was 200,000 animals. Before the intensive field study biologists had very little opportunity to study caribou during the rutting
Ecology and population dynamics of barren-groundcaribou in Canada
season. At that time “freeze-up” occurs on the tundra and travel becomes extremely difficult. Observations made during two successive rutting periods indicated that : 1. Rutting occurs from early October to mid-November with the peak of activity in late October. 2. Young males and females attain breeding condition latest and are the last to leave the rutting area on the tundra. 3.Bulls do not assemble and protect discrete harems, but one or more attach themselves to and travel with mixed bands of cows and their calves. 4.Bulls establish dominance over each other by assuming threatening postures, vocalizing (grunts), and by engaging in sparring and shoving contests with their antlers (Pruitt, 1960).
At the present time both laboratory and field studies of caribou are being conducted by the Canadian Wildlife Service. M c E w a n is engaged in a laboratory study of caribou nutritional physiology. Range and food studies initiated by Scotter‘are to be continued by D.R. Miller. This spring an intensive two-year study of the caribou of northern Manitoba and southern Keewatin began under the leadership of A. H. Macpherson. Objectives of that study are to determine movements, numbers, age and sex composition of the herd, and reproductive potential and success, nutritional state, and causes of mortality for various sex and age classes. It is hoped that ill provide the basis information from this study w for a regional caribou management plan.
Résumé Le caribou des toundras du
Canada
(A.G. Loughrey)
Depuis 1948, le Service canadien de la fafine étudie l’écologie et les variations de l’effectif des troupeaux de caribous [Rangifer tarandus groenlandicus (Linné)] qui vivent dans les toundras du nord du Canada. On évaluait autrefois à 2 ou 3 millions le nombre total de ces animaux. Des enquêtes aériennes effectuées en 1949 et 1955 ont indiqué respectivement des populations de 668 O00 et de 279 O00 têtes. Cette diminution est principalement due au nombre excessif des animaux tués par l’homme et à la destruction par le feu des pâturages d’hiver. Les accouplements ont lieu à la limite de la zone boisée vers la fin octobre, et la parturition dans la toundra, vers le milieu de juin, aux endroits traditionnels. Les animaux hivernent dans la taïga boisée, où ils se nourrissent surtout de lichens du genre Cladonia. L’épaisseur, la densité et la dureté de la couche de neige influent sur la quantité de lichens disponible, et donc sur les mouvements des troupeaux à travers les pâturages d’hiver.
A la naissance,on compte 106 mâles pour 100 femelles, mais le nombre des femelles adultes dépasse celui des mâles dans la proportion de 5 à 3. Les femelles deviennent sexuellement adultes à 17 mois, mais 50% seulement des femelles de 2 à 3 ans sont fécondées, contre 80 à 85% pour les femelles plus âgées. Les mâles sont sexuellement adultes à 17 mois, mais ne deviennent que plus tard de bons reproducteurs. Les femelles adultes pleines représentent 50% de l’effectif du troupeau ; en juillet, les nouveau-nés peuvent constituer 25% de cet effectif, mais il en meurt en moyenne 50% pendant la première année. Parmi les caribous adultes, la mortalité est due 2t la chasse, aux accidents,aux animaux prédateurs et aux maladies. L’étendue des forêts soumises au brûlage dans les pâturages d’hiver a triplé au cours du siècle dernier. On n’a pas encore évalué avec précision dans quelle mesure les changements écologiques résultant de cette pratique influent sur l’effectif du troupeau que la région peut nourrir.
Discussion T.AATI.I would like to point out that the Canadians know more about their caribou than we know here about our tame, domesticated reindeer in many respects. This is probably due to the fact that particularly in Finland and Sweden the people working on animal husbandry have neglected the reindeer until recent years, regarding it as too wild an animal whilst zoologists have regarded it as a domesticated animal. This is a typical case of a field of study lying
however, between two branches of science. In the U.S.S.R. the people have studied the reindeer thoroughlv. I believe that the study of reindeer industry will become much more important in the Scandinavian countries;but there is one difficulty:the language barrier. So much work has already been done in the U.S.S.R.and in Canada that we should not have to start from the very beginning, although this would mean that w e should all have to know Ruasian !
279
A. G. Loughrey and J. P. Kelsall
Bibliography
/ Bibliographie
ANDERSON, R.M. 1924. The Fresent status and future prospects of the larger mammals of Canada. Scottish Geog. Mag., vol. 11. BANFIELD, A. W . F. 1954. Preliminary investigation of the barren-ground caribou. Wildl. Mgmt. Bull., Ser. 1, no. 10A and 10B. (Canadian Wildlife Service.) --. 1955. A provisional life table for the barren-ground caribou. Canadian J. Zool., vol. 33. CLARKE, C. H.D. 1940. A biological investigation of the Thelon Garne Sanctuary. (Nat. Mus. of Canada, Biol. Series No. 25, Bull. 96.) HART, J. S.; HEROUX, O. ; COTTLE,W . H . ; MILLS(C. A. 1961. The influence of climate on metabolic and thermal responses of infant caribou. Canadian J. Zool., vol. 39. HOARE, W.H.B. 1926. Report of investigations affecting Eskimo and wildlife, District of Mackenzie, 1924-1925-1926. Canada Dept. Int., N.W.T.and Yukon Br. KELSALL, J. P. 1957. Continued barren-ground caribou studies. Wildl. Mgrnt. Bull., Ser. 1, vol. 12. (Canadian Wildlife Service.) --. 1960. Co-operative studies of barren-ground caribou 1957-1958. Wildl. Mgmt. Bull., Ser. 1, vol. 15. (Canadian Wildlife Service.) . (In press.) The migratory barren-ground caribou of I
280
Canada. (Canadian Wildlife Service, Monograph Series.)
_-. , LOUGRREY, A.G. 1955. Barren-ground caribou resurvey 1955. (Canadian Widlife Service Report C. 277.) LOUGHREY, A. G. 1952. Caribou winter range study 1951-52. (Canadian Wildlife Service Report C. 86.) . 1957. Interim report, caribou calving studies, Feld party II, June 10 to July 13, 1957. (Canadian Wildlife Service Report.) M C E W A N , E.H.1963. Reproduction of barren-ground caribou Rangifer tarandus groenlandicus (Linnaeus) with relation to migration. Ph. D. thesis, Dept. of Zool., McGill University, Montreal. PRUITT, W.O.,Jr. 1959. Snow as a factor in the winter ecology of the barren-ground caribou (Rangifer arcticus). Arctic, vol. 12, no. 3. . 1960. Behaviour of the barren-ground caribou. (Biol. Papers, University of Alaska, no. 3.) SCOTTER, G.W. 1964, Effects of forest &es on the winter range of barren-groundcaribou in northern Saskatchewan. ’ Wildl. Mgmt. Bull., Ser. 1, vol. 18. Canadian Wildlife Service. SETON, E.T. 1912. The Arctic prairies; a canoe journey of 2,000 miles in search of caribou. New York.
_-
__
Animal activity patterns under subarctic summer conditions V. A. Peiponen
Most terrestrial animals can be classified as dayactive or night-active,even when some species change their activity with the season. Thus the food habits of most passerine birds are strictly limited to the light period of the day, varying according to its length, whereas microtine rodents are on the move mainly, or only, during the dark period. The difference of light between night and day in the arctic and subarctic summer conditions is slight. Even in these conditions all the studied species of birds have been observed to have a daily period of activity interrupted by a rest period (see Haviland, 1926; Palmgren, 1935; Dunajeva and Kutscheruk, 1941; Franz, 1949; Karplus, 1952; Remmert, 1965). However, conceptions differ widely concerning the factors that control this rhythm. The daily rhythm of the small rodents in arctic and subarctic conditions has hardly been studied (e.g. Hansen, 1957; Pearson, 1962; Peiponen, 1962; Myllymaki et al., 1962). In this paper the daily activity of some subarctic passerines and microtine rodents will be dealt with mainly on the basis of the investigations carried out at Kilpisjärvi Biological Station (690 N.,20050' E.) in the summers of 1954-58and 1964-65.In the lighting conditions of the region,the continuous day, when the sun does not disappear below the horizon, goes on for over two months starting on 21 M a y and ending on 22 July. The so-called civil twilight, in which the centre-point of the sun does not fall more than 6 degrees below the horizon, still lengthens the light period by about two weeks at both ends (Fig. 1). In the middle latitudes a number of day-active animals are capable of activity both in the morning and evening in this kind of civil twilight. In spite of the proximity of the sea (50 k m from the Arctic Ocean) the climate at Kilpisjarvi is continental, because the region is situated east of the Scandinavian mountain range. During the summer
when these studies were carried out the average temperatures varied between the following limits : Month
Min.
May June
-1.50 C 3.80 C
July
8.80 C 7.40 c 2.90 C
August September
Mair.
2.30 C 6.80 C 12.80 C 10.90 C 5.60 C.
So the average temperature of the warmest month; July, is fairly low, which for homoiothermic birds and voles means a great consumption of energy to maintain the body temperature. The distribution of the average daily temperatures daring the s u m m e r months is shown in Figure 2. The studies that will be dealt with are only concerned with food and feeding activity of both birds and rodents, which has been obtained by means of various automatic recorders. It is, however, to be noticed that during the incubation period of the birds, the feeding breaks in fact include an active phase, incubation,which has its o w n influence on the daily routine.
FIG. 1. Luminosity on a clear day, 21 June in northern Finland (700 N, continuous line) and in southern Finland (600 10' N.,Helsinki, dotted line). From Lunelund, 1935.
281
V. A. Peiponen
I
I
-I
10-
IO
6
4
l
l
i
l
04
l
I
I
I
O8
1
I I
I I I I I I I 12 16
I
I
I
I
I
20
5
MT FIG.2. The hourly averages of temperature ("C) of different months in northern Finland (68036' N.,Ivalo, continuous line) and in southern Finland (Helsinki, dotted line). The
IO
information from Ivalo according to the unpublished observations by Dr.Aili Nurminen. 1 = 1959, 2 = 1961. MT = meridian time. Note the differences in June and July caused by the continuous sun radiation in north.
IO
5
5 04
oa
12
16
20
MT FIG.3. Average incubation activity of some passerine birds in June at Kilpisjärvi. Ordinate: recorded trips/houron average. A = Phoenicurusphoenicurus1957,9 days,B = Luscinia svecico 1955,9 days, C = L.svecica 1956,6 days, D = L.svecica 1958, 4 days, E = Phylloscopus trochilus 1956,8 days,F = P. trochilus 1964,7 days, G = Motacilla Pava 1964,4 days, H = Carduelis parnrnea 1957, egg laying, I = C. JEarnrnea 1957, incubation, 8 days, J =C.jarnrnea 1958,incubation in late July,7 days.
282
Animal activity patterns under suharctic summer conditions
RESULTS ON THE DAILY RHYTHM OF PASSERINE BIRDS Both incubation and feeding have been observed during a number of successive days in the field by means of the simple recorders installed on the nests, working on a 4.5 V battery and by contact only. The recorders registered the beginning and the end of the birds’ activity without disturbing them and the quantitative distribution of the activity in each 24-hourperiod. Altogether I have successfully observed the incubation period in five species of birds, during the feeding period of the young in ten. In addition to this I have also made use of the unpublished results, which have been obtained by Dr. R. G.B. Brown while studying some insect feeders in Finnmarken (69042’ N.,29025‘ E.), which he has kindly let m e have, as well as the information already published about the daily rhythm of the arctic and subarctic passerine birds during the periods mentioned above. INCUBATION PERIOD
The incubation period of most passerine birds occurs
in Lapland in a season when the difference of the light conditions of day and night is at its slightest (Fig.3). In spite of this the continuous incubation time (i.e. the continuous food lull) of certain passerine birds occurs here, too, remarkably symmetrically around midnight, corresponding in time to the change of lighting rather than that of temperature. This type is represented among others by the willow warbler (Phylloscopus trochilus). However, the influence of temperature on the incubation rhythm of this species is obvious. This is seen when the length of the incubation time in Lapland is compared to that in southern Finland. According to Kuusisto (1941)the continuous incubation in southern Finland goes on for about 4-5hours, whereas in Lapland, where the temperatureis from 50 to 70 C lower in June and July, the continuous incubation requires a considerably longer time, something like 7 hours. The bird seems to compensate its longer nightly feeding break by undertaking in daytime a greater number of shorttermed food visits in Lapland than in southern Finland. Next to the type of the willow warbler are the redstart (Phoenicurus phoenicurus), and the yellow wagtail (Motacilla java). Unlike these three the bluethroat (Luscinia svecica) again seems to have a continuous food rest and, at the same time, a long continuous incubation time clearly in the early morning hours, around the daily minimum of temperature. Its whole daily incubation and feeding rhythm is parallel to the changes of temperature. T h e daily rhythm of the redpoll (Carduelisj a m m e a ) represents a type of its own. The continuous feeding
break occurs in the broods of the early summer mainly late at night, ending about 1 a.m. clearly before the daily minimum of temperature. This is understandable on the ground of the special incubation habits of this species: the female is incubating almost constantly while the male is feeding it. Thus the redpoll is during its incubation freed from control by the external temperature. The activity curve illustrates chiefly the activities of the male. The redpoll is sensitive to the changes of the illumination : when studying the late broods at the end of July, when the nights are already growing darker, it has been found out that the feeding rest occurs fairly evenly around midnight. The fact that the continuous feeding rest lies in the late evening in the early summer is probably due to the physical fatigue of the bird, as its continuous activity period already begins after
midnight. T H E F E E D I N G P E R I O D OF T H E Y O U N G
W e noticed that during the incubation period the continuous feeding rest is, in the willow warbler for instance, considerably longer in Lapland than in southern Finland. W e concluded that this fact is connected with lower external temperature, making it necessary in Lapland to have a long, continuous incubation time at night. As to the feeding period of the young the case is reversed. The continuous feeding rest of the willow warbler in Lapland is considerably shorter,from 3 to 4 hours,than according to Kuusisto (1941) and, in southern Finland, from 5 to 6 hours. Most of the other species under study have, in the condition of continuous summerlight in Lapland, a considerably shorter daily resting period (from 3 to 5 hours) during the feeding, compared with what is k n o w n about the rest period of passerine birds farther south (Fig.4). Remarkable differences between species are to be seen in the timing of the daily feeding rest. W e can divide the investigated species into three main groups, as during the incubation period. Group 1. The feeding rest of certain species occurs symmetrically around midnight, suggesting that the rhythm is chiefly controlled by the light factor. A m o n g the primary seed feeders studied species belonging to this group are: Carduelis jamrnea and Fringilla montifringilla but not Ernberiza citrinella (Franz, 1949)and E.sehoeniclus, whose feeding break occurs mainly in the late evening. Of the twelve insect feeders studied, only three, Motacilla alba, M. j a v a and Phylloscopus trochilus, belong to this group. Group 2.The majority of the insect feeders studied as well as the majority of all species studied have their feeding rest concentrated or limited to the early morning hours, around the daily minimum of temperature, when the insect food, broadly speaking,is most difficult to get. This kind is represented by Phoenicurus
283
V. A. Peiponen
I
5-
I I
04
O8
20
12
B
15io :
-
:c-
-
I5 -
7 7
25 -
08
20 -
IO i5k
I
c
I
25 15k 10525 L
20-
-
E
-
15-
-
c -
-
Y
10-
5-
04
08
12
i6
20
r,
MT
20
15 IO 5
i
I
L 25 20 15
IO 5
04
08
i2
i6
20
FIG.4. Average feeding activity of nestlings in some passerine birds at Kilpisjärvi in July. (a) A = Carduellis JEamrnea 1957, 11 days, B = C. JEammea 1958, 3 days, C = Fringilla mont$ringilla 1958, 5 days, D = Emberiza schoeniclus 1958, 5 days, E = Turdus musicus 1957, 8 days. (b) A = Phoenicurus phoenicurus 1958, 5 days. B = Oenanthe oenanthe 1958, 6 days, C = Luscinia svecica 1954, 6 days, D = L. svecica 1955, 11 days. phoenicurus, Luscinia svecica, Oenanthe oenanthe and Anthus pratensis belonging to the species I have studied, with Saxicolu rubetra (Lennerstedt, 1964), as well as the swallows Delichon urbica (Lind, 1960), Riparia riparia (Brown, in litt.) and flycatchers, Ficectula hypoleuca (Brown, in litt.) and Muscicapa striata (Franz, 1949), which all belong to the most
typical insect feeders. Group 3. As to its daily routine, the redwing (Turdus musicus) differs from all the birds mentioned earlier. Its feeding rest occurs before midnight as Swanberg (1951) and Brown (1363) observed in Lapland too. The feeding activity is at its highest during the hours immediately after midnight, exactly at the same time as many insect feeders have complete feeding rest. The highest stage of activity (again, after Brown) corresponds to that of the food animals of the redwing. The main food of this species in summer consists of earthworms, snails and different terrestrial larvae, in other words largely nocturnal animals, which in continuous light are obviously most active when the relative humidity of the air is highest, around the minimum temperatureinthe early morning. The fieldfare (T.pilaris) has the same kind of daily routine in subarctic regions (Brown, 1963). The following features have been observed in the
284
daily rhythm of the subarctic passerine birds during the breeding period : 1. Continuous light tends to lepgthen the daily food and feeding activity period. However all the species studied have a fixed continuous feeding rest in each 24-hour period. 2. During incubation the low external temperature has the opposite effect, it tends to shorten the daily food activity period, and it makes it necessary to have a long continuous incubation time during or near the daily minimum of temperature. 3. The light conditions in the subarctic early summer and in the middle of the summer obviously do not prevent diurnal birds from eating at any time of the day and so their feeding rest generally does not correspond to the minimum light period of the day. In m a n y cases it seems to occur in each passerine at a time when its chances of getting food are least favourable.
RESULTS ON THE DAILY RHYTHM OF MICROTINE RODENTS As a kind of comparison I would like to present some results on microtine rodents, which as a group represent nocturnal type for their daily rhythm.
Animal activity patterns under subarctic summer conditions
20 15 IO 5
20 i5
25
-
---
- -
7
7
-
10-
I
I-
rill1
20
I I III I
151
16
20
24
MT
04
08
FIG.5. The daily activity rhythm of voles at Kilpisjfirvi during summer. (a) Clethrionomys rufocanus, male. A = June, B = July, C = August, D = September. (b) Microtus agrestis,a pair. A = June, B = July,C = August.
(c) Clethrionomys rutilis. A = June/July, one male, B = August, two males, C = September, two meles.
As is well k n o w n and as can also be seen from the following figures, the day for these animals is not divided in one activity period and one rest period as is the case with passerine birds. A short-termed basic rhythm, the so-called hunger-cycle of say from 2 to 3 hous, which means an interchange of feeding and sleeping periods independent of the time of day, is characteristic of microtine rodents (Miller, 1955; Saint Girons, 1960; Pearson, 1962). Normally the feeding periods of the voles at night are, however, more marked and the sleeping periods less marked
than in the day and so a daily rhythm is formed, ill be dealt with here. which alone w The studies have been carried out with the voles caught in the Kilpisjärvi area during the summers of 1957, 1958, 1964 and 1965. The cages of the animals were kept out of doors from June to September. Each cage was divided into a small feeding area and a large sleeping area, between which there was a stepping-board connected with a Siemens recorder. The number of the trips have been combined to represent the hourly average monthly.
285
Y.A. Peiponen
T w o Microtus species (M.agrestis and M. oeconomus) and two Clethrionomys species (C.rufocanus and rutilus) have been the main subject of study (Fig.5).
It is a feature,c o m m o n to these four voles, that night activity is clearly accentuated by the end of July and beginning of August, when the first dark nights come. It is first seen at the beginning of August as an increase in morning activity, which gradually, later in this month and in September, occurs before and after midnight and is at the same time still increasing. In continuous light conditions both Microtus species again distinctly differ from both Clethrionomys species for rhythm. The peak of the activity of Microtus n o w occurs in the early morning hours from midnight to 6 a.m., at a time when the daily temperature is at its lowest and the relative humidity of the air at its highest, whereas the activity of Clethrionomys in continuous light was arhythmically divided into a 24-hour period. The activity of Microtus and Clethrionomys species living in the middle latitudes clearly occurs in summer mainly during the dark period of the day (Calhoun, 1945; Miller, 1955; Ostermann, 1956; Saint Girons, 1960, 1961;Erkinaro, 1961). In addition to these subarctic rodents mentioned above the daily migration rhythm of the Norwegian lemming ( L e m m u s lemrnus) has been studied in the field at Kilpisjärvi. There is a sharp distinction in the rhythm between the two annual migration periods. During the autumn migration, which was studied in August, movements were limited to the time around midnight, and the migration was obviously nocturnal (Myllymäkiet al., 1962). During the spring migration, which was observed at the end of M a y and at the beginning of June, Norwegian lemmings were moving
between early morning and noon, with a peak from 6 to 9 a.m. (Koponen et al., 1961;Ah0 and Kalela, 1966). With this species the registering studies are not completed yet, but results reached suggest that the food activity in June and July is arhythmically divided into a 24-hour period, m u c h in the same w a y as with the Clethrionomys species.
CONCLUSIONS
Of the eighteen species of passerine birds and five species of microtine rodents studied in subarctic areas in three rodents only the feeding activity during the light summer period is arhythmically divided into a 24-hour period. All the others have a clear daily rhythm in food activity during the summer. C o m pared with the areas in which there is an interchange of the light and dark period during the summer, there are greater differences between the species in the daily rhythm in diurnal birds living in subarctic areas. The continuous feeding rest occurs in certain passerines symmetrically around midnight, in some in the late evening, and in others (most species studied) towards or only in the early morning hours. As is to be expected the light does not hold the same dominating position as a factor regulating the daily rhythm in subarctic summer conditions as farther south. Besides light, the external temperature, and probably, the relative humidity of the air too, regulate the daily activities, as ultimate factors, largely by means of food. The same kind of adaptation to subarctic conditions can also be seen in some nocturnal animals (microtine rodents) in the form that the peak of their activity period does not occur at midnight in the light summer.
Résumé Organisation de l’activité des animaux en fonction des caractéristiquesde I’été subarctique (V.A.Peiponen)
L a façon dont les animaux adaptent le rythme quotidien de leur activité à l’absence de nuit qui caractérise l’été dans les régions subarctiques ressort surtout d’une comparaison entre le comportement de deux groupes de vertébrés appartenant des types opposés en ce qui concerne la période pendant laquelle ils cherchent leur nourriture et celle pendant laquelle ils se reposent :les passereaux diurnes et les rongeurs nocturnes du type Microtus. D e s recherches sur ce point, visant à compléter des enquêtes plus anciennes,
286
ont été effectuées en Laponie finlandaise,à la station biologique de Kilpisjärvi (600 N), où le soleil reste au-dessusde l’horizon pendant deux mois. Malgré l’absence d’obscurité, les oiseaux subarctiques se reposent sans interruption pendant des périodes fixes, mais la longueur du repos varie selon le stade du cycle de reproduction, et le choix de la période selon l’espèce. Pendant l’incubation, en ce qui concerne Carduelis jlammea, Motacilla java, Phylloscopus trochilus, Phoenicurus phoenicurus et Luscinia suecica, la
période de repos dure de 7 à 8 heures en Laponie, alors qu’en Finlande du Sud (620 N), pour le Phylloscopus trochilus, par exemple, elle ne dure que 4 à
Animal activity patterns under subarctic summer conditione
5 heures. L’allongement de la période de repos dans les régions subarctiques est probablement lié aux basses températures extérieures, qui exigent une incubation plus intense. Pendant qu’ils abecquent leurs petits au nid, les oiseaux se reposent moins longtemps en Laponie que pendant la période d’incubation, la durée du repos variant entre 3 et 6 heures pour dix espèces étudiées et se plaçant, selon l’espèce, avant minGt, en partie avant et en partie après minuit, ou vers le matin.
Bibliography Ano, J. ; KALELA, O. 1966. The spring migration of 1961 in the Norwegian lemming, Lemmus lemmus (L.), at Kilpisjärvi, Finnish Lapland. Annal. 2001.Fenn., vol. 3, p. 53-65. BROWN, R.G.B. 1963. The behaviour of the willow warbler Phylloscopus trochilus in continuous daylight. Ibis, vol. 105, p. 63-75. CALHOUN, J.B. 1945. Twenty-fourhour periodicities in the animal kingdom, Part II, the vertebrates. J. Tenn. Acad. Sci.,vol. 20, p. 228-232; vol. 21, p. 208-216. DUNAJEVA, T.; KUTSCEERUK, O.1941.Material zur Okologie der Landwirbeltiere der Tundra des sydlichen Jamal. Mat. K. F. FI. Russlands, Neue Serie Zool., Sektion IV, p. 1-80 (In Russian.) ERKINARO, E. 1961. The seasonal change of the activity of Microtus agrestis. Oikos,vol. 12, p. 157-163. FRANZ, J. 1949. Jahres- und Tagesrhythmus einiger Vögel in Nordfinnland. 2.f. Tierpsychol.,vol. 6, p. 309-329. HANSEN, R.M . 1957. Influence of daylength on activity of the varying lemming. J. Mammalogy, vol. 38, p. 218-223. HAVILAND, M . D. 1926. Forest, steppe and tundra. C a m bridge. KARPLUS,M . 1952. Bird activity in the continuous daylight of the arctic summer. Ecology, vol. 33, p. 129-134. KOPONEN, T.; KOEKONEN, A.; KALELA, O. 1961. On a case of spring migration in the Norwegian lemming. Ann. h a d . Sci. Fenn. Ser. A, vol. 52, p. 1-30. KUUSISTO, P. 1941. Studien uber die Ökologie und Tagesrhythmik von Phylloscopus trochilus acredula (L.)mit besonderer Berücksichtigung der Brutbiologie. Acta Zool. Fenn., vol. 31, p. 1-120. LENNERSTEDT, I. 1964. Några drag i håckningsbiologin hos lövsångare, buskskvätta och sävsparv i mellersta Lappland. Fauna o. Flora, p. 94-123.
Dans chaque cas,la période de repos semble coïncider la recherche de avec celle qui se prête le moins nourriture. En ce qui concerne les cinq espèces de rongeurs Microtus étudiées en Laponie, l’activité, interrompue par de courts repos, est répartie plus OU moins régulièrement sur les 24 heures tant que les nuits restent claires. Lorsque reviennent les premières nuits obscures, l’activité nocturne s’accroît et ne tarde pas à prédominer.
i Bibliographie LIND, E.A. 1960. Zur Ethologie und Okologie der Mehlschwalbe, Delichon u. urbica (L.). Ann. Zool. Soc. ‘‘Yanamo”, vol. 21, p. 1-123. LUNELUND, H . 1935. Die Helligkeit in Finnland. Soc. Sci. Fenn. C o m m . Physicomathemat.,vol. VIII, p. 1-42. MILLER, R. S. 1955. Activity rhythms in the wood mouse, Apodemus sylvaticus and the bank vole, Clethrionomys glareolus. Proc. zool. Soc. London, vol. 125, p. 505-509. MYLLYYÄKI, A.; AHO, J.; LIND, E.A . ; TAST, J. 1962. Behaviour and daily activity of the Norwegian lemming, Lemmus lemmus (L.),during autumn migration. Ann. Zool. Soc. “Vanamo”, vol. 24, p. 1-31. OSTERMANN, K.1956. Zur Aktivität der heimischer Muriden und Gliriden. Zool. Jahrb. Allg. Zool., vol. 66, p. 355-388. PALMGREN, P. 1935. Uber den Tagesrhytbmus der Vögel im arktischen Sommer. Ornis Fenn., vol. 15, p. 65-69. PEARSON A. M. 1962. Activity patterns, energy metabolism and growth rate of the voles Clethrionomys rufocanus and C. glareolus in Finland. Ann. Zool. Soc. “Vanamo”, vol. 24, p. 1-58. PEIPONEN, V. A. 1962. Zur Aktivität der Graurotelmaus, Clethrinomys rufocanus, im Dauerlicht des arktischen Sommers. Arch. Soc. “Vanamo”, vol. 17, p. 171-178. REMMERT, H. 1965. Uber den Tagesrhythmus arktischer Tiere. L. Morphol. Okol. Tiere, vol. 55, p. 142-160. SAINTGIRONS,M.-Ch. 1960. Le rythme nycthéméral d’activité du campagnol roux, Clethrionomys glareolus (Schreber) 1780. I. Les males. Mammalia, vol. 24, p. 516-532. .1961. Le rythme nycthéméral d’activité du campagnol roux, Clethrionomys glareolus (Schreber) 1780. II. Les femelles. Mammalia, vol. 25, p. 342-357. SWANBERG, P.O. 1951. Till kännedom o m vissa fåglar i Lappland. Fauna o. Flora, vol. 46, p. 11-136.
-_
287
Les Iles Saint-Pierreet Miquelon,une enclave subarctique méridionale E.Aubert de L a Riie
En l’absence de toute définition précise du domaine subarctique, divers caractères physiques, climatiques et botaniques m’ont incité à y inclure le petit archipel de Saint-Pierre et Miquelon, situé par 470 lat. N, proche des côtes méridionales de Terre-Neuve. I1 marque sans doute l’une des avancées les plus accusées de la zone subarctique dans le domaine des régions tempérées, auxquelles il devrait normalement appartenir du fait de sa position géographique. Cette anomalie tient à son climat exceptionnellement froid pour la latitude,influencé par les basses températures de la mer environnante. Ces îles de peu d’étendue, n’occupant que 240 km2, sont en effet baignées par le courant du Labrador, le plus réfrigérant du globe. Leur climat sévère, très venteux, est caractérisé par de longs hivers et de brefs étés sans chaleur. Au nombre de trois principales Saint-Pierre, ces îles ont une faible Langlade et Miquelon élévation, leur point culminant ne dépassant pas 250 mètres. L’altitude n’intervient donc pas ici pour aggraver le climat. Des plaines, des plateaux et des collines composent la topographie peu accidentée de l’archipel. Langlade et Miquelon, les plus étendues mais les moins peuplées, sont unies par une longue chaussée de galets édifiée par la mer, sur laquelle le vent ne cesse d’accumuler du sable et de construire des dunes en partie fixées naturellement par des ammophiles et autres plantes herbacées. C e sont là, avec d’autres cordons littoraux, des talus d’éboulis et les sommets de certaines collines, les terrains les plus secs de l’archipel. D’une nature très complexe,il est formé de terrains précambriens et paléozoïques très divers. D e s roches métamorphiques, accompagnées d’intrusions granitiques et autres,y voisinent avec des laves rhyolitiques et basaltiques accompagnant des sédiments cambriens (schistes, phyllades, quartzites). Les calcaires font toutefois défaut dans cet ensemble. L e quaternaire
-
-
est représenté par d’importants dépôts glaciaires, en partie étrangers aux îles et apportés de Terre-Neuve par l’inlandsis et par des formations littorales. Celles-ci isolent de la mer un certain nombre de lagunes saumâtres. Saint-Pierre et Miquelon ont été entièrement submergées par la glaciation pléistocène et en portent les marques manifestes. Outre les dépôts morainiques, parfois très argileux, ailleurs formés d’éléments grossiers, ce sont des vallons en forme d’auge, des surfaces moutonnées avec de nombreux étangs, d’autres où la roche est polie et striée. L a présence d’un permafrost est l’un des traits caractéristiques des régions subarctiques pour certains biogéographes, ceux-là m ê m e s qui associent souvent la notion de toundra à la présence d’un sol gelé ne se ramollissant que pendant une brève période estivale. D’autres, cependant, admettent l’existence de la toundra s u r des sols ne gelant que superficiellement au cours de l’hiver. Tel est le cas, pour les phytogéographes, de la toundra australe des îles subantarctiques. Les îles Saint-Pierreet Miquelon, de m ê m e que la côte terreneuvienne proche, qui leur est très semblable du point de vue écologique, n’ont plus actuel. lement de permafrost. C’est peut-être ce qui les dietingue le plus des terres subarctiques situées à de plus hautes latitudes. Elles sont néanmoins le lieu de certains phénomènes caractéristiques de ces régions, le plus notable étant l’éclatement des roches sous l’action répétée du gel et du dégel. I1 donne lieu, dans les parties dénudées, à la formation de champs de pierres anguleuses. Ces fragments rocheux ont parfois une tendance à s’orienter pour former des cercles et des polygones plus ou moins nets. Signalons également l’existence de buttes gazonnées. En dépit de la violence du vent l’érosion éolienne est peu active, du fait de l’humidité du sol. Ses effets les plus marqués s’observent sur certaines tourbières mortes, très exposées
289 21
E. Aubert de La Rue
au sommet des falaises. L a glace filamenteuse (pipekrakes) désagrège progressivement la tourbe que le vent laboure et disperse. Les données climatiques rapprochent beaucoup l’archipel d’îles occupant une position plus boréale. L a température moyenne annuelle de Saint-Pierre, son chef-lieu,est de + 5,60. Trois mois seulement dépassent 100 : ce sont juillet, août et septembre, ayant respectivement 13,90, 15’90 et 13’50. Trois mois d’hiver ont une température moyenne inférieure à O0 : ce sont janvier (-2’10)’ février (-2,70) et mars (-1,20). Lors des hivers rudes,la moyenne de janvier et de février tombe à -40. I1 est intéressant de comparer ces valeurs avec celles, fort semblables, de Reykjawik par 640 lat. N,donc à 170 de latitude au nord de Saint-Pierre, mais influencé par le Gulf Stream; ces valeurs sont les suivantes pour les trois mois les plus chauds : juin (9,60), juillet (11’30) et août (10,60), alors que les trois mois les plus froids ont et pour moyenne : décembre (00)’ janvier (-0’60) I1 en résulte que les étés de Saintfévrier (-0,20). Pierre sont légèrement plus chauds et les hivers sensiblement plus froids. L’amplitude thermique annuelle y est de 18,60. On compte, au total, 132 jours de gelées SaintPierre, réparties entre le 15 octobre et le 30 mai. Pendant quatre mois seulement (juin à septembre) le thermomètre ne s’abaisse pas au-dessous de OO. Certains hivers connaissent des minimums absolus de 200 à 220.Ils sont rares dans les années présentes, où les plus grands froids sont de l’ordre de 150. Les maximums absolus de juillet et août ne dépassent habituellement pas 23 à 250. L e long des côtes, la température de la mer est voisine de O0 en janvier et février, époque où il lui arrive de geler de loin en loin, emprisonnant l’archipel dans la banquise. En d’autres années, où l’océan ne gèle pas en cet endroit, la banquise en déiive, provenant du nord ou du golfe du Saint-Laurent,vient parfois, entre janvier et avril, s’amonceler pendant une durée variable autour des îles. L e réchauffement de la m e r est lent, sa température n’étant encore que de 50 2 60 en juin. L e maximum, variant entre 120 et 150 suivant les années, est en août. Sa température n’est plus que de 110 en septembre et de 80 en octobre, après quoi elle décroît rapidement. Les étangs intérieurs sont pris par la glace de la fin de novembre à la fin d’avril,les lagunes côtières de janvier à mars. Les précipitations sont abondantes (1 424 mm par an), réparties sur 186 jours et de façon assez régulière au long de l’année. I1 se produit néanmoins une légère diminution de fréquence et d’importance entre juin et août et un m a x i m u m en automne (octobredécembre). Les plus fortes pluies, amenées par des vents humides et doux du sud-est, sont rares mais peuvent avoir un caractère torrentiel, donnant u n total de précipitations de 60 à 90mm en 24 heures. L e couvert de neige est un facteur important du
-
290
-
-
point de vue écologique et il est considérable dans le groupe. Les chutes de neige s’observent au cours de 65 jours par an, le plus souvent entre novembre et avril, mais il peut s’en produire en octobre et en mai. J’ai été témoin, Ie 12 juin 1941, d’une chute tardive qui a blanchi le sol durant une journée. On cite, c o m m e tout à fait exceptionnelle, une abondante chute de neige survenue le 15 juillet 1901. I1 est assez habituel que l’enneigement persiste pendant près de 5 mois, de décembre à la fin d’avril. I1 arrive m ê m e qu’il débute dès la fin de novembre. En d’autres années, toutefois, la neige ne recouvre le sol d’une façon permanente qu’à partir de la fin de décembre. En raison de la violence et de la constance du vent, la couverture neigeuse est très inégale. Absente ou presque de certaines surfaces très exposées, elle forme ailleurs de puissantes accumulations, très longues à disparaître, persistant dans certains ravins jusqu’au seuil de l’été. I1 arrive qu’à Langlade la neige n’ait pas encore totalement disparu au mois d’août au niveau de la mer. L e temps est d’une extrême instabilité et la persistance du vent d’ouest est l’un des traits dominants du climat. L’humidité relative est élevée, se maintenant toute l’année entre 82 et 84%. U n e source d’humidité atmosphérique considérable est la brume, dont on compte 120 journées par an. Elle est particulièrement tenace au printemps et en été, de mai à août, et relativement rare l’hiver. L a végétation de l’archipel n’est pas ce qu’elle devrait être normalement sous cette latitude. L e climat froid lui imprime un caractère nettement subarctique et elle offre de grandes similitudes avec celle de régions franchement plus nordiques, avec le Labrador notamment, cela malgré l’absence de permafrost. L. Arsene a montré que Ia flore des îles Saint-Pierre et Miquelon était effectivement très comparable à celle de la région de Hamilton Inlet, à près de 1 O00 km plus au nord, et cela dans une proportion de 64%. Un autre botaniste, le père C. Le Gallo, excellent connaisseur de la flore insulaire de l’archipel, a mis en relief l’importance des éléments subarctiques dans la composition de celle-ci. Au cours d’une traversée de la péninsule d’Ungava, faite en compagnie du professeur Jacques Rousseau, le long du 60e parallèle, et de reconnaissances dans la toundra forestière des environs de Fort Chimo (580 N), j’ai retrouvé des aspects de végétation très comparables A ceux des îles envisagées.En l’absence de permafrost, on trouve à Saint-Pierre et Miquelon des tourbières à sphaignes très développées, qui diffèrent peu, par leur composition floristique, de la toundra nordique, des tourbières boisées fort semblables à la toundra forestière et la taïga caractéristique, composée principalement d’Épicéa. U n e remarque s’impose ici, qui a trait à l’action de I’homme, responsable de l’introduction de plantes étrangères et de profondes dégradations du couvert
Les îles Saint-Pierreet Miquelon,une enclave subarctiqueméridionale
végétal, à Saint-Pierre principalement et, dans une moindre mesure, dans le reste du groupe. D e s pêcheurs bretons et normands fréquentèrent celui-ci dès l’an 1504 et s’y installèrent de façon permanente un siècle plus tard. L a population présente dépasse 5 O00 habitants’ établis en majorité sur l’île Saint-Pierre, aujourd’huià peu près totalement déboisée et dénudée. Ailleurs, bûcherons et cultivateurs ont apporté en divers endroits de profondes modifications à la couverture végétale originelle. Ils ont heureusement épargné des étendues trop ingrates et d’accès malaisé au centre de Miquelon,mais surtout à Langlade où la couverture végétale demeure encore en partie intacte. L’aspect de la toundra est offert ici par les tourbières à sphaignes occupant les plateaux de Langlade et les plaines de Miquelon, où elles reposent sur des argiles à blocaux très puissantes par endroits. Ces tourbières, constamment détrempées, constellées d’étangs, se composent de diverses espèces de sphaignes, d’hypnacées, de carex, d’Eriophorum, accompagnées de nombreuses éricacées ( L e d u m groenlandicum, Rhodora canadensis, Andromeda glaucophylla). Les vacciniacées
(V.uliginosum, V. vitisidaea, V. oxycoccos, V. macrocarpon) ont une place importante dans ce milieu palustre où abonde également une rosacée arctique : Rubus chamaemorus.
A la surface des tourbières s’étalent fréquemment des tapis de lichens (Cladonia rangiferina) et aussi d’une épaisse mousse laineuse blanchâtre (Rhacomitrium lanuginosum). Des espèces ligneuses rampantes, communes dans les tourbières à sphaignes,sont représentées par des épicéa, des mélèzes (Larix laricina),des saules,des aulnes, et des bouleaux nains. L’épaisseur de la tourbe est variable, allant de 1 à 3 mètres. On observe parfois, dans les niveaux profonds, des souches d’arbres, dans les positiops les plus diverses, indiquant l’existence d’anciens boisements dont la disparition ne semble pas toujours imputable i l’intervention de l’homme. Ces tourbières à sphaignes gèlent à partir de novembre et l’action du froid se fait progressivement sentir jusqu’à 80 c m de profondeur. Leur surface se ramollit à la fin d’avril, mais il n’est pas rare d’y trouver encore de la glace, en juin, à une certaine profondeur. D e nombreux étangs, riches en plantes aquatiques, possédant une microflore et une microfaune abondantes,parsèment les tourbières. On peut se demander, mais ce n’est là qu’une simple hypothèse, si les excavations qu’ils occupent ne correspondaient pas, primitivement, après la disparition de l’inlandsis quaternaire, à des lentilles de glace attardées parmi la moraine de fond abandonnée par celui-ci. Ces tourbières retiennent d’importantes quantités d’eau qui, jointes i celle des étangs, alimentent de nombreux ruisseaux. L’eau provenant du drainage imparfait de ces marais est fortement colorée en brun par des hydroxydes de fer et de manganèse.
11 est souvent difficile d’établir une démarcation nette entre les tourbières à sphaignes et la forêt de conifères, ces deux types étant étroitement enchevêtrés, et passant insensiblement de l’un à l’autre par des tourbières boisées occupées par une curieuse forêt naine. L a persistance de vents violents fait que celle-ci revêt souvent une forme particulière. Les conifères qui la composent, des spruces, des sapins, éventuellement des genévriers, sont des arbres tortueux, rampants, entrelaçant leurs branches au point de former des fourrés impénétrables. On peut raisonnablement qualifier de taïga la forêt des îles Saint-Pierreet Miquelon, car elle est essentiellement formée de conifères : des spruces noirs (Picea mariana) en majorité, qu’accompagnent des spruces blancs (Picea glauca) et des sapins baumiers (Abies balsamea). Les pins font totalement défaut. Les rares mélèzes présents (Larir laricina), ne dépassant pas quelques mètres, préfèrent les clairières très humides tapissées de sphaignes. L a taïga dense fuit ici les plaines et les plateaux, recherchant les vallées et les pentes des collines où elle est établie indifféremment sur des sols tourbeux ou morainiques, généralement très humides les uns et les autres. Dans les lieux les plus abrités,au fond des vallons et le long des rivières, des feuillus, en petit nombre, se mêlent par endroits aux conifères. Ce sont surtout des bouleaux (Betula papyrifera, B. lutea), des sorbiers, un petit érable (Acer spicatum), des aulnes (Alnus mollis, A. rugosa), etc. Là où elle est le mieux conservée, dans les vallées de Langlade, la taïga, n’est jamais d’une belle venue et ses arbres sont généralement chétifs, malmenés par le vent. L a forte humidité du sol est sans doute aussi responsable de cet état de choses. Les spruces les plus robustes ne dépassent pas 10 à 12 mètres de haut. Nombreux sont les conifères mutilés, le tronc fendu ou les branches cassées chaque hiver par le poids excessif de la neige ou, ce qui leur est plus nuisible encore,par l’énorme surcharge du verglas. Le sommet des collines rhyolitiques, sans être très escarpé, est généralement privé de végétation arborescente et partiellement dénudé. Sur ce substratum sec et très siliceux prospèrent par endroits des plantes trouvant là des conditions écologiques favorables. C. L e Gallo a montré que ce milieu était particulièrement favorable à la persistance de certaines espèces arctiques et subarctiques dont il donne une longue liste. Les basses températures du printemps régnant dans l’archipel sont responsables du départ très tardif de la végétation. L a période végétative ne débute guère avant la première quinzaine de mai et prend fin en septembre, durant environ 150 jours. Petites et pauvres, soumises à un climat inclément et froid,les îles Saint-Pierreet Miquelon ne disposent que de ressources naturelles très limitées. Celles-ci s’apparentent néanmoins aux ressources plus n o m 291
E.Aubert de La Riie
breuses et variées dont disposent des régions subarctiques plus favorisées. L a principale, à laquelle ces îles doivent leur peuplement déjà ancien, est la pêche à la morue, sa préparation et son exportation. Des tentatives ont été faites, mais sans grand succès, en vue de la chasse aux phoques, dont un troupeau fréquente l’étang salé du Grand Barachois. Les boisements naturels, sans valeur industrielle en raison de leur exiguïté et de la médiocre qualité du bois, ont en grande partie disparu. L a taïga est en effet exploitée depuis longtemps pour les besoins locaux, principalement pour les nécessités du chauffage et divers usages domestiques,en particulier pour l’établissement de clôtures protectrices contre le vent autour des cultures et des champs drainés et aménagés. Des incendies imprudemment allumés ont d’ailleurs également contribué à réduire l’étendue de la forêt de conifères. L’étroitesse du cadre,la nature du sol et les rigueurs du climat limitent les possibilités d’élevage et de culture. Ces deux formes d’activité rurale ont pourtant connu dans le passé un sort meilleur que maintenant. Les fermes de Miquelon et Langlade élevaient des bovins, des moutons et une race de petits chevaux robustes. L e peu de chaleur de l’été ne permet pas aux céréales de parvenir à maturité. Quelques cultures (pommes de terre, choux, carottes et autres) sont possibles là où le terrain a été aménagé. Celle des arbies fruitiers est vouée à un insuccès total dans ces îles. Des liaisons fréquentes et rapides avec le Canada, permettant un ravitaillement facile en produits frais, ont amené depuis une quinzaine d’années le déclin et l’abandon de la plupart des fermes.
Un élevage que favorise le climat est celui des animaux à fourrure. Des ranches pour l’élevage du renard argenté, du ragondin et du vison furent créés il y a une trentaine d’années et fonctionnèrent avec succès, mais ils ont été fermés depuis peu. Abondante et de bonne qualité en tant que combustible et pour d’autres usages, la tourbe n’a jamais été exploitée. Elle aurait pu fournir une excellente litière pour le bétail, servir de matière isolante pour la construction des habitations et surtout de combustible, ce qui eût permis d’économiser le bois. On peut citer c o m m e autre ressource celle que fournit la chasse très active faite aux oiseaux migrateurs, qui passent en grand nombre dans ces îles. Certaines espèces s’y attardent en hiver, d’autres au printemps, pour nicher. Leur chasse est autorisée toute l’année sans restriction. Les tourbières à sphaignes fournissent enfin à la population des plantes médicinales et surtout d’abondantes récoltes de baies comestibles. Les plus recherchées sont les fruits de Rubus chamaemorus, Vaccinium vitis-idaea, V. oxyCOCCOS, V. macrocarpon, V. uliginosum et V. pennsylvanicum.
CONCLUSIONS L’anomalie que représente le petit groupe insulaire de Saint-Pierreet Miquelon est intéressante à évoquer. Suffisamment de particularités climatiques, écologiques et botaniques permettent en effet de le rattacher aux régions subarctiques malgré la latitude quelque peu excentrique à laquelle il est situé.
Summary The St. Pierre and Miquelon Islands, an enclave of the southern Subarctic (E.Aubert de la Rue) This small archipelago, situated at 470 N. off the southern coast of Newfoundland, probably constitutes one of the deepest incursions of the subarctic zone into the temperate regions, to which it should normally belong by reason of its geographical position. This is due to the exceptionally cold climate for the latitude, under the influence of the low temperatures of the surrounding sea (Labrador current) and the flow of icy air from the North Pole. The annual mean temperature of the atmosphere is 5.630 C, and only during three months (July, August, September) does the mean rise slightly above 100 C. Snowfall is irregular, due to the persistent violent winds,but in some years the islands remain snow-coveredfor nearly five months. Precipitation is heavy, with a mean figure of 1,424 mm a year. It is spread over 186 days, on 65 of which it snows. There are 132 days of frost. 292
The morphology of the St. Pierre and Miquelon Islands has been greatly influenced by the persistency of the Labrador ice-cap,which entirely covered them in the Quaternary, leaving behind a thick layer of clayey moraines, the cause of the impermeability of the soil. The vegetation is of two principal types: forest of subarctic conifers (taïga), in which Epicea predominate and pines are absent, and sphagnum peat bogs. The forest-at least all that exists today, after excessive deforestation by the inhabitants for various domestic purposes-has suffered from the incessant storms and is sparse and stunted. The peat bogs, studded with pools, are encroaching on the forest. No use is made of the peat. Certain crops can be grown, but the inhabitants are showing less and less interest in them. The same is true of stockbreeding, which is on the decline. The archipelago’s principal resource remains codfishing, as in the past.
Les îles Saint-Pierreet Miquelon, une enclave subarctique méridionale
Discussion W. PRUITT. Apparently St. Pierre and Miquelon are quite similar to the nearby Avalon Peninsula of Newfoundland. Knowledge of these islands is important to our understanding of the concept “Subarc,tic”. It seems quite impossible to include such areas as northern Alaska, south and central mainland of Canada and St. Pierre and Miquelon in the same category. Thus w e must modify our definitions to suit nature. I might mention that such maritime snow covers appear superficially as “subarctic”, but upon detailed study are found to be quite different. This shows another reason for better understanding of snow cover,with a view to eventually being able to classify snow types in a more reasonable fashion than n o w possible. I must note that in such areas as Newfoundland and St. Pierre and Miquelon, which have been inhabited by m a n for hundreds of years, historical factors are of primary importance. In m a n y cases the original vegetatjon was quite different from that existing today and can only be understood b y careful perusal of historical records. T. AHTI. Please excuse m y rather lengthy comments but I a m doing this because I a m not going with you to Kevo. Dr. Aubert de la Rue’s lecture on St. Pierre and Miquelon Islands was very interesting from the point of view of the delimitation of the Subarctic. I worked for a few weeks on the Burin and Avalon Peninsulas in adjacent N e w foundland, which are apparently very similar to those French islands. T h e eastern North American coast from Nova Scotia up to Labrador is ecologically quite comparable to most parts of southern Greenland, Iceland, northern Norway, the Kuril Islands and south-western Alaska, where the highly maritime lowland areas have stunted tree growth or are quite treeless. These are treated in very different ways by different authors. Bioclimatically and floristically, for example, they present both more or less arctic and more or less temperate features. They are climatically characterized by strong winds, cool summers, long growing seasons and mild winters when compared to the adjacent interior areas.
Bibliography ARSENE, Bro. L. 1927. Contribution to the flora of the islands St-Pierre et Miquelon. Rhodora, vol. 29, no. 343, p. 117-133 ; no. 344, p. 144-158 ; no. 345, p. 173-189; no. 346, p. 204-221. AUBERT DE LA RUE, E. 1951. Recherches géolo,’q 01. ues et minières aux îles Saint-Pierre et Miquelon. Paris, Ofñce de la recherche scientifique outre-mer, 75 p., XVIII pl., carte.
T w o of m y colleagues and I have recently proposed that these coastal areas should be regarded only as highly oceanic sections of the adjacent transcontinental circumpolar zones and snbzones rather than southward extensions of the arctic tundra. Thus w e would speak about boreal maritime heaths and grasslands, the northern parts of which are subarctic and represent the transition of the real tundra. Thus St. Pierre and Miquelon are boreal but the mountain tops, at least, are extremely similar to the Subarctic-in whatever sense this term is used. In this connexion w e have also concluded that the presence of the stunted birch woodlands-you will see them at Kevo-are essentially an oceanic phenomenon, characteristic of the total boreal zone in western Eurasia rather than solely a subarctic feature. For historical and climatical reasons the coasts are almost devoid of conifers. This means that w e consider the continuous birch woodlands to be taiga rather than forest tundra. Only the uppermost, irregularly open parts of the birch woodlands would be forest tundra and even that only as a vertical zone, because w e do not accept any horizontal arctic zone in Fennoscandia, including the Kola Peninsula. All is vertical zonation, although it is difficult to recognize physiognomically in m a n y areas.
AUBERTDE LA RUE. Je préfère, pour les diverses raisons exposées dans m a communication, quant aux grandes analogies de composition floristique entre les tourbières des iles Saint-Pierreet Miquelon et la toundra subarctique de m ê m e entre leur forêt de conifères et celle de la taïga subarctique -désigner c o m m e “subarctique atténué” le caractère de ces îles, de préférence à la désignation de “highly oceanic seotion of the adjacent transcontinental circumpolar zone”. E n fait, il ne s’agit pas d’une avancée extrême de la zone subarctique, mais plutôt d’enclaves de caractère subarctique atténué dans la zone tempérée boréale, dans le cas des îles Saint-Pierre et Miquelon et d’autres anomalies semblables, en relation avec un climat anormalement froid pour la latitude.
-
/ Bibliographie --. 1965. Les tourbières des îles Saint-Pierreet Miquelon. C. R. Soc. Biogéographie, Paris, séance du 18 novembre, no 372, p. 141-147.
LE GALLO, Père C. 1949. Esquisse générale de la $ore vasculaire des iles Saint-Pierre et Miquelon. 60 p., 23 illustr.
(Contributions de l’Institut botanique de l’université de Montréal, no 65, 31 octobre.)
293
Forests and forestry in subarctic regions Peitsa Mikola
SUBARCTIC FORESTS The subarctic vegetation zone is defined as a transition between the boreal coniferous belt and the treeless arctic tundra. This transitional nature is properly characterized by the term “forest tundra”. Thus, the northern part of the subarctic zone consists mainly of tundra, with scattered trees or small stands in protected localities. The proportion of wooded areas increases towards the south, treeless areas diminish, and tree stands become denser and higher. W h e n closed forests grow everywhere where edaphic conditions permit, the subarctic transition zone has been passed. According to this definition, the subarctic zone is taken in a broader sense, i.e. it includes both the forest tundra proper and the open boreal woodland region (Hustich, 1948; Hare, 1950). Approximately the same definition has also been used in m a n y other papers at this symposium, e.g. in those by Rapp, Tedrow, and Hustich. The northern boundary of the Subarctic against the Arctic is usually distinct; it is the tree line. The only difficulty sometimes is to find out whether the tundra is native or anthropogenous. “Man-made tundra”, of course, does not belong to the Arctic. On the other hand, the boundary between the subarctic forests and the boreal coniferous forests proper is hard to define precisely. It is essential that subarctic forests always bear some features related to tundra. These features are the more pronounced the closer the proximity of the real arctic tundra. The climatic factors responsible for the complete treelessness of the tundra also exert their effect in the subarctic region, although in slighter degree, leaving their marks on the forests. The subarctic zone can also be thought of as a battlefield of the forest and the tundra, the latter having the advantage in its northern half (forest tundra) and the former prevailing in the south.
Properties characteristic of subarctic forests, distinguishing them from the boreal forests proper, are low density of the stands and small height and slow growth of the trees. The thickness of the tree stems m a y be as large as in more southern regions, and the trees often attain old age, becoming m u c h older than in their main range in the boreal forests. Thus, for instance,the dominant height of a mature pine stand, which on an average pine site in the boreal forests (southern and central Finland) is 22-26 m , is only 12-14m in the subarctic forests of Lapland and even less than 10 m at the outermost tree limit. Subarctic and subalpine forests have m a n y features in common, as is natural, since both represent a transition between closed forests and treeless plant communities. Factors limiting the subsistence of forests are mainly the same in subarctic and subalpine regions and likewise the problems connected with the protection and utilization of the forests are largely the same. It is true that subalpine areas have, in addition, their specific problems, like avalanches and soil erosion, which do not exist on the subarctic lowlands. In the mountains, in general, the limits of the subalpine vegetation against both the alpine zone and the forests proper are quite distinct, and the whole subalpine belt between the closed forests and the treeless regio alpina is usually quite narrow, a few kilometres only, whereas the width of the subarctic zone between the tundra and the boreal forests is hundreds of kilometres and its limits hard to define in nature. If there are mountains in subarctic regions, there also exists the alpine forest limit and even large treeless alpine areas. Towards the north the elevation of the alpine tree line decreases continuously. Thus, the tree line, which in the Central European mountains is found at an elevation of 1,500-2,000m , has sunk to approximately 400 m at the southern edge
295
P. Mikola
of the subarctic zone, and in northernmost Scandinavia is less than 200 m above sea level. The number of tree species in the subarctic forests is limited. According to various sources (Hustich, 1952a; 1966; Fowells, 1965, and others) the northern limit of coniferous forests is formed by Larix laricina and Picea glauca and P. mariana in North America, by Larix sibirica and L. gmelini in Asia, by Picea obovata in the European Soviet Union, and by Pinus silvestris in Scandinavia. The dwarf conifers Pinus pumila and Juniperus communis, which grow on forest tundra as low bushes, can be omitted, since they do not attain tree size. In northern Fennoscandia the range of a broad-leaved tree, Betula pubescens ssp. tortuosa, extends beyond the conifer limit, and there a zone of low open birch forest is characteristic of the forest tundra. Subarctic birch and aspen forests also grow in Iceland, southern Greenland, Kamchatka and some other areas. In general, however, conifers are the principal or only tree species of the subarctic forests. A rough estimate of the total area of subarctic forests in the world m a y be attempted. In northern Fennoscandia the width of the subarctic belt is 200-300km.In Labrador, according to Hustich (1949) and Hare (1950),the width of the corresponding zone (forest tundra plus open boreal woodland) amounts to 800 km and exceeds 1,000 k m in eastern Siberia (cf. the m a p of Tedrow in this symposium). Between these limits the width of the subarctic zone varies in both the old and the n e w continent. Tunstell (1956) writes on the Canadian forests: “Because of adverse climatic or edaphic conditions, more than two-fifths of the country’s total forested is as area is classified as non-productive-that land incapable of producing crops of merchantable timber.” Although the above non-productive forest area also includes some subalpine forests and bogs in the boreal zone, roughly two-fifths of the Canadian forests m a y be classified as subarctic. Probably the proportion is approximately similar in the Soviet Union (Zon, 1956). Only part of the subarctic zone is covered by forests, the other part consisting of treeless tundra, fens, and mountains. In Finland, for instance, the national forest inventories (Ilvessalo, 1957) reveal that in the northernmost drainage area, from which water runs to the Arctic Ocean and which as a whole belongs to the subarctic region, 45 per cent of the land area consists of forest land and 55 per cent of treeless wastes. Of the northernmost county of Norway, Finnmarken, (Landskogtakseringen,1930) only 9 per cent is covered by forests, 91 per cent being open tundra and mountains. W e c k (Weck and Wiebecke, 1961) estimates the total area of the boreal forests as 9.9 million km2. On the above grounds one-third to one-fourth of this area, i.e. 2.5-3 million km2, might be classified
296
as subarctic forests. Although, consequently, ;the timber volume and annual growth per unit area are low in subarctic forests, owing to the tremendous areas involved they comprise considerable timber resources and potential wood-producingcapacity. Regarding the timber resources, and growth of subarctic forests, the following examples m a y be mentioned. According to the Finnish national forest inventory (Ilvessalo, 1957), in the drainage area sloping to the Arctic Ocean, which approximately corresponds to the subarctic part of the Finnish territory, the average standing timber volume on productive forest land, which is mainly pine forest, is 53 m3/ha, and in the birch forests of forest tundra 15 m3/ha, including bark. The annual growth has been estimated at 0.8 and 0.4 m3/ha respectively. In Finnmarken of Norway (Landskogtakseringen, 1930) the standing volume has been estimated at 18 m3/ha in pine forests, and only 8-11 m3/ha in the subarctic birch forests, the annual growth being 0.2 m3/ha. Although the figures per hectare are low, however, they are compensated by the large extent of the areas. Thus, for instance,in Finland the timber resources of the above subarctic part of the country are estimated at 55 million m3 and the total growth at 870,000m3per annum, i.e. 2 per cent of the present annual growth of the Finnish forests. According to the estimate of W e c k (Weck and Wiebecke, 1961) the timber volume in mature boreal forests varies from 30 to 300 m3/ha and the annual growth from 0.5 to 5 m3/ha. The lower limits of these estimates m a y be applied to subarctic forests, i.e. 30 m3/ha for standing volume and 0.5 m3/ha for annual growth, figures which correspond well to the averages of the above Finnish and Norwegian estimates. Under these requisites the annual growth of all the subarctic forests in the world would be somewhere between 100 and 200 million m3. The following survey of the present use and prospects for future utilization of subarctic forests is divided into two sections, namely, use for wood production, and other forms of the multiple use of forests.
W O O D PRODUCTION Tunstell (1956) writes on the Canadian forests as follows: “Along the northern boundary of the Boreal Forests is an area of transition from the merchantable forests of the south to the treeless wastes of the Far North. The forests here are of no commercial value, although some have considerable local economic value, since they provide cover for fur-bearing animals and wood for fuel and buildings for the scattered inhabitants of the area.” According to the above statement, the Canadian subarctic forests have, today at least, no importance
Forests and forestry in subarctic regions
for commercial timber production, and the situation might be similar in m a n y other subarctic areas, too. This is mainly due to the small population of these regions and the long distances from industrial centres. The only exception is perhaps northern Fennoscandia, i.e. northern Finland and Norway, where, for such a northern location, the population has for centuries been relatively dense, and communications have existed with even more densely populated areas. Therefore, in northern Finland, Norway, and Sweden there is plenty of practical experience on the utilization of subarctic forests, and the possibilities of forestry in the far north have been discussed in several scientific papers. Factors limiting the utilization of the subarctic forests are both technical and ecological. Of the technical factors, the remoteness of the forests is the foremost; for this reason the distances for transportation to factories and marketing centres are very great. Furthermore, the rivers, the natural ways of timber transportation, mainly run northwards from subarctic forests to the Arctic Ocean, whereas factories and consumption centres usually lie to the south of the forests. Because of the sparse population,local labour is usually not available. The severe climate, i.e. low temperature and deep snow cover, also hamper forestry operations. A m o n g the technical drawbacks must also be included the small size and often poor quality of the trees. Much of the subarctic forests does not produce saw timber at all, but can only provide pulp-wood and fuel, and because of poor stocking even clear-cuttingyields little timber per area unit. For the above reasons the great majority of the subarctic forests of the world have so far remained outside the realm of commercial forestry and in the statistics of FAO,for instance,are classified as “inaccessible forests”. In recent years, however, because of the increasing demand for forest products, paper in particular, use has been made of more and more remote forest areas, and at the same time technical development has made possible logging operations and timber transportation under more difficult conditions. This development has made or in the near future will make even subarctic forests accessible for commercial forestry. Consequently w e are faced with a new problem, namely that of the continuity of the subarctic forests, i.e. h o w m u c h timber the subarctic forests can deliver on a sustained yield basis without danger to the existence of the forests and what kind of silvicultural methods should be used under subarctic conditions. Also the question arises as to what are the possibilities of increasing the wood production of the subarctic forests. The ecological factors which restrict the growth and even the existence of the forests are mainly climatic. The sh-,rt duration and low temperature of the growing season are the primary factors resulting in the slow growth and small size of the trees. Moisture
is usually sufficient, and in geologically young soils the nutrient status is fairly satisfactory. The paramount importance of temperature has been shown by several studies; a close correlation exists between the width of the annual ring and the summer temperature (e.g. Erlandsson, 1936; Ording, 1941; Hustich, 1945, 1948; Mikola, 1950, 1962 and others). In fact,tree ring analyses have been used successfully in studying the variation of summer temperatures in past centuries (Schove, 1954; Sirén, 1961, and others). Because of cold winters the soil temperature is low in the summer and the ground remains frozen long into or even through the summer (permafrost). Icy winds and heavy snow are also among the unfavourable ecological factors of subarctic regions; they cause breaking and deformation of trees and, in consequence,rot injuries. Therefore, the technical quality of the subarctic forests is usually very poor, the trees are old and small and, in addition, rotten or otherwise defective. The severe climate restricts the generative reproduction of the trees still more than their vegetative growth. To mature, the tree seed needs a certain amount of heat, an amount which is greater than &at necessary merely to support life and vegetative growth. In an average year the amount of heat is insufficient for tree seed to mature in subarctic areas ; thus, generative reproduction is possible only after summers that are warmer than average. The importance of the summer temperature for generative reproduction was already demonstrated by early investigators (e.g. Hagem, 1917; Wiebeck, 1920; Heikinheimo, 1921; Kujala, 1927; Eide, 1932, and others). The closer the real tundra,the more infrequent is the occurrence of favourableyears for seed maturing. Renvall (1912), for instance, concluded that at the outermost tree line in northern Scandinavia reproduction of pine occurred only once in a century, whereas on the southern edge of the subarctic zone climatic conditions would permit seed maturation once in ten years, on the average. According to more recent studies (Eide, 1932; Sirén, 1961) it is true, reproduction years at the timber line are not quite so uncommon, but nevertheless reproduction has been concentrated in certain periods of favourable years, and between these periods several decades m a y elapse without any regeneration. Since the maturing of seed depends only on summer temperatures, the continuous existence of forests in the northern timber-line region is determined by h o w often summers recur which are sufficiently w a r m for seed to reach maturity. Consequently, even slight climatic fluctuations m a y cause an advance or retreat of the forest limit. Thus, owing to a deterioration of the climate, the northern limit of the pine forests has retreated considerably in northern Europe in the last 3,000-4,000years, as has been shown by plantpalaeontologicalinvestigations. In the early twentieth century several scientists (e.g. Tanfiljev, 1911;
297
P. Mikola
Renvall, 1912) believed that the retreat was still continuing. This opinion was based on the fact that pine seedlings and young stands were rare or lacking in a broad zone along the tree limit. Meteorological data indicate, however, that for at least a hundred years (1850-1950)the northern European climate has been growing warmer (Keränen, 1952), and consequences of this trend are visible in both the fauna and vegetation (Hustich, 1952b). Thus, since 1920 pine seed has matured in several years up to the northernmost tree line and at the present time thriving young seedlings are c o m m o n in the Scandinavian forest tundra (Hustich, 1958). Natural reproduction of pine forests has also proved succesful in northern Norway (Bergan, 1961). Recent investigations suggest that the forest limit has advanced in the few last decades (Ve, 1951; Mikola, 1952; Andrejev, 1956; Hustich, 1958 and others). This assumption is also supported by the fact that the trees of the northernmost pine stand of the world (Stabursdalen, Norway, 70040’ N.) are thriving and there are numerous young seedlings (Kierulf, 1953). As was stated above,the commercialuse of subarctic timber resources has so far been negligible. Northern Scandinavia has been perhaps the only region where subarctic forests have been logged on a considerable scale. The arctic coast of Norway has long been relatively densely populated, and timber has been needed for building houses and boats. Since the local forest resources were scarce, trees were cut from the northernmost pine forest inland and floated to the coast along rivers running to the Arctic Ocean, such as the Tana River and its tributaries and the Neiden River. According to old records, alongside the Polmak and Neiden rivers, for instance, in the middle of the nineteenth century closed pine forests grew in areas where between 1910 and 1920 only stumps and solitary cull-trees were found (Renvall, 1919). Illegal utilization of forests continued in northern Lapland until the early twentieth century, resulting at least locally in a retreat of the forest limit. Reindeer grazing and-partly as a consequence-forest fires have also had a detrimental influence on the subarctic forests of northern Scandinavia. Therefore, although the climatic trend has been unfavourable in the postLitorina period, Renvall (1919) regarded h u m a n activity as the main reason for the retreat of the forest limit which has taken place in the last few centuries. The same factors, namely excessive grazing and careless logging, have destroyed subalpine forests in several mountain areas. Forest fires are reported to have resulted in permanent treelessness in some subarctic regions, e.g. Alaska. Because of the sporadic and uncertain occurrence of generative reproduction, the subarctic forests are very vulnerable and their recovery after destructive cuttings or natural calamities is slow. Therefore, in
298
the northern countries special laws have been passed to protect endangered forests and to prevent the retreat of the northern forest limit. Such forestprotection Acts were passed in Norway in 1893, in Sweden in 1903, and in Finland in 1922. As the basis of this legislation, scientific research has been necessary, and therefore in subarctic regions, at least in northern Europe, investigations were started relatively early to study the vital conditions and natural regeneration of the forests. Thus, for instance,when the Finnish Forest Research Institute was established in 1918, one of its first commissions was to survey the timber-lineforests and outline their future utilization (Heikinheimo, 1921). The leading principle of forest-protection laws is usually strict conservation. Measures which might endanger natural regeneration are prohibited. Accordingly, cuttings are restricted to a minimum, and when they are necessary for the support of the local population, according to the Finnish forest-protection law trees to be cut must be marked by a skilled forester. For the protection of these forests, state ownership has been preferred. If cuttings are conducted in subarctic forests, great caution is obliged by the forest-protection laws. The same principle has been adopted in the silvicultural instructions which have been prepared for the management o£ the respective areas (e.g. Renvall, 1919; Heikinheimo, 1921; Oinonen et al., 1959). Since natural regeneration is wholly dependent on occasional exceptionally w a r m summers, reproduction cuttings, i.e. cuttings in which the old tree generation is removed, are permitted only where a sufficient young generation already exists. Even seedling stands are more susceptible to various climatic and biotic injuries than in the south. Therefore seedling stands can not be regarded as established until the seedlings have grown well above the winter snow cover, i.e. have reached a medium height of 1-2 m. There still remains the alternative that forests are regenerated artiticially, i.e. by seeding or planting. The use of artificial reforestation, howevei, also depends on climate, since local seed can be obtained only after exceptionally w a r m summers, i.e. in the same years when conditions are favourable for natural reproduction, too. For this reason artificial reforestation in subarctic regions suffers for chronic shortage of local seed. The use of a foreign seed source is hazardous, as has been shown by bad experiences. T o guarantee the supply of local seed, seed orchards for subarctic regions can be established in more southern localities where climatic conditions permit seed maturation almost every year. T w o such seed orchards were established in Finland a few years ago, to provide northern Lapland with pine seed of local provenance. The use of artificial reforestation in subarctic regions, however, is also restricted by economic
Forests and forestry in subarctic regions
reasons. Because of the slow growth of the trees a n d long rotation, investments in seeding or planting are hardly profitable. According to Norwegian estimates (Braathe, 1966), artificial reforestation can profitably be practised only under such climatic and edaphic conditions, w h e n the average yield for the whole rotation is at least 2.5 m 3 / h a per year. A s w a s stated above, according to the national forest inventories the average annual growth of subarctic forests is 0.4-0.8 m3/ha in Finland a n d 0.2 m3/ha in northern Norway. These figures, however, represent the present growth, not the productive capacity of the sites. Because of old age and poor stocking the present growth is relatively low, as it also is in subarctic forests elsewhere. According to the yield tables of Ilvessalo (1937), in central North Finland, i.e. 100-200 k m south of the real subarctic forests, with a 120-year rotation the yield of pine forest is 1.3 m 3 / h a per year o n dry lichen heathland, a n d 3 m3 o n the best pine sites. In the subarctic region growth probably remains 25-50 per cent below these figures and rotations would have to be longer. R u d e n (1949)has estimated that the present growth of subarctic pine forests in northern N o r w a y corresponds to only one-tenth of their productive capacity which, accordingly, would be about 2 m3/ha per year (0.3-3.5 m3/ha, depending o n the site). Occasionally annual growth as high as 4 m3/ha has been measured in s o m e pine stands in northernmost N o r w a y (Eide, 1932). Consequently, although the productive capacity of the forests is m u c h higher than their present growth, artificial reforestation will prove profitable in exceptional cases only, o n the best sites in protected localities. In spite of this, seeding a n d planting have long been practised in the forest tundra area of northern Fennoscandia, first on a n experimental scale but lately to a larger extent. These afforestations have primarily aimed at restoring pine forests in areas, where they have been destroyed by irresponsible logging or natural calamities. Planting of spruce north of its natural range has also been tried successfully in Norway. Afforestation trials in northern N o r w a y were started as early as 1861, and several plantations there date from the past century; the results of Norwegian afforestations have been reported by Eide (1932) and others. T h e first direct seedings of pine in the Finnish Subarctic were m a d e in 1911-14. Their results have been reported in several papers (Kalela, 1937; Nuorteva, 1948; Mikola, 1959). Encouraged by positive experience, afforestations have been continued o n a n expanding scale. Thus, in the northernmost district of the Finnish State Forest Service (Utsjoki), which as a whole belongs to the subarctic region, pine has hitherto been s o w n on a n area of 3,400 hectares and planted o n 1,200 hectares. In recent years (1960-65)the rate of afforestation
(seeding or planting) there has been 300-400hectares per year, o n the average.
O T H E R USE OF FORESTS A s w a s stated above, in most subarctic regions the use of forests for timber production has so far been negligible. On the other hand, they have been used for other purposes, a n d these uses can also be developed. Hunting and reindeer-grazing are the oldest forms of forest utilization. Although hunting is a rather extensive a n d ineffective form of land use, giving a low profit per unit area, the subarctic forests of Canada a n d the Soviet Union still produce considerable amounts of furs annually for the world market. Reindeer husbandry is a n ancient industry which for centuries has supported nomadic people in the arctic and subarctic parts of Europe and Asia. Reindeer usually graze in the s u m m e r time o n treeless tundra a n d mountains, but the northernmost coniferous forests are their m a i n winter pastures. Reindeergrazing constitutes a m u c h m o r e effective land use than hunting. In northern Finland, for instance, there are about 250,000 reindeer, m o r e than a half of t h e m grazing in the subarctic region, and reindeergrazing plays a n important role in the e c o n o m y of the northernmost part of the country, as well as in northern N o r w a y and Sweden. Likewise in the Soviet Union, reindeer-grazing is considered a n effective utilization of the natual resources of the arctic a n d subarctic regions and, therefore, great attention is paid to the promotion of this industry. T h e vast subarctic forests of North America also would offer good natural opportunities for reindeer-grazing (Hustich, 1951; Scotter, 1965). Very divergent opinions have been expressed o n the d a m a g e which reindeer-grazing m a y cause to forests. In the older Scandinavian literature numerous papers can be found describing the reindeer as one of the worst enemies of the forests, whereas authors of s o m e other papers claim that reindeer are even beneficial in promoting the natural regeneration of forests. Without further discussion of the possible benefits and d a m a g e caused by the reindeer, it is to b e kept in mind that both reindeer-grazing and forestry are suitable natural industries for the subarctic regions and, therefore, both of t h e m should be promoted in such a w a y that they can exist side by side. Alongside the above ancient forms of forest utilization, a n e w one has appeared in recent years, namely the recreational use, which has its origin in m o d e r n industrial a n d social development. T h e northern subarctic regions are, at least in Europe, almost the only extensive wilderness areas where virgin forests can still b e seen. Old virgin forests have great scenic and recreational value and are important tourist
299
P. Mikola
attractions. This has been generally recognized and, therefore, in all the northern countries large national parks and other recreational areas have been established in subarctic regions. The typical nature of the subarctic forests makes them very suitable for various recreational purposes. Open old stands, where sturdy trees have for centuries resisted the elements and where the ground surface is smooth and devoid of obstructive brush vegetation, are an ideal terrain for hiking in the summer and for skiing in the winter, particularly if hills and valleys, and lakes, rivers and open fens bring variation to the scenery. Subarctic forests also offer excellent opportunitiesfor recreational hunting and fishing. The recreational use of subarctic forests is rapidly expanding. Modern trends, i.e. industrialization, urbanization, and the increase of the h u m a n population, tremendously increase the need for recreational areas, while at the same time rising standards of living, increasing leisure, and improving communications enable larger numbers of people than ever before to satisfy their desire for recreation in the virgin nature of the subarctic wildernesses. Thus, for instance, the number of tourists coming to Finnish Lapland from the densely populated countries of
Central Europe grows from year to year. The promotion of tourism consequentIy forms an essential part of plans of the future utilization of subarctic forests. Exploitation of timber resources spoils or decreases the recreational value of virgin forests. S o m e subarctic forests are already at the point where logging operations are approaching the very areas where new large national parks and wilderness areas should be established and protected to satisfy the requirements of tourism and other recreational purposes. Tourism and other recreational uses also constitute the economic utilization of forests and, therefore, in planning the management of subarctic forests great care is necessary to decide which areas should be reserved for recreation and which devoted for wood production or to what degree the same areas can serve both purposes. Thus, the future management of subarctic forests must be based on the principle of the multiple use of forests, i.e. all the different forms of utilization -wood production, reindeer-grazing, recreational uses, etc.-should be equally taken into consideration. In this w a y the subarctic forests can best benefit mankind tomorrow.
Les forêts et la sylviculture (P.Mikola)
demi-mètre cube par hectare et par an. L a capacité réelle de production,toutefois,est sans doute plusieurs fois supérieure. L’utilisation à des fins économiques et l’aménagement des forêts subarctiques sont limités par 17incertitude de la reproduction naturelle. L a maturation des graines n’a lieu que lorsque l’été est beaucoup plus chaud qu’en moyenne. Les expériences de reboisement artificiel ont en général donné des résultats encourageants. Mais les semis et plantations à grande échelle sont limités par la pénurie chronique de graines locales. En outre, étant donné la lenteur de la croissance et la longue durée de la rotation, le reboisement artificiel peut difficilement être rentable. Les forêts subarctiques présentent un grand intérêt pour les loisirs, et leur utilisation à diverses fins récréatives s’accentue constamment. I1 importe de tenir compte de ce fait dans les projets d’aménagement des diverses régions. L’élevage du renne est également une activité qui convient aux régions subarctiques.
dans les régions subarctiques
Les forêts subarctiques,c’est-à-direcelles de la région de la toundra proprement dite, et de la région découverte, selon la définition de Hustich (1949)et de Hare (1950), s’étendent sur une bande de 300 à 800 k m de large à la lisière nord des forêts boréales. L a superficie totale des forêts subarctiquesdu monde entier couvre, d’après les évaluations, de 2,5 à 3 millions de kilomètres carrés. En raison de difficultés techniques (éloignement, population clairsemée,manque de moyens de c o m m u nication, faible calibre et qualité médiocre du bois, etc.), la plupart des forêts subarctiques n’ont pas encore été exploitées commercialement.L a pauvreté et la lenteur de croissance de ces forêts s’expliquent par la rigueur du climat. En raison de leur immense étendue,les forêts subarctiques contiennent cependant des réserves de bois considérables et seraient susceptibles de devenir des régions de production. L a croissance actuelle des forêts vierges est en moyenne d’un
300
Forests and forestry in subarctic regions
Discussion J. BLUTHGEN. There are two pecularities of tree growth quite near the tree limit which need biological explanation: (a) the tall narrow growth of some (not all) species of spruce trees and to a minor degree also pine trees just near the tree limit; and (b) the torquated spiral growth of most of the trees in subarctic forests which makes them unfit for m a n y economic uses. The origin of this peculiarity seems to be still doiibtful.
P.MIKOLA. The reason for the narrow growth habit of trees m a y be genetic. Snow is an important factor in the Subarctic., and snow breaks are c o m m o n in forests. The narrower the crown the smaller the weight nf snow and the risk of breaking, therefore trees with narrow crown form are favoured in subarctic forests. The winding growth habit is not restricted to subarrtic forests only. It is characteristic of trees at a very old age. In boreal forests, trees are usually cut before they attain the age of winding growth. The reasons for the winding growth are discussed in textbooks of tree physiology; m a y I refer to Physiology of Trees by Kramer and Kozlowski, for instance.
C. O. TAMM.I wish to express m y gratitude to Professor Mikola for the excellent way in which he has summarized Scandinavian experience in this field. I would only like to comment a little on the statement that soil nutrient status is relatively satisfactory in the subarctic region. This I believe is true only as regards minerals, not nitrogen. Supply of nitrogen yields, according to our experience, a strong growth response on a relative basis, even if the absolute yield remains low. W e believe that we in Sweden have an increasing deficiency in total and available nitrogen,from south to north. The reasons for this trend are not quite understood,but it is interesting that a similar trend appears in the supply of ammonia and nitrate-nitrogen with precipitation, according to the network for rain-water analyses,
Bibliography ANDREJEV, V. N. 1956. The contemporary afforestation of the tundra. In: B. A. Tikhomirov (ed.), The vegetation of the Far North ofthe U.S.S.R. and its utilization,I. MoskvaLeningrad, Akad. N a u k SSSR.(Orig. Russian.) BERGAN, J. 1961. E n undersmgelse av naturlig gjenvekst av furu etter en del foryngelsehogster i Pasvik. Tidskr. f. Skogbr. vol. 69. BRAATHE, P. 1966. Metsien uudistaminen Norjassa (Reforestation in Norway). Metsüt. Aikakl., vol. 83. EIDE, E. 1932. Furuens vekst og foryngelse in Finnmark (The growth and regeneration of the pine forests in Finnmark), Medd.fr. d. norske Skogjors0ksv.,vol. 4. ERLANDSSON, S. 1936. Dendrochronological studies. (Data 23 fr. Stockholms Högsk. Geokron. Inst.) FOWELLS, H . A. 1965. Silvics of forest trees of the United States. (U.S. Dept. of Agric., Agr. Handbook No. 271.)
established for the IGY. As biological fixation of nitrogen appears to be almost absent in taiga forests, rain-water nitrogen m a y well be responsible for the building up of a nitrogen store in a forest ecosystem, a process extending over centuries. Studies on atmospheric supply of bound nitrogen would be welcome also from other parts of the subarctic and taiga zones. P. MIKOLA. I agree with you. W h e n I said that the nutrient status in subarctic forest soils usually is satisfactory, I only wanted to stress the importance of the temperature factor and also to refer to the fact that subarctic soils generally are young and therefore their mineral composition little altered.
N. S~YRINKI.I would ask Professor Mikola’s opinion about the forest limit in the Subarctic-whether it was originally continuous or if there were isolated, single-tree specimens growing beyond the forest limit. There are two opinions about the alpine forest limit and tree limit in the mountains. M a n y scientists agree that the alpine forest limit has naturally been continuous: where the climatic conditions are suitable for a single tree to grow, a whole forest could exist too. W h a t is the situation in the Subarctic ? W a s there a separate tree limit in front of the forest limit originally, or has the forest limit been continuous ?
P. MIKOLA. W h e n the forest limit is advancing today we can find solitary pine seedlings and young pine trees far beyond the present tree line. This suggests that the arctic forest limit is not continuous. W e have to consider the long distance transport of seed on the snow surface, and reindeer, too, can carry pine seeds in their fur, and these seeds can germinate and trees grow in edaphically favourable spots. Thus there is a belt between the forest and the tundra where solitary trees and small groups of trees can exist.
i Bibliographie HAGEM, O. 1917. Furuens og granens frmsaetning i Norge. Medd. V e d .forstl. formkst. vol. i. HARE, F. K. 1950. Climate and zonal divisions of the boreal forest formations in eastern Canada. Geogr. Rev.,vol. 40. HEIKINHEIMO, O. 1921. Suomen metsänrajametsät ja niiden vastainen käyttö (Die Waldgrenzwälder Finnlands und ihre künftige Nutzung). Comm. Inst. Forest. Fenn., vol. 4. HUSTICH, I. 1945. The radial growth of the pine at the forest limit and its dependence on climate. Soc. Scient. Fenn., Comm. Biol., vol. 9, no. 11. . 1948. The Scotch pine in northernmost Finland and its dependence on the climate in the last decades. Acta Bot. Fenn., vol. 42. . 1949. On the forest geography of the Labrador Peninsula. Acta Geogr.,vol, 10.
_-
__
301
P. Mikola
_- . 1951. The lichen woodlands in Labrador
and their importance as winter pastures for domesticated reindeer. Acta Geogr., vol. 12. . 1952a. Barrträdsarternas polara gräns på norra halvklotet. C o m m . Inst. Forest. Fenn., vol. 40. .(ed.). 1952b. The recent climatic fluctuation in Finland and its consequences. Fennia, vol. 75. -. 1958. O n the recent expansion of the Scotch pine in northern Europe. Fennia, vol. 82. . 1966. O n the forest-tundraand the northern tree-lines. Ann. Uniu. Turku, ser. A (II), vol. 36. ILVESSALO, Y. 1937. Pery-Pohjolan luonnon normaalien metsiköiden kasvu ja kehitys (Growth of natural normal stands in central North Finland). C o m m . inst. Forest. Fenn., vol. 24. . 1957. The forests of Finland by the main water system areas. C o m m . Inst. Forest. Fenn., vol. 47. KALELA, A. 1937. Zur Synthese der experimentellen Untersuchungen über Klimarassen der Holzarten. C o m m . inst, Forest. Fenn., vol. 26. KERÄNEN, J. 1952. Temperature changes in Finland during the last hundred years. Fennia, vol. 75. KIERULF, T. 1953. Verdens nordligste furuskog. Tidskr. f. Skogbr.,vol. 61. K U J A L A ,V. 1927. Untersuchungen über den Bau und die Keimfähigkeit von Kiefern- und Fichtensamen in Finnland. C o m m . Inst. Forest. Fenn., vol. 12. LANDSKOGTAKSERINGEN. 1930. Takseringen av Norges skoger. XI. Finnmarks fylke. Oslo. MIKOLA, P. 1950. Puiden kasvun vaihteluista ja niiden merkityksestä kasvututkimuksissa (On variations in tree growth and their significance to growth studies). C o m m . Inst. Forest. Fenn., vol. 38. . 1952. Havumetsien viimeaikaisesta kehityksestä metsänrajaseudulla (On the recent development of coniferous forests in the timber-line region of northern Finland). C o m m . Inst. Forest. Fenn., vol. 40. . 1959. Metsynviljelyn mahdollisuuksista pohjoisella metsänrajalla (Afforestation trials at the northern timherline). Metsüt. Aikakl., 76. -._ . 1962. Temperature and tree growth near the northern timber line. In: T.T. Kozlowski (ed.), Tree growth. New York, The Ronald Press Co.
___-
--
--
--
302
NUORTEVA, M . 1948. Metsänviljelyä männyn metsänrajan pohjoispuolella (Forest cultures north of the forest limit of pine). Metsät. Aikakl., vol. 65. OINONEN, E.; SARVAS, R . ; SIREN, G. 1959. Lapin suojametsìen käsittelyohjeet (Instructions for the management of protection forests in Lapland). Finnish Forest Research Institute. ORDING, A. 1941. Arringanalyser på gran og furu (Annual ring analyses of spruce and pine). Medd. fr. d. norske Skogforsrksu., vol. 7. RENVALL, A. 1912. Die periodischen Erscheinungen der Reproduktion der Kiefer an der polaren Waldgrenze. Acta Forest. Fenn., vol. 1. .1919. Suojametsäkysymyksestä, I-VI(On the problem of protection forests). Acta Forest. Fenn., 11. RUDEN,T. 1949. Trekk fra Nord-Norges skoger. Det Norske Skogselskap 1898-1948. Oslo. RUDEN, T. 1949. Trekk fra Nord-Norges skoger. Oslo. Det Norske Skogselskap 1898-1948. SCHOVE,D. J. 1954. Summer temperatures and tree-rings in North-Scandinavia, A. D. 1461-1950. Geogr. Ann., vol. 36. SCOTTER,G. W. 1965. Reindeer ranging in Fennoscandia. Jour. Range Man., vol. 18. SIREN,G. 1961. Skogsgränstallen s o m indikator för klimatfluktuationerna i norra Fennoskandien under historisk tid. C o m m . Inst. Forest. Fenn.,vol. 54. TANFILJEV,G. I. 1911. Die polare Grenze des Waldes in Russland. Odessa. ~ U N S T E L L ,G. 1956. Canada, In: S. Haden-Guest, J. K. Wright and E.M . Teclaff. A world geography of forest resources. New York, The Ronald Press Co. VE, S. 1951. Stig skoggrensa ? Tidskr. f. Skogbr., vol. 59. WECK, J. ; WIEBECKE, C. 1961. W'eltforstwirtschaft und Deutschlands Forst- und Holzwirtschaft. München. WIEBECK, E. 1920. Det norrländska tallfröets grobarhet (Die Keimfähigkeit des norrländischen Kiefernsamens). Medd.st. Skogsförsöksanst,vol. 17. ZON,R. 1956. The Union of Soviet Socialist Republics. In: S. Haden-Guest, J. K. Wright and E.M. Tecla5, A world geography of forest resources. New York, The Ronald Press Co.
National parks and nature reserves in subarcticregions R. Kalliola
The subarctic and, particularly, arctic areas of the earth have until n o w generally baffled all h u m a n attempts to conquer the virgin nature and to build a landscape. These circumpolar zones are, for the main part, unpopulated and natue remains quite untouched. As far as the habitat of the arctic fauna is concerned it is also usually in its original state. Unfortunately however, there is only very little left of the initial number of animals because hunting in the arctic zone began as early as the seventeenth century and excessive shooting has taken place in m a n y areas. Although the Siberian tundra, the American barren grounds and the northern seas m a y have seemed inexhaustible, wild-life began to disappear from one place after another. S o m e species have entirely vanished (Steller’s sea cow, N e w foundland wolf and great auk), others are almost extinct (sea otter, Greenland whale and other big whales, walrus, several seal species). Even populations of typical arctic species such as wild reindeer,caribou and polar bear are n o w only a small fraction of what they were in earlier times. Unrestricted shooting by people w h o came from southern areas endangered the living of the arctic people. Indians, Eskimoes, Lapps, Tsukts, Korjuks, Kamtsadals, etc. are entirely dependent on hunting and fishing, so that, the governments concerned had to establish hunting regulations. Nowadays management of wild-life is just as important in the Arctic as anywhere else. As far as I k n o w national parks, in the proper meaning of this concept, do not exist in the arctic zone ; but there are several large refuge areas for game and bird sanctuariesin the arctic areas of America and Asia where hunting is strictly regulated. Perhaps the laws are also better enforced n o w because travelling is restricted in some circumpolar areas which are guarded for military reasons. Extinct species can never be brought back, but
otherwise the future of northern wild-life seems promising. There are good possibilities to preserve this type of fauna, because the wild-life habitat-as I mentioned above-is still almost in its virgin state. The treeless arctic zone is shown by stippling in Figure 1. T h e boreal coniferous zone to the south extends, in Fennoscandia, from the arctic region to the 60th parallel of latitude and, in the continental parts if Asia and North-America,to the 50th and in some places even to the 40th parallel. The greater part of the coniferous zone is thinly populated and the northern or subarctic areas of this zone are even now, to a great extent, in as natural a state as it ever was. As mentioned by professor Mikola in the preceding paper, the great majority of the subarctic forests of the world have so far remained outside the realm of commercial forestry, the most remarkable exception being perhaps northern Fennoscandia; but it is also suggested that this state of affairs is changing very rapidly. The increasing demand for forest products, paper in particular, and the technical development which have made forestry and timber transportation possible even in more and more remote forest areas and under more difficult conditions, will make subarctic forests accessible for commercial forestry, in the near future. Knowing this situation w e must give sufficient attention to it and in time to preserve nature in the Subarctic. It seems to m e that there are good possibilities for this to be achieved and quite a lot has already been done. In those subarctic areas farming is not usually possible and the productivity of the forests will in any case be smaller than in the greater part of the coniferous zone; in addition the forest protection is also important to prevent the retreat of the forest limit. Moreover, the areas near the tree limit and subarctic mountains are of special scientific interest. They also offer varied and beautiful scenery and, as such, the areas
303
R. Kalliola
FIG.1. National parks and nature reserves in the circumpolar zones. The stipple shows the arctic region, between it and the broken line the boreal coniferous region. 1, Northern Fennoscandia; 2-8, European and Asiatic Russia: 2. Pechera-Ilych (7,143 kmz), 3. “Denezhkin kamen” (1,467 km2), 4. Altai (9,148 km2), 5. “Stolby” (470 km2), 6. Barguzin (2,482 km2), 7. “Kedrovaya pad” (151 km2), Sudzukhe (1,400 km2) and Suputinka (159 km2), 8. Kroroki (9,640 km2); 9-17,North America: 9. Mt. McKinley (7,878 kmz), 10. Olympic (3,429 km2) and Mt.Rainier (978 km”, 11. Jasper (10,920kmz), Banff (6,666 km2): Yoho (1,318 km2), Kootenay (1,411 km2), Glacier (1,349 km2), Mt. Revelstoke (260 km2), Elk Island (195 km2) and Glacier (Montana) (4,041 kmz), 12. W o o d Buffalo (44,807 km2), 13. Yellowstone (9,027kmz) and Grand Teton (1,208 km2), 14. Prince Albert (3,889 kmz), 15. Isle Royal (542 km2), 16. Fundy (2,074 km2) and Cape Breton Highlands (954 km2), 17. Terra Nova (406 km2); 18. Iceland: Thingvellir (40 km2).
304
National parks and nature reserves in subarctic regions
are particularly suitable for the increasing demands of tourism and outdoor recreation. The International Union for Conservation of Nature and Nature Resources (IUCN)is preparing a list on the national parks and equivalent areas of the world; but this seems more difficult a task than had been thought, because the names and legislations vary greatly from one country to another. The preliminary list, however, is available and in accordance with it the national parks of the coniferous zone have been marked o n the m a p (Fig.1) and classified according to their size. (The wild-life refuges and bird sanctuaries, however, are not marked on our small map, partly because of the insufficient information available.) National parks almost reaching or exceeding 10,000 km2 include Altai in central Asia, Koroki in Kamchatka and, in North America, the Jasper, W o o d Buffalo and Yellowstone parks. It is not possible to give here a description of the different parks. Their purpose is the preservation of nature,including mainly scientific,social and economic objectives. These are usually concurrent, but their mutual emphasis m a y be different. There are also differences between different countries and cultures
regards motivation and practice. In North America the social aspect of national parks is very important. These parks are highly valuable for tourism and outdoor recreation with all kinds of facilities and the number of visitors is enormous. Several hundred thousand, and in m a n y cases more than a million people, visit these areas each year. In the U.S.S.R. conservation is connected more than elsewhere with practical and economic aims, e.g. with the management of fur-producing animals and fishing and this is also reflected in national park policy. In Fennoscandia the purpose of nature reserves is-at least it has been-primarily scientific and purely idealistic. The preservation of representative samples of virgin nature and of primitive types of cultivation is regarded as a national duty. In this connexion it might be mentioned that the conception of national parks is in a w a y contradictory: a national park is an area, in which nature is to be kept as untouched as possible, while at the same time it is hoped that m a n y tourists and travellers will visit there. Although the visitors strictly obey regulations, they are bound to leave some marks in nature. Moreover roads,buildings and various devices
FIG.2. National parks and nature reserves in northern Fennoscandia.The broken lines show the geobotanical regions,arctic, subarctic and boreal, the alpine areas are stippled.According to Hustich 1960 and Ahti et al., 1961. Norway (so far only a a proposal): Bsrgefjell (c. 1,000 kmz), Saltfjell (1,100 km2), 0 DI (= Ovre Dividal) (600 km2), AN (=Anderdalen) (31 km2) 0vre Anarjokka (c. 1,000 km2),ST (= Stabbursdalen) (65 kmz), 0 PA (=0vre Pasvik) (60 km2); Sweden: PE (= Peljekaise) (146 km2), MU (=Muddus) (492 kmz),Sarek (1,900 km2), Stora Sjöfallet (13,800km2), Abisko (50 km2),VA (=Vadvet jåkko) (25 km2); Finland: RU (=Runkaus) (16 km2), PI (=Pisavaara) (50 km2),Oulanda (107 km2),PY (=Pyhätunturi) (30 km2), MA (=Maltio) (147 kmz), SO (= Sompio) (181 kmz), Pallas-Ounas(500 km2), LE (=Lemmenjoki) (385 km2), KI (=Kivach) (103 km2), K A (=Kandalaksha) (203 kmz), Imandra Malla (30 km2)and Kevo (342 kmz); U.S.S.R.: (=1.apland) (1,583 km2).
305 22
R. Kalliola
must be constructed in the area for them. This contradictory situation is of course more marked when the area in question is smaller, and vice versa. In Scandinavian countries, Norway, Sweden and Finland, where natural landscapes still prevail, it has been said sarcastically, that the worst that can happen to the nature of a n area is for that area to be declared a national park; a n d similarly cynical is the consolation put forth, that the advantage of a
FIG.3. National parks (black circles) and strict nature reserves (white circles) in Finland. Figures show the size in square kilometres.
306
national park is that destruction of nature caused by tourism can be concentrated o n certain places, while nature elsewhere is saved. This confliction between touristic and scientific use is a problem which one has tried to solve in different ways. T h e most c o m m o n method is to divide the park into different zones. There can be a strictly protected zone for scientific purposes, and another for visitors, etc. In Finland we have tried to solve this problem-with the future in view-in another way, i.e. by having t w o kinds of national parks. O u r national parks in the proper sense of the word have been established primarily as public displays of the Finnish natural treasures. The public has free access to these areas. Roads and facilities have been built for tourists travelling and staying in the area. The other type, strictly nature reserves, has been established primarily for scientific purposes. Nature is protected there from the unavoidable changes which occur in the national parks open to visitors. Access is, in most cases, allowed only by special written permission. Figure 2 shows the location and size of the national parks and nature reserves in the northern part of Fennoscandia. These areas fairly well include all types of forest, peatland, alpine and tundra vegetation of the subarctic zone. As far as N o r w a y is concerned, the national parks are under consideration but have not yet been established. The largest national parks (900-2,000km2) in Fennoscandia are the Sarek and Stora Sjöfallet in Sweden, and the Imandra (Lapland) Park in the Kola Peninsula in U.S.S.R. If N o r w a y obtains its national parks as expected, there will be three new parks of 1,000 k m 2 : Bsrgefjell, Saltfjellet and 0vre Anarjokka. In Finland there is n o park of that size. T h e location and size of the Finnish national parks and nature reserves can be seen in Figure 3. W e hope that we can enlarge our Lemmenjoki Park so that it would form one really vast park area together with the Norwegian Anarjokka. There are several examples of international nature reserves (e.g. the Tatra National Park o n the border of Poland and Czechoslovakia), so why not in Fennoscandia also ? Of course the value of a nature reserve is not only dependent on its size; large areas are needed, however, especially for the preservation of the original fauna of wild-life areas. Keeping in mind that tourism and camping are growing very fast in northern Europe also, it seems that the establishment of large national parks like those in North America would be one of the most economic and wisest forms of use of the natural \ resources.
National parks and nature reserves in subarctic regions
Résumé Parcs nationaux et réserves naturelles dans les régions (R.Kalliola)
subarctiques
La plupart des régions subarctiques n e sont pas peuplées et la nature y demeure intacte. Malheureusement, il n e reste que peu d’animaux parmi ceux qui y vivaient initialement. I1 est certes impossible de faire revivre les espèces éteintes, mais l’avenir de la faune septentrionale semble cependant prometteur ; il reste e n effet possible de la préserver puisque son
habitat demeure presque à l’état vierge. Un certain n o m b r e de parcs nationaux ont été créés dans la zone des conifères (fig. i). En outre, il existe des refuges pour la faune et des sanctuaires d’oiseaux. Etant donné q u e le tourisme et le camping se développent très rapidement e n Europe septentrionale, il semble que la création de vastes parcs nationaux analogues à ceux de l’Amérique constituerait l’une des formes les plus économiques et les plus judicieuses d’utilisation de ses ressources naturelles.
Discussion F. E. ECURDT.Vous avez évoqué divers problèmes de législation liés à la protection de la nature. Pourriez-vous m’indiquer dans quelle mesure il y a actuellement un effort en vue d’uniñer cette législation dans les divers pays qui possèdent des territoires dans la région arctique et subarctique du globe ?
Thus, to m y knowledge, there is no area of “lowland spruce taiga” in all North America that is protected for biological research b y being an inviolate sanctuary. There is a clear and critical need for the establishment of large inviolate areas to function as “control areas” for biological research.
R.KALLIOLA. Concerning the whole world, the International Union for Conservation of Nature and Natural Resources (IUCN)maintains a Commission on Legislation, which until n o w has collected the existing legislation. Regarding Scandinavia, next September there will be a meeting of the national nature conservation officers which will pay attention to this question.
R. KALLIOLA. I fully agree.
W. PRUITT. According to the m a p the coniferous forest of North America has protected areas. This is a misconception due to misnaming of areas. For example, W o o d Buifalo “Park” has a buffalo management programme with slaughtering facilities; Mt. McKinley National Park in Alaska has had until not long ago, an active wolf-poisoning programme. These areas cannot be considered “natural” in any sense of the word.
N. S~YRINKI. In looking at the m a p shown b y Dr. Kalliola one is surprised to realize h o w few national parks there are in the Subarctic outside of Fennoscandia. There are only a couple in the U.S.S.R.and the situation is no better in North America either. T h e reason is perhaps that the Subarctic scenery in those parts of the world is not touristically very attractive and that there has not been any real threat to nature yet. W e k n o w however in Finland h o w soon an area which is not protected can be changed b y man. I would suggest therefore that this symposium make a recommendation to all countries in the subarctic region to establish sufficiently large national parks or other reserves for scientific and other purposes.
Bibliography / Bibliographie AHTI, T.; HAMET-AHTI, L.;
JALAS, J. 1961. LuoteisEuroopan kasvillisuusvyöhykkeistä ja kasvillisuusalueista. Luonnon Tutkija, vol. 68, p. 1-28. HUSTICH, I. 1960. Plant geographical regions. A geography of Norden. Oslo. J. W.Cappelens forlag, p. 54-62. ANON. 1965. Natur- og nasjonalparker i Nord-Norge. Norsk Natur, 1965, p. 21-27.
INTERNATIONAL UNIONFOR CONSERVATIONOF NATURE AND NATURAL RESOURCES. 1962. United Nations list
os national parks and equivalent reserves, 1-11. Morges
(Switzerland), I U C N . PEDERSEN, A . 1956. Das Tierlaben in der Arktis. Die letzten Oasen der Tierwelt. Frankfurt a.M., Umschau Verlag, p. 202-205. TROTZ, E.P. 1956. Naturschutzparks in der Sowjetunion. Die letzten Oasen der Tierwelt. Frankfurt a.M., Umschau Verlag, p. 212-215.
307
Problems of the conservation of relict arctic and subarctic species in Britain I. G.Simmons
INTRODUCTION The British Isles are not usually considered a subarctic environment. Nevertheless there are areas where altitude combined with various climatic elements constitutes an environment for plant life which in its rigours resembles in some ways that of the Subarctic proper. Areas approaching and above 3,000ft (915 m) m a y have 100 days with snow cover at O900 h, and at this altitude it is likely to fall in any month of the year: there are a small number of snow patches which persist through the year (Manley, 1944, 1952). Green (1964) suggests that “the combination of high winds and snowfall, and also of frequent alternation of freezing and melting, make the climate of the Scottish Highlands m u c h more severe for plants than, for example, the Norwegian fells and thus m u c h poorer in plant species and vegetation generally”. Pearsall (1950), writing of the temperature régime above 2,000 ft (610 m), compares the climate with the areas just above sea-level in South Iceland.
PHYTOGEOGRAPHIC BACKGROUND It is perhaps not surprising therefore that the Highlands of Scotland and the upland areas of Northern England and northern Wales have m a n y plant species in c o m m o n with more northerly or more alpine parts of Europe. An enumeration by Polunin (1954) produced 309 species of vascular plants c o m m o n to the British Isles (c. 30 per cent of the Arctic vascular flora) and the Arctic sensu (Polunin, 1951), and presumably extension to the Subarctic would increase the list. The recognition of these affinities (and others) in the British flora was categorized by Matthews (1955)when he recognized phytogeographical groups
in the British flora, and it is with the representatives of four of these-the Arctic-Subarctic,the Northern Montane, the Arctic-Alpine and the Alpine-that this paper is concerned. The Arctic-Subarctic group contains such plants as Artemisia norvegica, Deschampsia alpina and Rubus chamaemorusl of which the latter is one of the only two plants to occur in more than squares of 100 km2 of the Atlas of the British Flora (Perring and Walters, 1962): the average for this element is about 30 and only one species has one occurrence (Diapensia lapponica); these are plants having affinities with boreal floras. T h e Northern Montane affinity lies with subarctic and montane situations elsewhere in Europe and m a y be represented by S a h phylicifolia, a willow of mountain streams. This has 50-100 occurrences in the “Atlas”, the average for the group is 41 and three species have only one occurrence. The smallest group is the selfexplanatory Alpine element, e.g. Gentiana verna, found in two regions (average for the element 24) and the largest the Arctic-Alpine element with five species having only one occurrence, average 49. Betula nana, Myosotis alpestris, Silene acaulis, Alehemilla alpina, Dryas octopetala, and Empetrum nigrum will serve as examples. Table 1 summarizes the main numerical features of the distributions. (See also Figures 1 and 2.) These elements were recognized by Matthews (1955) because of their distinctive and often restricted distributions but there are m a n y other species of subarctic affinities which are widespread wherever there is a suitable environment, Betula pubescens, Vuccinium vitis-idaea, Eriophorum vaginatum and Armeria maritima being examples. But Matthews’ “elements” contain the taxa which are the most 1. The nomenclature used throughout is that of Clapbam, Tutm and Warburg. Flora of the British Isles, 2nd ed., Cambridge, 1962.
309
I. G. Simmons
Problems of the conservation of relict arctic and subarctic species in Britain
I. G. Simmons
312
P r o b l e m s of the conservation of relict arctic a n d subarctic species in Britain
I. G. Simmons
TABLE 1. Main numerical features
of four phytogeogra-
phical groups Group
Arctic-Subarctic Northern Montaae Alpine Arctic Alpine
No.of species
No. with five occurrences
Average occurrencen for group
21 31 8 74
8 6 3 14
30 41 24 49
ECOLOGY AND PRESERVATION
Source: Matthews (1955); Perring and Walters (1962).
restricted in distribution (many of them have less than 100 occurrences)and which have therefore excited the most interest in botanists. They are of course largely mountain and moorland plants and the least c o m m o n are often plants of open habitat: calcareous flushes and mires, block fields, slopes undergoing solifluction or other mass movement, stream and lake shingle are all typical habitats of Britain’s rarer plants of subarctic affinity. A number occur in closed communities with 100 per cent vegetation cover and it m a y be noted that they are usually among the commonest; Vaccinium vitis-idaea, Salix herbacea, Antennaria dioica all have 100 occurrences. The history of these species is outlined by Godwin (1956) w h o shows that m a n y of the taxa were widespread during either Full and/or Late-Glacial times (i.e. Weichselian or Late-Weichselian)and indeed for the elements considered there are Weichselian or LateWeichselian (zones 1-111)records for m a n y of them (Table 2).
.TABLE 2. Record of taxa during Full and/or Late Glacial times Species in Group
group
Northern Montane Alpine Arctic-Subarctic Arctic-Alpine
31 8 27 14
With records Full Glacial Late Glacial
3 1 1
12 2
19
24
5
Source: Godwin (1956) and personal communications.
These records usually include localities far to the south of the present distribution of the species. The suggestion is that the species have undergone contraction of range due to the extension of forest during the Postglacial when the numerous open habitats of Late Glacial times were reduced to a small coterie of river-banks, cliff-ledges and similar unstable areas and that the growth of ombrogenous blanket peat in the mid-Postglacial has further reduced the number of habitats for the northern species. It is, however, not inconceivable that some of the plants with very few occurrences belong to populations which have
314
arrived since deforestation and have established themselves in either natural or man-assisted open habitats. There is no palaeo-ecologicalevidence for the sustained presence at a particular site of any of the restricted species throughout the Postglacial.
Irrespective of their origin, the distribution and dynamics of several taxa with northern affinities are n o w the concerns both of lay conservationist bodies with botanical interests such as the Botanical Society of the British Isles (BSBI)and the County Naturalists’ Trusts, and those with statutory obligations like the Nature Conservancy, a member of the Government’s Natural Environment Research Council, and it is of interest to consider some of the influences on the species and the relation of conservation policies to those dynamics. Changes such as that of climate cannot of course be altered in any w a y by m a n , but it seems that m a n y of these plants have survived the hypsithermal interval of the Postglacial period and should therefore be in little danger on climatic grounds, especially as since 1950 there has been a trend away from oceanicity (Lamb, 1965) although this could be a short-term fluctuation, and because m a n y of the species grow anyway in the more rigorous climatic zones of the islands. Few if any of the ecosystems of Britain remain unaffected by man, and it is probably significant that m a n y of the taxa with which this paper is concerned occur in the least affected systems. The mountain rock-ledge inaccessible to domesticated animals, especially the sheep which have at one time or another chewed their w a y across Britain about 0.5 c m from the ground, is a c o m m o n habitat for m a n y of the northern species, often in what McVean and Ratcliffe (1962) have called their “tall herb nodum”. In Scotland at any rate, red deer are likely to graze such noda but the taller, ranker forbs and grasses appear to receive the brunt of the grazing and the restricted plants are, with exceptions, less affected and indeed doubtless benefit from the reduction of competition. S o m e species do occur in mountain grasslands (especially on calcareous soils) which are heavily grazed by sheep,and here it must be assumed that the reduction in competition affords suitable conditions for the continuance of the relict species. In the “turfy marshes” of Upper Teesdale in northern England, Pigott (1956) admits that “the present structure of these marshes and the related abundance of the rare species (Juncus alpinus,
K o bresia simpliciuscula, ToJieldia pusilla, Bartsia alpina, Primula farinosa, Saxifraga azoides) is largely the result of cattle grazing”. In this region also, sheep and rabbits are known to keep open (and
Problems of the conservation of relict arctic and subarctic species in Britain
eroding) the turf around outcrops of the “sugar limetone” (a limestone altered by contact with an intrusive volcanic rock). Thus Minuartia verna survives but one of the few English localities for Dryas octopetala becomes increasingly liable to erosion. Another result of h u m a n activity which can cause change is the use of total herbicides,under such trade names as “Dalapon” and “Paraquot”. It is not likely that these will be applied to mountain ledges but some habitats such as Calluna moor and mountain grassland are particularly suitable for the “kill everything and then re-seed” type of agriculture which is promoted by them. The last locality of Chamaepericlymenum suecicum in Yorkshire is close to an area subjected to this treatment and experiments at Moor House National Nature Reserve at over 1,000 ft (305 m) in the Pennines have demonstrated that good grassland can be nurtured in this environment so that in the absence of any legislative control the practice m a y spread. A most pervasive influence of Homo britannicus on his flora is of course the collector: not merely the botanist wishing a specimen for his herbarium but the gardener wanting at least one or two ArcticAlpines for his rock-garden.Although not thoroughly documented these practices must have been the cause of m u c h diminution of the populations of some of the more restricted plants. The general public, too, will pick visually outstanding flowers by the handful: Gentiana verna and Trollius europaeus are obvious candidates for this sort of treatment. It is rather surprising therefore that the country has apparently lost only one species of vascular plant during the last century (Carex davalliana from near Bristol) although m a n y localities have been lost of species not yet extinct. Protection of the restricted species under consideration has been undertaken in some cases by the Nature Conservancy which in some of its larger National Nature Reserves includes the habitats of a proportion of them: the Upper Teesdale, the Cairngorms, the Bein Eighe, and the Caenlochan National Nature Reserves are examples, and other sites are designated as Sites of Special Scientific Interest (SSSI) some of them with management agreements where the owners agree not to alter the present land use system, e.g. Ben Lawers. A difficulty, however, is that the Conservancy do not o w n all their reserves, holding some under Nature Reserve Agreements and frequently the grazing rights will be retained by the owner. The SSSI status means only that consultation with the Nature Conservancy has to take place before any development of the site. The role of man’s activities which have affected the survival of this relict flora can make management difficult. Public access can be restricted sometimes but often not especially in mountainous areas, and the
state of education which would lead people to avoid damage to the plants is in a very inchoate stage in Britain. Without doubt the more indirect effects of man, especially through grazing, cause the most difficulty for it appears that the relict species of grazed areas rely for their survival on absence of public pressure (which is not likely to be maintained for very long in areas which, for example the Cairngorms, are being developed for skiing) and on moderate grazing pressure, and this latter could naturally be tipped either w a y by changes in economic conditions. Cessation of grazing m a y initially produce spectacular results in the production of herbage and flowers by the relict species and their associates, but it m a y be suspected that grazing maintains an early seral stage and that release from the pressure allows succession to occur and m a y well result in the loss of the relict species, particularly where they are small plants likely to be shaded out. S o m e of the restricted plants therefore owe their present survival to their presence in man-dominated ecosystems; this is exactly analogous to m a n y of our “weeds” which were widespread in the Late Glacial, restricted during the forested period of the Postglacial and then rife again after the influence of m a n had become strong (e.g. R u m e x acetosa, Plantago spp., Artemisia vulgaris). It might be noted that the Arctic-Alpine species Minuartia verna is quite frequent on old lead mine spoil heaps in the northern Pennines (Eddy, 1963).
AN EXAMPLE The case of Upper Teesdale, in the northern Pennines near Durham, is instructive. Here there is a unique floral assemblage not only of northern species (usually at their southern limit), but of southern species also, often at their northern limit, some taxa have their only English locality (e.g. Betula nana G. vena), are its only British site (Minuartia stricta), and others are quite localized, e.g. Potentilla fruticosa. At least fifteen species of higher plants are present which are generally classed as “rare” or “very rare”. They occur in an assemblage of habitats ranging from mandominated systems such as hay meadows to rocky cliffs which are ungrazed (Valentine, 1965; Pigott, 1956). Half the interesting area is a National Nature Reserve although one in which sheep grazing continues, and the other half an SSSI where sheep are also grazed andland is managed as grouse (Logopus scoticus scoticus) moor for sport. S o m e land within the nature reserve is experimentally enclosed but so far the short period of the experiments has led to nothing but an increased production of herbage and flowers by the associated species. The influences at work on this area are listed below (Bellamy, 1965). Agricultural development. M u c h of the area of the
315
I. G. Simmons
SSSI could be converted at the whim of the landowner to reseeded grassland,using a total herbicide. Further, drainage schemes, e.g. to produce more dry heather moor for grouse m a y dry out some of the interesting flushes and calcareous marshes. Visitor pressure. Both botanists seeking rarities (at least one species has been lost here within living memory) and the public in search of outdoor recreation are attracted to the waterfall of Cauldron Snout or by the spring display of Gentiana verna. No quantitative information, it m a y be noted, exists on damage done in this w a y and this might well be a suitable objective of further research. Sport. Heather-burning is the major land-management practice for grouse production and if overdone this leads to peat erosion and even soil erosion which together change the drainage régime in the direction of flashier run-off. Industrial development. The large industrial conurbation of Teeside (population c. 350,000) is about 40 miles (65 km) away and is dominated by Imperial Chemical Industries. They require to increase their water supply from 25 to 50 million gallons (c. 1.12.2 million hectolitres per day) between 1965 and 1970 (Cooper,1966) and the Tees Valley and Cleveland Water Board have put before Parliament a Bill to enable !hem to construct a reservoir in Upper Teesdale of 700 acres (c. 285 hectares) which would drown 20 acres (8 hectares) of special botanical interest, cause damage to other areas during construction and possibly exterminate some species because of the microclimatic effects associated with a large water body. BSBI have fought the Bill, which has been favourably reported upon by a Select Committee of the House of C o m m o n s and n o w awaits passage through the House of Lords and the Royal Assent. At the time of writing (August 1966) these stages were not assured but seemed likely to be passed successfully. This area is perhaps at present the most extreme case of difficulty in the preservation of a relict flora, and most of the other sites are in Scotland where the pressures are less at any rate for the present. DISCUSSION The cost of keeping our relict floras in terms of foregone agricultural or industrial alternatives is not known. Listed below in general term5 are some reasons often advanced for protection. 1. As part of a genetic pool. But are any of these populations sufficiently large to contain enough diversity to be of any use for eventual breeding, especially since most of them are abundant elsewhere in the world ? Britain merely has the “scragend” of their distribution by virtue of her location.
316
2. Ecological balance. If it is argued that they occupy a specialized niche which no other plant can fill then can one locality properly be called a niche ? D o any of these plants represent balance points within ecosystems without which the systems would degrade ? More research is doubtless needed here. 3. Scientific research. The most convincing scientific argument is probably that too little is known about the biology of these plants in this setting (usually at the limits of their ranges)-some are genetically different from their continental relatives-and that they should be kept until the country can afford to investigate them. 4.Diversity for aesthetic satisfaction. Is it possible to balance this against the material returns from agriculture and industry ? Can it not be argued that it is easy to travel and see nearly all these plants where they are relatively c o m m o n and that one or two species apart, they do not contribute to landscape diversity at any scale ? This paper raises more questions than it answers. It m a y be suspected that the attachment to those relict plants which raises thousands of pounds for their defence is primarily emotional and aesthetic, although on scientific grounds there is a case for their preservation on the grounds that so little is known about them. Primarily it would seem that in a small but industrial country such as Britain a major problem of conservation is to keep up the general quality of our environmental resources by having as much biological and landscape diversity as possible. Whether this problem will ever be solved without regard to the other half of any resource problem equation-the population of the unit concerned-is difficult to say, and in the long run perhaps only the pegging or even the reduction of Britain’s population w ill allow them to survive.
CONCLUSION F r o m the perspective of a country which is very marginally Subarctic but densely populated, the Subarctic proper would seem to the present writer to have one of its greatest values as a resource (in terms of European and North American population and culture) as a wilderness area to which people from the heavily populated and “tamed” regions to the south can have resort (Simmons, 1966). It follows that “development” of the Subarctic will not often lead to an increase in the quality of it as a wilderness area and so it is suggested that its future be viewed keeping in mind the idea of the Subarctic as an element of diversity and cultural enrichment for those w h o seek it, rather than merely a biological unit whose productivity must needs be exploited.
Problems of the conservation of relict arctic and subarctic-speciesin Britain
Résumé Conservation des survivances botaniques arctiques et subarctiques en Grande-Bretagne (I. G. Simmons)
C’est surtout dans les régions montagneuses des îles Britanniques, et notamment dans les Highlands d’Ecosse et les landes du nord de l’Angleterre, qu’on trouve des survivances des formes végétales évoluées arctiques et subarctiques,encore que certaines plantes appartenant à cette catégorie se retrouvent plus au sud, notamment la Vaccinium vitis-idaea, à Dartmoor, dans le sud-ouest de l’Angleterre. Quant aux espèces présentes sur le mont Cairngorn, il ne s’agit pas tant de vestiges que de plantes vivant encore dans des conditions subarctiques. J. R. Mattews a défini et classé c o m m e ensembles floristiques quatre principaux groupes de survivances botaniques. Tous comprennent des plantes généralement répandues en Europe mais qui, dans les îles Britanniques, sont peut-être à leur limite extrême. Certaines d’entre elles ( E m p e t r u m nigrum, Vaccinium vitis-idaea, par exemple) sont fort communes, mais beaucoup d’autres le sont peu, ou sont m ê m e rares. Ce sont pour la plupart des vestiges du quaternaire (phase de la Vistule), époque à laquelle elles constituaient des formes végétales courantes. Dans le dynamisme actuel des formations végétales à survivances arctiques et subarctiques entrent des facteurs tels que les modifications climatiques, les changements intervenus dans
Bibliography BELLAMY, D.J. 1965. Conservation and Upper Teesdale. In: D. H. Valentine (ed.), The natural history of upper Teesdale. Newcastle upon Tyne,p. 59-65. The vegetation of Scotland.
BURNETT,J.H . (ed.). 1964.
Edinburgh. CLAPHAM, A.R.; TUTIN, T.G., WARBURG, E.F. 1962. Flora of the British Isles, 2nd ed. Cambridge. COOPER,J. A. 1957. Britain’s water supply. ICI Magazine, vol. 44, no. 328, p. 67-69. EDDY,A. 1963. Conservation and land use in Moor House NNR, Westmorland. In: E. Milne-Redhead (ed.), The conservation of the British flora. London. ELKINGTON, T.T. 1963. Biological flora of the British Isles: Gentiana vernu L. J. Ecol.,vol. 51,no. 3, p. 755-767. . 1964. Biological flora of the British Isles: Myosotis alpestris F.W.Schmidt.J. ECO~., vol. 52, no. 3,p. 709-722. __. , WOODELL, S. J. R. 1963. Biological flora of the British Isles: Potentilla fruticosa L. J. Ecol., vol. 51, no. 3, p. 769-781. GODWIN, H.1956. The history of the British flora.Cambridge. GREEN,F.H.W. 1964. The climate of Scotland. In: J. H.Burnett (ed.), The vegetation of Scotland, p. 15-35. JONES, V.; RICHARDS, P.W.1962. Biological flora of the British Isles. Silene acaulis (L) Jacq. J. Ecol., vol. 50, no. 2, p. 475-487.
_-
l’utilisation des terres, tantôt favorables, tantôt défavorables aux espèces rares (exemple :le pâturage), l’évolution démographique du lapin, l’emploi des herbicides totaux tels que le paraquot,et l’assèchement des marécages. L e déclin de certaines espèces est imputable aussi aux cueilles faites par les botanistes et les amateurs de jardinage ainsi que par le grand public. Grâce à des mesures de conservation,certaines plantes ont pu être mises à l’abri dans des réserves naturelles nationales gérées par une Commission de la protection de la nature. Bien d’autres, en revanche, ne survivent que parce que personne ne sait qu’elles existent ou parce qu’elles sont inaccessibles. Un cas intéressant est celui de la région d’Upper Teesdale, où l’on trouve de vastes peuplements de plantes rares. I1 est question d’y construire un grand barrage, qui submergerait une bonne partie des habitats de ces plantes. Aussi la Société botanique des îles Britanniques s’opposet-elle à ce que le Parlement vote la loi autorisant la construction du barrage. Cet exemple pose une question d’ordre général :un pays tel que la Grande-Bretagne doit-il se donner beaucoup de peine et dépenser beaucoup d’argent pour sauvegarder des espèces qui sont fort répandues dans d’autres parties d’Europe, et quelles sont, en fait, à échéance lointaine, les chances de sauvegarder les espèces botaniques rares dans un pays industriel tel que l’Angleterre où ces formes croissent non loin des agglomérationsurbaines ?
I Bibliographie LAMB, H.H. 1965. Britain’s changing climate. In: C. G. Johnson and L.P. Smith (eds.), The biological signi$cance of climatic changes in Britain. London.(InstituteofBiology Symposia, no. 14). MANLEY, G.1944. Topographical features and the climate of Britain. Geogr. J.,vol. 103,no.6,p. 241-263. -. 1952. Climate and the British scene. London. MATTHEWS, J. R. 1955. Origin and distribution ofthe British $ora. London. D.A. 1962. Plant communities MCVEAN, D.N.; RATCLIFFE, ofthe Scottish.Highlands.London.(Nat.Cons.Monog.,no. 1.) PEARSALL,W . H.1950. Mountains and moorlands. London. PERRING, F.; WALTERS, S.M. (eds.). 1962. Atlas of the British flora. London,Edinburgh. PIGOTT, C. D.1956. The vegetation of Upper Teesdale in the North Pennines.J. Ecol., vol. 44, no. 2, p. 545-586. POLUNIN, N. 1951. The real Arctic: Suggestions for its delimitation, subdivision and characterization.J. Ecol., vol. 39, no. 1, p. 308-315. . 1954. Vascular plants common to the Arctic and the British Isles: enumeration of species. Wutsonia, vol. 3, no. 2, p. 92-100. SIMMONS, I.G. 1966. Wilderness in the mid-20th century U.S.A.T o w n Planning Review, vol. 36, no. 4, p. 249-56. VALENTINE, D.H. (ed.). 1965. The natural history of Upper Teesdale. Newcastle upon Tyne.
__
317
Subarctic peatlands and their utilization R a u n o Ruuhijärvi
SUBARCTIC P E A T L A N D S So far n o attempt has been m a d e to define a subarctic peatland vegetation zone, although regional differences in peatland and forest vegetation have proved to run largely parallel. In the Northern Hemisphere the transition from boreal to arctic peatland vegetation is best represented by the “palsa” peatland zone in the northern boreal taiga and forest tundra. T h e n a m e “palsa” is derived from Lappish and has c o m e into fairly wide use as a scientific term. Palsas constitute the most striking and characteristic feature of these peatlands, but they are by n o m e a n s the sole distinction from the aapa fens that occur farther south or from the polygon peatlands of the Arctic proper (cf. Katz, 1948; P’javcenko, 1955; Ruuhijärvi, 1960). T o regard the palsa peatlands as subarctic is consistent, for instance, with Hustich’s (1960)opinion concerning the position of this zone. B u t other demarcations of the subarctic zone have been suggested. For instance, Hare (1954)a n d Sjörs (1963b) call the northernmost part of the boreal forest vegetation zone subarctic. Its counterpart o n peatlands would be the most tropical part of the aapa fen zone in northern Europe as well as Canada. On the other hand, Ahti, Hämet-Ahti and dalas (1964)have suggested that the whole of Fennoscandia, including the mountain birch forest area lying to the north of the coniferous forest zone, should be considered as belonging to the northern boreal zone as subalpine regions. According to the latter workers, only the vegetation beyond the continuous forest belt is orohemiarctic (which seems to be the nearest category to subarctic, taken in a strict sense). In Fennoscandia it is characterized by small birch forest islets or solitary trees and possibly by s o m e of the dwarfshrub heaths o n the treeless mountains. Palsas still occur in this altitude
belt too, and cannot be distinguished from the palsa peatlands of the subalpine mountain birch forests proper. E v e n though it is the palsas which I consider to be subarctic peatlands, I shall start with a short review of the peatland zones lying to the south and north of t h e m in the northern hemisphere.
REGIONAL PROBLEMS
IN B O R E A L P E A T L A N D S Peatlands m a y be subdivided into t w o m a i n ecological groups: bogs and fens. T h e life of bogs is sustained by rain; they are therefore extremely oligotrophic or ombrotrophic. In fens, the nutrients c o m e from the soil, and they accordingly have a wide nutrient amplitude ranging from poor to rich. T h e boundary between bog and fen, the fen plant limit, can be seen from the vegetation. This is the most important ecological boundary in peatland vegetation. In the circumpolar-boreal zone this ecological sub-division of peatlands is also appropriate for use as a basis w h e n the regional distribution of peatland complexes is studied. T h e southern parts of the coniferous forest belt adjoin a distinct raised-bog zone; while in the central and northern parts there is a n aapa-fen zone. “Aapa” is a w o r d used in northern Finland to denote a kind of peatland; it w a s first introduced in to the literature by A. K. Cajander (1913), and it has enjoyed increasing international use in recent years. E v e n in the aapa-fen zone raised bogs m a y be formed, but this is only possible at topographically and hydrologically suitable sites -especially o n watersheds, riversides and lake shores. As a rule they are only established at sites from which the flood-water is rapidly drained. On the banks of the big rivers of the northern
319
R. Ruuhijärvi
FIG.1. The regional division of peatlands in Finland.
taiga these northern raised bogs, which differ from those of more southerly location and represent a complex type of their own, m a y even attain quite considerable dimensions (Sjörs, 1959, 1963a ; Ruuhijärvi, 1960, 1963). There is hardly any other kind of vegetation so closely identical, for instance in northern Fennoscandia, northern West Siberia and the Hudson B a y Lowlands in Canada. In all these areas, however, minerotrophic aapa fens clearly predominate on the central and northern taiga. The powerful spring in the central and northern parts of the coniferous forest zone m a y indeed be considered the must important factor contributing to the establishment of minerotrophic aapa fens. But it has to be admitted that other climatological factors,such as the low evaporation and short growing season, all act in the same direction. On the other hand, the formation of raised bogs is restricted by soil and bed-rock rich in nutritients, even in places which climatologically and hydrologically would otherwise be suitable. Finland is a rewarding country for regional peatland
320
3
research,for it extends all the w a y through the boreal coniferous forest belt up to the subarctic regions. In fact, in recent years, Finnish botanists have been engaged in clarifying regional peatland problems (Ruuhijärvi, 1960; Havas, 1961; Eurola, 1962) and w e have worked out a new regional division based on the differences in peatland vegetation caused by the climate (Fig.1). The raised-bog zone of southern Finland extends to about latitude 630 N. and even farther north in the maritime coastal belt of the Gulf of Bothnia. Aapa fens thrust farthest towards the south in watershed areas, which are rich in peatlands and have a less favourable climate. The greatest width of our aapa-fen zone is about 600-700 km. In the north it reaches to approximately the northern limit of the pine forest. Within this zone, too, there are distinct differences in the direction from north to south. I have divided it into the southern,main and northern subzones. The southernmost aapa-fen zone is characterized by comparatively dry peatland. Their morphology-strings and hollows-is poorly developed,
Subarctic peatlands and their utilization
as a result of relatively minimal water flow and comparatively mild freezing phenomena. At higher latitudes the proportion of hollows increases,but it is only in the next aapa-fen zone that, with narrow strings, they are seen to form the entire open fen. In this main aapa zone we already find ourselves fairly close to the subarctic regions which constitute our topic. The most typical aapa-fen scenery is encountered here. In level terrain, the fens often attain vast dimensions. The most extensive complexes in northern Finland and in Sweden have an area of about 400 km2.M a n y localities have more peatland than forest. In the northernmost aapa-fenzone,in the northern parts of the coniferous forest zone, the most typical features of the aapas already begin to be less marked. The strings are not so clearly orientated but form a kind of incoherent network or are discontinuous. The proportion of strings on the fen area is far higher than in the south. This is a consequence of powerful freezing phenomena and of the persistence of frost in the strings over the summer period. As a rule, aapa fens in Fennoscandia have no frost extending to a level below the water-table during the growing season. In Finnish Forest Lapland, in the northernmost aapa-fen zone, the peatland is usually bordered by expanses of S p h a g n u m fuseum called “pounikko” in the local vernacular. These are peatlands with high and steeply sloping hummocks, where the vegetation is dominated by dwarf-shrubs, Betula nana and L e d u m palustre. Shaped by powerful freezing phenomena, they are actually counterparts of the pine bogs in more southerly regions, but they are no longer clearly ombrotrophic. Northern raised bogs are also very typical and extensive. L o w solitary palsas also exist in this region. Frost frequently persists over the summer period in the peat hummocks of this zone, and coniferous trees do not thrive on the peatlands longer. It is primarily owing to the permafrost occurring sporadically in the peatlands that the vegetation of the h u m m o c k surface takes on boreal characteristics farther south than the vegetation of the hollows or of the forests in the same region. This northernmost aapa-fen zone, which extends approximately to the pine-forest limit in Finland, already has the distinct character of a narrow transition zone, in which features of the subarctic peatland complex type, the palsa peatlands, begin to appear. A zonal distinction of the peatland vegetation similar to that observed in Finland is also encountered, according to the literature,-in Sweden and according to the Russian authors Boc and Jurkovskaja (1964) in the Karelian Soviet Republic. F r o m the Pechora region in the Soviet Union, Katz (1930, 1948) and KorEagin (1940) described an aapa-fen area located to the north of the raised bogs. Farther east, in the
the river Usa region, aapas have recently been reported by Bo; (1963). Sub-divisionof the extensive S p h a g n u m peatland zone and the corresponding peatland areas of central and eastern Siberia on the basis of ombrotrophy-minerotrophy would also be desirable elsewhere in Soviet territory. Conditions are fairly similar in large parts of Canada. In the regions which have been investigated, the zonal distribution of vegetation seems to be m u c h the same as that found in Fennoscandia. For instance, this applies to the Ungava-Labrador region according to Wenner (1947) and Hamelin (1957) and according to Allington (1959, 1962). Kalela (1962, 1963) has established a distinct correspondence to the Finnish forest and peatland vegetation zones in the Canadian Clay Belt region south of James Bay. The same is suggested by observations of Hustich (1957), Sjörs (1961, 1963a) .and Ahti (1964) from Ontario. The zonality of the peatlands thus appears to be similar at least o n both sides of the Atlantic. F r o m elsewhere, too, more detailed regional studies would be desirable. With aircraft and aerial photographs it is no problem nowadays to make such surveys.
MORPHOLOGICAL FEATURES It is possible to recognize aapa fens by the surface pattern of strings and hollows (Finnish: rimpis, Swedish: Jlarks). The genesis of this formation has been amply discussed,and several more or less speculative theories have been presented. Since hollows and strings are not absent on subarctic peatlands either, ill be necessary in this connexion, their consideration w too. The global distribution of rimpi hollows is restricted to regions having a short summer and a winter with abundant snow but not necessarily with severe cold. On the contrary,occurrence of permafrost in peatland tends to be a factor hampering the development of the normal hollows. Within a peatland complex, the presence of hollows depends on the hydrotopography. W h e n rimpi hollows are formed, the uniformly sloping peatland surface is replaced by a series of steps, i.e. by hollows. The starting point in this process is the competition on the peatland surface between plant communities of different degrees of wetness. As a result of the different rates of peat production of these communities, the differences in wetness are developed and accentuated. The size and shape of the hollows is determined primarily by two factors: the slope of the peatland area and the quantity of water passing through. Not until wet hollows and intermediate, drier, string surfaces have been formed do differences in frost formation begin to influence the existing conditions. In the first place they accentuate the string structure, although this, too, only occurs where a string with S p h a g n u m
321 23
R. Ruuhijärvi
fuscum hummocks has been formed for some other
reason. The complex of hollows and strings is usually best developed when the peatland area has a low gradient. W e have seen no indications of the Fennoscandian peatlands to suggest that the hollows have been formed after the t h a G n g of ancient permafrost, nor are there any signs of solifluction. The stratigraphy both in the hollows and in the strings is the result of a gradual unbroken course of development (cf. Sjörs, 1965 ; Aartolahti, 1965).
PALSA PEATLANDS In the aapa-fen region of Fennoscandia, the occurrence of permafrost is mostly limited to the high h u m m o c k y strings. Only in the northernmost aapa complexes do w e have solitary palsas or transient frost lenses in rimpi hollows. In the palsa peatland region, on the other hand, it is the rule that, in addition to the latent frost of dry h u m m o c k surfaces, active frost occurs in the wet parts of the peatland that have a thick peat layer. The occurrence in hollows of frost lenses persisting for a few years at most is a general phenomenon. The formation of palsas in the present-day climate is rare but possible. Stratigraphic palsa studies and radiocarbon datings appear to indicate that in Fennoscandia palsas m a y have been formed ever since the late part of the Sub-boreal period, that is, 3,000-3,500 years ago, when climate, especially the winter climate, changed to one of sufficient continentality in a thermal respect. The majority of the present palsas are thought to be comparatively old formations. In the present climate they are rather more likely to be in process of thawing. Numerous signs of collapsed palsas are to be seen in the palsa peatland region. Comparison of the stratigraphies and pollen analyses of palsas and rimpi hollows and their radiocarbon datings indicate that the palsas have come into existence as a result of a normal frost-heaving process (Lundqvist, 1951; Ruuhijärvi, 1962). T h e highest k n o w n palsas in Fennoscandia have a height of 7 m (Fries, 1913; Ruuhijärvi, 1960). Higher figures are reported from central Siberia, where the highest palsas m a y reach 10 m (Katz, 1948). The height of the palsas depends not only on the winter climate, but also on the thickness of the sedimentary layer that is subject to frost and the water content of the peatland. In addition to those on a peat substrate, palsas m a y also become established on sites where fluvial or lake sediments have accumulated. Fine mineral sediments also frequently contribute to the formation of palsas. In northern Fennoscandia, palsas are regularly found in clusters in the parts of the peatland area where the peat is thickest, while in extensive areas such formations are lacking. The typical aapa morphology is absent or
322
only very weakly developed. This is the result of the prolonged persistence of sporadic frost in hollows during the growing. The extensive S. fuscum “pounikkos” are often the most characteristic part of the peatland complex. Palsas are rather rare in Finland in the coniferous forest belt. Their principal occurrence there coincides with the subalpine birch forest regions. In Fennoscandia a northern limit also exists for palsa peatlands, because the winter climate in the coastal region of the Arctic Ocean is too mild for their formation. This area is again characterized by aapa fens. In Soviet Union territory, in north-eastern Europe, a zone of high palsas is especially characteristic of the forest tundra and the northeinmost part of the coniferous forest zone. In the thermally highly continental climate prevailing,for example,in western and central Siberia and in the Hudson B a y Lowland of Canada, palsas also occur in the coniferous forest zone far to the south, even as far as the central taiga. However, these peatlands are typical aapa fens in respect of their other characteristics, although in some instances, they are northern raised bogs (see Sjörs 1961, 1963a), and they should not be included in the circumpolar palsa-peatland zone. In delimiting this zone, other characteristics should also be taken into account. Most important among these are the absence of pine bogs and spruce swamps, at least in Fennoscandia (the black spruce is another matter), and the occurrence of subarctic and arctic plant species and communities on hummocks and in hollows. The high-palsa zone defined in this manner primarily comprises the forest tundra and peatlands on the northern margin of the taiga. However, it is still rather closely connected with the boreal aapa-fen zone, owing to the fact that there is no continuous permafrost. The next more northerly peatland zone or subzone is the so-called low-palsa zone in the area of continuous peatland permafrost. Here, owing to the frozen layer, the upward growth of the palsas is inhibited and the hummocks frequently remain as islets only 1 m high. These peatlands are often very extensive. In Europe, the occurrence of low palsas starts on the tundras east of the Kanin Peninsula and on Kolgujev Island. On the European side they are found in the southern tundra zones, whereas they occur already in the northern part of the forest tundra of western Siberia (Katz, 1948). In the last-mentioned regions, mountain ranges and total permafrost disturb the formation of regular peatland zones. Katz (1948) places the limit of palsa peatland zones :a the Anabar river in the east, but, according to P’javcenko (1955), complexes of the same kind occur in the Verchojansk region and in the Kolyma river valley, and probably also elsewhere in the northern parts of eastern Siberia. In the peatland zone with low palsas, Carex apuatilis fens are very typical, and these usually constitute
Subarctic peatlands and their utilization
the paths for the passage of flood-waters. As with low palsas, peat polygons are already encountered on the peatlands of this zone; it is not until the next, distinctly arctic zone of polygon peatlands is reached that such formations are properly CharacterisJic of the scenery (Katz, 1948;Andreev, 1955; P’javcenko,
1955). The occurrence of palsa peatlands is similar in North America. Polygon peatlands are also encountered. However, the reports are so sporadic that no precise regional sub-division can be presented.
PALUDIFICATION IN THE SUBARCTIC REGION In the subarctic region extensive peatland areas are found wherever paludification is not prevented by the topography. However, the peatlands do not attain quite the same extent as in the boreal zone. A very distinct difference is observed in the thickness of the peat layer. In the palsa-peatland zone of northern Fennoscandia, for instance, the peat layer is usually 2-3 m thick at most, but in the overwhelming majority of the peatlands the peat is less than 1 m thick. Similar thicknesses are reported from Soviet territory. According to personal investigations in northern Finland and northern Norway, paludification mostly started more or less simultaneously in the Preboreal or Boreal period, 8,000-9,500years ago. The greater part of the peat layer, often threequarters of it, had already been formed by the end of the Atlantic period. Paludification proceeded exceedingly slowly and has frequently been completely arrested during the last 3,000 years. This is also the period to which the frost phenomena in the peatlands can be attributed. Consistent findings indicating the relict character of the palsa peatlands have been reported from m a n y parts of the Soviet Union (e.g. P’javcenko, 1955; Neustadt, 1966). It is quite obvious that prolonged persistence of frost in the peatlands and a short,cool growing season are factors limiting the paludification and in particular the growth in height of the peat.
PEATLAND CLIMATE Right up to the present day the belief has persisted even in scientific circles that peatlands are plant habitats cooler than their surroundings, constituting veritable frost reservoirs. However, studies on peatland climate in the northernmost aapa fens in Finland (Franssila, 1964) have clearly demonstrated that extensive peatlands possess a more favourable climate than the surrounding forests. The daytime air temperatures remain somewhat lower on peatlands, owing to the heat consumed by evaporation. This
difference is reduced, however, by the advection of w a r m air, which is clearly evident in open country in daytime. At night the minimum temperatures are considerably higher on peatland (often by as m u c h as 20-30 C) because heat is released into the air from water in the hollows which has been warmed up in the course of the day. The mean temperature of wet peatlands, too, is more favourable than that of forests, and thanks to their high night temperatures they have a distinct frost-resistingcapacity. However, conditions are different on dry peatlands that have a h u m m o c k y surface. These m a y even be inferior to mineral soils. The climate of a peatland area also frequently becomes more extreme after drainage. Likewise closely connected with peatland climate is the phenomenon of frost-heaving, which is an important ecological factor affectingboth the peatland vegetation and also the utilization of peatlands.
UTILIZATION OF SUBARCTIC AND RESTRICTING FACTORS IN FINLAND PEATLANDS,
E C O N O M I C SIGNIFICANCE IN U N D R A I N E D CONDITION
In the following review of the utilization of subarctic peatlands, I shall primarily confine my comments to Finnish conditions. The degree of paludification on the northern boundary of Finland’s coniferous forest zone and in the palsa-peatland region is less than half the value in the aapa-fen region of northern Finland, chiefly on account of topography. The peatlands here form only about 20 per cent of the land area. Mostly, they are completely treeless wet fens or permafrosted hummocks. Their economic exploitation is negligible. The peatlands are used by the reindeer (Rangifer tarandus) for summer grazing. It should be noted that in summer the reindeer does not eat m u c h lichen but prefers vascular plants. Particularly in the early part of the summer, peatland sedges, Eriophorurn species, Menyanthes and willows provide an important source of nutrition for reindeer. Moreover, on extensive, treeless peatlands, as in open mountain areas there are usually few gnats or gadflies, owing to the prevailing winds. The higher the nutrient of a peatland area and the higher its land-use quality, the better it serves as reindeer pasture (Ahti, 1961). Peatland areas located close to villages have also been used for cattle pasture. In the past, Carex aquatilis peatlands in particular used to be cut for fodder. Extensive peatland areas also possess significance as wild-life areas. The bird density is often very high on wet subarctic peatlands. Most important from the 323
R. Ruuhijärvi
viewpoint of hunting are the willow grouse (Lagopus lagopus), geese and ducks (Anatidae), e.g. the bean goose (Anser arvensis) and the Anas species. Until quite recently grouse-trapping w a s a n important winter source of livelihood in subarctic regions. Peatlands frequently attract the ordinary m a n only at the cloudberry (Rubus chamaemorus) season. T h e subarctic regions in Fennoscandia have famous berrybearing peatlands to which people c o m e from great distances. In good berry years the cloudberry constitutes a valuable source of income for the local inhabitants. This berry is marketed even in foreign countries. M o r e recently a n entirely n e w method of exploiting the northern peatlands has entered the picture, namely, their conversion to artificial lakes. T h e water systems in northern Finland possess very few lakes; the spring inundations are consequently powerful, and the discharge conditions are very variable. W i t h the requirements of power e c o n o m y in mind, large storage basins have therefore been planned at the sources of the Kemijoki water system. In the next f e w years peatlands totalling about 800 km2in area will be submerged in these basins, including Finland’s most extensive peatland area, Posoaapa, which covers about 400 km2. D R A I N I N G F O R SILVICULTURAL P U R P O S E S
If the subarctic zone is circumscribed to contain only the area of palsa peatlands, n o signgcance can be assigned to its peatlands in respect of the most important kind of utilization, draining for silvicdtura1 purposes. T h e northern limit of draining activities passes from about 680 N. in the western parts of Finland, in a n oblique direction to a point o n the eastern border north of the 67th parallel. There is hardly a n y other part of the world where drainage for silvicultural purposes is practised so far up in the northern taiga. Drainage of waterlogged mineral soils having a thin peat layer, or conduction of surface flow with the aid of water furrows, m a y also be considered at higher latitudes w h e n the a i m is to counteract the tendency to paludification and to prevent d a m a g e to growing tree stands as a result of waterlogging. If a relative figure of about 100 is chosen to represent the average growth of tree stands o n drained peatlands in southern Finland, the corresponding figure at the northern limit at which draining is practised, close to the subarctic region, is less than 35 (Heikurainen 1959). It is s o m e w h a t questionable whether such undertakings can be considered profitable in these conditions. T h e risk of failure here is high, the chief danger being excessive draining. T h e actual purpose of drainage is to promote the mobility of water in
324
the peat. On the northern taiga, spruce s w a m p s on sloping ground, where water flows, are often characterized by better tree growth than surrounding mineral soils. Application of phosphorus and potassium fertilizers in connexion with draining and planting of seedlings m a y essentially increase the silvicultural potential of peatlands in northern regions. Intensive research aimed at improvement of forests is in progress, but has not yet yielded results. P E A T L A N D CULTIVATION Reclamation of peatland has taken place farther north than silvicultural draining, in fact virtually all over Finland. T h e subarctic region, too, has a considerable wealth of peatlands suitable for clearing a n d containing nutrients. N o data are available o n the area of cultivated peatlands in the Subarctic. However, n o great amounts are concerned, seeing that the total area of arable land in the northernmost Finnish c o m m u n e s , Inari, Utsjoki and Enontekiö, is only 2,000 hectares. T h e most important cultivated plants are barley, potatoes, and timothy grass (Phleum pratense). T h e northern limit of barleygrowing coincides with the spruce-forest Iimit. T h e potato, being a long-day plant, gives high yields if it escapes frost damage. Timothy is the most important crop o n reclaimed peatland. Cultivation o n a peat substrate is restricted only by the lower temperature of this substrate in comparison to mineral soils, and by susceptibility to frost caused by low location. L o w temperature also exerts a detrimental effect o n microbial activity and, hence, o n nitrogen rnobilization in the peat. W I N N I N G OF PEAT Peat-cutting for fuel is restricted in the subarctic regions to places where w o o d is not available in sufficient quantity. This situation is chiefly m e t with o n the Norwegian coast of the Arctic Ocean. T h e practice of using peat as building material for huts is becoming rare. In the s u m m e r one m a y still occasionally meet a few Lapps grazing their reindeer and living in a peat hut.
Any kind of peatland utilization is so strongly restricted by climate and frost in the subarctic region that the greater part of the peatlands in this region will continue in their natural state. In the course of the process which is bringing nearly all the peatlands in the boreal zone within the scope of silvicultural draining, the subarctic peatlands will increase in significance from the viewpoints of peatland science, teaching, nature conservation, wandering and tourism.
Subarctic peatlands and their utilization
Résumé Les tourbières subarctiques et leur utilisation (Rauno
l’absence de marécages à pins ou à sapinettes, et l’apparition de communautés subarctiques et arctiques sur les bosses et dans les creux. L a zone à hautes palses ainsi définie est formée principalement de la toundra forestière et de tourbières à la limite septentrionale de la taïga, dans la région de pergélisol discontinu. Les complexes de tourbières à hautes palses dans la ceinture de conifères des régions situées très au sud, par exemple en Sibérie et dans les basses terres de la baie d’Hudson, sont des tourbières réticulées ou, parfois, des tourbières hautes septentrionales. Plus au nord, la zone voisine de tourbières subarctiques est la zone dite à basses palses de la région de pergélisol continu à tourbières. L’existence de la couche gelée entrave la croissance en hauteur des palses, qui demeurent fréquemment à l’état de monticules isolés de 1 mètre de hauteur. En Europe, on les trouve exclusivement dans les toundras méridionales ; alors qu’ils apparaissent déjà dans la partie nord de la toundra forestière de Sibérie occidentale et m ê m e dans des régions de latitude plus basse de Sibérie centrale et orientale. I1 existe également des zones de tourbières analogues en Amérique du Nord. Mais leur description est si incomplète qu’il n’est pas possible d’en faire une analyse régionale précise. On se préoccupe aussi d’utiliser les tourbières de la zone subarctique, bien que leur exploitation à des fins économiques soit d’une portée très limitée.
Ruuhijärvi) Dans l’hémisphèrenord, la transition de la végétation dss tourbières boréales à celle des tourbières arctiques est particulièrement bien représentée par la zone des tourbières à palses, dans le nord de la taïga boréale et la toundra forestière. Les palses constituent la principale caractéristique de ces tourbières, mais elles ne sont nullement le seul trait qui les distingue des tourbières réticulées (aapa) que l’on trouve plus au sud ou des tourbières polygonales plus arctiques. Les tourbières réticulées constituent la plus grande partie des tourbières boréales. En Finlande cette zone a de 600 à 700 kilomètres de largeur. Au nord, elle confine à la limite septentrionale de la forêt de pins. Du sud vers le nord, elle présente des différences marquées. Je l’ai subdivisée en une sous-zone méridionale, une sous-zoneprincipale et Line sous-zone septentrionale. L a sous-zone principale, située en Finlande du Nord, est la plus typique ; elle est caractérisée par de vastes étendues de crêtes et de dépressions. Dans l’étroite sous-zone septentrionale commencent i apparaître les traits distinctifs des tourbières à palses. Les palses de l’extrême-sud de cette région se trouvent dans des complexes dont les autres caractéristiques sont celles des tourbières réticulées. Pour définir les limites des tourbières à palses, il faut également tenir compte d’autres éléments particuliers. Parmi ceux-ci,les plus importants sont
Bibliography
/ Bibliographie
AARTOLAHTI,T. 1965. Oberflächenfornen von Hochmooren und ihre Entwicklung in Südwest-Häme und NordSatakunta.Fennia, vol. 93, no. 1,p. 1-268. AHTI, T. 1961. Poron ravinnosta ja laitumista (Synopsis: On food and pastures of the reindeer). Lapin Tutkimusseuran Vuosikirja,vol. 2, p. 18-28. . 1964. Microlichens and their zonal distribution in boreal and arctic Ontario, Canada. Ann. Bot. Fenn., vol. 1, no. 1, p. 1-35. -. , HAMET-AHTI, L.;JALAS, J., 1964. Luoteis-Euroopan kasvillisuusvyöhykkeistäja kasvillisuusalueista.Luonnon Tutkija, vol. 68, p. 1-28. ALLINGTON, K. 1959. The bogs of Central Labrador-Ungava. A n examination of their physical characteristics, p. 1-89. (McGillSub-Arctic Res. Pap.,7.) . 1962. The bogs of Central Labrador-Ungava. A n examination of their physical characteristics. Geogr. Ann., vol. 43, p. 401-417.
_-
__
ANDREEV, V. N. 1955. Desifrirovanie PO a6rofotosnimkan razliEnyh tipov tundr i ih a6rovizual ’naja harakteristika PO moroznoj treihovatosti. Geogr. Sbornik, vol. 7, p. 103-120. Bo;, M.S. 1963. O b aapa-bolotahna severo-vostoke Evropejoskoj Easti SSSR. Bot. Zurn., vol. 48, p. 1818-1822. ; JURKOVSKAJA, T.K. 1964. Sopostavlenie bolotnyh rajonov Karelii, Kol’skogo poluostrova i Finljandii (Summary: Comparison of bog regions of Karelia, Kola Peninsula and Finland). Bot Zurn., vol. 49, p. 980-988. CAJANDER, A.K.1913. Studien über die Moore Finnlands. Acta Forest. Fenn., vol. 2, no. 3, p. 1-208. DRURY, W.H. 1956. Bog fiats and physiographic processes in the upper Kuskokwim river region, Alaska, p. 1-130. (Contr. Grey Herb. Harvard Univ., 178.) EUROLA, S. 1962. Uber die regionale Einteilung der südfinnischen Moore. Ann. Bot. Soc. “Vanamo”, vol. 33, no. 2, p. 1-243.
-
325
R. Ruuhijärvi
FRANSSILA,M . 1964. O n the temperature conditions in a large aapa bog area in Finnish Lapland, p. 1-21. (Finnish Meteorol. O 5 c e Contr., 53.) FRIES, T.C.E. 1913. Botanische Untersuchungen im nördlichsten Schweden. Vetensk. och prakt. unders. i Lappland anord. af Luossavaara-Kiirunavaara Aktiebolag, p. 1-361. HAMELIN, L.-E. 1957. Les tourbières réticulées du QuébecLabrador subarctique: interprétation morpho-climatique. Cahiers Géogr.,Québec,vol. 3, p. 87-106. HARE, F.K. 1954. The boreal conifer zone. Geogr. Stud., no. 1, p. 4-18. HAVAS, P. 1961. Vegetation und Ökologie der ostfinnischen Hangmoore. Ann. Bot. Soc. “Vanamo”, vol. 31, no. 1, p. 1-188. HEIKURAINEN, L. 1959. Tutkimus metsäojitusalueiden tilasta ja puustosta. (Ref.: über waldbaulich entwässerte Flächen und ihre Waldbestände in Finnland.) Acta Forest. Fenn., vol. 69, no. 1, p. 1-279. HUSTICH, I. 1957. On the phytogeography of the subarctic Hudson Bay Lowland. Acta Geogr.,vol. 16, no. 1, p. 1-48. . 1960. Plant geographical regions. A Geography of Norden, Oslo, p. 54-62. KALELA, A. 1962. Notes on the forest and peatland vegetation in the Canadian Clay Belt Region and adjacent areas. I. C o m m . Inst. Forest. Fenn.,vol. 55, no. 33, p. 1-14. . 1963. Notes on the forest and peatland vegetation in the Canadian Clay Belt Region and adjacent areas, II. Comm. Inst. Forest. Fenn., vol. 57, no. 5, p. 1-19. KATZ, N.J. 1930. Zur Kenntnis der Moore Nordost-Europas. Beih. zum Bot. Centralbl., vol. 16, no. 2, p. 279-394. .1948. Tipy bolot SSSR i Zapadnoj Evropy i ih geografic‘eskoerasprostranenie. Moskva, 316 p. KOREAGIN, A. A. 1940. Rastitel’nost’ severnoj poloviny PeEoreko-IlyEskogo sapovednika. Trudy Pec‘.-Ylye‘. Gosud. Sapov., no. 2, p. 1-416.
__ _-
__
326
LUNDQVIST, G. 1951. E n palsmyr sydost om Kebnekaise. Geol. För Förh., vol. 73, no. 2, p. 209-225. NEUSTADT, M.I. 1966. Prinzipien zur Rayonierung der Moore in der USSR.Ann. h a d . Scient. Fenn.,ser. A (III), vol. 89, p. 1-29. P’JAVCENKO,N . I. 1955. Bugristye torfjaniki.Moskva, 280 p. RUUHIJÄRVI, R. 1960. Uber die regionale Einteilung der nord-finnischenMoore. Ann. Bot. Soc. “ Vanamo”, vol. 30, no. 1, p. 1-360. . 1962. Palsasoista ja niiden morfologiasta siitepölyanalyysin valossa (Ref.:aber die Palsamoore und deren Morphologie im Lichte der Pollenanalyse). Terra, no. 2, p. 58-68. . 1963. Zur Entwicklungsgeschichte der nordfinnischen Hochmoore. Ann. Bot. Soc. “Vanamo”, vol. 34, no. 2, p. 1-40. SJÖRS, H. 1959: Bogs and fens in the Hudson Bay L o w lands. Arctic, vol. 12, no. 1, p. 1-19. . 1961. Forest and peatland at Haeley Lake, Northern Ontario. Nat. Mus. of Canada Bull.,no. 171, p. 1-31. . 1963a.Bogs and fens on Attawapiskat River, Northern Ontario. Nat. Mus. of Canada Bull., no. 186, p. 45-133. . 1963b. Amphi-Atlantic zonation, nemoral to arctic. In: A. L O V E and D.LOVE (eds.), North Atlantic biota and their history, p. 109-125.Oxford. . 1965. Regional ecology of mire sites and vegetation. Acta Phytogeogr. Suecica, vol. 50, p. 180-188. W E N N E R , C.-G.1947. Pollen diagrams from Labrador. Geogr. Ann., vol. 24, p. 137-374. ZINSERLING, G.II., 1932. Geograíija rastitel’nogo pokrova severozapada Evropejskoj Easti SSSR. (Ref.:Die Geographie der Vegetationsdecke des Nordwestens des europäischen Teils der U.d.S.S.R.) Trudy Geomorf. Inst., vol. 4, p. 1-377.
_-
--
_-
__ _-
The rational use of forests and bogs in view of comparative observations in north-westernCanada and northern Finland ~
Eckart Ehlers
Both Canada and Finland are countries in which extensive agricultural colonization is still being carried out today. In Finland the area of n e w farming land is restricted mainly to the northernmost provinces of Oulu and Lapland but extends far north of the Arctic Circle. The Canadian agricultural frontier is centred upon two pioneering areas: the Great Clay Belt of Ontario and Quebec and the Peace River Country of north-western Alberta. The northern margin of agriculture in Canada can be delineated approximately at 590 N. Thus, agriculture and forestry are carried out in both northern Finland and north-westernCanada under subarctic conditions. However, the extent of bogs and forests, their use, and the methods and techniques of their cultivation, are quite different. The characteristic trait of Canadian agriculture is that the aim of the Canadian farmer consists in placing as much land as possible under the plough. In order to achieve this he not only cultivates those soils covered with worthless bush and scrub, but also clears all forests and trees on his land. Forestry of any kind is alien to the Canadian farm economy and nowhere to be found as a part of the agricultural land use. Thus, there is a distinct division between privately-handled agriculture and company-owned forestry. Agriculture in the Canadian north-west, i.e. first of all in the Peace River Country, has long been a northward extension of prairie agriculture, wheat being the primary crop grown. It was mainly owing to several economic setbacks, poor harvests and the research work carried out in the government experimental stations of Beaverlodge and Fort Vermilion that the predominance of wheat has ceased in recent years. Nowadays the agriculture of the Canadian North m a y be more aptly described as a mixed agriculture based on the symbiosis of grain, livestock
and special crops. It is obvious, from this list of farm-income sources, that in none of the farming areas is forestry of any economic importance. It is also worth emphasizing that, as a source of farm income, dairying is of very limited importance and restricted to a few special areas. However, it should be pointed out that the expanding cultivation of grasses. legumes, flax and rape for the purpose of seed production is of increasing importance for the economy of the Peace River Country. T h e achievements of the last twenty years,resulting in a greater variety of farm products and farm economy being expanded on a wider basis are, however, only an intermediate stage on the w a y towards an “ideal” form of land use in north-western Canada. According to the results of research carried out at the Fort Vermilion Experimental Station, serving north-western Canada from 570 N. to the shores of the Great Slave Lake, a grass and livestock economy with an 8-year rotation cycle would represent the best form of agricultural land-use.It should be based on the cultivation of grass and legumes in the first four years, flax and rape in the fifth, and coarse grains in the sixth to eighth year. T h e greath advantages of this rotation cycle are first that the proposed land use is very well suited for the climatic and pedologic conditions of this area, and secondly that farmers are able to practise a diversified agriculture with possibility of refining and finishing the crops. A natural supplement to this form of land use is to breed more beef-cattleand hogs. It is a promising and economical enterprise in so far as grasses and legumes, after having been cleaned for the purpose of seed production, still provide excellent fodder. This is true also for barley and oats; which do not ripen annually,owing to marginal climatic conditions, but which always show sufficient growth to be used as green fodder for livestock.
327
E. Ehlers
The above-mentioned proposals and considerations correspond with that of Canadian agriculturists and agronomists w h o also stress the point that a decisive reduction of grain acreage and of fallow land, which in some parts of the Peace River Country comes to 40 per cent of the total cultivated land area, is desirable. As these acreages are reduced, the area of forage crops should be proportionally increased, i.e. by approximately 100 per cent. These changes lead to a major increase in both seed production and live-stockbreeding with beef as the main product. Special attention should be given to the importance of forage-,flax-and rape-seedproduction in the Peace River Country of north-western Alberta. For some varieties of forage seeds the Peace River Country is Canada’s greatest producer: in 1961 about 77 per cent of Creeping Red Fescue seeds, 43 per cent of broom grass seeds and 36 per cent of Alsike clover seed were produced in the Peace River Country. Also the acreage for growing rape has increased enormously: it expanded within three years from 121,000 acres (1962-63)to 445,000 acres (1965). The reason for the extraordinary importance of forage and oil seeds is to be found partly in the favourable ecological conditions of north-western Canada for the cultivation of these cfops and partly in their commercial interest for the Canadian farmer w h o is very adaptive to economic trends and whose farm is not a farm in the European sense of the word, but a “viable commercial venture”. The favourable natural conditions and the ecological suitability of the Peace River Country for the cultivation of these crops is stressed by the fact that in Fort Vermilion, i.e. 590 N. at the northernmost margin of continuous agriculture in Canada, a seed farm has been established. This commercial enterprise,covering more than 8,000acres, came into operation in 1955 and, within eight years grew into a “seed factory” with ten year-round employed workers, three tractors for the continuous breaking of n e w arable land, its o w n seed-cleaning plant and air strip from which the seeds are flown to all parts of Canada, to the United States and even to Europe. Forestry in north-western Canada, despite its remarkable potential, is of minor importance only in regard to the whole economy of the area. The main reason is that traffic conditions did not offer the possibility of developing forest resources. These conditions have changed considerably in recent years. A number of pulp and paper mills have been or are in the course of being built in north-western Canada. Several smaller saw-millsare operating in the Peace River Country. All these developments, however,have no, or only limited,influenceon the economy and ecology of the whole region. Their only impact and influence on the agricultural development of the Canadian north is the fact that the wood-processing companies offer jobs as forest workers to a few hundred
328
pioneer settlers during the winter-months. Practically all company-owned timber stands are very isolated and at considerable distances from the agricsltural areas. Most of the forests still belong to the provincial government and their present and potential values have hardly been explored. Presently their only value is their very large extent, of little use except to a few trappers and hunters w h o cross these forests during the winter months. In contrast to the Canadian farm economy the Finnish form of agriculture must be described as a symbiosis of forestry and farming. The Finnish farm therefore represents the exact opposite to the Canadian farm. This contrast, however, is not only economic, but is important also as regards the physiognomy of the Finnish agricultural landscape and the techniques or methods of agricultural colonization. While the Canadian farmer w ill try to put the plough to as m u c h land as possible and strives to clear all the forest land on his farm, the Finnish farmer will tend to preserve his forests to a large degree. Instead of turning the sometimes sandy and very poor forest-covered mineral soils to agriculture, he prefers to clear the boggy areas with eutrophic organic surface material for cultivation. But even where wooded soils contain enough nutrients for agriculture to be successful, the Finnish farmer-in contrast to his Canadian colleague-will likely prefer to clear boggy areas in order to preserve the forests which are of such great value to the Finnish farm economy. These differences in technique and method of land reclamation and improvement result in the amalgamation of forests, arable and waste lands in Finland and in a distinct separation of forests and arable lands in Canada. Agriculture in northern Finland is markedly limited to the cultivation of grass for the purpose of producing hay. About 70 per cent of the arable land in northern Finland, and 58 per cent in the average of the whole country are cultivated with grass. In northern Finland the second crop in acreage is barley with 12.1 per cent, then oats with only 5.2 per cent of the total arable area. These few figures m a y be sufficient to show the extremely narrow basis on which agricultural production of northern Finland is founded. Grass and coarse grains are grown for lhe purpose of maintaining a comparatively intensive dairy industry in northern Finland. As a matter of fact, dairying is by far the most important source of farm income, as far as this stems from agriculture. Comparing 65 sample farms in the area of the Oulu Agricultural Society and 53 sample farms in the area of the Peräpohjolan Agricultural Society, covering the area north of it, m a y prove the following statement. According to these samples, which were taken in 1963-64, the farmers derived in Oulu 62.3 per cent and in Peräpohjola 70.5 per cent of their agricultural farm income from the sale
The rational use of forests and bogs
of milk and milk products; another 24 per cent in Oulu and 17.4 per cent in Peräpojola from other animal products, especially from the sale of cattle. Thus, the remainder only about one-eight of the total agricultural farm income, is derived from the sale of all kinds of crops vegetables or berries. It is especially noteworthy that neither grain production nor special crops are of any agricultural importance. The almost complete absence of cattle- and pig-breeding with the purpose of meat production is also a typical trait of farm economy in northern Finland; the uniformity of the agricultural production shows that this farm economy is extremely subject to economic fluctuations. Also, the fact that the acreage of arable land amounts to an average of only 25-50 acres per farm makes it easy to understand that all these farms must have a complementary source of income. This economic complement is supplied by the forest with which almost every Finnish farm is equipped. The economic importance of forestry is twofold: first, the farmer-owned forests annually provide a certain amount of additional income by the sale of wood, which can reach 25 per cent of the total farm income; secondly, the state-owned forests supply most of the Finnish farmers with an annual off-farm income also amounting to about 20 or 25 per cent. During the winter months there is an increased need for forest-workers to ensure cutting, logging and transportation of the wood. Thus, on the whole, it seems that forestry is an ideal complementary addition to farm economy in Finland, especially under the subarctic conditions of Finnish Lapland. Investigations in Finnish Lapland showed that 14 sample farms derived 42 per cent of their total income from agriculture, 27 per cent .from forestry and 25 per cent from secondary enterprises, i.e. mostly from forestry work. Although these figures give a fair impression of the average situation, the importance of complementary forestry varies in different parts of the country, and especially according to the size of differentfarms.Thisis true particularlyin Lapland,the only area of Finland in which large-scale agricultural colonization is being carried out at present. Investigations in three recently-startedsettlement areas revealed that forestry is the only source of farm income during the first years after setting up the n e w farms. For instance, in Puolakkavaara, an area settled with 40 farms in 1962 about 20 miles east of Sodankylä in Finnish Lapland, none of these farms had in 1965 any remarkable income derived from agriculture. All the settlers stated that forestry and secondary enterprises were the main sources of the farms’income. The same is true for the settlement area of Jouttiaapa,40 miles south-east of Rovaniemi, also settled in 1962 where seven out of nine farmers considered forestry to be still more important than agriculture. In Kuparivaara,however, an area settled with 45 farms in 1956,
approximately 20 miles from Kuusamo, almost all farmers considered forestry as a necessary, although not primary,source of farm income. W e m a y therefore conclude that forests are of decisive importance for the development of n e w farms in subarctic northern Finland. Later, forests and forestry provide a second but still necessary source of income and it is only after m a n y years of rational farm management that agriculture m a y become the sole source. In northern Finland, however, the sale of wood or forest work will probably always be a necessary secondaryenterprise to provide the farmers with an economic standard equal to that in other parts of the country. Before evaluating the advantages and disadvantages of the present use of forests and bogs in both northern Finland and north-western Canada, the main differences should be pointed out and summarized. 1. In north-western Canada the development of agricultural farm land is restricted to those areas covered with mineral soils and consequently with a more or less thick forest cover. The forests are cleared but boggy and swampy areas are hardly thought to be valuable potential for farming. In Finland, on the contrary, forests are spared wherever possible, and cultivation of n e w land is directed mainly towards boggy soils, which have to be drained and broken. 2. F a r m economy innorth-westernCanada is unilaterally based on agriculture, while in Finland the symbiosis of agriculture, forestry and forest work is the necessary basis for sound farm management. 3. Agriculture itself is characterized by a large range of crops grown in north-western Canada. Besides cereals of all kinds, m a n y farms have specialized in various crops such as grass, forage, flax and rape seeds. Cattle and pigs are kept for meat production, dairying is of minor importance only. In Finland, on the contrary, agriculture has a very narrow basis, almost the only source of farm income being milk. Grasses are grown to produce hay; other crops or beef-cattle and pig breeding are hardly worth mentioning. It is against this background that possible proposals should be made for amendments within the boreal forest zone and for the rational use of forests and bogs within the Subarctic proper. However, one should keep in mind that these suggestions are of a somewhat restricted value in so far as boreal and subarctic north-western Canada still holds a potential arable land approximating 15-20million acres and a reserve of untouched forest for exceeding that of the arable land. In Finland,on the contrary, only comparatively small areas of unused forests and arable land are left so that the planning and use of the renewable resources has to be somewhat different. Nevertheless the comparison between both areas gives some indication as to h o w to improve the use of forests and bogs in both north-western Canada and northern Finland
329
E. Ehlers
and h o w to broaden the economic basis of both these regions. As set forth in the beginning of this paper, forests and bogs are hardly used in north-western Canada, where bogs are considered as waste lands and forests cut d o w n along the agricultural frontier to be turned into arable land. The tremendous land potential of Canada hardly justifies at present the expensive cultivation of boggy and swampy areas. However, the example of northern Finland shows that a more rational use of forests could raise the whole economy of a region. This should be especially true for large parts of the Peace River Country in Alberta and British Columbia where each year m a n y pioneer settlers have to leave their homesteads during the winter months, because of the lack of industries which could provide them with additional earnings. The absence of wood-processing factories where the settlers could sell the timber cleared from their land instead of burning it, is also a disadvantage in large areas of the Canadian north-west. These facts alone show the necessity to increase industrialization in this region, for which forestry and wood-processing industries would be the natural foundation. Owing to these conditions increased industrial use of forests along the agricultural frontier of the Canadian north-west would be desirable. Pulp and paper mills would not only meet the steadily rising demand of the international market, but would also utilize the latent wealth of the forests. Along with these activities, careful reforestation and forest conservation would offer jobs to m a n y thousands of settlers and provide them with additional sources of income. These proposals do not advocate the s y m biosis of forestry and agriculture that is so typical of Finland, but stress the need for a rational use of forests. This would not only mean an economic b o o m for the areas in question, but would first of all enlarge the field of h u m a n activities under the subarctic conditions of the Canadian north-west. Agriculture in the Canadian north is remarkably varied. Without doubt its main characteristic is its conscious turn towards highly specialized crops such as grass and forage seeds. Export to foreign markets in recent years has been very important to Canada with large quantities of forage seeds being sold not only to the United States, but also to the United Kingdom, Sweden and central Europe. The Peace River Country has a big share in these markets. It is to be expected that north-western Canada will increase this type of agriculture, which is so aptly suited to the ecological conditions of subarctic regions. The situation of boreal and subarctic northern Finland seems m u c h more complicated. The rational use of forests and bogs is a vital qqestion for the economy of the whole country, especially as there are only small land reserves left. Comparatively, land
330
use in Finland should be described as extremely intensive in regard to the extent of the cultivated land. Even in the far north of Finland agriculture and forestry are being carried out wherever possible. ,Thus,according to Ilvessalo (1960),in the communities of Kittilä or Codankyla productive forests cover over 50 per cent of the total land area, in Inari even over 70 per cent. Only in the hilly parts of Enontekiö, Inari and Utsjoki are the forests reduced to less than 10 per cent of the total land area. The agricultural production of the admittedly small fields is restricted to the cultivation of grass, oats and barley, the products of kitchen gardens and also beets and rape (Ohlson, 19606). In northern Finland although the actual potential of forests and bogs is limited, and although agriculture is restricted mainly to swampy and boggy areas, the comparison between north-western Canada and northern Finland reveals some possibilities, the adaptation of which should greatly increase the value and possibilities of subarctic Finnish agriculture. These possibilities are first of all: (u) the increased production of grass and forage seeds for export; and (b) a distinct decrease in milk production in favour of an increase of meat production by breeding beef-cattle and pigs. As said before, grass is the most widely cultivated plant in Finland with over 50 per cent of the total arable area; nevertheless, this is insufficient to meet the requirements of the Finnish market. Production is obviously sufficient to supply the necessary winter fodder for the cattle, but hardly sufficient to supply the necessary grass seeds. According to Westermarck (1956),it is one of the main mistakes of the Finnish farmers to grow grasses and other fodder crops, sqch as alfalfa for too m a n y years on the same ground. Thus productivity decreases after three or four years and consequently Westerrnarck (1956) recommended a tri-annualrotation of the alfalfa-grass crop combination. As a pre-conditionfor the change he advocated a strong increase in the production of grass seeds, especially of alfalfa. The situation in Finland has hardly changed since 1956. Despite its favourable physiogeographical conditions for grass production Finland still imports seeds instead of being an exporter as it normally should be. In 1962, the country imported flax and rape seeds worth over 9 million Finnmarks and grass seeds worth over 1.3 million Finnmarks. This seems an abnormal situation and is worth examining in regard to a more rational use of the arable lands. As reported by several farmers and district agriculturists in northern Finland, it is both agricultural advice and capital investment which are the causes of the minor importance of seed production. Its increase would not only require appropriate instruction of the farmers, but also the establishment of seed-cleaning plants and cooperative market associations. The installation of
The rational use of forests and bogs
seed farms, for instance in the ideally suited experimental area of Teuravuoma near Kolari, m a y well favour Finland’s change from a grass-seed importing country to an exporting country, for which at least European countries should be a stimulating market. Increasing the area for grass-seedproduction would, of course, reduce that of hay production. Possibly, this would result in reducing the number of cattle, a reduction which should be used first of all to bring about a decisive restriction of dairying. This applies especially to subarctic northern Finland, a region in which the recent expansion of milk production is typical. This trend, already stressed by Smeds (1962), has increased during the last couple of years and m a n y n e w dairies and milk depots have been established in northern Finland. In several communities of the north, milk production has increased enormously within one year: for example in Sodankylä by over 54 per cent in 1961-62,in Kemijärvi by about 45 per cent in 1962-63.Likewise, the production of cheese and butter shows a strong increase. Thus, in recent years northern Finland has become an area with an over production of dairy products,for which milk powder is the only outlet. Factories in Haapavesi, in Kuusamo and a n e w one in Tornio, manufacture milk powder for which market conditions, however, are fairly difficult. Breeding beef-cattle and pigs in preference to dairy cattle seems rational. The argument that there is no market for meat is no longer true. In the Oulu province, meat factories try to stimulate pig breeding by giving young pigs to the farmers with the charge of feeding them. Presently the market forpork is good and another advantageofpig breeding would be to allow dairying to concentrate on the production of refined and finished products because of the great distance to market, while the milk residues could well be used to feed the pigs. Thus, on the whole, it will seem that both intensification and differentiation of agriculture in northern Finland are desirable. The comparison with northwestern Canada shows that specialization in certain high-value crops along with rational crop rotations and farmer-owned market co-operative societies are a sound basis for developing a broadly based and export-orientated agriculture, even under subarctic conditions. The conscious turn of northern Finnish agriculture to improved and refined agricultural products should be of economical value at least in respect of the vast European market. As regards forestry in northern Finland the comparison with north-western Canada is a very favourable one. Proof of this statement is found in the exemplary and careful use of foreststhe great number of scientific experiments and observations which aim at increasing the value and potential of forests and the efforts to increase the acreage of productive forest land in Finland. There is no doubt that for large stretches of land in the highly boreal and subarctic regions of
the Northern Hemisphere,forestry w ill turn out to be the predominant and most rational form of land use (cf. the contributions by Mikola and Ruuhijärvi). The results of research and of practical experiments in Finnish forestry, especially the results obtained from annual drainage of large stretches of boggy and s w a m p y land with the purpose of expanding forest land and results of fertilizer experiments, m a y well be of future value for a more intensive use of north Canadian woodlands. T o s u m up,the comparison shows that both northwestern Canada and northern Finland still have possibilities of improving the effectiveness of their agricultural and silvicultural production. The comparison reveals that northern Finland could first of all intensify and diversify its agriculture, while a more intensive use of forests as it is practised in Finland could well improve the economic stability of the settlement fringes in northern Canada. By extending the results of the aforementioned comparison to the subarctic regions in the proper sense, it becomes obvious that in both northern Finland and north-western Canada the range of cultural plants is comparatively narrow. This is even more restricted in the Subarctic proper and, in evaluating the potential agricultural use of subarctic lands, one should consider the observations and results of the experimental stations set up within subarctic regions. According to Nowosad and Leahey (1960) five points have to be reckoned as a pre-conditionfor the successful development of agriculture in Canadian subarctic regions. 1. At least sixty frost-free days per season. 2. Mean July temperature above 500 F. 3. Good soils. 4.The demand factor. 5. The production cost factor. However, taking for instance point 3, “good soil”, it is obvious that these demands are of a somewhat limited validity in regard to subarctic agriculture in general in so far as the term “good soil” is very m u c h influenced by the land-potentialfactor. For example, nowadays Finland has only small potentially productive land reserves for agriculture. For subarctic northern Finland, therefore, the rational utilization of all available land reserves is a vital question. Canada, on the other hand, has a land potential which is hard to estimate exactly. W h e n comparing it to Finland it is evident that even the estimate of 15-20 million acres of potentially arable land in north-western Canada is just a very low one. By employing the methods and techniques of Finnish agricultural colonization, the acreage of the arable land of north-western Canada in particular, and northern Canada in general, could easily be increased to m u c h higher figures. Apart from these restrictions, which have to be
331
E. Ehlers
made because of the great differences in land potential, the comparison between Finland and Canada enables us to make some general suggestions in regard to the rational use of potential agricultural lands in subarctic regions. Without sharing the great optimism of V. Stefansson (1923) about the richness of the Subarctic, the enormous land potential of subarctic regions in Canada and in the Soviet Union is a fact and a possible important source of food to supply the steadily rising needs of the world’s population. In 1933 Albright published two papers, one bearing the title “Gardens of the Mackenzie”. The title and contents of this paper try to show the richness and variety of agricultural production in the Canadian north-west.In the meantime research has been carried out and agricultural experimental farms have been established throughout the Canadian Subarctic. Thus-according to Nowosad (1959)-in the Y u k o n Territory at mile 1019 of the Alaska Highway the coarse-grain oats and barley have ripened almost annually. T h e suitability of broom grass is termed as outstanding, with other grasses also doing well. Also the raising of cattle and fowls is easily possible. Similar results have been reported from the NorthW e s t Territories, the Ungava B a y area, where sheep and geese have been raised with great success. The fact that even wheat is mostly being cultivated successfully in parts of the Canadian Subarctic reveals its latent potential. Also good results have been obtained with the cultivation of different kinds of vegetables and potatoes; turnips, cabbage, peas, cauliflower and different berries especially have ripened well and procured good harvests. These good results are the more remarkable in that they have been obtained either from small field plots, especially subject to killing summer frosts because of their “Lochschlag” character (Geiger, 1961), or on soils lying over permanently frozen ground. These disadvantages, however, are only of secondary importance in so far as they will vanish as soon as more land is under the plough or as the soil is cultivated for a consecutive number of years. For instance, in the Peace River Country of north-western Canada m a n y pioneer settlers are n o w cultivating coarse grains on their small and isolated newly-broken field plots, knowing that their future crop, after more land is broken will be wheat of a less frost-resisting variety. As to the influence of permafrost and its changes under the impact of agricultural land use, Nowosad (1960)gives interesting data from the Canadian Subarctic. In Inuvik, east of the Mackenzie and approximately 680 N.,a n e w experimental station was set up. After clearing the land, removing the surface layer of organic material and preparing it for cultivation, in June 1956, frozen soil was to be found at less than 3 inches depth. T w o months later, however, the frozen horizon had receded to approximately 25 inches. In the following
332
year, permafrost was to be found at more than 45 inches and year later at a depth of nearly 60inches. Apart from the first year, agricultural productivity of the land increased remarkably in three consecutive years. One must therefore always bear in mind that, along with a more intensive agricultural use of larger stretches of land within the Subarctic, the climatic and especially the microclimatic conditions will improve; thus furthering the suitability of these regions for agriculture. That the actual potential use of agricultural lands in the Finnish Subarctic is approximately the same is clearly seen from Ohlson’s investigations in the community of Enontekiö in northernmost Finland. Here also, grass is by far the most cultivated crop, followed by coarse grains and some special crops such as flax, rape and others. Comparing the actual land use of the marginal areas of present permanent land use and the potential use of subarctic regions proper it is evident that coarse grains and, above all different kinds of grasses are the most promising crops under subarctic conditions. Therefore one m a y conclude that the best potential use of subarctic agricultural lands would be to cultivate these crops on which a livestock economy would be based with the object of producing meat or, where market conditions are favourable, even dairy products. Besides, seed production seems to be, even under subarctic conditions, a promising form of land use. The advantage of this proposed form ot land use is not only the adaptation of agriculture to subarctic conditions but when the necessities require it, it m a y result in a distinct shift of agriculture towards the north, leaving the present fringe areas of agricultural production to a more intensive land use for the purpose of producing cereals or oilseeds. T o s u m up, one might finally try to give an outline of the future development of subarctic areas in both Finland and Canada. For Canada the future will probably not bring too much expansion of agriculture or forestry into the Subarctic proper. The land reserves of the northern boreal-forestzone are still big enough and easier to cultivate than those of the remote subarctic areas. Small, cultivated islands, especially along the Y u k o n and Mackenzie rivers and their tributaries, m a y develop here and there; however, ill be restricted to growing the agricultural their use w products already mentioned, and directed towards the necessities of the small isolated settlements of the Canadian Subarctic. Neither will Finland expand its agriculture on a large scale, but forestry m a y experience a notable expansion resulting from the drainage of boggy and swampy areas. The motives, however, are quite different and, in Finland,seem to stem from a shortage of exploitable land reserves. Thus, in northern Finland,rationalization of land use is of great importance. In view of the economic problems of agriculture it seems that the best agricultural use of the bogs in northern subarctic Finland
The rationaluse of forests and bogs
will be to introduce the intensive cultivation of grass which seems to guarantee the best yields under the ecological conditions of the Subarctic. Intensified cultivation of grass and its refining by seed production or cattle feeding would not only be a h u m a n response to the adverse conditions of the Subarctic,but would also seem to be rational in view of the great distances to market which afford an intensive use of agricultural lands.
On the whole,the comparison between northern Finland and north-western Canada shows that, through careful and rational utilization, the land potential in subarctic regions is still considerable. Therefore it is quite possible that agriculture and forestry of subaictic regions m a y develop in the future and steadily increase in importance for the economy and nourishment of the world’s population.
Résumé Utilisation rationnelle des forets et des marécages d’après des observations comparées efiectuées dans le nord-ouest du Canada et le nord de la Finlande (Eckart Ehlers)
en vue de la production de graines. L’élevage de bœufs de boucherie joue un rôle important dans l’économie de cette région, tandis que la sylviculture mérite à peine d’être mentionnée.
L a Finlande et le Canada sont deux pays où se poursuit encore un effort intensif de mise en culture des terres aux limites de la forêt boréale. Fondée sur des études effectuées en 1962-1963 dans la région de la Rivière de la paix (nord-ouest du Canada) et en 1965 dans certaines régions de peuplement du nord de la Finlande la présente communication vise à exposer les tendances générales de l’utilisation des terres dans chacune de ces régions. Dans le nord-ouest du Canada, tout au long de la forêt, cette utilisation est caractérisée par l’effort des colons pour cultiver le plus de terre possible. fitant donné l’immensité des superficies disponibles, on considère que seules les régions qui peuvent être défrichées et mises en culture à bas prix conviennent 5 l’agriculture. C’est pourquoi la colonisation ne porte que sur les régions boisées à sol minéral. Les marécages et les fondrières sont considérés c o m m e indéfrichables. Sur les terres mises en valeur, on cultive surtout les céréales et certaines plantes spéciales, notamment le lin, le colza et diverses graminées,
Les exploitations finlandaises, au contraire, sont caractérisées par la symbiose de l’agriculture et de la sylviculture. Les forêts couvrent en général la plus grande partie de l’exploitation,et les nouvelles terres mises en culture sont surtout celles dont la couverture forestière est faible ou inexistante. Presque toutes les terres exploitées sont consacrées à la culture de graminées dont les graines permettent de nourrir le bétail fournissant des produits laitiers. L a principale source de revenus agricoles est l’industrie laitière, la sylviculture ne venant qu’ensuite. L a comparaison de ces deux façons, presque diamétralement opposées, d’utiliser la terre dans des conditions physio-géographiquesanalogues a pour but d’évaluer les avantages de ces deux formes d’exploitation. Elle doit permettre de déterminer les méthodes propres à intensifier l’utilisation de la terre dans les deux régions et de suggérer une exploitation sans doute plus rationnelle des forêts et des marécages du nord de la Finlande et du nord-ouest du Canada.
Discussion P. MIKOI.A. Dr.Ehlers mentioned that raising animals for meat production plays an insignificant role in the farm economy in northern Finland. This is true in more southern localities. In the subarctic area, however, reindeer husbandry is an importantindustry,and the relativeimportance of meat production increases towards the north. H e also mentioned that Finland has little possibilities of improving the land use for forestry.May I point out that approximately 2 million hectares of wet peatlands have been drained for forest growth and there are another 5-6million hectares of wet peatlands where drainage for forestry is considered profitable. Today some 200,000 hectares are
drained annually. This work is not done in the subarctic part of the country,but very close to it in northern Finland. W e are now facing the situation that virgin bogs and fens are going to disappear in other parts of the country except the Subarctic.
W. PRUITT. I must comment to the effect that northern agriculture, as presently practised,is severely limited in its scope. There are many pitfalls that only show up after the land has been cleared for some years.The energy input of the region is quite low and thus energy output is low. Northern% agricultural research in North America has
333
E. Ehlers
generally been characterized by a remarkable lack of imagination. It consists primarily of attempts to “force” traditional temperate-zoneanimals and plants into the subarctic environment. There are a number of species of animals and plants native to the environment which have considerable potential if their yield were improved. I took up this problem in a short paper in Arctic a few years ago, if you would like further information and references.
Bibliography ALERIGHT, W.D. 1933a. Gardens of the Mackenzie. Geogr. Rev., vol. 23, p. 1-22. . 1933b. Crop growth in high latitudes. Geogr. Rev., vol. 23, p. 608-620. BENTLEY, C. F. 1965. Present and future agriculture production in Alberta’s Peace River Country. The changing frontier, p. 24-31. (Proceedings Conference at Peace River, Alberta.) BOWSER, W.E.; ODYNSKY,W. 1959. Alberta’s land resources. Agricultural Institute Review, vol. 14, no. 1, p. 9-11, 56. CAJANDER, A. IC. 1909. Über Waldtypen. Fennia, Helsinki, vol. 28, no. 2. EHLERS, E.1965.Das nördliche Peace River Country,Alberta, Kanada. Genese und Struktur eines Pionierraumes im borealen Waldland Nordamerikas. (Tübinger Geogr. Studien, 18.) . 1966a. Landpolitik und Landpotential in den nördlichen kanadischen Prärieprovinzen. Zeits. f. Ausl. Landwirts.,no. 5, p. 42-55. . 1966 b. Pohjois-Suomen ja Pohjois-Kanadan asutusrajan vertailua. Terra, Hensinki, vol. 78, p. 11-20. GEIGER, R. 1961. Das Klima der bodennahen Luftschicht, 4th ed., Brounschweig. HILLS,G.A. 1960. Soils of the Canadian Shield. Agricultural Institute Review, vol. 15, no. 2, p. 41-43. HUSTICH, I. 1952. Agriculture production in Finland and the recent climate fluctuation. Fennia, Helsinki, vol. 75, p. 97-105. . 1960. Plant geographical regions. In: A. S o m m e (ed.), A geography of Norden.,p. 54-62.Oslo. ILVESSALO,Y. 1960. Suomen metsät kartakkeiden Valossa. Comm. Inst. Forest. Fenni., vol. 52, no. 2. Helsinki. JAEGER, F. 1946. Die klimatischen Grenzen des Ackerbaus. Denkschr. Schweiz. Naturfors. Gesells., vol. L X X V I , no. 1. Zürich. JÄTZOLD, R. 1962. Die Dauer der ariden und humiden Zeiten des Jahres als Kriterium für Klimaklassifikationen. u. Wissmann-Festschrift, Tübingen, p. 89-108.
--
__
__
334
E. EHLERS. I a m verv grateful for Dr. Pruitt’s remark and fully agree that present land-useof the northern fringe areas of agriculture is far from being ideal. M u c h more research and international co-operation seems to m e to be necessary in order to achieve better results of animal and plant adaptation to subarctic conditions. This research is without doubt the prerequisite of a possible large-scale rise of subarctic agriculture.
/ Bibliographie LEAHEY, A. 1961. Appraisal of Canada’s land base for agriculture. Resources for tomorrow., vol. I, p. 49-54. (Proceedings Ottawa Conference.) NOWOSAD, F. S. 1959. Farming in the Subarctic. Agricultural Institute Review, vol. 14, no. 1, 1960., p. 11-14 and 53. . LEAHEYA. 1960. Soils of the arctic and subarctic regions. Agricultural Institute Review, vol. 15, no. 2, p. 48-50. OHLSON, B. 1960a. Sirkka, an agricultural village in the coniferous region of western Lapland. Fennia, Helsinki, vol. 84, p. 5-20. . 1960b. Settlement and economic life in the parish of Enontekiö in northernmost Lapland. Fennia, Helsinki, vol. 84, p. 21-46. OTREMBA, E. 1960. Allgemeine Agrar- und Industriegeographie. 2nd ed., Stuttgart. ROWE, J. S. 1959. Forest regions of Canada. (Canada Department of Northern Affairs and National Resources, Ottawa, Forestry Branch, Bulletin 123.) RUUHIJÄRVI, R. 1960. Über die regionale Einteilung der nordfinnischen Moore. Ann. Bot. Soc. “Vanamo”, Helsinki, vol. 31, no. l. SCHOTT, C. 1941. Die Agrarkolonisation und die Holzwirtschaft der nordischen Länder. Lebensraumfragen Europäischer Völker, vol. 1, p. 150-213.Leipzig. SMEDS, H . 1960. Post-war land clearance and pioneering activities in Finland. Latest advance of the Finnish pioneer fringe. Fennia, Helsinki, vol. 83, p. 1-31. . 1962. Recent changes in the agricultural geography of Finland. Fennia, Helsinki, vol. 87, no. 3. STEFANSSON, V. 1923. Länder der Zukunft,vol. 2. Leipzig. TROLL, C. 1964. Karte der Jahreszeiten-Klimate der Erde. Erdkunde, vol. 18, p. 5-28. VALMARI, A. 1957. Über die edaphische Bonität von Mooren Nordfinnlands. Acta Agralia Fennica, Helsinki, vol. 88, no. 1-2, 1957. WESTERMACH, N. 1956. Die Finnische Landwirtschaft. Helsinki.
--
-
-_
On the string formation in the aapa moors -
and raised bogs of Finland Erwin Schenk
INTRODUCTION String formation and frost structures in the high moors and aapa moors of Finland are part of the characteristics of the circumpolar and subarctic forest belt around the Earth. G a m s and Ruoff (1929) showed for Fennoscandia,as Katz (1948) and finally Frenzel (1959, 1960) did for Eurasia (Fig. i), that these areas are not part of the permafrost territory but precede it. This is true for pals moors also. The moors developed directly upon soil freed from the ice (Ruuhijärvi, 1960; Vasari, 1965). There permafrost existed first, just as it exists today, ahead of glaciers and in the arctic tundra. The fact that there are no string moors here and none developing anew in territory freed of permafrost puts their origin into the climatically labile zone along the border line of
FIG. 1. The distribution of palses,pingos and string bogs in Eurasia according to Frenzel, Katz, and others.
SEE
Paises, pingos
11~111111111111Il1~1111String
permafrost which is the northern border of the Subarctic. In this area with an annual mean temperature of -20 to -60 C permafrost can appear and disappear again. Obviously, the development of the strings is connected to the disappearance of permafrost (Schenk, 1964). U p to this time, scientific research mostly reported by botanists (Rancken, 1912; Aario, 1932; Tanttu, 1915; Auer, 1920; Cajander, 1913; Paasio, 1933; Sjörs, 1948; Drury, 1956; Lundqvist, 1958) led to the belief that biological factors were the cause of the formation of strings, along with solifluction, regulation and water flow. Sjörs (1948, 1959), Ruuhijärvi (1960) and Eurola (1962) also found in their research of recent years that in the fight for utilizableland at the time of annual flooding,a sideways grouping of individual bulges created a certain order between dry bulges and wet hollows.
\ Boundary
of permafrost
O
1 O00 km
d
bogs
335
E. Schenk
In string moors near Jokkmok, I observed in 1958 that strings contain cores of permafrost and that not only the surface of the moors but also their layers are generally slanted toward the direction of the drainage. The down-stream side of the strings shows that their spacing was caused by a flexure. My continued research in Canada and Alaska in 1959 not only verified my observations but also proved the solution I had found for the problem: an up-stream tilt of either frozen or tightly felted pieces of peat with edges showing above water when melting and settling of melt water from underground had occurred (Schenk, 1963, 1964, 1966). The deformation through settling of the original surface of the moors proved to be the prime cause for string formation. Therefore, the Finnish moors in the string formations studied by several authors hold a very special interest. In July 1965, Professors Ruuhijärvi, Vasari and Eurola were kind enough to take m e on a journey from south to north to show m e special string formations in moors. Ruuhijärvi (1960)and Eurola (1965) in connexion with a survey of vegetation by Professor Kalela, began a very detailed analysis of the moors of Finland according to their types. Each of these types of moor shows specific forms of strings and hollows.
STRING FORMATION IN THE RAISED BOGS The raised bogs of Schären, Finland, which are described by Eurola (1962) do not show clearly defined formations of strings (=Kermi) and schlenken. The small moors, especially, have developed into heather bogs which have wide, shallow rises and hollows still showing schlenken characteristics. The raised bogs in a plateau of this area, however, show recognizable shapes of schlenken and strings. Both are broad and shallow. The raised bogs of southern Finland, however, along the coast as well as inland are dominated by very definite formations. Large; longish mounds and broad strings,lined by humps and grooves and edged with bulges connect horizontally; they rise steeply one or two metres above the water level of the small schlenken which are filled with turfy mud. According to this very characteristic appearance they are called kermi in contrast to the strings of aapa moors. As w e move away from the isles in the Gulf of Finland and from the coast, these forms become even more distinct towards the interior. The schlenken become wider and dryer and m a y even be covered
Also due to frost are the pronounced bulges along the edges, between schlenken and kermi, as well as a surprising number of frost-structure shapes in the schlenken. Bore holes produced evidence that the S. fuscum peat covers of the kermi have a minerotropic foundation and history. In the border areas of the aapa moor area the minerotrope plants of the schlenken move in on the S.fuscum. In these schlenken a whitemoor vegetation develops, above which the S.fuscum of the kermi rises and thereby keeps out of reach of the mineral layer’s water-level.The S.fuscum layers are between 1 and 2.5 m thick, the layers of minerotrophic peat between 3 and 4 m.It must be mentioned that S.fuscum peat of today’s kermi formed during the subatlantic phase. Along the coast, Aario (1932) and Sauramo (1954) recognized a subatlantic age. The formation of the kermis in southern Finland therefore belongs to the subatlantic period.
STRING FORMATION IN THE AAPA MOORS According to Ruuhijärvi (1960), Rancken (1912)and Cajander (1913),the minerotrophic vegetation of the moors that developed on layers of Seggen peat accounts for the characteristic basic formation. This is true, also, for the basic layer of the S.fuscum strings. As ombrotrope vegetation, it covers the humps and bulges. The size and height of these bulges increases according to latitude and frost effect. A connexion between these humps and bulges becomes more evidently related to the string formation, especially where they show growth of underbrush and trees. The border areas which turn into brown moors with wet rimpi and dry strings show this most clearly. The ability of the peat to absorb and store water and the favourable climate of the region cause a faster growth of rimpi and formation of peat layer so that vast, treeless, white moors of rimpi develop. The sloping moors of this group of formations accomplish the numerous varieties of moors which Ruuhijärvi (1960) groups into Pohjanmaa-, Peräpohjola- and W o o d Lapland-moors(Fig.2). Just as in Finland,these are found in North America as well as in Siberia, covering an area of m a n y hundreds of kilometres from north to south (Frenzel, 1959). I have studied and observed them in Scandinavia, Canada and Alaska, from the Atlantic to the Bering Strait and have seen them from the air in Labrador and N e w foundland.
with S p h a g n u m fuscum.
POHJANMAA AAPA M O O R S
Along the edges of the raised bogs in the wet lagg (which can be up to 30 m wide but averages a width of 5-7m) and in the birch forests the tussocks and mounds appear clearly as active frost structures.
The characteristic shape of the kermi in the moors of southern Finland is very clear; but the border line toward the aapa moors of the Pohjanmaa region in
336
O n the string formation in the aapa moors and raised bogs of Finland
FIG. 2. The regional distribution of the Finnish raised bogs and aapa moors, Ruuhijärvi (1960).
the north is not distinct. T h e transition is a gradual one. According to the length of the frost periods, the effect of the frost in shaping the small relief of strings and laggs a n d the transition in the border areas from strings into schlenken a n d rimpi shows m o r e and m o r e from south to north. There is n o doubt that the relief formation of the strings, which continues on into wide stretches of rimpi, is caused by the large water content of the rimpi. A process of decomposition of the strings, starting at the edges by small and large frost welts often causes miniature rimpi along the strings. This interchange of very tiny rimpi and strings m a y m a k e
the surface of the m o o r look like a cover of vegetation rippled by water waves. E v e n giant areas of rimpi are typical for the Pohjanmaa-aapa moors. A t times they are without string formation, at other times richly veined by strings. These strings rise from the rimpi in the f o r m of long and relatively dry areas. Sometimes, they s h o w n o directional tendencies at all. Sometimes, however, they run perpendicular to the direction of the valley or parallel to the foot of the hills. T h e single strings are 1-4 m wide but their complex zone is perhaps 50 m wide. T h e y rise 20-30 c m above the water-level
337 24
E. Schenk
of the rimpi. In the southern part of the Pohjanmaa region a carpet of moor vegetation often covers the strings (Ruuhijärvi, 1960, p. 211 f.). Slightly defined shallow rises and single and island-likehumps in the moor and along the strings attest to the effect of frost. There the S. fuscum settles on soil with very little depth. Its distribution is therefore limited. The transition from string to rimpi is very gradually effected. A more flat or steep border line of the string is better recognized by the sharp line downvalley and a curved line up-valley formed by hydrophytic vegetation. The water level of the moor is relatively high. Precipitation and evaporation favour fast growth and peat formation. Accordingly, the peat is thick and naturally able to store m u c h water. At this point, the effect of frost is relatively even and minimal. Once humps and strings have developed further, the frost effect will be different. Ice lenses form in the string area and define it more clearly. As the thickness of the peat diminishes and the water level sinks, the growing frost effect toward the north raises the string formation to about 50 cm. All this shows the connexion of events and effects and explains the directional tendencies. In the north, one m a y find ice lenses in the core of these strings and humps which last throughout the summer and must be considered as permafrost. Here, then, is the transition to the next type: Peräbojohla-aapa moors.
accordingly, and this difference in frost effects has already been noted by Tanttu (1915), Auer (1920), and Ruuhijärvi (1960). Altogether this again shows the string formation of the moors in the Peräbojohla zone (Rancken, 1912) in relationship to frost effect and diminishing peat thickness. FJELD LAPLAND-AAPA M O O R S
The border line of this territory follows the tree line, north as well as south. Vegetation and surface morphology limit the extension of this type of moor. Evaporation is minimal, but pyecipitation almost equals that of more southerly areas, so that there is a good water-supply.This favours the growth of plant life, especially of S p h a g n u m , and thus the extension of the moors. Frost is very deep and string structures are allowed to develop attaining the unusual height of 1-1.30m otherwise found in southern Finland only. The shapes of the strings show high above water level. S. fuscum is above ground-water level and grows unusually well; the rimpis are very deep. Living conditions for vegetation are often so unfavourable that open pools of water (blänken) and schlenken, are devoid of rimpi vegetation. With growing height, the string formations show partitions. They always have a core of ice which, in melting, destroys the protection cover of S p h a g n u m leaving holes which sometimes line up in basins 1 m wide along the strings.
PERÄBOJOHLA-AAPA MOORS
THE PALSA M O O R S
As plant life progresses towards the north, growing conditions become less favourable because of the climate,so that there is also less peat formation.Thus, frost can penetrate more deeply. In drainage ditches through such moors one sometimes recognizes the shape of large humps and bulging cuts through the string formation. On the surface of the moors there are string formations across the axis of the valley or running parallel to the edges of the moors. They often show hummocks with S. fuscum rising 5070 c m above the rimpi surface which is more than 10 m wide. In between, there are cuts with hydrophytic vegetation. Often, small palses have formed on the strings or big ones as islands in the rimpi. Here,ice cores last throughoutthe summer. The whole area of the Peräbojohla-aapamoors, therefore, must be considered as a zone of discontinuous or sporadic permafrost. At the time of the spring thaw, the frost-caused, strong, micro relief of the strings and rimpis causes uneven thawing and settling. This influences the circulation of the flooding waters, as does the stopping effect of the high-lying strings. At the low points, erosion occurs, often creating deeper pools which hold open water until winter. Because of this a larger amount of rise is exposed to frost right around the immediate area. The spring thaw looks different,
338
In Fjeld- and Fjord-Lapland,finally, frost develops the most extreme type of moors. These are the pounikkos and palses (pingos) which all have permafrost. The conditions for their development are the same and only their height depends on the amount of water that they receive from their peat and mineral bases to form the ice lenses. They need no further water. Peat layers and the clay of the moors with their large content of organic residue and the water films of the soil particles give off enough water. Palses (and pingos also) are usually centrally located in the moors. Their greater depth and greater water content again differentiates the permeation by frost; it is slower than along the edges. Here the moor is frozen through, while the centre-in the deepest partcontinues to form ice lenses, causing the rise of palses and pingos. Along with this rise goes the development of a system of crevices caused by dehydration of soil layers throughout winter and summer. The evolution of these large frost bulges is younger than the permafrost-caused humps; this is shown clearly by the fact that the giant polygons of the tundra with their net of ice wedges are being deformed by the pingos. It should also be said that in some areas of Pohjanmaa and Peräbojohla, and in Fjord-Lapland too, the string formations appear like nets similar to the
O n the string formation in the aapa moors and raised bogs of Finland
netting of ice wedges. N o real accordance however has yet been proved. In this area of palses and pounikkos permafrost m a y be continuous. String bogs do not occur or are infrequent. FJORD
-
LAPLAND MOORS
In the fjelds of Fjord-Lapland in Norway, however, and in the border zone of Fjeld- and W o o d Lapland in Finland there occur white bogs with initial strings. Sometimes they are recognizable only by an alteration of height and by a dark green colour, sometimes by single, real strings which cross the bogs in the form of arcs opened down-stream which rise 0.10-0.30m above the water-levelup-stream and 0.5 m and more down-stream. Polar birch and polar willow grow upon these strings which are covered by S p h a g n u m and xerophytic plants. I observed that this initial structure of string bogs is the same near Jokmokk (Sweden), in Canada and in Alaska.
RESULTS
A cut through the moors of Finland from the Baltic Sea to the south and as far as the Barents Sea reveals the c o m m o n basic principles of the formation of the strings,kermis, rimpis and schlenken. There is asymmetry in the formation of all types of moors and strings, because the greatest depth of water m a y be found on the steep, up-stream side, and the lowest depth on the shallow, down-stream side (Eurola, 1962; Ruuhijärvi, 1960). In the initial string moors of Fjord-Lapland,however, the strings are flat up-stream and steep down-stream; this was also observed in Canada (Schenk, 1966, Table 3, fig. 6; Table 5, fig. 13). All the strings in the moors of Finland show evidence of very severe frost effects. They are a tight body of pounikkos which are separated by small grooves. Their formation and rise was caused by the segregation and the aggregation of ice. This was especially favoured along the edges of the rimpis because there the water is plentiful for ice formations. The string edges are then steep. If the water is not very deep, new peat bulges erupt ;these line the edges of the strings;strings and kermis become wider, rimpi and schlenken narrower. Often, the strings appear to be a settlement of very narrow string formations. If the water is deep, however, the annual repetition of freezing and refreezing m a y cause in extreme cases the formation of single strings, wall-high.They retain again more water, making it deeper and forming small lakes with individual strings. The vegetation differs according to the climatic and hydrological conditions of the landscape influenced by frost and peat formation.
In the area finally freed of permafrost and ice in Fjord- and Fjeld Lapland w e find string moors in statu nascendi in the Grosseggen Weissmoores. All over the country, this particular type of moor represents the first phase of moor formation (Vasari, 1965 ; Ruuhijärvi, 1960) in valleys and on slopes. These moors developed on permafrost soil. With the setting and flowing motion in the thawing moors the rising edge develops by tilting against the flow. Here the frost can attack in depth. By forming pounikkos, finally the edge reaches above the minerotrope area. Here, S. fuscum n o w finds its first possibility to settle and, beyond the strings, rimpis develop. Perhaps this can n o w explain the invasion of S. fuscum into northern Finland which until n o w was unexplained (Ruuhijärvi, 1963, p. 1). The frost effect in W o o d Lapland is often so strong that strings rise like thick bulges, acting as dams so that water drowns the vegetation of the rimpis, creating open blänken and schlenken. It is not rare that the strings should appear then as isolated walls in the open water. In Peräpojohla the area behind the strings raised by the frost is filled by extenses of rimpi. There seems to be a balance between the vegetation, which favoured by the climate, has caused formation of peat and high moors, and the water retained through the effect of frost. On both sides of the initial strings the peat bogs have grown and form a well-divided string, high above water-level. In Pojanmaa, however, where floods often remain for a long time, rimpi moves over on to the strings SO that these lie as a flat, wide and relatively dry zone above water level. M a n y strings have probably sunk below water level and the living area for S. fuscum is thereby considerably limited. In southern Finland the water condition is less favourable for aapa moors because evaporation is considerably greater. The ground-water level shows considerable fluctuations over the course of a year. Larger peat settlements follow. If strengthened by frost, the strings appear much more clearly and offer better living conditions to S. fuscum. This reaches over to the rimpi on both sides of the schlenken edges. The strings widen, finally narrowing the rimpi to a deep groove (schlenke). If S.fuscum reaches the border line of its best conditions for living above moor water-level, it finally dries up. In the schlenke, it continues to grow. In Schären Finland this leads to the change from kermis to schlenken and from schlenken to kermis (Eurola, 1362). The development reaches its climax in the formation of heather bogs which in wet hollows merely suggest the former schlenken. Thus, the basic structure of string moors appears to bave a variable influence on the water content of the moors in summer and on the long-lastingeffect of frost in winter throughout the whole range of moor types 339
E. Schenk
in Finland. In view of this it is possible to understand the formation or nature of the Finnish moors, a fact which has already been explored by Ruuhijärvi, Eurola and Vasari. Their division into characteristically constructed zones of vegetation represent a historical picture of development in the lowlands which is a picture of co-existence of plants. This development is not as previously believed, only dependent upon changes of climate and its conditional blanket of vegetation. It depends largely on the effect of frost and its co-effect of bulge formations which were not considered important before. Frost, causing peat bulges with apparent rises, influences the widening of the initial string which had developed at the tilt of the peat during settling of the moors when permafrost thawed. The annual effects of frost in this primeval string formation caused vegetation to separate into xerophile and hydrophile associations which again, effected water storage. Growth possibilities together with frost are the important factors. These processes are those which overcame the original string forms and surface structures. In the wide open moors, with their differentiating vegetation, there is no reason whatsoever to acknowledge the effect of frost on low moors or rimpis. There are floods even in areas with arctic or moderate climates but no string formations could be observed although the fight for utilizable space on which to survive, which is to cause the string formation, is very much the same as in the Subarctic.
ACKNOWLEDGEMENT In 1957, 1959 and 1965 I visited the arctic and subarctic regions of North America (Alaska, Canada), in 1958 and 1965 those of Europe (Sweden, Norway, Finland), which I saw for the first time in 1936 (Spitzbergen). In 1961, 1963 and 1965 I took part in the congresses at Lafayette, Indiana (United States), Warschaq, and Boulder, Colorado (United States). This was made possible only by the grants of the Deutsche Forschungsgemeinschaft. I was able not only to study arctic and subarctic features, but also to contact m a n y investigators of Quaternary geology. For all this I want to express deep gratitude to the Deutsche Forschungsgemeinschaft and also to offer special thanks to Professor Dr. Troy L. Péwé (University College of Alaska, n o w University of Arizona), Dr. M a x Brewer (Arctic Research Laboratory, Barrow, Alaska), Professor Dr. Clyde Wahrhaftig (now University of California,Berkeley), Dr.David M. Hopkins (United States Geological Survey, Menlo Park, California), Professor Dr. E. Mückenhausen (Institut für Bodenkunde, Universität Bonn), Professor Dr. Rauno Ruuhijärvi (Botanisches Institut, Universität Helsinki) and last but not least to Dr. Zellmer, Director of the Zweckverband Oberhessische Versorgungsbetriebe, Friedberg, Hessen.
Résumé L’origine
des tourbières réticulées [aapamoors]
(E.Schenk) L a formation des tourbières réticulées (string bogs, strangmoors) dans la zone forestière boréale de l’hémisphère nord ne résulte ni de la solifluction,ni d’une différenciation initiale de la végétation dans les marais ; elle est due à l’effondrement du pergélisol dans la couche subsuperficielle des marais. L a formation du pergélisol (sol congelé pendant plus d’un an) a produit une stratification de lentilles de glace et de couches de sol déshydratées. Dès que le gel eut cessé de pénétrer la croûte terrestre pendant la dernière glaciation, le flux thermique provenant de l’intérieur de la terre commença à faire remonter vers la surface l’isotherme zéro degré. Lorsque celui-ci eut atteint la couche subsuperficielle des marais gelés en permanence, les couches et les lentilles de glace fondirent. I1 en résulta d’abord une stratification de couches de sol déshydratées et de nappes d’eau.
340
L e sol était si sec qu’il ne pouvait absorber que difficilement et lentement l’eau provenant de la fonte des lentilles de glace. Mais les particules de sol situées dans les zones de contact entre l’eau et le sol furent saturées d’eau et devinrent glissantes. En outre, l’eau provenant de la fonte des lentilles de glace était soumise à une forte pression par la couche gelée située au-dessus. Elle fut donc inévitablement poussée à s’échapper en totalité. Sous l’effet de ces phénomènes, les couches superficielles encore gelées furent disloquées en sections discontinues, qui s’affaissèrent par suite de la disparition de l’eau sous-jacente,non pas verticalement, mais en subissant un léger mouvement de bascule et un déplacement latéral sous l’effet de succion des eaux souterraines. Les bords relevés des sections gelées formèrent ainsi les crêtes (strings), tandis que les parties inférieures de ces sections furent inondées et transformées en étangs peu profonds (Flarken et Schlenken), qui déterminent la configuration initiale des tourbières réticulées (aapamoors).
On the string formation in the aapa moors and raised bogs of Finland
Celles-cipeuvent également se former lorsque le marais est si plat que l’effet saisonnier du gel et de la fonte atteint la base du marais.
Origine des tourbières réticulées (aaparnoors)
Les structures de l’effondrement de la couche gelée déterminent donc essentiellement la formation et la différenciation ultérieures de la végétation qui recouvre
les tourbières réticulées (aapamoors, muscegs, string bogs). C e phénomène sera illustré par des photographies prises e n Finlande e n 1965. Avec leur étrange configuration, les tourbières réticulées constituent donc la caractéristique de l’effondrement du pergélisol dans cette zone périglacière pléistocène boréale (taïga et grandes forêts de conifères) qui ceinture la terre e n traversant la Sibérie, la Russie septentrionale, la Scandinavie, le Canada et l’Alaska.
Discussion C. O. TAMM.A satisfactory explanation of the structure of “aapa moors” must take into account: (a) the elaborate and to some extent self-regulating hydrological system, retarding the water flow as m u c h as possible; (b) the range of distribution of “strings” and “rimpis” perpendicularly to slope, avoiding very wet condition; and (c) the fact that the explanation must be based on known biological and geological processes. T h e similar orientation of the hollows of raised moors represents a closely connected problem. I do not think that Dr.Schenk‘s theory can fulfil the first condition, at least not without further assumptions. On the other hand, I think professor Ruuhijärvi’s model requires only very slight further elaboration in order to fulfil all three conditions. If w e assume as a starting point an irregular system of tussocks spread out over a slightly sloping mire surface, the water will flow between the tussocks. T h e Sphagnum species and vascular plants growing on the sides of the tussocks will grow faster than those on the front and back of the tussocks, because of the better supply of nutrients with the flowing water. This m a y eventually lead to tussocks growing together perpendicularly to the slope, i.e. the formation of “strings”.
E. SCHENK. T h e aapa moor with its system of strings and rimpis is not a self-regulating hydrological system, neither by the properties of water nor b y the properties of plants or plant associations, nor by both working together. Without frost activity on and along the first original string into the recent strings there would be no string bogs and no raised bogs with kermis. If the aapa moor were to represent such a system of self regulation, then aapa moors or similiar moors would necessarily develop everywhere, in all the places where minerotrophic moors exist or originate. But such moores do not show even a slight trend to develop
Bibliography AARIO,L.1932. Pflanzentopographischeund paläogeographische Mooruntersuchungen in N-Satakunta. Fennia, vol. 55, no. 1, p. 1-179.
ridges or anything like strings, for example the muskegs in Alaska over frozen ground, the very vast rimpis in Pohjanmaa (Finland), bogs in the raised area along the seashore of Finland and Sweden on never frozen ground, or in northern and southern Germany where bogs developed through primary moor formation (=primäre Vermoorung in the meaning of Ruuhijarvi (1960) and Auer (1920)), or b y drying out. There are m a n y moors which show the required conditions of being flooded and having the effect of slowing the waterflow b y tussocks, but in which no aapa moors develop. T h e surface features of the aapa moor and string bogs, in a belt 500-1,000 k m wide along the permafrost boundary which w e observe today, cannot therefore be explained b y recent processes of water movement and plant growing on the surface of these moors, but only by including the original pattern of the subsoil settling and raising in sequence of underground drainage. This original structure with an elevated edge is a geological result of underground drainage and is being observed only in the far north where permafrost is either collapsing or ephemeral. In all southern parts of the subarctic area biological processes develop according to the climatic conditions which include frost activity and hydrological effects according to the existing soil and its features. T h e cross-sectionthrough the Finnish aapa moors from north to south makes it obvious that the overwhelming presence of the primary strings through plant and peat and peat growth seems to be influenced by the up-stream and down-stream position of the string, in a way so different that today the primary structure is reversed. This can be ascertained or increased only b y boring into the deepest layer of the moor to get the samples for pollen analysis and 14C data. In order to do this one has to traverse strings and rimpis with bore holes very close together.
/ Bibliographie AUER, V. 1920. uber die Entstehung der Stränge auf den Torfmooren. Acta Forest. Fenn., vol. 12, no. 2, p. 1-145. CAJANDER, A . K. 1913. Studien über die Moore Finnlands. Acta Forest. Fenn.,vol. 2, no. 3, p. 1-208.
341
E. Schenk
DRURY, W.H. 1956. Bog Pats and physiographic processes in the upper Kuskokwin river region, Alaska. p. 1-130. (Contr. Grey Herb. Harvard University, 178.) EUROLA, S. 1962. Über die regionale Einsteilung der südfinnischen Moore. Ann. Bot. Soc. “Vanamo”, Helsinki, vol. 33, no. 2. --. 1965. Beobachtungen über die Flora und Vegetation a m südlichen Ufersaum des Saimaa-Sees in Südostfinnland. Aquilo, Ser. Botanica, vol. 2, p. 1-56. (Societas Amicorum Naturae Ouluensis, Separatus.) FRENZEL, B. 1959. Die Vegetations- und Landschaftszonen Nord-Eurasiens während der letzten Eiszeit und während der postglazialen Wärmezeit. I. Teil: Allgemeine Grundlagen. Abh. math.-nat. Kl. Akad. Wiss. u. Lit. Mainz., Wiesbaden, no. 13, p. 938-1099. . 1960. Die Vegetations- und Landschaftszonen NordEurasiens während der letzten Eiszeit und während der postglazialen Wärmezeit. II. Teil: Rekonstruktionsversuch der letzteiszeitlichen und wärmezeitlichen Vegetation Nord-Eurasiens. Abh. math.-nat. KI. Akad. Wiss. u. Lit. Mainz, Wiesbaden, no. 6, p. 290-453. GANS, H.; RUOFF, S. 1929. Geschichte, Aufbau und Pflanzendecke des Zehlaubruches.Monographie eines wachsenden Hochmoores in Ostpreussen. Srhr. Phys. -ökon. Ges. Königsberg Pr.,vol. 66, no. 1, p. 1-193. KATZ, N.J. 1948. Tipy bolot SSSR i sapadnoi Ewropy i ich geograjitscheskoje rasprostranenije. Moskwa (After Ruuhijärvi, 1960 and Frenzel, 1959.) LUNDQVIST, G. 1958. Beskrivning till jordartskarta över Sverige. SV. Geol. Unders.,vol. 17, p. 1-106. PAASIO, I. 1933. Über die Vegetation der Hochmoore Finnlands. Acta Forest. Fenn., vol. 39, no. 3, p. 1-190. RANCEEN, H.1912. Lapin suomaiden kehityksestä.Suomen Suoviljelysyhd. Vuosik.,vol. 3, p. 238-274.
__
342
RUUHIJÄRVI, R. 1960. Über die regionale Einteilung der nord-finnischen Moore. Ann. Botan. Soc. “ Vanamo”, Helsinki, vol. 31, no. 1. . 1963. Zur Entwicklungsgeschichte der nord-finnischen Hochmoore. Ann. Bot. Soc. “ Vanamo”, Helsinki, vol. 34, no. 02. SAURAMO, M . 1954. Myöhäisjääkautisesta kasvistosta ja kasvillisuudesta, erityisesti metsähistoriasta. Luonnon Tutkuja, vol. 58, no. 5, p., 130-135.(After Eurola, 1962.) SCHENK, E. 1963. The origin of string bogs. Proceedings, International Conference of Permafrost, Lafayette (USA), p. 155-159. (National Academy of Science, National Research Council, Publication no. 1287.) .1964. Entwicklung und Zusammenbruch der Strukturen des Dauerfrostbodens. Report of the VIth International Congress on Quaïernary, Warsaw 1961, vol. I V Periglacial Section, Lódi. .1966. Zur Entstehung der Strangmoore und Aapamoore der Arktis und Subarktis. Zeitschrijt f. Geomorphologie, Berlin. SJÖRS, H. 1948. Myrvegetation i Bergslagen. Acta Phytogeogr. Suec.,vol. 21, p. 1-299. (Summary: Mire vegetation in Bergslagen, Sweden.) . 1959. Bogs and fens in the Hudson Bay lowlands. Arctic, vol. 12, no. 1, p. 1-19. TANTTU, A. 1915. Über die Entstehung der Bülten und Stränge der Moore. Acta Forest. Fenn., vol. 5, no. 2, p. 1211. VASARI,Y. 1965. Studies on the vegetational history of the Kuusamo district (North East Finland) during the Late Quaternary period. III. Maanselänsuo, a Lateglacial site in Kuusamo. Ann. Botanici Fennici, vol. 2, p. 219-235. IV. The age and origin of some present-day vegetation types. Ann. Botanici Fennici, vol. 2, p. 248273.
_-
__
Remarks on current silviculturalresearch in the Subarctic of Finland Gustaf Sirén
T h e subarctic area of Finland can be regarded as a natural reserve of land which did not c o m e under the modifying influence of civilization until the historical period. Except for disturbances caused by the extensive exploitation of the forests for the production of bark-bread during the eighteenth and nineteenth centuries, a n d the destruction of the vegetation by fires lighted in the course of intertribal warfare, the oscillations of the tree and forest limit have largely been determined by the climate. As a complement to, a n d to s o m e extent in illustration of the admirable expositions given earlier in this symposium, m a y I touch briefly u p o n s o m e ecological prerequisites for the practice of the extensive form of forestry which in this part of the Subarctic permits the population to utilize rationally the f e w natural resources available. First, m a y I c o m m e n t o n the manifestations of climatic fluctuation and o n their occurrence as stress factors. T h a n k s to the w o r k of palaeobotanists a n d quaternary geologists, using fossil plant remains a n d sediments as indicators, the mapping of post-glacial climatic fluctuations has been carried as far as the historical period. T h e method of dendrochronology
is a valuable a n d relatively precise complement to these methods, bridging the gap between, for instance, the recurrence surfaces of the palaeobotanist a n d the m o d e r n instrumental data of the meteorologist. W h e r e the Fennoscandian Subarctic is concerned, this implies that the record m a y b e carried b a c k by dendrochronological m e a n s as far as the last indisputable recurrence surface, and thus allow the climatic changes during the historical period to be m a p p e d with greater accuracy than is attainable for example, by the use of pollen analysis. T h e standardized series of annual rings available for the regions adjacent to the Fennoscandian forest limit (see Fig. 1) demonstrate that the trees’ radial increment has fluctuated remarkably. T h e analogical comparison of these series with meteorological data for the years 1903 to 1960 permits one to d r a w the natural conclusion that a l o w s u m m e r temperature retards radial increment, while a high s u m m e r temperature for several years results in increased increment a n d marked flowering and seed-setting, followed by the regeneration of the forest. T h e negative aspect of these climatic fluctuations is, however, manifested otherwise than by poor increment
Time of reproduction
1.000 X Log M(g)
II I I Ill Ill Il I 111 I III IlIl ( O
FIG. 1. Pattern of reproduction of pine in the forest limit region and its relation to tree-ring fluctuations.
-=,n,1181
1300
I400
1500
.
1600
1700
1800
1900
1960
343
G. Sirén
and a failure to set seed. A systematic examination of the areas o n each side of the present limit of pine shows that large areas beyond this limit were forested, probably even at the end of the sixteenth century. Windfalls from the beginning of the sevenby the teenth century, w h e n the climate-judging narrowness of the annual rings-was severe, were without exception very shallow-rooted throughout the area investigated. T h e cause of the retreat of the forest, whether it w a s the onslaught of occasional violent storms, accompanied by a n unfavourable climate for regeneration, whether it w a s the temporary formation of permafrost in the forest, or whether it w a s the inadequacy of assimilation, which prevented the necessary root activity-I at least cannot ascertain. B u t the fact remains that even as late as the sixteenth century, the tree a n d forest limit in places lay 30 k m further north in the districts of Enontekiö, Inari and Utsjoki in northern lapland, than it does today.
Plant species' iolerance
insu5cient
Stress factors: type, strength, duration
sufficient
All the above-mentioned components of the reaction pattern m a y b e traced, to various degrees, in the zone behind the farthest limit of individual trees of both pine and birch. That n e w subspecies or provenances, or both, emerge as a result of the stress should be apparent from the fact that, for instance, the pine population in the neighbourhood of the forest limit is notably heterogeneous. As further evidence of this m a y b e cited the occurrence of the virescens form, 344
Unspeaking, but eloquent witnesses of this today are the massive, overthrown stumps of trees which grew up in the twelfth and thirteenth centuries. B u t it might b e mentioned that the improvement of the climate since the beginning of the twentieth century has given rise to a n advancing front of y o u n g trees, which is still progressing towards the earlier forest outposts (Fig.2). In this context, it m a y be of interest to discuss the climatic fluctuations as stress factors with biological and even phytogeographical consequences. If the constellation of ecological factors which has long dominated a fairly large forest area is suddenly thrown out of balance, for instance, in respect of one of its climatic components, the forest stand reacts against the disturbance. Depending o n whether the disturbance exceeds the limits of tolerance of the constituents of the plant community, the reaction follows the following preliminary schematic pattern:
Plant species' reaction
Population dies, or is reduced (a) Population exterminated (b) Population is strongly reduced (c) Population suffers damage to individuals (d) Natural functions of individuals reduced Population adapts; resistant subspecies arise by genetic adaptation.(e.g.mutation) and occupy space made available Population adapts through selection; reduction of individuals leads to a resistant population which occupies the available space The expansion of the unchanged population is limited, or there is a concentrated retreat from the disturbance The unchanged species and population survives the disturbance in situ The natural functions of the individuals are redeployed for storage or reproduction, sufficient to m e e t still more severe disturbances The individual reproduces normally, both sexually and asexually, or both
with its genetically determined delayed chlorophyl formation, which appears even in large plants because of the small degree of individual competition. This special case might also be interpreted as a normal, environmentally determined chlorosis, this in itself being a result of stress. T h e dominant stress factor at the forest limit is the scarcity of energy. As regards the light conditions at the forest limit, it should be emphasized that in
Remarks on current silvicultural research in the Subarctie of Finland
this respect, the area is not especially unfavourable during the s u m m e r . However, the light cannot b e utilized because of the l o w temperatures which prevail even in June. T h e physiologically significant heat s u m 1 here is for this reason very low, as is s h o w n clearly by a regression analysis of the relation between the temperatwe s u m s for different parts of Finland (according to Keränen) and the increment of the forest (according to Ilvessalo) for sites in comparable regions (see Fig. 3). It m a y be mentioned that the scatter of the primary data w a s greatest in the case of the dry sand areas, where the importance of the water factor is almost as great as that of the s u m m e r temperature. Figure 3 illustrates a further point; namely, that the regression lines reach zero, i.e. the forest limit, at a temperature s u m of about 5500 C, for all forest types with the exception of spruce s w a m p s , the northern boundary of which is determined by spruce, a n d of tall herb forests, whose itself production indicates that by fertilization-in
certainly uneconomic-the forest limit could be pushed a f e w miles further north. After this cursory review of the ecological background to forest production near the forest limit, it might be appropriate to deal with purely silvicultural problems. It has already been indicated that there are few seed years per century. However, relative to the local rotation length, the interval between seed years appears less alarming. Add to this the fact that over large areas, the condition of the ground surface for regeneration is good, or at least satisfactory, a n d one can understand that within the economic framework of extensive forestry, the m e t h o d of natural regeneration is acceptable as the first choice of regeneration method. n 1. As deñned by Kerinen, viz., T
=
.Z
(tm- 5), m = l where n = number of days with m e a n temp. 5 + 50 C, tm = the m e a n temperature of these days. snd + 50 C = the threshold value for temperature which permita measurable growth. +
x x x o x Northern limit of pine forest (Hiilivir to, 1936)
fi&&%%
Northern limit of individual pine trees and the area of scattered isolated stands and trees Northern limit of individual pine seedlings and scattered seedling stands (densi~y>ltreehaJ in year 1958.
FIG.2. Result of pine-seedlinginventory in the area of western Utsjoki in year 1958.
345
G. Sirén
FIG. 3. Forest growth as function of temperature sum (as defined by Keränen) in Finland.
Annual growth cub.d/year
. Herb
---~
type,
Myrtillus w p e a
ROO
700
600
1 O00
900
1100
1200
Temperature sum
Since the technique m a y be improved by skilful burning of the competing vegetation, the choice becomes still clearer. As a n example, the numbers of plants per unit area in a n experiment at Aksujärvi (see Fig. 4) a n d the forest limit in general (Fig. 5) m a y be compared. As the diagram shows, the n u m b e r of plants increases with every increase in the n u m b e r of seed trees, up to a broad optimal level, beyond
Seed trees por hectare
/ /
/
t / / O /O /
/
/ O
/ / I
I
I
I
I
i0,ooo
I
I
l
I
1 20,000
Nurnbcr of seedlings per hacloie
FIG.4. Number of seedlings per area unit under canopies of varying density in prescribed-burned experimental plots.
346
which both the n u m b e r of plants a n d their vigour is reduced. Experiments at these latitudes and o n sites of these quality classes have s h o w n that regeneration m a y be considered acceptable, and the spacing normal, w h e n the n u m b e r of plants exceeds 800 per hectare. W h e n the plants are too closely spaced, natural thinning is extremely slow. It might be mentioned that the occurrence of various types of frost d a m a g e to pine can be reduced if the “competing vegetation”, for instance, the mountain birch, is cleared in such a w a y that the remaining birch trees form a sheltering stand. T h e climatic data, and those for d a m a g e to plants, obtained from the ecological station at Aksujärvi, m a y be quoted as a n example. On the area from which all birch w a s removed (A), the temperature o n a cold July night in 1962 varied between -20 C a n d -50 C, while the temperature o n the neighbouring control plots (B,C, D),o n which the birch stand w a s retained to various degrees, lay at about O0 C. Injury to the pine shoots w a s appreciably m o r e frequent on plot A , (i.e. greater than 60 per cent of the total n u m b e r of plants), than o n plots B, C and D, where f e w plants were d a m a g e d under the protective shelter of the birch. E v e n in frost-free growing seasons, plants less than 2.5 m tall o n the plot from which birch w a s completely cleared (A), grew notably less well than those which grew under the shelter of the thinned birch. Because of its extremes of climate, the forest limit is especially suitable as a site for forestry experiments designed to investigate, for instance, the resistance of the plant material tested to the exigencies of
Remarks on current silvicultural research in the Subarctic of Finland
FIG. 5. Principal interrelationship between number of seed trees and seedlings per area unit and dominant seedling development.
Height of dominant seedlings (cm)
4001
__BOO Number of seedlings per hectare
climate, both in winter and in s u m m e r . Such experiments, employing various sizes and ages of plant, various hybrids and provenances and different methods of treatment of roots a n d shoots, have been laid out in ten experimental areas, in accordance with a m o d e r n design (> 200 ha), and incorporating different types of soil preparation and treatment of the birch shelter stand, as follows (Fig. 6):
~
I. II. III.
Pine provenances, and methods of clearing. Planting methods. Pine provenances. IV. Natural regeneration after burning, with different numbers of seed trees retained (demonstration, n o replications). Seeding following various degrees of clearance and forms of soil preparation (e.g. burning).
~~~~
FIG. 6. General m a p of regeneration experiments carried out at Aksujärvi, northern Finland.
347
G. Sirén
Effectof plant size,and of various root and shoot treatments at different altitudes (220-350m above sea level). The importance of planting depth, screefing and size of plant. IX.The importance of time of planting compared for various planting methods. X.Importance of plant size, age, root and shoot treatments with various provenances. The experiments were laid out between 1953 and 1964. The ecological station has been in operation since 1957 (to a reduced extent since 1962). The investigations have shown so far that: 1. The larger rooted plants are clearly superior where survival and growth are concerned. 2. The pine provenances of northern origin are preferable to those of southern origin. 3. The reduction of the plant’s needle mass by half does not harm its ability to grow; for the purposes of mechanized planting, it is important to k n o w that a narrow crown is not disadvantageous to the plant. 4. The importance of planting depth has been exaggerated-in conventional planting, at least two-thirds of the root should be below the soil surface, but plants will tolerate burial to the base of the youngest whorl. 5. The method of planting becomes less important when planting is done carefully. 6. Rooted plants will tolerate even very careless planting; this is an important consideration in mechanized planting. 7. Plants survive better when planted early or late in the season,i.e. when they are dormant.
Remarques sur les recherches effectuées en sylviculture dans la région subarctique de Finlande (Gustaf Sirén)
L’auteur rend compte de recherches effectuées en sylviculture, au cours des années 1953-1964, dans dix zones expérimentales et portant sur des plants d’âges et de tailles diverses, sur des sujets de provenances variées et sur des hybrides. I1 décrit également les méthodes employées dans le traitement des organes aériens et souterrains. Ces expériences ont démontré que les plants à système racinaire très développé sont nettement supérieurs en résistance et en développement, que les pins originaires du nord sont préférables à ceux du sud, que le rognage de la moitié de la couronne n’a pas d’incidence sur la croissance et qu’on a donné trop d’importance à la profondeur de plantation. On a
348
8. The production ofinterprovenancehybrids appears promising;the cross northern provenance x southern provenance, although the scatter is wide, gives offspring of superior growth. 9.Seeding is less successful than planting, and natural regeneration is easy to obtain after seed years. 10. The opportunities for using silvicultural methods for environmental control should be exploited. In addition to these experiments, since 1964 the relation between increment and various components of the climate has been investigated with the aid of registering dendrometers, which continuously record, to an accuracy of 10 p, the radial fluctuations of pine. Such pairs of instruments are set up at latitudes 610, 620, 630, 640, 670, and 690 N. This work is carried out partly in collaboration with M. Leikola, w h o has published a paper on the subject. Before closing this discussion of the climate as a stress factor, I should like to draw the attention of the symposium to the influence of other such factors on the forest limit.Here at Aksujärvi, w e can see h o w the birch geometer has devastated the birch forest, causing what is effectively a retreat of the birch-forest limit, and one which will probably also affect, in the course of time, the tree limit of pine. In summary, it m a y be said that the Subarctic should be considered, from the point of view of forestry, as a vast but inexpensive laboratory, with almost unimaginable opportunities for studying the interesting relationships between plants and their environment.
également pu constater que la méthode employée a moins d’incidence que le soin dans l’exécution et que les plants bien racinés pouvaient m ê m e supporter Une plantation défectueuse. Les plants survivent mieux si on les met en terre soit tôt, soit tard dans la saison, c’est-à-dire à l’état dormant. B o n nombre des remarques ci-dessus sont d’un intérêt particulier pour l’étude des mécanisations de plantations. Les croisements entre plants du nord et du sud semblent pleins de promesses avec, toutefois, une dispersion très marquée. L e semis donne de moins bons résultats que la transplantation et la régénération naturelle peut être facilement obtenue dans la période qui suit une année de fructification. L’auteur termine en souhaitant que tous les efforts soient faits pour mettre au point les méthodes permettant d’améliorer les conditions écologiques.
International scientific co-operationin conservation with special reference to peatlands T.Pritchard
THREATS TO WETLANDS
INTERNATIONAL PROGRAMMES
Wetlands, whether they be peat bogs, marshes, lakes or other wet habitats, are increasing in importance for research in ecology, peat stratigraphy, freshwater biology and other environmental sciences. Such habitats are exceptionally vulnerable to damage, and sites of great importance for research, education and wild-life conservation are disappearing at an alarming rate, especially in some parts of Europe. Similar sites in those parts of the world which have so far remained relatively unaffected by mankind are likely to be exploited or polluted in the very near future. Peatlands are being exploited or modified by peat extraction, drainage schemes and reclamation for agriculture and forestry. Freshwater lakes are being used increasingly to supply water for domestic and industrial uses and for recreational purposes and are becoming polluted by m a n y kinds of organic and inorganic substances. Such exploitation of natural wetlands often leads to irreversible changes in the habitats, and sites well known to scientists and well documented scientifically over the years have become either useless for further research and education or greatly reduced in value for these purposes. Quite apart from the losses to science-in that classical and other valuable but lesser known sites are being damaged or destroyed-many areas of great amenity value to the public are being adversely affected. However, I a m not going to consider this broader problem of environmental conservation. I will concentrate on the theme of international co-operationamongst scientists and practical conservationists directed towards the promotion of a programme to take care of representative examples of wetlands of international importance for research in ecology and other environmental sciences.
Amongst the international agencies and societies which are interested in this kind of task, IUCN and IBP are the two organizations most intimately concerned. IUCN has promoted the three wetland conservation programmes listed below, in recent years. PROJECT M A R
This project, under the supervision of Dr. Luc Hoffm a n of France, has been in operation for about five years and has resulted in the publication of a list of temperate 'wetland sites of special ornithological importance. More detailed information about this project can be obtained from IUCN Headquarters in Morges (Switzerland). PROJECT A Q U A
This is operated jointly by the Societas Internationalis Limnologiae (SIL), IUCN and IBP/PF, under the supervision of a committee whose Chairman is Professor Hans Luther of Helsinki. The initial object of Project Aqua is to draw up an annotated list of inland waters of international interest to science for research and for education and training. It is proposed that inland waters of scientific importance should be classified as follows: Part A: Habitats in a natural state or only very slightly modified. Part B: Habitats already altered or entirely created by man. Both can be subdivided into: 1. Classical sites, well studied and well documented. 2. Sites (u) which are important as having been used for past or current research; (b) where extensive
349
T. Pritchard
research is planned; or (c) which are regarded as of high potential research value. Part B will contain a further group: 3. Sites subject to rapid change, e.g. natural waters subject to accelerated eutrophication and m a n made lakes. PROJECT T E L M A
(sometimes known as PROJECT PEATLAND) The basic object in the case of peatland sites is very similar to that in the case of freshwater sites. The aim is to obtain data to enable a world list of the most important peatland sites to be prepared. This project was considered by IUCN and IBP at a meeting in Lucerne in July 1966, and it was agreed that these two international bodies would co-operate to carry it through. It was also agreed that Project Telma should be announced and discussed at the Unesco Symposium held at Helsinki, and IUCN and IBP asked m e to undertake this task and to set up an informal steering committee for the project.
ACTION ON PEATLANDS-THE FUTURE F O R PROJECT TELMA
An informal meeting was held during this symposium at Otaniemi, of scientists w h o are interested in peatlands. This ad hoc steering committee recommended that an international committee of peatland specialists should n o w be established,under the auspices of IBP and I U C N , so that Project Telma can be planned in detail. It is not possible to say precisely ill be carried out, but, in the light h o w the project w of discussion held here and elsewhere, I would like to give a tentative outline of what is involved. De3nition of the objectives
It is suggested that the objectives will be: 1. The preparation of a world list of sites which are of international importance to science and the promotion of their conservation. 2. The encouragement of communication and collaboration amongst research scientists investigating peatlands.
The geographical areas to be covered Although a world list should eventually be aimed at,
350
emphasis in the first instance should perhaps be placed on documentation of sites north of 500 N. Techniques for implementing the project
The main tasks will be to design an international system for gathering and evaluating data about northern peatland sites of scientific importance so that an internationally agreed world list can be prepared and published as an IBP/IUCNHandbook. ill involve holding at least one international This w meeting of experts in the near future. It is envisaged that Site Report Forms will be prepared for the systematic recording of data, much of which can be transferred to punched cards. Clearly, in some countries considerable progress has already been made in distinguishing and mapping major peatland types, as, for example, has been demonstrated by Dr. Ruuhijärvi of Helsinki in his paper, read earlier. It is very important that the techniques used and the results of research should become more widely k n o w n and applied. The Subarctic is a vitally important region for peatlands, and the papers given and views expressed at this symposium are going to be invaluable in the development of n e w ideas and techniques for conservation. W e have already heard about the proposal to include certain resolutions about conservation in the proceedings of the symposium. It seems that the conservation of peatlands should be carefully considered, not only in the context of their importance in studies on the ecology of subarctic regions but also because of their value to scientists further south. Dr. Pruitt has referred to the heavy responsibility of all countries with subarctic conditions to ensure conservation of places of importance for education and research for the benefit of those in the temperate and tropical regions. This responsibility lies in the first instance with the scientists and conservationists in the Subarctic. They can certainly depend upon what help IBP and IUCN can give them, and I feel sure that other influential international agencies will also be anxious to assist. Finally, I would like to say that it is anticipated that more detailed discussion on the full range of northern peatlands can be held next year. The IBP, in association with IUCN and the Natural Environment Research Council of Great Britain, hopes to hold an International Technical Meeting on the Conservation of Peatlands in Northern Regions.
General discussion and conclusions
Discussion générale et conclusions
Closing session at the University of Turku
Séance de clôture à l’université de Turku
Discussion leader: F. E. Eckardt
Animateur de la discussion : F. E. Eckardt
Au cours des séances de travail à Helsinki,nous avons pu entendre de nombreux et excellents exposés
Tel était, entre autres, le sens du terme adopté lors du premier colloque internationalsur les écosystèmes, tenu à Copenhague en 19651. L e géographe, c o m m e l’écologiste, se trouve devant un fait :la complexité de la biosphère. Pour décrire et comprendre la structure et le fonctionnement de cette biosphère, il est donc nécessaire, en plus de l’observer en tant que totalité à l’échelle de la planète, de l’étudier à une échelle plus petite davantage compatible avec les moyens dont dispose la science actuellement dans le domaine de la mesure et du stockage de l’information. L e géographe, pour cette raison, la subdivise en régions dont les dimensions varient en fonction du thème de recherches choisi : bassin de sédimentation, villes, cultures, etc. L’écologiste moderne, lui, préfère, dans la grande majorité, 1’ “écosystème” c o m m e unité de base. L’écosystème, dans l’acception courante du terme, peut être défini c o m m e une subdivision de la biosphère caractérisée par une certaine individualité structurale et fonctionnelle, une subdivision dont la surface coïnciderait, par exemple, avec celle, entière ou partielle, d’une forêt, d’une prairie, d’un marais, d’un lac, d’une rivière, d’un estuaire, d’un océan ou d’une autre entité semblable. Autrement dit, il s’agit d’un système fonctionnel plus ou moins stable dans le temps, possédant une étendue suffisante pour être caractérisé du point de vue de l’homogénéité et comprenant l’ensemble des organismes vivants présents et le milieu physique du globe avec lequel ils échangent de l’énergie et des substances.
traitant divers problèmes en rapport avec les régions arctiques et subarctiques et allant de la géomorphologie à la pédologie en passant par la physiologie et la climatologie. Un ensemble de sujets aussi vaste ne pouvant que difficilement faire l’objet d’une discussion générale valable,je m e permets de proposer, en conformité avec le titre et le but du colloque, que cette discussion soit, de préférence, centrée sur des problèmes d’écologie en faisant la part qu’il convient aux problèmes futurs d’organisation de la recherche dans les régions arctiques et subarctiques. I1 ne m’appartient certes pas d’imposer une définition particulière des divers concepts utilisés au cours de la discussion. Dans une réunion internationale il est naturel que les opinions concernant le sens à donner à tel OU tel terme divergent. C’est donc uniquement afin d’éviter la confusion que je voudrais d’abord,si vous le voulez bien,faire quelques commentaires, en particulier à l’intention de ceux qui ne sont pas eux-mêmes directement engagés dans des recherches dans le domaine de.l’écologie, sur la terminologie de cette science. D’abord en ce qui concerne le terme “écologie”, qui est utilisé, on le sait, dans des acceptions multiples. Pour certains, en effet, l’écologie est la science qui traite des rapports existant entre les êtres vivants et le milieu qui les entoure ou parfois de ce milieu seulement. Pour d’autres, c’est l’étude de l’être vivant en tant que totalité, le terme “physiologie” étant réservé à l’étude du comportement des organes, des tissus, des cellules ou d’autres fractions de l’organisme. Or, actuellement, une certaine tendance se dessine en faveur de l’emploi du terme dans le sens de “l’étude de la structure et du fonctionnement de la biosphère ou de ses composants, les écosystèmes”.
1. Functioning of terrestrial ecosystems at the primary production level. Proceedings of the Copenhagen SymposiumlFanetionne~ntdes écosystèmes terrestres au niveau de la production primaire. Actes du colloque. de Copenhague (ed. F. E. Eckardt). Paris, Unesco. (Natural reBources research/Recherehes sur les reBmources naturelles. V.) 2. En Union soviétique, toutefois, le terme **biogéocÉnose” est utilisé p o w désigner un concept biologique similaire.
351
General discussion and conclusions/Discussiongénérale et conclusions
~
’
C o m m e pour la région géographique la dimension de l’écosystème varie en fonction du thème de recherches ou du degré d’homogénéité ou d’individualité structurale et fonctionnelle demandés par le chercheur. Pour un zoologiste étudiant les grands ongulés de l’Ouest africain, l’écosystème comprendrait de vastes étendues de savanes, pour un mycologue, au contraire, il pourrait ne comprendre qu’une parcelle de forêt. Ce qui, cependant, caractérise toujours l’écosystème, c’est d’abord une certaine autonomie, une strate autotrophe, essentiellement des plantes vertes, assurant la fixation de l’énergie solaire nécessaire au fonctionnement de l’ensemble, ensuite une certaine faculté d’autorégulation,créée lors de sa genèse, lui permettant de conserver sa structure fonctionnelle face à la tendance de dispersion de l’univers. L a création du concept d’écosystème répond donc à un besoin à la fois théorique et pratique :théorique, en facilitant l’étude du degré de rationalité d’organisation de la biosphère, pratique, en permettant le choix le plus judicieux de terrains d’expériences. En adoptant cette manière de concevoir l’écologie, l’écologie végétale et l’écologie animale deviennent donc les sciences gui étudient respectivement la plante et l’animal en tant qu’éléments intégrés de l’écosystème. L’homme, m ê m e l’homme appartenant à des civilisations primitives, joue un rôle particulièrement important dans la biosphère. Mais, c o m m e d’une part son activité transgresse de manière considérable les limites de ce type d’écosystèmes qui intéressent le naturaliste,et c o m m e d’autre part ses moyens d’action reposent sur des connaissances accumulées pendant des générations, il convient généralement d’étudier cet aspect sous un chapitre à part. Pour des raisons de commodité et pour miesx mettre en relief les rapports réciproques existant entre l’organisme humain, ou la société humaine, et le milieu ambiant qui nous intéresse, on peut le désigner sous le n o m d’écologie humaine, bien que cette appellation ne soit pas acceptée par tous les chercheurs. En ce qui concerne la discussion qui va débuter,je pense qu’il serait souhaitable, après en avoir terminé avec les sujets proprement écologiques, d’aborder certains problèmes annexes dont dépend le succès de tout projet de recherches intégré à long terme dans les régions arctiques et subarctiques. L a recherche écologique dans les régions arctiques et subarctiques implique la prise en considération d’un nombre extrêmement élevé de données. Pour permettre la meilleure collection de ces données en vue de leur traitement ultérieur, il faut faire une large place à l’étude du traitement de l’information. Afin d’assurer, pour les générations futures, la préservation de l’information génétique contenue dans la biosphère, i1 convient d’aborder les problèmes relatifs à la conservation de la nature, autrement dit
352
la préservation de divers types d’écosystèmes fonctionnant normalement et dans lesquels peuvent trouver refuge des organismes vivants menacés d’extinction. Je m e permets donc de proposer que la discussion suive dans les grandes lignes le schéma suivant :
L’écosystème arctique et subarctique Structure
i
Basse atmosphère Couverture végétale Substrat édaphique
i
Fonctionnement
Ecologie humaine Traitement de l’information Conservation de la nature Conclusion
Dans ce schéma j’ai séparé la structue du fonctionnement de l’écosystème.Bien que, certes, la structure et le fonctionnement de l’écosystème ne constituent que deux faces d’une m ê m e chose, les liens entre les deux pouvant être révélés par l’observation de la biosphère à la lumière de sa genèse, leur étude distincte semble encore pratique au stade actuel du développement de l’écologie.
N. POLUNIN. Although I came to this symposium merely to listen, as a start to my rehabilitation to northern work after too m a n y years in hot climates, I have been prevailed upon by our Chairman to open this discussion on the types of ecological studies that could most usefully be undertaken by a network of ecological research stations such as should surely be established by and for international collaboration. T o my mind the various ecosystems cry out for concerted investigation. Although the subarctic and arctic biosphere m a y be distinguished very approximately by the presence or absence, respectively, of arborescent growth, it seems unnecessary and even undesirable in this context to separate them, but to visualize some such initial series of ecosystems and/or habitats as the following: Land from colonization
Nudation areas (glacial, emergence, etc.) Barrens (more or less permanent) Fellmark Tundras Marshes and mires Bogs Scrub Heath Taiga Forests (coniferous,broadleaf,mixed)
General discussion and conclusions/Discussiongénérale et conclusions
Freshwater
Marginal Benthic Planktonic and pleustonic (pelageal) Profundal Cryophytic Marine
Marginal (including tidal) Benthic Neritic planktonic Pelajic planktonic Profundal Abyssal Cryophytic Special
Bird-cliffs and their autotrophic ancillaries Other manured habitats and their ancillaries Man-cleared areas S n o w patches Flower slopes Saline and alkaline areas Cliffs, screes, etc. Besides all these, the various seres and seral areas should be investigated, and a fertile field should be that of transplant experimentation between stations located in different climatic belts.
F.E. ECKARDT.U n e des premières tâches de l’écologiste qui désire étudier la région arctique et subarctique est de différencier, au sein de la biosphère, par des approximations successives, des surfaces suffisamment homogènes et individualisées pour servir de base à ses recherches. I1 est secondaire au départ, que les surfaces délimitées correspondenteffectivement à des écosystèmes. Les unités proposées par le professeur Polunin ne peuvent peut-être pas être toutes considérées c o m m e des écosystèmes mais elles présentent l’avantage d’être repérables dans la majorité sur des images réaliséesàpartirde plates-formesaéroportéeset méritent certainement d’être retenues. Elles pourraient sans aucun doute faciliter considérablement l’organisation des premières recherches à entreprendre en c o m m u n par les pays participants. P.L. Johnson. During the presentations at this symposium w e have heard many fine reviews of subarctic research. It has often been difficult, however, to focus on the trends of these investigations in terms of original suggestions for the future. I should like, therefore, to invite speculation by the assembled experts on the appropriate areas to be emphasized in future ecological research. Toward this purpose I should like to discuss briefly four points. First, if it is still possible to consider the Subarctic and the Subalpine as a transition zone between forest and tundra, then several
problems are evident. The biogeography of the transition zone suggests that there are no organisms that m a y be called “subarctic”, but it is important to understand their distribution across this ecotone. Further it is necessary to k n o w the structure of interspecific populations, their cytology and ecotypic variation, if w e are to explain their occurrence. The more extensive use of reciprocal transplant gardens is recommended. The morphological and physiological adaptations of organisms to their habitat is poorly elucidated. In this regard measurements of canopy geometry, albedo and pigment content of vegetation would also be of direct aid to the application of remote aerial sensing. T o couple plants with their environment requires meaningful climatic measurements, especially net radiation, but to understand regional ecosystems more measurements of atmospheric dynamics are desired. For example, the atmospheric circulation of CO, and chemical pollutants such as DDT needs attention, Recently DDT has been found in m a n y migratory waterfowl nesting in the Subarctic, although this chemical was applied in temperate ecosystems. The second point is the need for an increase in experimental ecology, not only physiological measurements of processes in single species populations and community metabolism, but the experimental manipulation of organisms and environments. Ecologists should not shackle their investigations solely to climax communities ; in fact, it is conceivable that entirely n e w and synthetic ecosystems will be realistic solutions for areas irreversibly disrupted by man. It is feasible to experimentally disrupt an ecosystem for the purpose of discovering h o w it functions. Thirdly,it would seem that the IBP programme and its theme, productivity in the service of man, is especially pertinent to the Scandinavian countries. The dependence of m a n on his resource base is everywhere evident and appreciated in Finland. Perhaps circumpolar ecology is the best context to rally the international point of view. Toward that end I would urge that effort be directed toward construction of models of typical as well as abstract arctic ecosystems, such that eventually such techniques as general systems theory and systems analysis m a y hasten our feedback to the solution of applied problems. My fourth point concerns the training and exchange of information to accomplish these objectives. Technology including the tools of remote sensing radioecology and information theory is developing at an accelerated rate. This background information is important as the foundation for progressive research and is often difficult to read in m a n y languages. Perhaps the most direct and efficient means of approaching these problems is by exchange of young graduate students and investigators between research centres. An increase in the world-wide exchange of visiting scholars at the several centres of arctic and
353 25
General discussion and conclusions/Discussiongénérale et conclusions
subarctic research is indicated and hopefully the International Biological Programme among others, will make this possible.
F. E. ECKARDT. Suivant le climat, la nature du substrat géologique et l’information génétique disponibles, les écosystèmes qui se sont développés au cours de la genèse de la biosphère ont atteint un degré de complexité structurale et fonctionnelle très différent. Dans les forêts denses, en particulier dans les forêts équatoriales, les espèces qui les composent sont fortement interdépendantes. L’étude d’un organisme isolé, extrait de cette forêt, est donc difficile, le comportement de l’individu ne pouvant être compris que dans le contexte du fonctionnement de tout l’écosystème. L’effet direct du milieu physique extérieur à la végétation sur les plantes et les animaux est faible. Dans les écosystèmes arctiques, par contre, c o m m e d’ailleurs dans les écosystèmes désertiques, le contact entre les organismes vivants et le milieu physique est très intime. L e problème de l’étude de l’évolution des caractères d’adaptation se pose donc de manière très différente pour ces deux types d’écosystèmes. Les deux types d’écosystèmes se distinguent également par leur degré de stabilité très différent. L e mécanisme homéostatique autorégulateur est bien plus développé dans la forêt équatoriale que dans les écosystèmes des régions arctiques, où les populations animales et végétales peuvent varier considérablement d’une année à l’autre. A cela s’ajoute que le climat des régions arctiques et subarctiques est lui-même caractérisé par une très grande instabilité,en particulier pour ce qui concerne la température estivale, dont le rôle sur le développement des organismes vivants peut jouer un rôle décisif. I1 s’ensuit que les régions arctiques constituent un champ de recherches particulièrement favorable à ce qu’il convient d’appeler l’autoécologie téléologique, autrement dit l’étude des caractères d’adapta~ionde l’individu végétal ou animal, considéré isolément, en sous-entendant que l’adaptation chez l’animal peut être de nature éthologique. C’est pour cette raison que la proposition du Dr Johnson d’accroître le nombre des jardins de transplantation est tout particulièrement intéressante. Les problèmes susceptibles d’être étudiés dans de tels jardins sont très nombreux. On peut mentionner, en plus de ceux déjà évoqués, l’action de la température sur la consommation respiratoire des substances de réserve chez les végétaux, la phénologie des plantes chez lesquelles l’initiation florale précède d’une année l’éclosion des bourgeons floraux, etc. Les résultats obtenus dans ces jardins pourraient faciliter considérablement les recherches sur l’évolution des divers caractères adaptatifs, dont l’intérêt transgresse le cadre de la région arctique et subarctique. 354
I1 convient d’ajouter que la fluctuation des populations animales, qui n’est autre que le << pompage» du système de régulation de l’écosystème, constitue un sujet d’étude très important en soi, bien que très souvent négligé. H.ODIN.In connexion with what has just been said, I would like to add a few words about meteorological stations and methods used for meteorologicalmeasurements. First, the network of weather stations now operating in the Subarctic should be extended considerably. This might be achieved by using automatic stations. Secondly, intensive studies should be made of the climate within limited areas (local climate), so as to elucidate the influence of topography and vegetation. In our o w n research centre w e use stations which permit the recording of temperature both at standard and at lower levels as well as the measurement of precipitation. These stations are further equipped with simple and effective devices for measuring the depth of frost in the soi1.l The risk of frost damage evaluated from measurements made by means of minimum thermometers and depth of snow cover are mapped more or less precisely from measurements made along various transects. Thirdly, when the reaction of the vegetation to the ecological complex of which climate is a part is to be studied,different types of measurements are required. W h e n describing the environment, great attention must be paid to the heat and water exchange between the soil and the atmosphere. These exchanges can at present be measured directly, but they m a y also be calculated more roughly by conventional methods from profile measurements of wind, temperature and humidity, as well as from radiation measurements. At the Royal College of Forestry, profile equipment is used near the forest limit and temperaturemeasured from a depth of 1 m in the soil to a height of 10 m above ground. It is the intention later to use very small thermocouples at zero level at the ground as well as just above and just below it; the thermocouples must be smallin order to reduce error due to radiation. For measurements at higher and lower levels,platinum resistance meters are employed. Cup anemometers are used for measuring wind velocity and hair hygrometers for humidity-these latter, however, prove to be entirely unsatisfactory. When presenting my paper in Helsinki, I gave examples of radiation instruments. Here I must also mention the Australian net radiometer (Middleton & Co. Pty. Ltd.), which is a rather good inSIrument. All instruments (about fifty) are connected to a digital recorder, which enter data automatically on punched tape. 1. These measurements are carried out by DI.Anderson at the Royal College of Agriculture in Uppsala.
General discussion and conclusions/Discussiongénérale et conclusions
H.AHLMANN.I would only like to say that I agree with what Dr. Odin said. W e need more meteorological stations and this especially at different altitudes. W e must do all w e can to enlarge the number of meteorological stations. W e still k n o w very little about climate but w e have n o w learned that excellent instruments have been devised. Please do all you can to develop meteorological investigations. W.O. PRUITT. Snow
and winter ecology is actually
in its infancy;there is m u c h old-fashionedexploration necessary. At this stage of development there is danger in premature stagnation of ideas, nomenclature and techniques. There is a need for a number of simple stations, operating throughout the year, in areas where the fauna, flora and environment are protected from disturbance and where detailed, longterm studies of winter ecology can be carried out. There is a need for n e w and better instrumentation and techniques. There should be exchanges of instruments and ecologists between Eurasia and North America. When new instruments and techniques have been thoroughly evaluated they should be standardized to ensure comparable data. There is a great need for emphasis on training students to be active in winter field work. There is a need for large, regional snow surveys, carried out from a biological point of view. Surveys fgr hydrological, engineering or synoptic weather purposes, as presently designed, are virtually useless for biological purposes. There is an especial need for fund-grantingagencies to appreciate the value of winter field studies.
F. E. ECKARDT. I1 est certainement possible de faire des observations très précises et utiles sans disposer d’un matériel de mesure complexe. Un chercheur expérimenté, connaissant bien sa région, c o m m e le Dr Pruitt lui-même, peut obtenir de très nombreux renseignements précieux à l’aide de méthodes simples. I1 y a peut-être actuellementune tendance exagérée à considérer c o m m e seules valables Ies données enregistrées automatiquement sur cartes perforées, à surestimer les techniques métrologiques modernes. A cet égard, il importe sans doute de trouver le rapport le plus judicieux entre la nature de l’observation et le degré de perfection de l’instrument de mesure, en se rappelant,toutefois, que les avantages des méthodes simples sont souvent contrebalancées par les difficultés que présente l’exploitation des résultats. C. TAMM. I would first like to emphasize that I a m not entending to recommend any standard methods for intensive pedological studies of special problems. In m a n y cases the pedologist will have to devise new methods himself or to modify existing ones with a view to solving particular problems.
It would be however of the greatest value, I think,
if at least some soil data could be included in the site descriptions made by biologists, geographers, geologists and other field scientists. Although soil properties are structural features, it would be highly desirable to concentrate on those properties which are most directly related to the functioning of the ecosystems. These properties can be divided into three groups: 1. Morphological profile characteristics. 2. Physical characteristics. 3. Chemical and physico-chemicalcharacteristics. Group 1 covers visual observations such as the occurrence of soil horizons, mottling, depth of impermeable layers (bed-rock,permafrost). This type of observation should always be made, if possible. Group 2 includes a number of properties related to the water and heat régimes of the soil. Actual water content in itself is of course of interest,but it is desirable to relate it to the water-holding capacity of the soil (“field capacity” and 15 atm value). By doing this, information is also obtained concerning pore space and aeration. Soil texture is relatively easy to determine, while no universally accepted method for aggregate structure (if any) can be recommended. Soil texture data are somewhat difficult to deal with in extensive studies,but in some cases it m a y be useful to take the diameter of the particle at the “lower quartile” of the distribution as parameter (the 25 per cent of the particles smaller than the parameter have at least as m u c h importance for the soil properties as have the 75 per cent larger particles). Group 3 concerns characteristics important for plant nutrition. They are, however, also related to the soil-formingprocess. Soil acidity and conductivity can be measured in the field or in a simple field laboratory. Determinations of total and “available” plant nutrients require better laboratory facilities and are usually made on dried samples. Organic matter content (or percentage of carbon) as well as total nitrogen are importact characteristics and should be expressed both as percentage of the dry weight (much more time-consuming) and per unit area. Total phosphorus and potassium are particularly important when dealing with peat samples. “Available” nutrients determined by various extraction methods m a y be useful under certain circumstances but the results are often difficult to interpret. Finally I would like to call attention to the possibility of measuring organic matter turnover by radiocarbon dating, although this method is n o w becoming more difficult to apply due to atmospheric contamination with CI4. Water movement can under certain circumstances be studied by tritium measurements.
F.E.ECKARDT. Nous avons entendu des interventions ayant trait surtout à l’étude de la structure des éco-
355
General discussion and conclusions/Discussiongénérale et conclusions
systèmes. Permettez-moi d’ajouter quelques commentaires sur le fonctionnementdes écosystèmes. L e maintien de l’écosystème en tant qu’entité organisée est assuré grâce à deux fonctionsessentielles : la photosynthèse et la minéralisation, dont le rôle est de permettre respectivement l’élaboration de composés organiques complexes à partir d’éléments simples empruntés au milieu ambiant et le retour de ces éléments au monde minéral. L’importance de ces deux processus, soit l’un par rapport à l’autre, soit en valeur absolue, détermine la quantité de biomasse pouvant être présente dans une localité donnée. L a minéralisation a déjà été traitée par le D*T a m m . Nous savons que les éléments minéraux sont rendus plus ou moins accessibles aux racines suivant les modalités de la pédogénèse, c’est-à-dire suivant le mode de stockage des substances en voie de décomposition, lui-même dépendant de nombreux facteurs physiques et biologiques. Dans les régions froides, la minéralisation semble souvent, c o m m e nous l’avons vu, constituer un facteur limitant. L e processus photosynthétique nécessite l’apport, . au niveau de la chlorophylle, de photons porteurs de quantités d’énergie convenables et de CO,, ainsi qu’un ensemble de conditions thermiques, hydriques et nutritionnelles favorables. Suivant les régions du monde, la production de matière végétale résultant de ce processus peut donc être limitée par un ou plusieurs de ces facteurs. Dans la région arctique et subarctique qui nous intéresse,la déficience minérale et la basse température semblent responsables pour une large part de la faible production, mais cette supposition demande à être étayée par des expériences sur le terrain. Pour mieux comprendre le fonctionnement des écosystèmes arctiques et subarctiques,il faudrait donc d’une part déterminer la production primaire réelle en matière sèche des écosystèmes - ou la fixation du carbone - d’autre part, chercher les causes qui la rendent si faible. Parmi les méthodes permettant l’étude de la production primaire de l’écosystème en tant que totalité, il convient de mentionner celles basées sur l’analyse verticale du flux de CO, au sein de la basse atmosphère immédiatement en contact avec la surface de la couverture végétale. Elles présentent l’avantage de ne pas entraîner de modifications au sein de l’écosystème pendant les mesures, mais supposent connue la quantité de CO, dégagé par les organismes hétérotrophes, en particulier dans le sol. L’une de ces méthodes est fondée sur le principe dit “de similarité”: le mécanisme de transport vertical par diffusion turbulente de gaz carbonique,de vapeur d’eau, de chaleur, de quantité de mouvement horizontal est semblable dans certaines conditions de stabilité atmosphérique. Lorsque ces conditions sont satisfaisantes, il suffit de connaître la densité de flux vertical d’une de ces entités physiques, ainsi que les
356
gradients correspondants,pour calculer la densité de flux des autres. Connaissant, par exemple, l’évaporation réelle, la concentration en CO, et en vapeur d’eau à deux niveaux différents au-dessus de la couverture végétale ainsi que la respiration du substrat édaphique, on peut en déduire la photosynthèse apparente. L’évaporation peut être déterminée en établissant le budget de chaleur de l’écosystème. U n e autre méthode, très simple dans son principe, est fondée sur la mesure instantanée de la composante verticale de la vitesse du vent, ainsi que de la concentration en CO, de l’air déplacé. I1 est peu probable, cependant, que ces méthodes, excellentes dans leur principe, puissent convenir aux recherches dans la région arctique et subarctique, sans être combinées avec d’autres méthodes, la production primaire et, par conséquent, le flux vertical de CO, étant trop faibles. I1 ne reste donc que des méthodes consistant à déduire le fonctionnement de l’écosystème d’après le comportement individueldes plantes qui le composent. Parmi ces méthodes on peut mentionner la méthode gravimétrique et la méthode des enceintes climatisées. L a méthode gravimétrique est basée sur la détermination de la variation de biomasse en fonction du temps. En présence de plantes faciles à extraire du sol, le procédé consiste normalement à peser des échantillons de végétaux entiers prélevés au hasard à intervalles de temps régulier. Par contre, en présence d’une forêt, il convient généralement de cueillir seulement les pousses annuelles, la quantité de substance accumulée dans le bois étant évaluée par des mesures répétées de l’épaisseur des tiges intactes. Les résultats de telles mesures peuvent être généralisés en établissant une fois pour toutes les corrélations existant entre la production totale et un paramètre facile à déterminer,tel que par exemple l’épaisseur des troncs à hauteur d’homme. Basée sur un pro tocole d’échantillonnage approprié, la méthode gravimétrique permet une excellente détermination de la variation de la biomasse aérienne en présence. Cette variation de biomasse aérienne, cependant, ne fournit que des renseignements peu précis sur la production primaire totale nette de l’écosystème, la production radiculaire et la perte de substances par chute de feuilles et de racines, par l’action des herbivores, etc., étant souvent difficile ou impossible à évaluer. L a méthode des enceintes consiste à enfermer un rameau ou une parcelle de végétation dans une enceinte climatisée et à étudier les échanges gazeux en fonction de diverses combinaisons de facteurs climatiques (température,humidité, teneur en CO, de l’air, vitesse du vent, rayonnement, etc.) créées artificiellement. Connaissant le microclimat réel, il est donc possible d’en déduire les échanges gazeux tels qu’ils s’effectuent dans les conditions naturelles. En présence d’une végétation haute, nécessitant
General discussion and conclusions/Discussiongénérale et conclusions
l’emploi de plusieurs enceintes disposées à différentes hauteurs, une généralisation des résultats est possible en déterminant le flux vertical de CO, au sein de la couverture végétale au moyen des méthodes basées sur le principe de similarité déjà décrites. Cette méthode des enceintes climatisées,apparentée à celle des phytotrons, présente l’avantage de fournir de nombreux renseignements non Seulement sur les échanges gazeux réels, mais aussi sur les facteurs susceptibles de limiter ces derniers. Elle implique l’emploi, cependant, d’un appareillage complexe nécessitant la surveillance d’un personnel hautement qualifié, mais devrait être utilisable assez facilement dans les régions arctiques et subarctiques, qui toutes dépendent de nations industrialisées. I1 n’est guère possible, en l’état actuel du développement de la méthodologie éco-physiologique, de recommander l’une des méthodes proposées de préférence aux autres, le choix le meilleur dépendant du type de végétation soumis à l’étude et de la précision des mesures demandées. D e manière générale, c’est en combinant plusieurs méthodes qu’on parvient aux résultats les plus fatisfaisants. Par de telles combinaisons, on peut en particulier estimer la respiration des racines et des organes chlorophylliens exposés au soleil,paramètres actuellementimpossibles à déterminer directement.
J. MALAURIE. AU terme de ces journées de discussion, qu’il soit permis à un géographe de la délégation française de formuler d’abord ses vifs remerciements pour les organisateurs de ce premier et précieux colloque d’écologie subarctique, l’Unesco et le gouvernement finlandais très particulièrement. Ensuite d’ajouter quelques remarques générales qui m’ont été demandées et qui peuvent avoir u n intérêt méthodologique. je le pense - laissé Nos divers débats ont apparaître la difficulté de définir, de façon satisfaisante pour tous, les notions d’arctique, hémiarctique et subarctique. Qui plus est, ils ont fait découvrir que cet effort n’était pas essentiel, les problèmes de spécificité se situant sans doute à un niveau différent. Ambigus, les concepts géographiques c1assZcatoire.s traditionnels nous sont apparus trop vagues pour être opérants,. L’observation est aussi valable pour les écologistes que pour les géologues et géomorphologues ; j’oserai dire aussi pour les administrateurs responsables du développement économique de ces territoires. Nos communications ont souligné que pour s’en tenir à la seule géomorphologie, formes actuelles et vives, formes anciennes coexistent si intimement dans au moins dans les secteurs que j’ai l’Arctique particulièrement étudiés (le Nord-Ouest groenlandais : terres d’hglefield et de Washington) qu’il n’est pas toujours possible de discerner ce qui revient aux unes ou aux autres. Seules les formes en matériel sableux très meuble sont, dans le nord-ouest du
-
-
-
Groenland, exclusivement postglaciaires. En roche résistante, le polygénisme du modelé n’est pas douteux. L’identité arctique des formes vient seulement de la localisation géographique, du lieu de résidence de ces reliefs. C’est dire l’équivoque de ces qualifications, d’autant que I’on ne s’accorde pas toujours sur la spécificité des agents et systèmes en cours. On sait les erreurs néfastes commises à cet égard. Par exemple les pingos, reconnus un peu vite c o m m e formes caractéristiques de pays à permafrost, ont été découverts en Belgique. I1 en est de m ê m e pour les craquelures de gel, et pour la gélifraction, qu’une analyse correcte révèle beaucoup moins active en haute latitude qu’on ne le supposait. Vues seulement grossières ? U n e meilleure connaissance pourrait toujours les corriger. Et insister sur ces erreurs descriptives serait d’un médiocre intérêt. I1 y a plus :la définition d’un système arctique est d’autant plus malaisée en géomorphologie que l’observateur ignore aussi bien l’essentiel de la structure profonde des processus en jeu que la finalité morphologique des forces en présence. I1 est ainsi possible de concevoir qu’au sens géodynamique, il n’est pas de système, de faciès arctique, mais des formes, des types composites, conséquence de la pluralité d’épisodes dont nous ignorons l’échelle et la perspective à partir desquelles ils deviennent cohérents. Sans cesse sollicité par les sciences anthropoloau giques et zoologiques, le géographe humain moins dans l’Arctique et le sub-Arctique ne répond guère par l’analyse globalisante attendue de lui, mais, ali titre des questions physiques, par des physiographies sommaires et des typologies descriptives. L e site, le théâtre, la région, objets d’interrogation passionnée, restent de la sorte des cadres inertes, ne devenant jamais supports ni explications. Reste au géographe et bien à lui d’établir par des faits concrets la réalité d’un itinéraire, d’un lieu, d’une région. L a notion de territoire, si riche en éthologie animale, de site et d’habitat en ethnologie, appelle de sa part des réponses concrètes. L’unité d’un espace tel que les bêtes, les chasseurs s’y soient regroupés et, au moins pour les hommes, maintenus pendant des générations - et ce, dans les mêmes sites (une centaine pour les Esquimaux), pratiquement les mêmes aires d’action répond de facteurs corrélatifs, notamment de caractère tellurique, hydrologique et atmosphérique. M o n intervention vise donc à souhaiter que, polir l’immense et exceptionnel horizon que constituent les espaces arctiques, la géographie parvienne à briser les cadres de son propre académisme. Si nos idées en anthropologie hyperboréale sont pauvres, c’est qu’elles sont appauvries par la division forcée d’un substrat phénoménal unique et source de quiproquo et d’incompréhension. I1 serait souhaitable qu’un tel colloque soit l’occasion d’admettre que l’analyse géohistorique dans les régions arctiques devrait relever du m ê m e
-
-
-
-
-
357
General discussion and conclusions/Discussiongénérale et conclusions
régime d’indivision que l’expérience, par exemple, psychologique... L’histoire des autochtones hyperboréaux est c o m m e la composante de faits naturels tels la biogéographie, les changements climatiques, la topographie ayant, avec les faits humains, u n rôle tout à la fois dominateur et dialectique.
F. E. ECKARDT. Les plantes et les animaux se sont adaptés au cours de leur genèse, non pas en vue de supporter l’action d’un ou de quelques facteurs mésologiques agissant séparément,mais pour pouvoir vivre et se reproduire dans un milieu complexe où l’action spécifique de chaque facteur est inséparable de celle des autres, les chaînes de causalités étant intimement interconnectées. C’est la raison pour laquelle beaucoup de chercheurs mettent en question actuellement le primat de l’étude causale en écologie. Au lieu d’axer leurs recherches sur la détermination du comportement des organismes vivants en fonction de l’action d’un ou de quelques facteurs du milieu ambiant considérés isolément, ils s’efforcent d’élucider comment les diverses plantes et animaux se trouvent insérés organiquement dans l’écosystème. En présence d’écosystèmes simples, c o m m e il en existe dans les régions arctiques, où par surcroît un facteur,le froid,joue un rôle déterminant,l’étude causale garde sa valeur. En présence d’écosystèmes plus complexes, comprenant un nombre plus élevé d’espèces,il devient nécessaire par contre de considérer l’environnement physique et biotique c o m m e un tout. Les difficultés méthodologiques s’accroissent donc en fonction de la complexité de l’écosystème. Elles deviennent presque insurmontables lorsque l’homme en est considéré c o m m e un élément intégré. En écologie humaine, en effet, il faut ajouter, à la complexité du milieu ambiant, la complexité de l’homme qui est à la fois objet et sujet, fini et infini, physique et métaphysique, réel et symbolique. L’écologie humaine consiste donc essentiellement en l’étude d’un ensemble d’interactions résultant de la présence sur le globe terrestre de deux systèmes OU structures fonctionnelles, celle de la biosphère et celle du cerveau humain. Cette manière de voir, dans son essence structuraliste, adoptée avec succès dans le domaine de l’anthropologie et de l’histoire,s’avérerat-elle utile en écologie humaine ? I1 est difficile de se prononcer actuellement, l’élaboration d’une méthodologie propre à cette science étant à peine ébauchée. Un autre problème très lié à l’écologie est celui de l’avenir des populations des régions isolées du monde, c o m m e la région arctique et subarctique, qui nous intéresse. Ce problème se réduit actuellement à une option entre deux solutions : isoler les groupes ethniques afin d’éviter la ruptiire des structures sociales qui leur sont propres en préservant le milieu dans lequel ils vivent et dans lequel cette structure autrement dit en les empêchant de tirer a été créée
-
3.58
profit des avantages de la civilisation moderne - ou bien accélérer leur intégration dans la civilisation moderne avec les risques d’échec qu’elle implique. Dans un avenir pas trop lointain, on peut espérer que le problème pourra être résolu grâce au développement des méthodes modernes d’enseignement. En effet, la faculté de communication symbolique ayant pris naissance pendant des périodes reculées de l’évolution humaine ne devrait pas, aujourd’hui, varier sensiblement d’un groupe ethnique à l’autre : cette faculté ne permet plus actuellement d’augmenter le taux de procréation ou les chances de survie de l’individu. L a faculté de communication symbolique de l’homme est seulement exploitée de manière très inégale par les différentes civilisations. Dans les civilisations dites avancées,l’information transmise d’une génération à l’autre est préparée de manière à permettre une meilleure mise en valeur des aptitudes innées de l’enfant. Dans les civilisations moins développées, par contre, il y a en général sous-exploitation de ces aptitudes. I1 y a donc tout lieu de croire que, par le développement des méthodes de préparation de l’information, on peut parvenir à rationaliser l’activité intellectuelle et rendre l’enseignement plus universel. Un bon exemple de cette possibilité de rationalisation de l’activité intellectuelle est fourni d’ailleurs par la création de ces nouveaux outils mathématiques, souvent évoqués pendant ce colloque, que sont la théorie des ensembles et la théorie de l’information, dont les principes de base semblent bien intelligibles aux enfants en bas âge.
F. DI CASTRI. L’exploitation de données recueillies lors de la réalisation d’un programme de recherches scientifiques ne constitue pas une opération mécanique pouvant reposer sur des règles simplistes. I1 s’agit d’un processus complexe s’appuyant sur une base théorique très solide comportant au moins quatre étapes :collecte,traitement,transmission et stockage, sous forme codée, des données. C’est un problème qui se pose de façon très critique pour la recherche scientifique et que ce colloque illustre parfaitement bien. En effet, un nombre impressionnant de résultats intéressants a été présenté, mais qui nous donne unë information très difficile à interpréter et à exploiter. I1 serait fort malaisé d’en déduire des conclusions générales, faute d’une optique de synthèse et d’un langage communs capables de faciliter la communication entre spécialistes travaillant dans des domaines de recherche éloignés. Les deux premières étapes,collecte et traitement des données, sont bien connues, en particulier par les chercheurs qui désirent mettre l’accent sur l’aspect quantitatif de leurs mesures. I1 convient, toutefois, de faire ressortir certains points : 1. U n e planification des expériences doit précéder la
General discussion and conclusions/Discussiongénérale et conclusions
phase de récolte des données sur le terrain et être soumis à l’examen de statisticiens et spécialistes en informatique,car le traitement de l’information se montre souvent difficile ou impossible lorsque le statisticienn’a pas p u exprimer son avis au moment de la planification des expériences. 2. Cette planification préalable des expériences ne doit pas avoir pour unique but d’obtenir des données se prêtant à la vérification mathématique, mais aussi de permettre l’expression des données sous une forme facilitant leur transmission et leur codification ultérieures. 3. L a nécessité de la planification expérimentale préalable se fait davantage sentir lorsque les recherches ont pour objet un thème faisant appel à différents spécialistes (physiciens, mathématiciens, zoologistes, botanistes, etc.), qui voient chacun les problèmes avec leur optique et les expriment dans leur langage. Un tel travail d’équipe constitue certainement la formule la plus favorable pour aborder l’étude des écosystèmes. 4.Pour de telles recherches écologiques quantitatives, la manière dynamique et historique d’aborder les problèmes convient généralement bien, et cela m ê m e lorsqu’on désire y appliquer les diverses méthodes statistiques et les principes tirés de la théorie de l’information. En pareilles circonstances, onpeut effectuerles récoltes sur le terrain suivant des transects comprenant des zones à maturité et complexité variables, et étudier aussi les chaînes de sols (catenas), des successions écologiques,des régressions par suite de modifications zoo-anthropogenes (disclimax), etc. 5. I1 faudrait enfin attirer l’attention sur le fait que l’échantillonnage dans la zone subpolaire doit parfois être basé sur une méthodologie fort différente de celle qui s’applique aux régions tropicales et tempérées. Cela, non seulement en raison des différencesde degré d’homogénéitébiotiquemais aussi à cause des fluctuations très marquées qui caractérisent les biocénoses arctiques et antarctiques. En ce qui concerne la nécessité de rendre les travaux écologiques pIus accessibles et compréhensibles pour les non-écologistes, c’est surtout par la synthèse mathématique que l’on peut espérer y parvenir. Pour établir un tel trait d’iinion interdisciplinaire,il suffit de mettre en évidence les fonctions unissant les phénomènes en s’appuyant notamment sur les théories relatives à l’analyse de systèmes (cybernétique, théorie de la communication, théorie de la décision, théorie des jeux). Par ce procédé on peut parvenir à établir un modèle d u système étudié. A cet égard, la structure simple des écosystèmes subpolaires présente l’avantage de faciliter l’établissement d’un modèle qui, par surcroît,pourrait servir pour élucider d’autres problèmes fondamentaux se présentant dans des zones caractérisées par une plus grande complexité écologique.
En dernière analyse, ce point de vue synthétique est très important également pour codifier l’infarmation en vue de son stockage. Bien entendu, il s’agit d’un stockage dynamique pouvant être comparé analogiquement à une mémoire active. Or cet aspect ne peut être envisagé de manière efficace que dans le cadre d’une large coopération internationale. atant donné que les écosystèmes des régions arctiques et antarctiques présentent de grandes similitudes structurales et fonctionnelles, il serait préférable d’englober dans un seul programme international l’ensemble des recherches écologiques des zones subpolaires, cela surtout afin de pouvoir normaliser les techniques et le symbole de cartographie,de standardiser les instruments de mesure, d’établir une planification des recherches capables de faciliter la comparaison des résultats et leur diffusion finale.
F. E. ECKARDT. L a réalisation d’un programme de recherches écologiques à l’échelle de la région arctique et subarctique nécessite l’établissement d’un inventaire très vaste de paramètres physiques et biologiques. Cet inventaire reposant en grande partie sur des échantillonnages, une certaine normalisation des observations, c o m m e vient de le dire le professeur di Castri, est certainement indispensable :il faut que les observateurs utilisent un langage bien défini leur permettant de comparer leurs résultats et d’effectuer des synthèses au moyen de méthodes statistiques. I1 conviendrait donc,je pense, d’envisager l’établissement pour la région arctique et subarctique d’un code écologique, analogue à celui réalisé par exemple en France par le CNRC. Un tel code consiste tout simplement en une liste de paramètres écologiques avec des chiffres correspondants,arrangés de manière pratique et permettant le stockage facile de l’information sur des cartes ou des bandes perforées. Un des avantages évidents d‘un code écologique est d’assurer la comparabilité des observations : les observations météorologiques seront effectuées dans les mêmes conditions, les relevés floristiques et faunistiques établis par des procédés identiques. A cet égard, il pourrait donc contribuer à faciliter la compréhension mutuelle entre chercheurs de nationalités différentes désirant travailler en collaboration. T. PRITCHARD. I a m not a specialist in any aspect of subarctic research and will venture only as far as commenting on some of the conservation and administrative problems relating to subarctic ecology. It seems to m e that this symposium should not be terminated before at least two questions have been posed, namely: 1. Which natural resources do the scientists require to be conserved in order to satisfy their requirements for field research in the forseeable future, where should they be located and roughly what is the total area of land and water involved? 359
General discussion and conclusions/Discussiongénérale et conclusions
2. H o w is it going to be possible to obtain security for such resources, h o w are they to be managed and w h o is going to manage them? S o m e might argue that there is yet no pressing urgency to deal with the conservation of land, water and wild-life for scientific purposes because huge areas of the Subarctic are relatively free from threat. But is this true? If so, h o w long is the situation going to remain as it is now? In so far as some parts of the Subarctic are concerned, spectacular changes in the environment are already taking place by the application of modern technology and other forms of h u m a n interference and demands for more intense and extensive utilization of natural resources for a variety of purposes are mounting. For instance, more and more water is required for industrial purposes, more timber is being extracted, more grazing land is needed, and finally above all, the Subarctic is becoming a recreation ground for larger numbers of tourists than ever before. In an age of increasing affluence and shortening distances, the socio-economic significance of the Subarctic as an area for outdoor recreation is becoming a major factor in conservation. People in the densely populated urban areas of central and western Europe, with increasing wealth, mobility and leisure time, are seeking more sophisticated experience in life as a means of relaxation and are placing a high premium on the aesthetic, cultural and recreational values of wilderness. This demand can and is being met in such places as national parks but conservation of landscape for recreational purposes can in itself be in conflict with the needs of research workers unless appropriate attention is paid at an early stage to methods of integrating the management of natural resources to meet both types of requiiement. In the light of current trends in the pattern of demand for natural resources, it is not too early for scientific workers to survey, assess, designate and state their requirements. If the Subarctic is to function fully as a natural environment for scientific research and education, then the following needs must be met. Nature reserves for research
A circumpolar network of two types of reserves should be prepared by a team of leading experts in the major scientific disciplines drawn from all the countries involved. This network should include two major classes of reserves-outdoor museums, which are the best examples of major types of environment to be strictly protected for reference purposes, and research and experimental reserves to provide suitable working ground for scientists. The series of reserves in such a network must be designated on the basis of extensive surveys and assessment, and the degree of international co-operation amongst scientists will be a decisive factor governing the comprehensiveness and completeness of the plan. Such a network of
360
reserves cannot be secured on a haphazard basis; it is essential to have a far-sighted outlook and an aggressive operational programme. Nature reserves for education Execution and training in the environmental sciences are becoming more international in scope. The value of cool temperate environments for studies by students from other parts of the globe has already been discussed at the meeting, mainly by Dr. Pruitt. M a y I suggest that it is perhaps timely to put forward the idea of an annotated list of potential educational resources which m a y be used in teaching, for demonstration and experimental studies. Such reserves, once designated, should be accompanied by descriptive literature and other interpretative services, and equipped for educational purposes by means of routes, nature trails and other devices to enable students to obtain the m a x i m u m possible benefit from the use of such areas. I a m sometimes horrified by tales of junior overseas expeditions fumbling ineffectively as a result of inadequate or total absence of suitable guidance either from their o w n leaders or from scientists resident in the countries they visit. Building up a list of cool temperate educational sites should not prove too arduous a task and would, I a m quite sure, lead to a substantial raising of teaching standards and m u c h easier exchange between countries of more accurate information of educational value. Research and education centres
Like the site conservation problem, the development of field stations should surely be tackled, at planning stages, on an international level. This symposium has shown the value of co-operationbetween scientists in different parts of the Subarctic. It has also shown the need to make the best possible use of scarce scientific manpower and supporting staff as well as of expensive scientific equipment. Moreover, the need for top class multidisciplinary teams of environmental scientists is also obvious, to cover hydrology, the earth sciences and meteorology, to mention only a few of the disciplines involved, and to offer scope for specialist studies without perpetuating the spread of compartmentalism and isolationism which could so easily become a characteristic of some specialist work, especially when it is undertaken in remote areas. Needless to say, there is tremendous scope for enlarging scientific manpower and capital investment in the Subarctic and it seems to m e that this could be facilitated if a comprehensive long-term programme for ecologicalresearch could be prepared and the cost of implementing it estimated. Is this symposium not an excellent platform for promoting joint action by countries in the Subarctic towards long-termplanning, co-ordination of effort, pooling of resources and practical co-operation?The interest currently shown
General discussion and conclusions/Discussiongénérale et conclusions
by international agencies in promoting environmental research,especially by Unesco,IBP and I U C N ,
2. Etude du fonctionnement des écosystèmes
provides a good background of opinion for doing something along these lines. I will conclude, wearing my conservation hat, by saying that vigorous action by scientists and conservationists is n o w urgently required in subarctic regions to ensure that existing and projected natural laboratories are not damaged or destroyed. I have already referred to Project Peatland in my paper. That project, however successful it turns out to be, will after all cover only one type of environment. Similar action will have to be taken to deal with a range of other habitats before one can begin to feel reasonably satisfied that environmental conservation is adequately taken care of.
mesure de la variation de biomasse végétale en fonction du temps, avec des déterminationsdes échanges gazeux au niveau des divers organes végétaux et, dans certains cas, au niveau de la basse atmosphère. (b) fitudier le comportement de l’écosystème en fonction de modifications provoquées artificiellement de la structure de la biomasse (coupes, traitementschimiques)et du milieu physique des organismes (arrosages, installations d’enceintes climatisées, etc.). (e) Utiliser des capteurs aéroportés en vue de compléter et généraliser les résultats de mesures éco-physiologiqueseffectuées au sol. (d) Estimer la production secondaire en mettant l’accent sur la fluctuation des populations animales (e) Encourager des études pédologiques tendant à élucider le fonctionnementdes écosystèmes. 3. $tude des caractères d’adaptation des organismes
W.O. PRUITT. I would like to add some comments Dr. Pritchard’s intervention in the discussion. 1. There should be more research on productivity and
to
sustained yield of subarctic fauna and flora, i.e. energy flow. 2. There is a need for establishment of protected “control areas”; not recreational parks but areas for standards against which management programmes can be evaluated. 3. There is a great need for a better understanding of cyclic fluctuations of subarctic mammals and birds. This is the prime problem of boreal terrestrial ecology. 4.There is an immediate need for international treaties protecting migratory birds, whales, seals, etc. This need is critical. 5. There should be closer co-operation and exchange between scientists investigating the Subarctic.
F. E.ECKARDT. L e professeur Kallio, dans son discours de clôture, résumera les résultats essentiels qui se dégagent de l’ensemble du colloque et formulera des recommandations quant au mode d’organisation future de la recherche dans la région subarctique. Qu’il m e soit donc permis ici de rappeler seulement les recommandations essentielles évoquées lors de la discussion générale. Ces recommandations, pouvant être groupées sous cinq rubriques,étaient les suivantes: 1. Description du terrain et du climat (a) Entreprendre, essentiellement au moyen de plates-formes aéroportées, une prospection de l’ensemblede la régionarctiqueet subarctiqueen vue de délimiter des surfaces d’expérimentation appropriées. (b) Installer dans ces surfaces des stationsmétéorologiqueséquipées de dispositifs perm ettant l’enregistrement automatique de données sous une forme facilement exploitable par ordinateur. (c) ,[Améliorer le réseau d’étude de la neige.
(a) Evaluer la production primaire en combinant la
vivants (a) fitablir des jardins de transplantation et des
jardins phénologiques.
(b) Construire des phytotrons fixes et mobiles. 4. Conservation des écosystèmes (a) Faire précéder toute tentative de mise en valeur
agronomique et industrielle des régions arctiques et subarctiques par un examen de ses conséquences sur l’équilibre des structures sociales et sur l’équilibre de la biosphère tout entière. (b) Protéger ou rétablir des écosystèmes susceptibles de servir de refuge aux organismes vivants menacés d’extinction. Préserver des écosystèmes intacts pour permet(c) tre,dansle futur,des études de la structureet du fonctionnementde la biosphère. 5. Organisation de la recherche (a) Encourager la collaboration avec des statisticienslors de l’établissementde programmes de recherches. (b) lhablir un inventaire codifiable de paramètres biologiques et physiques à étudier. (c) Standardiser,ou étalonnerréciproquement,certainestechniqueset méthodes de base à condition que cette mesure n’entrave pas l’activité créatrice des chercheurs. (d) Faciliter la collaboration interdisciplinaire, nationale et internationale,dans le domaine de la recherche arctique et subarctique par la création de laboratoires appropriés. A ces recommandations il convient d’ajouter celles tendant à encourager le développement de nouvelles méthodes plus spécialement adaptées à la recherche écologique dans les régions polaires et subpolaires.
361
/.
General discussion and conclusions/Discussiongénérale et conclusions
General conclusions P.Kallio The special problems raised during this symposium on subarctic ecology have been grouped under ten different topics based on the particular climatic conditions prevailing in the area concerned. The dynamics of geomorphological features and soil formations,the main characteristicsof vegetation and animal life, reflect the particular energy relations prevailing in the Subarctic. Although the definition of the concept “subarctic” is to a great extent subjective, some criteria have been outlined for practical reasons.The northern tree line appears to be a suitable boundary of the region, which thus includes the forest tundra. The subarctic belt is well characterized by the effects of “macroclimate” even under winter conditions, is easily distinguishable in the landscape and m a y be mapped by aerial photography. In m a n y respects this circumpolar subarctic zone is rather homogeneous. Local features, however, m a y vary from one point to the other owing to absence of continuity of temperature and light climates, differences in the bed-rock and in geomorphology and different effects of the ocean. Observations made in different parts of the zone should therefore be compared more intensively than before. During the lectures and discussions, the following views have been emphasized: in order to reach a better understanding and classification of the relations between climatic boundaries, distribution of geomorphic forms, and frozen-ground features, as well as of plants and animals considered individually and in communities, the net of research stations in the Subarctic should be made denser. It is highly desirable to have accurate meteorological observations,similarly planned, from all these stations,for periods of several years, because of the great annual variations in the Subarctic. Particular attention should be paid to microclimatic measurements including measurements of soil temperature and temperature under the snow cover. Standardization of measurements as well as of instruments and equipment for routine work would make results more easily comparable. In all stations, inventory and mapping of the same
362
features and the same basic phenomena should be performed accurately. At the same time, m a n y parameters e.g. soil formations, geological and geomorphological features, distribution of permafrost and of plants and animals as well as their communities, should be observed. Basic routine material from these stations should be freely available, and in the same form, to all investigators. The two groups of data-distribution of organisms and their “environmental response”would make possible a synthetic subarctic total ecology which would be of importance also for landuse and planning. Experimental ecophysiology stressing dynamic aspects, particularly adaptability of organisms to the environment,requires well-equippedlaboratories with modern instruments and facilities for cultivating plants, e.g. for physiological, genetical, phenological and similar studies. The stations might also be of value for teaching purposes, especially for students from southern regions. Increased communication and exchange of information and of research workers,including young students, is necessary. Alist,kept up to date,containinginformation about subarctic stations in different countries, available housing, laboratory facilities and published work, is needed. One step in a large-scale planning of subarctic research would consist in the establishment of large protected areas-control areas- for long-termstudies. Because subarctic animals often migrate far to the south, there is a strong need for international agreements to ensure their protection. At this stage of social and economic development,ii appears that there are very few, if any, clearly profitable ways of exploiting the subarctic regions. As, however, new demands, and means of satisfying them, m a y arise, it would be advisable to maintain an attitude of great reserve towards current plans for reclamation of these areas and to concentrate n o w on research which m a y help to lay a sound foundation for future activities of m a n in the Subarctic.
’
List of participants / Liste des participants
AARIO, L.Department of Geography,University of Helsinki, Helsinki (Finland). AHLMANN, Hans. Stockholm University, Geografiska Institutionen, Drottningsgatan 120, Stockholm (Sweden). AHTI, Teuvo. Botanical Museum, University of Helsinki, Unioninkatu 44, Helsinki 17 (Finland). AUBERTDE LA RUE, E. 2, Chemin des Combes, 1009 PullyLausanne (Switzerland). HAMET-AHTI, Leena. Botanical Museum, University of Helsinki, Unioninkatu 44, Helsinki 17 (Finland). BELL, Frances (Miss). Botany School, Cambridge (United Kingdom). BELLAMY, D.J. Department of Botany, Durham University, Durham City (United Kingdom). B L ~ T H G E N , Joachim. Professor, University of Münster, Goebenstr. 8, 44 Münster (Federal Republic of Germany). BRITTON, M a x E. Code 414, Office of Naval Research, Washington, D.C. 20360 (United States of America). BRONNY, Horst. Geographisches Institut der Ruhr-Universität Bochum, Roomerscheide 71, 463 Bochum (Federal Republic of Germany). BROWN, R.J. E. National Research Council, Division of Building Research, Ottawa, Ontario (Canada). BÜDEL, Julius. Professor, Geographisches Institut der Universität Würzburg, Klinikstr. 3, 87 Würzburg (Federal Republic of Germany). CORTE, A. Instituto Antártico Argentino, Buenos Aires (Argentina). DI CASTRI, Francesco. Professor, Universidad de Chile, Casilla 5681, Santiago de Chile (Chile). DOLGIN, Isaak Markovich. The Arctic and Antarctic Research Institute, Fontanka 34, Leningrad D. 104 (U. S.S.R.). ECRARDT, F. E. Institut botanique et CNRS de Montpellier, 34 Montpellier (France). EHLERS, Eckart. Department of Geography, University of Tübingen, 74 Tübingen (Federal Republic of Germany). EINARSSON, Eythor. Director, Department of Botany, Museum of Natural History, P.O. B o x 532, Reykjavik (Iceland). EVTEEV, S. Natural Resources Research Division, Depart-
ment for the Advancement of Science, Unesco Secretariat, Paris-7e (France). FLOWER-ELLIS, Jeremy. Department of Ecology, Royal College of Forestry, Stockholm 50 (Sweden). FORMOZOV, Alexandr. Institute of Geography, Academy of Science, Stazomonetny pez 29, Moscow v-17 (U.S.S.R.). HAVAS, Paavo. Assistant Professor, Institute of Geography, University of Oulu, Torikatu 15, Oulu (Finland). HULTEN, Eric. Professor, Riksmuseurn, Stockholm 50 (Sweden). HUOVILA, Seppo. Finnish Meteorological Office, Box 10503, Helsinki 10 (Finland). HUSTICH, Ilmari. Professor, Hollantilaisentie 1, Helsinki 33 (Finland). JAHN, Alfred. Professor, University of Wroclaw, P1. Universytecki 1, Wroclaw (Poland). JOHNSON, Philip. Cold Regions Research and Engineering Laboratory, Box 282, Hanover, New Hampshire (United States of America). JOHNSTON, G. H. National Research Council, Division of Building Research, Ottawa, Ontario (Canada). JOURNAUX, André. Professeur, Université de Caen, 19, rue Isidore-Pierre,14 Caen (France). KALELA, Olavi. Professor, Department of Zoology, University of Helsinki, P. Rautatiek 13, Helsinki 10 (Finland). KALLIO, Paavo. Professor, The Sub-arctic Research Station of University of Turku, Turku (Finland). KALLIOLA, R. Unioninkatu 40 A, Helsinki (Finland). KELSALL, J. P. Canadian Wildlife Service, Edmonton, Alberta (Canada). LENT, Peter. University of Basutoland, R o m a , Basutoland. LOUGHREY, Alan George. Canadian Wildlife Service, Eastern Region, 293 Albert Street, Ottawa 4, Ontario (Canada). LUTHER, Hans. International Union for Conservation of Nature and Natural Resources, Djurgårdsvillan 8, Helsinki 53 (Finland). MALAURIE, Jean. Directeur, Centre d'études arctiques et finno-scandinaves, Ecole pratique des hautes études (VIe Section), Sorbonne, Paris (France). MELA, Pirkko.Ministry of Education,Department for Foreign Affairs, Korkeavuorenkatu 21, Helsinki 13 (Finland).
363
List of participants/Liste des participants
MIKOLA, Peitsa, Professor, University of Helsinki, Department of Silviculture, Unioninkatu 40 B, Helsinki 17 (Finland). MYSTERUD, Ivar. Zoological Laboratory, University of Oslo, Postbox 1050,Blindern, Oslo 3 (Norway). ODIN,Hans. Meteorologist, Skoghögskolan, Stockholm 50 (Sweden). OFILSON, Birger. Department of Geography, University of Turku, Turku (Finland). OSTBYE, Eivind. Zoological Laboratory, University of Oslo, Postbox 1050, Oslo (Norway). PEIPONEN, Valto. Department of Zoology, University of Helsinki (Finland). PÉwÉ, Troy L. Department of Geology, Arizona State University, Tempe, Arizona (United States of America). PERTTU, K . Skoghögskolan, Stockholm 50 (Sweden). PICARD, Alice. Professeur, Collège universitaire de Brest, 26, rue Voltaire, 29 N Brest (France). PICARD,Elie. Professeur, Collège universitaire de Brest, 26, rue Voltaire, 29 N Brest (France). POLUNIN, N.Professor at the University of Ibadan (Nigeria). PRITCHARD, T. International Union for Conservation of Nature and Natural Resources, Morges (Switzerland). PRUITT, William. Department of Biology, Memorial University of Newfoundland, St. John’s, Newfoundland (Canada). RAPP,Anders. Department of Geography, University of Uppsala, Uppsala (Sweden). REED,John. Arctic Institute of North America, 1619 N e w Hampshire Avenue N.W., Washington, D.C. 20009 (United States of America).
364
RUUHIJARVI, Rauno. Assistant professor, Department of Botanv, University of Helsinki, Helsinki (Finland). SALMI,Martti. Geological Survey of Finland, Bulevardi 9 A 22, Helsinki 12 (Finland). SARVAS, Risto. Forest Research Institute, Unioninkatu 40 A, Helsinki 17 (Finland). SCHENK,Erwin. Geologische Forschungsstelle,Niddaer Str.2, 6303 Hungen, Hessen (Federal Republic of Germany). SIMMONS, Ian. University of Durham, Science Laboratories, South Road, Durham Citv (United Kingdom). SIREN,Gustaf. Professor, Royal College of Forestry, Stockholm 50 (Sweden). SORSA, Kalevi, Secretary-General,Finnish National Commission for Unesco, Korkeavuorenkatu 21, Helsinki 13r (Finland). S~YRINKI, N. Professor, Rector of the University of Oulu (Finland). SVENSSON,Harald. Department of Geography, Sölvegatan 13, Lund (Sweden). TAMM,Carl Olof. Professor, Department of Ecology, Royal College of Forestry, Stockholm 50 (Sweden). TEDROW, John. Professor, Rutgers University, N e w Brunswick, N e w Jersev (United States of America). TICKLE, W.M. (Miss). Department of Botany, Durham University, Durham City (United Kingdom). TIKHOMIROV, Boris. Professor, Komarov Botanical Institute Academy of Sciences, Leningrad (U.S.S.R.). VIERECK, Eleanor. Forestry Sciences Laboratory, B o x 430, College, Alaska (United States of America). VIERECK,Leslie A., Forestry Sciences Laboratory, Box 430, College, Alaska (United States of America).