The Comparative Roles of SuspensionFeeders in Ecosystems
NATO Science Series A Series presenting the results of scientific meetings supported under the NATO Science Programme. The Series is published by IOS Press, Amsterdam, and Springer (formerly Kluwer Academic Publishers) in conjunction with the NATO Public Diplomacy Division.
Sub-Series I. II. III. IV.
Life and Behavioural Sciences Mathematics, Physics and Chemistry Computer and Systems Science Earth and Environmental Sciences
IOS Press Springer (formerly Kluwer Academic Publishers) IOS Press Springer (formerly Kluwer Academic Publishers)
The NATO Science Series continues the series of books published formerly as the NATO ASI Series. The NATO Science Programme offers support for collaboration in civil science between scientists of countries of the Euro-Atlantic Partnership Council. The types of scientific meeting generally supported are “Advanced Study Institutes” and “Advanced Research Workshops”, and the NATO Science Series collects together the results of these meetings. The meetings are co-organized by scientists from , NATO countries and scientists from NATO s Partner countries – countries of the CIS and Central and Eastern Europe.
Advanced Study Institutes are high-level tutorial courses offering in-depth study of latest advances in a field. Advanced Research Workshops are expert meetings aimed at critical assessment of a field, and identification of directions for future action. As a consequence of the restructuring of the NATO Science Programme in 1999, the NATO Science Series was re-organized to the four sub-series noted above. Please consult the following web sites for information on previous volumes published in the Series. http://www.nato.int/science http://www.springeronline.com http://www.iospress.nl
Series IV: Earth and Environmental Series – Vol. 47
The Comparative Roles of Suspension-Feeders in Ecosystems edited by
Richard F. Dame Marine Science Department, Coastal Carolina University, Conway, SC, U.S.A. and
Sergej Olenin Coastal Research and Planning Institute, Klaipeda University, Klaipeda, Lithuania
Proceedings of the NATO Advanced Research Workshop on The Comparative Roles of Suspension-Feeders in Ecosystems Nida, Lithuania 4–9 October 2003
A C.I.P. P Catalogue record for this book is available from the Library of Congress.
ISBN-10 1-4020-3029-0 (PB) Springer Dordrecht, Berlin, Heidelberg, New York ISBN-13 978-1-4020-3029-1 (PB) Springer Dordrecht, Berlin, Heidelberg, New York ISBN-10 1-4020-3028-2 (HB) Springer Dordrecht, Berlin, Heidelberg, New York ISBN-10 1-4020-3030-4 (e-book) Springer Dordrecht, Berlin, Heidelberg, New York ISBN-13 978-1-4020-3028-4 (HB) Springer Dordrecht, Berlin, Heidelberg, New York ISBN-13 978-1-4020-3030-7 (e-book) Springer Dordrecht, Berlin, Heidelberg, New York
Published by Springer, P.O. Box 17, 3300 AA Dordrecht, The Netherlands.
Printed on acid-free paper
All Rights Reserved © 2005 Springer No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed in the Netherlands.
Table of Contents List of Contributors…………...………………………………………………………...…..…vii Preface…………………………………………………………………………………...…..… xi 1. Modelling particle selection efficiency of bivalve suspension feeders P Zemlys and D Daunys……………………………………………………….……...…..….1 2. Field measurements on the variability in biodeposition and estimates of grazing pressure of suspension-feeding bivalves in the northern Baltic Sea J Kotta, H Orav-Kotta and I Vuorinen…………………………………………………...….11 3. Can bivalve suspension-feeders affect pelagic food web structure? T Prins and V Escaravage……………………………………………………….…...…….31 4. Motile suspension-feeders in estuarine and marine ecosystems D Bushek and D M Allen………………..………………………………………...………..53 5. Impact of suspension-feeding nekton in freshwater ecosystems: patterns and mechanisms H Ojaveer………………………………………………………………..………….……….73 6. Influence of eastern oysters on nitrogen and phosphorus regeneration in Chesapeake Bay, USA R I E Newell, R R Holyoke and J C Cornwell…….…………………….…………………..93 7. How does estimation of environmental carrying capacity for bivalve culture depend upon spatial and temporal scales? P Duarte, A J S Hawkins and A Pereira………..……………………….…………..……..121 8. Impact of increased mineral particle concentration on behavior, suspension feeding and reproduction of Acartia clausii (Copepoda) N Shadrin and L Litvinchuk………………………………………………………...….…137 9. Suspension-feeders as factors influencing water quality in aquatic ecosystems S A Ostroumov…………………………………………………………………...…….….147 10. Neoplasia in estuarine bivalves: effect of feeding behaviour and pollution in the Gulf of Gdansk (Baltic Sea, Poland) M Wolowicz, K Smolarz and A Sokolowski……………………………….…...…….….165 11. Bivalves as biofilters and valuable by-products in land-based aquaculture systems M Shpigel……………………………………………………………………..…………..183 12. Significance of suspension-feeder systems on different spatial scales H Asmus and R M Asmus………………………………………………………...……...199 13. Invaders in suspension-feeding systems: variations along the regional environmental gradient and similarities between large basins S Olenin and D Daunys…………………………………………………………………..221
v
vi
14. Contrasting distribution and impacts of two freshwater exotic suspension feeders, Dreissena polymorpha and Corbicula fluminea A Y Karatayev, L E Burlakova and D K Padilla................................................................239 15. Functional changes in benthic freshwater communities after Dreissena polymorpha (Pallas) invasion and consequences for filtration L E Burlakova, A Y Karatayev and D K Padilla….……………………….………...…..263 16. Does the introduction of the Pacific oyster Crassostrea gigas lead to species shifts in the Wadden Sea? A Smaal, M van Stralen and J Craeymeersch………………………………………...….277 17. One estuary, one invasion, two responses: phytoplankton and benthic community dynamics determine the effect of an estuarine invasive suspension-feeder J K Thompson…………………………………..………………………………………...291 18. Development of human impact on suspension-feeding bivalves in coastal soft-bottom ecosystems W J Wolff………………………………………………………………………………...317 19. Oyster reefs as complex ecological systems R Dame…………………………………………………………………..…………….....331 20. Synthesis/Conclusions………………………………………………………….………..345 Index…………………………………………………………….…………………………….355
List of Contributors (Mailing Addressses) Drs. Harald and Ragnhild Asmus Alfred-Wegener-Institut für Polar- und Meeresforschung Wattenmeerstation Sylt Hafenstraße 43 25992 List/Sylt Germany
Dr. David Bushek Haskin Shellfish Research Lab Rutgers University Port Norris, NJ 08349 United States
Dr. Lyubov Burlakova Department of Biology Stephan F. Austin State University SFA Station Nacogdoches, Texas 75962 United States
Dr. Richard Dame (Co-Director NATO) Coastal Carolina University P.O. Box 261954 Conway, SC 29528 United States
Dr. Darius Daunys Coastal Research and Planning Institute Klaipeda University, Manto 84 LT-5805 Klaipeda Lithuania
Dr. Pedro Duarte Universidade Fernando Pessoa Praça 9 de Abril, 349 4200 Porto Portugal
Dr. Alexander Y. Karatayev Department of Biology vii
viii
Stephan F. Austin State University SFA Station Nacogdoches, Texas 75962 United States
Dr. Jonne Kotta Estonian Marine Institute Marja 4d 10617 Tallinn Estonia
Dr. Roger Newell University of Maryland Center for Environmental Studies Horn Point Laboratory PO Box 775, Cambridge, MD 21613 United States
Dr. Sergej Olenin (Co-Director NATO-Partner) Coastal Research and Planning Institute Klaipeda University, Manto 84 LT-5805 Klaipeda Lithuania
Dr. Henn Ojaveer Estonian Marine Institute Maealuse 10a 12618 Tallinn Estonia
Dr. Sergei A. Ostroumov Department of Hydrobiology, Faculty of Biology, Moscow State University Moscow 119899, Russia
Dr. Theo Prins National Institute for Coastal and Marine Management/RIKZ POBOX 8039 4330 EA Middelburg The Netherlands
ix Dr. Nikolay Shadrin Institute of Biology of the Southern Seas 2, Nakhimov Ave. Sevastopol, 99011 Ukraine.
Dr. Muki Shpigel Israel Oceanographic and Limnological Research; National Centre for Mariculture PO Box 1212 88112 Eilat Israel
Dr. Aad Smaal Shellfish Research Centre, RIVO-DLO Korringaweg 5, P.O. Box 77 4400 AB Yerseke The Netherlands
Dr. Jan Thompson U.S. Geological Survey MS496 345 Middlefield Rd. Menlo Park, CA 94025 United States
Dr. Wim J. Wolff Dept. of Marine Biology Groningen University P.O. Box 14 9750 AA Haren The Netherlands
Dr Maciej Wolowicz Laboratory of Estuarine Ecology, University of Gdansk Al. Pilsudskiego 46 81 378 Gdynia Poland
x Dr. Petras Zemlys Coastal Research and Planning Institute Klaipeda University, Manto 84 LT-5805 Klaipeda Lithuania
PREFACE Animals are a major link between the water column (pelagic) and the bottom (benthic) habitats in most shallow systems. This coupling is dominated by active processes such as suspension-feeding in which the organism actively uses energy to pump water that is then filtered to remove suspended particles that are consumed while undigested remains are deposited on the bottom. As a result of this feeding on and metabolism of particles, the animals excrete dissolved inorganic and organic waste back into the water column, and thus, become major components in the cycling and feedback of essential elements. With relatively high weight specific filtration rates of 1– 10 liters/hour/gram dry tissue and a propensity to form large aggregated populations (beds, reefs, schools and swarms), these organisms can play an important role in regulating water column processes Although estuarine bivalve molluscs such as oysters and mussels dominate the suspension-feeder literature, other groups including plankton and nekton that are found in estuarine as well as other aquatic systems are also potentially important removers of suspended particles. Thus, a significant part of the NATO Advanced Research Workshop focused on suspension-feeders as controllers of plankton abundance, biomass and diversity, system metabolism, nutrient cycling and scale dependency. Systems dominated by suspension-feeders are typically impacted by human activities including recreation, aquaculture, human and industrial pollution, and bilge water from shipping. Suspension-feeders are often impacted by fisheries and over-exploitation. These impacts commonly result in changes in ecosystem structure either through the food chain concentration of harmful substances or diseases, the introduction of alien species of suspension-feeders, or the instability of suspension-feeders systems through species displacement or phase shifts in the dominance between different suspension-feeding components such as nekton or zooplankton. These issues were addressed near the close of the workshop along with conclusions and syntheses developed by the working groups. In the almost 10 years since one of us (RFD) led a NATO ARW in The Netherlands on bivalve filter feeders, interest in suspension-feeders as major influences on aquatic ecosystem processes has grown dramatically. This development is particularly evident in freshwater systems, yet the communications between the freshwater and the estuarine-marine scientific communities are weak (probably because of scientific, societal and funding agency structure and habitat separation). Thus, one of our major goals was to balance process orientated topics with presentations from the three general aquatic environments, freshwater, estuarine and marine. An additional overarching aim was to bridge the geographical distribution of NATO and partner countries. Our workshop proposal is timely and compliments
xi
xii
NATO’s new approach involving partner countries because many partner countries have mainly freshwater and brackish systems while most NATO members also have large estuarine-marine components. In addition to chronicling the current status of suspension-feeder research, we believe that this workshop has and will foster greater communications between the various groups and support the cross-fertilisation strategy has been shown to have a strong positive effect on the generation of new scientific approaches, theories and knowledge. The participants are grateful for the financial and logistics support and guidance provided by the NATO staff. We also thank the Kluwer editorial staff for their timely and constructive support.
MODELLING PARTICLE SELECTION EFFICIENCY OF BIVALVE SUSPENSION FEEDERS
Petras Zemlys, Darius Daunys Coastal Research and Planning Institute, Klaipeda University, H.Manto 84, Klaipeda 5808, Lithuania
Abstract: The choice of an appropriate index to adequately describe the efficiency of preingestive organic material selection is important for modelling the material flux within the suspension feeding process. Recently, a new selection efficiency index was suggested by Zemlys et al. (2003) which simplifies the quantification of the selection activity. A simple equation with interpretable parameters calculates the selection efficiency index using literature values of uptake rate and food quality. This analysis suggests the possibility of developing more general and biologically interpretable models. Keywords: Pre-ingestive selection efficiency; suspension feeding bivalves; modelling food processing
INTRODUCTION Bivalve suspension feeders reject part of the food they filter as pseudofeces depending on seston concentration. An important process accompanying pseudofeces production is the selection of particles that result in an increase in the organic material fraction of the ingested food (Fig. 1). Although the physiological regulation of feeding and selection of particles in particular is rejected by some authors (Jørgensen 1996), a number of in vitro and in situ investigations (Kiørboe and Mølenberg, 1981; Hawkins et al. 1996; Defossez and Hawkins 1997; Ward et al. 1998; Hawkins et al. 1998; Schneider et al. 1998; Baker et al. 2000) confirm the selective feeding by bivalves. The selection efficiency of organic matter can be as high as 60% under certain conditions (Hawkins et al. 1998). Therefore, this efficiency may considerably change the organic and inorganic material ratio of ingested food and biodeposits. Although the pre-ingestive selectivity of particles by bivalves is generally recognised, the factors influencing preferential ingestion remain 1 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 1–9. ©2005 Springer. Printed in the Netherlands.
2 uncertain. Various proposed criteria include particle size, shape, motility, density, and chemical cues such as algal ectotrines (for review see Defossez and Hawkins 1997). The organic material selection activity defined as selection efficiency, however, can be quantitatively described by experimental relationships that already have been determined for some species of marine bivalves (Hawkins et al. 1996, 1998). These findings strongly suggest that particle selectivity should be included in models of bivalve feeding.
Food uptake by filtration
Preingestive selection, rejection
Suspended particles
Increasing the fraction of low quality particles
Pseudofaeces
Decreasing the fraction of low quality particles Ingestion
Digestion
Fig. 1. A conceptual model of food processing by suspension feeders exhibiting active particle selection.
When selection activity is absent the quantitative description of the allocation of seston fractions (organic, inorganic material, etc.) in pseudofeces and ingested food is easily expressed by the difference between food uptake rate and pseudofeces production or ingestion rate. In the case of selection activity there are two options. One option is to model the rejection (ingestion) rates of different seston fractions separately (e.g. Hawkins et al. 2002; Scholten and Smaal 1999). In this case, the selection efficiency index serves an auxiliary role and can be used as output variable characterising the selection activity. Alternatively, the selection efficiency index can be used to determine by mass balance the different seston fractions beforehand and to estimate the allocation of seston fractions in pseudofeces and ingested food (e.g. Bendell-Young and Arifin 2004). Problems with the last approach arise when the seston is fractionated into organic and inorganic material (Zemlys et al. 2003). They found that the choice of an appropriate selection efficiency index is critical. The most widely used selection efficiency index is that defined in Bayne et al. (1993) and based on the comparison of organic content
3 in pseudofeces and seston. However, Zemlys et al. (2003) demonstrated that using a different selection efficiency index based on comparison of organic content in seston and ingested food could result in series of advantages for food processing modelling. These advantages can be summarised as follows (see Zemlys et al. (2003) for details): x The traditional selection efficiency estimate is generally applicable to obtain organic content in pseudofeces and ingested food, however in some cases it may lead to analytically unsolvable equations. The introduction of an alternative selection efficiency considerably simplifies the solution and never requires iterative methods; x The analysis of constraints originating from mass balance determinations revealed an advantage of employing the alternative selection efficiency estimate. These constraints are more straightforward and obvious for alternative selection efficiency, for example the traditional selection efficiency is limited by certain value that is less than one while alternative index is limited by number one only (see formula (4) in the text below for more details); x Utilising the response surface approach for estimating the alternative selection efficiency as a function from food uptake rate and seston organic content is expected to produce a monotonously increasing function. This function might also have interpretational and analytical advantages in comparison to the traditional bell-shaped response surface. The aim of the study is to compare the response surfaces of traditional and newly defined selection efficiency index for three bivalve species Mytilus edulis, Cerastoderma edule and Crassostrea gigas and to determine an analytical expression for the alternative selection efficiency index. Data from the experimental studies of Hawkins et al. (1998) are used in this paper.
CALCULATION OF RESPONSE SURFACES Traditionally, the selection efficiency (SE) is defined as (Bayne et al., 1993) SE
(1
/
)
(1)
where FPOM M is organic content of seston (fraction particulate organic material), i.e. food quality; FPOMPF F is organic content of pseudofeces (for complete list of variables and parameters used see Table 1). An alternative
4 Table 1. List of notations with explanations. Notation a ase bse cse dse FPIM FPOM FPOM M0 FPIMING FPOMING FPOMPF IRMAX r SE SE1 SEMAX UPR
Explanation Parameter in selection efficiency equation (7) Parameter in equation (3) Parameter in selection efficiency equation (3) Parameter in selection efficiency equation (3) Parameter in selection efficiency equation (3) Inorganic content of seston Organic content of seston (fraction) The value of seston organic content at which the selection activity starts Inorganic content of ingested material Organic content of ingested material Organic content of pseudofeces Maximal ingestion rate, g day-1 Coefficient of proportionality in equation (6) Traditional selection efficiency Alternative selection efficiency Maximal possible selection efficiency value Seston uptake rate, g day-1
definition of selection efficiency based on organic content of ingested food (Zemlys et al. 2003) was defined as SE1 (
) /(1
)
(2)
where SE1 is alternative selection efficiency; FPOMING G is organic content of ingested food. SE1 is similar to selection index reported in Hawkins et al. (1998) but it is normalized by maximal value of difference FPOMING FPOM instead of FPOMING. The equation (2) can be expressed in form similar to (1) but in terms of inorganic content: SE1 1 FPIM / FPIMING
where FPIM is inorganic content in seston; FPIMING is inorganic content in ingested material. The advantages of the alternative index were demonstrated (Zemlys et al. 2003) however equations for evaluation of selection efficiency as a function of food uptake rate and seston organic content exist only for the traditionally defined selection efficiency (Hawkins et al. 1996; Hawkins et al. 1998). The limitations of these empirical equations due to complex response surfaces were also determined by Zemlys et al. (2003). They hypothesized that the response surface shape of newly defined selection efficiency should be considerably simpler. As the first step the response surfaces for both selection efficiency definitions should be constructed. Traditional selection efficiency depends on M). For three food uptake rate (UPR) and organic content of seston ((FPOM
5 bivalve species M. edulis, C. edule and C. gigas the following regression equation was proposed by Hawkins et al. (1998): SE
ase bse / FPOM
cse UPR dse UPR / FPOM
(3)
where ase, bse, cse and dse are parameters (see Hawkins et al. (1998) for numeric values). Unfortunately this equation is based on the rather narrow range of organic content and does not contain any theoretical information about possible shape of response surface outside this range. The analysis of material balance revealed (Zemlys et al., 2003) that traditionally defined selection efficiency is not increasing with UPR and FPOM M but is limited by certain value (SEMAX) X defined by formula (Zemlys et al., 2003) SEMAX
[(1
)/
]
/(
)
(4)
where UPR is seston uptake rate by bivalve; IRMAX X is maximal ingestion rate. SEMAX X is the value of SE E which corresponds to the saturation condition FPOMING 1 (Zemlys et al., 2003). The response surface for SE was constructed assuming SE min i ^SEMAX , SET ` , where SET T are selection efficiency index calculated by equation (3). The empirical equations for the calculation of UPR and IRMAX X that depend on organic content and seston concentration are taken from Hawkins et al. (1998). Does the saturation take place in real conditions or not the question is still open and needs further experimental investigation. However, the shape of SEMAX X surface gives enough information about possible shape of SE E surface because SE E is always less or equal to SEMAX X (Zemlys et al. 2003). The obtained SE surfaces for all three species are shown in Fig. 2a, 2c, 2e). The response surface for SE1 can be easy obtained from SE E surface using the following relationship between SE1 and SE E (Zemlys et al., 2003): SE1 [
/(( /(1
))]
/
1 SE
(5)
The results of these calculations are presented in Fig. 2b, 2d, 2f. For all three species the response surface of SE E is asymmetric bellshaped surface with the maximum along FPOM M axis which makes it difficult to approximate by simple interpretable analytical expression. In contrast, the SE1 surface is an increasing function with regard to both arguments monotonically approaching the value 1 and therefore might be approximated by traditional saturation functions, like Michaelis-Menten function or similar. Of course, the numerical analysis of three species only support the hypothesis but does not let to conclude that SE1 will be increasing function for other bivalve species, however as it is shown by Zemlys et al. (2003) the sufficient
6
Fig. 2.The SE E and SE1 response surfaces for three bivalve species: a), b) – Mytilus edulis; c), d) – Cerastoderma edule e), f) – Crassostrea gigas.
condition to be SE1 monotonically increasing is the monotonic increasing of relative pseudofeces production. It seems that this condition is realistic for majority of suspension feeding bivalve species at least regard to the uptake rate.
APPROXIMATION FUNCTION The description of SE1 directly as a function of UPR and FPOM is very important in order to use the alternative selection efficiency for modeling of food processing. We will demonstrate here that such a function could be
7 obtained using simple assumptions. The simplest assumption that seems to be acceptable for analyzing SE1 for all three species is the linearity of the relationship between SE1 changing rate regard to FPOM M and SE1 that can be formulated by following equation: wSE1 wF FPOM F
r((
) (1
(6)
)
where r(UPR) is coefficient of proportionality depending on uptake rate. Assuming also linearity for r(UPR), i.e. r (UPR ) a UPR after integration the following expression is obtained SE1 max(0 max(0, 1 exp(
(
0
)
))
(7)
0 , a is a parameter and is assumed that SE1=0 at where SE1(( 0) FPOM FPOM 0 . While SE1 and SE E equals to zero simultaneously the equation for FPOM0 can be obtained by assuming SE 0 in (3) and solving it regard to FPOM, M what results in FPOM0((
)
bse dse UPR ase cse UPR
(8)
The equation (6) has only one unknown parameter a which can be interpreted as multiplier determining the rate of SE1 approach to saturation condition (value 1). Together with (8) the equation (6) was used to approximate the surfaces given in fig. 2a, 2c, and 2g by minimum square method. The estimated values of parameter a and mean square error are given in Table 2.
Table 2. The results of approximation SE1 by equation (6) a
Mean square error
Mytilus edulis
7.89
0.034
Cerastoderma edule
10.08
0.066
Crassostrea gigas
3.59
0.159
Bivalve species
The estimated values of parameter a allows us to compare the ability of the different species to increase ingested food quality by means of selection
8 activity when seston food quality and uptake rate are increased. The highest value a 10.08 for C. edule (Table 2) shows that this species has the highest capability. Less, but still comparable ( a 7.89 ) capability has M. edulis while C. gigas is able to exploit less than half of selectivity potential estimated for other two species (Table 2).
DISCUSSION AND CONCLUSIONS The pre-ingestive food selection by bivalves is an important phenomenon determining the organic content in pseudofeces and ingested food, simultaneously controlling the energy fluxes inside the organism and between the organism and environment. In this paper we parameterize the newly proposed alternative selection efficiency index. We show that the response surface for the alternative selection efficiency index for three bivalve species has a simpler shape that is more proper for approximation by analytical expressions. An analytical expression based on simple assumptions for alternative selection efficiency index is proposed and we demonstrate that it can satisfactorily approximate the values of selection efficiency recalculated from traditionally defined selection efficiency for three bivalve species. The proposed relationship contains a parameter that enables the comparison of selection activity of different bivalve species, i.e. the capability to exploit the increase of food quality and food uptake to improve the ingested food quality. Our results lead us to believe that modeling of the allocation of organic and inorganic material in pseudofeces and in ingested food can be based on mass balance and alternative selection efficiency index that can be expressed by simple functions with interpretable parameters. This approach may make food processing models more general than models consisting of purely site and species specific regression equations. It is important to note that equation (7) could be more simplified. As can be seen from Fig. 2, the selection efficiency index depends much more on seston organic content than uptake rate, thus the dependence on uptake rate can be neglected (at least in some cases). The right side of equation (8) is a constant in this case. The traditional selection efficiency index as a function of seston organic content only was used Bendell-Young and Arifin (2004). In some cases other fractions of seston than organic and inorganic material are necessary. For example, the organic material was divided to phytoplankton and non-phytoplankton organics (Hawkins et al. 2002). The further development of the approach considered above for this case is an important task for future investigations.
9 REFERENCES Baker SM Levinton JS Ward JE 2000 Particle transport in the Zebra Mussel, Dreissena polymorpha (Pallas). Biol Bulll 199: 116-125 Bayne BL Iglesias JIP Hawkins AJS Navarro E Héral M Deslous-Paoli JM 1993 Feeding behavior of the mussel Mytilus edulis L.; responses to variations in both quantity and K 73: 813-829 organic content of seston. J Mar Biol Assoc UK Bendell-Young LI Arifin Z 2004 Application of a kinetic model to demonstrate how selective feeding could alter the amount of cadmium accumulated by the blue mussel (Mytilus ( trossolus). J Exp Mar Biol Ecol 298::21– 33 Defossez JM Hawkins AJS 1997 Selective feeding in shellfish: size-dependent rejection of large particles within pseudofaeces from Mytilus edulis, Ruditapes philippinarum and Tapes decussatus. Mar Bioll 129: 139-147 Hawkins AJS Bayne BL Bougrier S Héral M Iglesias JIP Navarro E 1998 Some general relationships in comparing the feeding physiology of suspension-feeding bivalve molluscs. J Exp Mar Biol Ecoll 219: 87-103 Hawkins AJS Duarte P Fang JG Pascoe PL Zhang JH Zhang XL Zhu MY 2002 A functional model of responsive suspension-feeding growth in bivalve shellfish, configured and validated for the scallop Chlamys farreri during culture in China. J Exp Mar Biol Ecoll 281: 13-40 Hawkins AJS Smith RFM Bayne BL Héral M 1996 Novell observations underlying the fast growth of suspension-feeding shellfish in turbid environments: Mytilus edulis. Mar Ecol Progr Ser 131: 170-190 Jørgensen CB 1996 Bivalve feeding revisited. Mar Ecol Prog Serr 142:287-302 Kiørboe T Mølenberg F 1981 Particle selection in suspension-feeding bivalves. Mar Ecol Prog Ser: 5: 291-296 Schneider DW Madon SP Stoeckel JA Sparks RE 1998 Seston quality controls zebra mussel ((Dreissena polymorpha) energetics in turbid rivers. Oecologia 117: 331-341 Scholten H Smaal AC 1999 The ecophysiological response of mussels (Mytilus edulis) in mesocosms to a range of inorganic nutrient loads: simulation with the model EMMY. Aq Ecoll 33: 83-100 Ward JE Levinton JS Shumway SE Cucci T 1998 Particle sorting in bivalves: in vivo determination of pallial organs of selection. Mar Bioll 131: 283-292 Zemlys P Daunys D Razinkovas A 2003 Revision of pre-ingestive selection efficiency definition for suspension feeding bivalves: facilitating the material fluxes modelling. Ecol Modell 166: 67-74
FIELD MEASUREMENTS ON THE VARIABILITY IN BIODEPOSITION AND ESTIMATES OF GRAZING PRESSURE OF SUSPENSION-FEEDING BIVALVES IN THE NORTHERN BALTIC SEA
Jonne Kotta1, Helen Orav-Kotta1,2 and Ilppo Vuorinen3 1
Estonian Marine Institute, University of Tartu, Mäealuse 10a, 12618 Tallinn, Estonia Institute of Zoology and Hydrobiology, University of Tartu, Vanemuise 46, 51014 Tartu, Estonia 3 Archipelago Research Institute, University of Turku, SF-20500 Turku, Finland 2
Abstract: Functional relationships between environmental variables, the biodeposition and clearance rates of Dreissena polymorpha and Mytilus edulis were estimated in the northern Baltic Sea. The biodeposition and clearance of the bivalves increased with ambient temperature. In more eutrophicated regions biodeposition and clearance rates increased curvilinearly with ambient concentrations of chlorophyll a and leveled off at high food concentrations. In less eutrophicated conditions a linear model gave the best fit suggesting that saturation level was not obtained. Additional variation in the biodeposition and clearance was explained by the interaction of water temperature, current velocity and chlorophyll a. Salinity had a significant effect on the biodeposition and clearance of D. polymorpha. The population of suspension-feeders cleared daily on average from 3 to 2426% of overlaying water in the littoral area constituting an important sink for primary production. Keywords: Baltic Sea, benthic grazing, Dreissena polymorpha, Mytilus edulis
INTRODUCTION Owing to elevated nutrient levels and consequent phytoplankton blooms a dramatic increase of dense populations of benthic suspension feeders has been recorded world-wide (Barnes and Hughes 1988; Kautsky 1995; Dame 1996). At high densities the suspension-feeders are capable to deplete phytoplankton (Cloern, 1982 Fréchette and Bourget 1985) and therefore control the standing stock and production of primary producers and limit via competition the growth of pelagic herbivores and fish (e.g. Officer et al. 1982; Møhlenberg 1995). Consequently, suspension-feeders are considered to play a key role in the stability of coastal ecosystems (Herman and Scholten 1990). 11 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 11–29. ©2005 Springer. Printed in the Netherlands.
12 In situ studies quantifying broad-scale effects of suspension-feeder populations are scarce and usually they are based on indirect evidence and modelling approaches (e.g. Cloern 1982; Møhlenberg 1995). It has been demonstrated that laboratory measurements are often difficult to interpret and compare (Riisgård, 2001) and they may overestimate the filtration rate by 1300% (Doering and Oviatt 1986; Cranford and Hill 1999). It suggests that suspension-feeders in nature exploit their full clearance capacity for short periods and more often feed at a much-reduced rate (e.g. Cranford 2001). Water temperature, salinity, the quality and concentration of seston coupled with flow regime have significant impact on the activity of suspension feeders (Bayne et al. 1977; Kiørboe et al. 1980; Widdows 1985; Fréchette at al. 1989; Asmus and Asmus 1993) and may account for the major variability of in situ feeding behaviour. Hence, there is a need for field measurements of the feeding behaviour of suspension-feeders combined with the measurement of those environmental variables in the near-bottom layer. These functional relationships have to be estimated for different areas and different times of the year, to assess the importance of suspension feeder grazing to the coastal ecosystem. Suspension-feeders derive their food by filtering the water column and retaining particulate matter on their gills. Clearance rate refers to an amount of water that is cleared per time unit by animal or biomass. Biodeposition is defined as the production of faeces and pseudofeces. An in situ biodeposition approach has been used to evaluate the variations in the feeding behaviour of mussels (Kautsky and Evans 1987; Hawkins et al. 1996; Cranford et al. 1998; Cranford and Hill 1999). By applying an in situ trap technique, the biodeposition was quantified in terms of carbon and nutrients. However, as phytoplankton is considered to be the prime food for benthic suspension-feeders, we chose, in contrast to these previous studies, chlorophyll a (Chl a) as a tracer. The present study focuses on the grazing impact of suspension-feeders on the pelagic algal community in the northern Baltic Sea. The blue mussel (Mytilus edulis Gould) and the zebra mussel ((Dreissena polymorpha (Pallas)) were selected as experimental species due to their ubiquity and, hence their significant potential contribution to phytoplankton removal. The functional relationships between ambient temperature, salinity, current velocity, phytoplankton biomass and the biodeposition of the suspension-feeders were estimated at five sites differing in their eutrophication level during different times of the year. Based on these functional relationships algal grazing by the mussel populations was estimated in multiple areas taking into account the data on ambient temperature, salinity, Chl a concentration, mussel abundance and size distribution.
13 MATERIALS AND METHODS M. edulis and D. polymorpha are the most conspicuous suspensionfeeders in the northern Baltic Sea. The species are most prevalent on hard bottoms above the halocline where, owing to low predation and high input of nutrients, they often form extensive multilayered mats (Segerstråle 1957; Kautsky 1981; Kautsky 1995; Öst and Kilpi 1997). D. polymorpha dominates at salinities less than 5 psu and M. edulis in more saline environments (Kotta, 2000). The study was carried out on three transects in the littoral zone of the Gulf of Riga (GOR) and two transects in the Gulf of Finland (GOF) during one year period between 1996 and 2002 (Fig. 1). Sampling was performed during ice-free period in spring (T=2–15ºC), summer (T>15ºC) and autumn (T=2–15ºC). Northern GOR was characterised by a wide and sheltered coastal zone with diverse bottom topography and extensive reaches of boulders. Depending on the salinity, a scattered population of M. edulis or D. polymorpha occurred on the boulders. The southern transect had a narrow and exposed coastal zone. Coarse sandy substrate prevailed down to a depth of 4 m being replaced by boulders at greater depths. The boulders harboured a dense population of D. polymorpha. Hard substrate prevailed at the northern GOF site. The coverage of M. edulis was almost 100 % along this transect. The southern GOF was characterised by a mixture of sand, pebbles and boulders above 3-m depth. Deeper down only sandy substrate is found and, hence, the area was practically devoid of suspension-feeding bivalves. As a result of the differences in exposure to deep waters, the frequency of upwelling was higher in GOF than GOR sites. Due to high riverine load and moderate water exchange the nutrient concentrations were on average twice as high in GOR than in the Baltic Proper. Northern GOR sites were moderately eutrophicated and southern GOR site was highly eutrophicated. The southern GOF site was moderately eutrophicated due to municipal pollution load of Tallinn City. The concentration of nutrients in the northern GOF site was similar to the values of the Baltic Proper and, hence, representing the least disturbed environment in terms of eutrophication (Astok et al. 1999; Hänninen et al. 2000). In each season the abundance, biomass and size-frequency distribution of the suspension-feeders were estimated along the five abovementioned and an additional transect in a more exposed part of northern GOF. Samples were collected from the seashore down to 12 m depth at steps of 1 m. Metal frames of 20u20 cm surface area were placed randomly on the bottom by a diver. All suspension-feeders within the frame were collected. Three replicates were taken at each location. The length of the bivalves was measured to the nearest 0.1 mm using vernier callipers. In situ grazing rates of M. edulis and D. polymorpha were estimated by quantifying the defecation of Chl a by the mussels at 1 m in each transect
14 during different seasons. Bivalves of 9–31 mm shell length were collected by a diver in the vicinity of deployment. Three individuals were placed on the net of the funnel allowing biodeposits to sediment to the collecting vial below. The near-bottom temperature and salinity were monitored at the beginning and at the end of the deployment using CTD profiling. Each incubation lasted 4 hours. In each season we performed at least five incubations replicated three times.
Fig. 1. Study area. The transects of M. edulis are indicated by crosses and that of D. polymorpha by open circles.
Except for northern GOF plaster balls were used to estimate the water currents in near-bottom layer. The method is recognised as a simple and inexpensive tool for measuring integrated water motion over a wide range of flow rates. The dissolution rates of plaster balls are mainly a function of water velocity and less influenced by salinity and temperature within a range of our study (Thompson and Glenn 1994).
15 After deployment the shell lengths were recorded, the sedimented material in the vials was sorted under a dissecting microscope; faeces were collected with a pipette and filtered on Whatman CF/F filters within 4 h of retrieval. Filters were extracted in dark in 96% ethanol overnight. Chl a was quantified fluorometrically correcting for phaeopigments (Pha) (Strickland and Parsons 1972). The values of Chl a equivalent or total Chl a (Chl a eq) were calculated as Chl a eq = Chl a + 1.52 u Pha. During deployment water for Chl a measurement was daily sampled by a diver at near-bottom layer along the whole transect at steps of 1 m. Additional samples were taken at a distance of 25 cm from the cages in connection with retrieving biodeposits (i.e. in every 4–12 h). Hence, the average concentration of Chl a sampled at the start and end of an incubation was used as a measure of food concentration during incubation. Filtration and extraction of these samples were carried out within 1 h after sampling. The water samples were filtered onto Whatman GF/F filters. Chl a and Pha were measured as noted above. In order to estimate the loss of Chl a during gut passage separate experiments were carried out aboard ship. The mussels were incubated in 5 l buckets for 4 h. Buckets without experimental animals served as controls. The animals were fed natural sea water. At the end of the incubation the biodeposits were cleaned from the buckets by careful pipetting and water samples for Chl a were taken. The content of Chl a and Pha were estimated in biodeposits and water samples as described above. The loss of Chl a during gut passage was estimated as the ratio of the loss of Chl a in water to biodeposit production taking into account the algal growth and sedimentation in the control bucket. Clearance rate by the mussel population was calculated from the estimates of biodeposition. The functional relations between biodeposition and environmental variables were determined after correction for loss of Chl a during gut passage. The data on ambient temperature, salinity, Chl a concentration, mussel abundance and size distribution were taken into account when estimating population grazing potential in multiple areas. Annual population grazing potential is defined as the average of the calculated clearance rates of each incubation by transect and depth interval. The minimum and maximum values represent the extremes of the calculated clearance rates. Grazing by individuals of different size (Gl) was scaled by shell length, i.e. Gl = G20 u l2/202, where G20 is the grazing rate of 20 mm individuals and l the shell length (Kiørboe and Møhlenberg, 1981). We assumed no significant spatial variation in current velocity and complete vertical mixing along the transect. At low current velocities water exchange was likely not sufficient to supply the local suspension-feeder communities with phytoplankton. Hence, the grazing potential tends to overestimate the impact of mussels on phytoplankton communities when the water exchange is low.
16 The biodeposition and clearance of the mussels were analysed by factorial ANOVA including transect and season as the main effects. We employed linear and second-order polynomial linear regression analyses to describe the relationships between the biodeposition and ambient environmental variables. Polynomial regression results are only reported if significantly better fits were achieved using this method compared with the linear model. Table 1. The mean values r S.E. of water temperature, salinity, Chl a eq (Pg l-1) and current velocity (cm s-1) at the study sites in the Gulf of Riga (GOR) and the Gulf of Finland (GOF) during different seasons. Site Seili (GOF-N)
Kakumäe (GOF-S)
Kõiguste (GOR-N)
Audrurand (GOR-N)
Saulkrasti (GOR-S)
Year 1998
Season summer
Temperature 16.4 r 0.2
Salinity 5.9 r 0.1
Chl a 4.9 r 1.4
1998
autumn
8.2 r 0.2
5.8 r 0.1
5.0 r 1.2
1999
spring
2.4 r 0.2
5.8 r 0.1
1.2 r 1.1
2002
spring
12.6 r 0.4
6.1 r 0.1
9.4 r 2.5
Current not measured not measured not measured 0.1r0.0
2002
summer
20.6 r 0.2
4.9 r 0.1
5.7 r 1.6
29.6r4.3
2002
autumn
4.4 r 0.5
5.2 r 0.2
8.0 r 3.5
43.0r2.5
1996
spring
5.9 r 0.2
5.5 r 0.1
19.2 r 1.0
12.1r0.3
1996
summer
16.2 r 0.1
5.7 r 0.0
4.0 r 0.7
0.1r0.0
2002
spring
13.3 r 0.4
3.2 r 0.1
1.7 r 2.5
1.3r0.8
2002
summer
24.0 r 0.3
3.1 r 0.1
6.5 r 1.9
42.9r7.6
2002
autumn
2.0 r 0.5
5.4 r 0.2
14.9 r 3.5
61.3r3.0
1996
spring
4.6 r 0.1
5.0 r 0.0
65.0 r 0.9
0.1r0.0
1996
summer
16.0 r 0.2
5.1 r 0.1
14.4 r 1.0
30.4r5.6
RESULTS In summer water temperature was higher in southern GOF and northern GOR site of D. polymorpha. Salinity values were slightly higher in GOF than in GOR sites. All means were less than 7 psu. The values of maximum water Chl a eq (i.e. a measure of eutrophication level) were higher in GOR than in GOF sites. Highest Chl a eq values were measured in the
17 southern GOR site during the spring bloom. The summer values in the southern GOR site were in the same magnitude as the spring bloom values in other studied sites. In general, current velocities were lowest in spring, intermediate in summer and highest in autumn. The values varied between sites being lowest in the northern GOR site of M. edulis, intermediate in southern GOF and southern GOR site of D. polymorpha and highest in the northern GOR site of D. polymorpha (Table 1). Table 2. The models of multiple linear regressions describing the biodeposition and clearance rates of Mytilus edulis and Dreissena polymorpha. The abbreviations are as follows: T – temperature, S – salinity, Chl – Chlorophyll a eq, Curr – current velocity, multiple terms indicate their interaction. P values of the regressions are lower than 0.001. Site Seili (GOF-N)
Kakumäe (GOF-S)
Kõiguste (GOR-N)
Audrurand (GOR-N)
Saulkrasti (GOR-S)
All sites
All sites
Species Mytilus
Mytilus
Mytilus
Dreissena
Dreissena
Mytilus
Dreissena
Model Biodeposition
Model terms T, T2, Chl2T2
R2 0.85
Clearance
TChl
0.72
Biodeposition
TChl, TCurr, Chl2T2
0.92
Clearance
TChlCurr
0.92
Biodeposition
T, T2Chl
0.85
Clearance
T2Chl
0.73
Biodeposition
TCurr
0.91
Clearance
Chl, ChlCurr
0.92
Biodeposition
SChl, T2Chl, T2Chl2
0.91
Clearance
T, TChl2, T2Chl, ST2Chl2
0.91
Biodeposition
T, T2, Chl, Chl2, TChl, Chl2T
0.57
Clearance
T, T2, Chl, Chl2, T2Chl, Chl2T2
0.57
Biodeposition
TChl
0.89
18
Fig. 2a. Biodeposition rate (Pg ind-1 h-1) of D. polymorpha as a function of ambient temperature, salinity and Chl a eq.
19
Fig. 2b. Biodeposition rate (Pg ind-1 h-1) of M. edulis as a function of ambient temperature, salinity and Chl a eq.
The clearance rates (l ind-1 h-1) increased curvilinearly with ambient temperature. There was a significant interaction between temperature and Chl a eq. The effect of Chl a eq varied between sites and seasons. The clearance rate of D. polymorpha decreased with increasing salinity. In southern GOF
20 and northern GOR site of D. polymorpha current velocity interacting with temperature and Chl a eq had significant effect on biodeposition (Table 2, Fig. 4a). M. edulis had significantly higher clearance rates than D. polymorpha (Fig. 4b). Similarly to the biodeposition values the clearance values were higher at GOF sites than at GOR sites (ANOVA: F1,0.4 = 77.76, p < 0.001) and increased from spring to autumn (ANOVA: F2,0.2 = 39.03, p < 0.001) (Fig. 5). Biodeposition rate (µg Chl a eq ind-1 h-1) was mainly a function of ambient temperature and Chl a eq. The biodeposition values increased curvilinearly with temperature and ambient Chl a eq. The effect of temperature interacted with Chl a eq. The biodeposition of D. polymorpha decreased with increasing salinity. There were statistically significant differences in the regressions between different basins, sites within a basin and seasons. In southern GOF and northern GOR site of D. polymorpha current velocity interacting with temperature had significant effect on biodeposition (Table 2, Fig. 2a, b). The two studied bivalve species did not differ in their biodeposition rates. In general, the biodeposition values were higher at GOF sites than at GOR sites (ANOVA: F1,5 = 24.75, p < 0.001). Biodeposition was lowest in spring, intermediate in summer and highest in autumn (ANOVA: F2,4 = 21.12, p < 0.001) (Fig. 3).
Fig. 3. Mean biodeposition rate (± 95% CI) of M. edulis (solid line) and D. polymorpha (broken line) in relation to region and season.
21
Fig. 4a. Clearance rate (l ind-1 h-1) of D. polymorpha as a function of ambient temperature, salinity and Chl a eq.
22
Fig. 4b. Clearance rate (l ind-1 h-1) of M. edulis as a function of ambient temperature, salinity and Chl a eq.
23
Fig. 5. Mean clearance rate (± 95% CI) of M. edulis (solid line) and D. polymorpha (broken line) in relation to region and season.
Fig. 6. Annual averages of population grazing potential of M. edulis and D. polymorpha along each transect (% of overlaying water cleared m-2 h-1). The minimum and maximum values represent the extremes of the calculated clearance rates of each incubation.
24 The major variability in population grazing potential (% of overlaying water cleared m-2 h-1) was due to the spatial differences in the density of bivalves. The grazing potential was highest at 2–6 m. The lack of hard substrate indirectly reduced the grazing pressure in the shallower areas of southern GOR and in the deeper areas of northern GOR and southern GOF (Fig. 6). When averaged over the transect the suspension-feeders removed daily on average from 3 to 2426% of phytoplankton stock in the coastal area. Population grazing decreased with increasing Chl a eq i.e. eutrophication level (Fig. 7).
Fig. 7. Relationships between annual averages of water Chl a eq and population grazing potential of bivalves in the littoral zone (0-12 m).
DISCUSSION Results of this study indicate that temperature and phytoplankton biomass were the major causes for temporal and spatial variations in biodeposition and clearance rates of mussels. Salinity was important factor for D. polymorpha. Current velocity affected biodeposition and clearance rates at sites where mussels were confined to the shallower depth of the transect (GOF-S, GOR-N D. polymorpha site). The relationships varied significantly between species, sites and seasons. The models predicted between 57 and 92% of the variability in the biodeposition or clearance rates of the bivalves. Regressions with low coefficient of determination were described in the areas where food conditions were unstable due to upwelling (GOF-N, GOR-N M. edulis site). At low temperatures (< 8 ºC) the biodeposition of the studied suspension-feeders was lower regardless of food conditions. Similar results were found earlier for M. edulis and D. polymorpha in the GOR and in the northern Baltic Sea (Kautsky and Evans, 1987; Kotta and Møhlenberg, 2002). Low pumping rates at low temperatures can be caused by temperature induced
25 changes in ciliary beat frequency and an increased viscosity of the water (Jørgensen et al., 1990; Loo, 1992). An active regulation of suspensionfeeding through food concentration (Newell and Bayne, 1980) is less likely. However, as low temperatures and low food levels coincide in many areas, low temperature can be considered a favourable condition during such periods of food shortage, reducing high costs of suspension-feeding when concentration of food particles is low. At higher temperatures the relationship between suspension-feeding and the water temperature was not constant but depended on the food supply and features of the studied basin. Suspension-feeding increased with rising ambient concentrations of Chl a and levelled off at high food concentrations. The saturation reduction occurred above 5–10 µg Chl a eq L-1. It is known that high algal concentrations may lead to reduced valve gapes and a reduction of the filtration rate (Riisgård and Randløv, 1981; Riisgård, 1991, 2001). Such high Chl a values were prevailing in GOR suggesting that fitted polynomial functions between suspension-feeding and the food supply reflected the saturation at high Chl a concentrations in GOR (Kotta and Møhlenberg, 2002). In the northern GOF site, however, a linear functional response between Chl a eq and clearance showed that such a reduction due to saturation was never reached. D. polymorpha is found in a wide range of salinities but each of its subspecies has narrow and different tolerance to salinity. In the Baltic Sea D. polymorpha can live in salinities up to 6 psu (Karatayev et al., 1998). In accordance to these findings the biodeposition and clearance of D. polymorpha decreased with increasing salinities at the southern GOR site. Although, salinity varied within the same range at the northern GOR site, the effect of salinity was not significant and interactions among other factors had a much greater impact on D. polymorpha. We may assume that differences in functional responses are due to food availability. Owing to high riverine load the events of low salinity and high Chl a eq coincided at the southern GOR site (correlation analysis: r = -0.37, p = 0.006) whereas Chl a eq increased with salinity at the northern GOR site (r = 0.78, p < 0.001). The comparison of the clearance rates of M. edulis showed about three times lower suspension-feeding activity in GOR than in the northern GOF site despite of similar temperatures and Chl a concentrations. Hence, temperature and Chl a concentrations per se can not explain the behavioural variation in filtering activity. Due to the difference in exposure the current velocities are likely higher in the northern GOF than in GOR sites. The vertical mixing of the water column increases the amount of available food and promotes the filtration activity of mussels (Dolmer 2000; Newell et al. 2001). In southern GOF and northern GOR site of D. polymorpha the inclusion of the current velocity interacting with temperature and Chl a eq significantly improved the models of biodeposition and clearance rate of M. edulis and D. polymorpha. Besides, the quality of the seston affects the filtration activity of mussels (Asmus and Asmus 1993). The GOR sites are
26 characterized by higher variability of phytoplankton communities than GOF sites. The share of diatoms is higher in GOR than GOF sites (HELCOM 1996, 2002). At the cleaner GOF site, where Chl a concentrations were more stable, the bivalves seemed to use their full filtration capacity regardless of the ambient food level. Due to the aggregated distribution the grazing rate of the populations of suspension-feeders varied significantly within and between study areas. The grazing rate was orders of magnitude higher in GOF than in GOR. The northern GOR has extensive shallow areas and moderate water exchange. Thus, the daily removal of 14% of Chl a on average in GOR may be sufficient for the benthic control of phytoplankton in the area. This was also reflected by the depletion of Chl a at the near-bottom water of mussel beds. In GOF, owing to a relatively small share of coastal area and significant water exchange, it is difficult to estimate the effect of suspension-feeders grazing on phytoplankton population at the regional scale. Nevertheless, the accumulation of biodeposits through mussel filtration was an important process at GOF sites, especially in the sites, that were more exposed to deep waters and housed higher densities of suspension-feeders. The utilisation rate of phytoplankton by suspensionfeeders varies with benthic boundary-layer flow conditions (Fréchette et al. 1989; Dolmer 2000; Newell et al. 2001). On the population level the biomass of suspension-feeders increases with the current intensity and frequency (Gili and Ballesteros 1992; Lesser et al. 1994). It is likely that in the outer archipelago stronger vertical mixing increased the amount of food available to the suspension-feeders and hence, supported higher biomasses. Lower values of population grazing in the middle archipelago might be attributed to the lower wave energy input to the system. In shallower areas phytoplankton production is low due to high turbulence and dense macrophyte assemblages whereas the growth of suspension-feeders is controlled by ice scouring. In deeper areas the suspension-feeders are food limited due to low current velocities. Consequently, the suspension-feeders have highest biomasses and impact at intermediate depths unless substrate is not limiting their distribution. The inverse relationship between water Chl a concentration and population grazing of bivalves indicated that under eutrophicated conditions the impact of suspension-feeders on pelagic communities is small relative to cleaner environments. The studied bivalves exploit their filtration capacity to about 5 µg Chl a l-1 (Clausen and Riisgård 1996). Above that level a considerable reduction of filtration rate is observed presumably caused by overloading of the alimentary canal (Riisgård 1991). Besides, very high sedimentation and concentration of inorganic particles often observed in eutrophicated areas are detrimental to suspension-feeding bivalves (Kiørboe et al., 1980). In this study we have observed the feeding behaviour of suspensionfeeders under a wide range of environmental conditions. We have demonstrated that an important share of phytoplankton is grazed by benthic
27 suspension-feeders in the northern Baltic Sea. There were significant interactions between temperature, phytoplankton biomass and current velocities when affecting the biodeposition and clearance rates of the bivalves. The values of clearance rates measured in this study were in accordance with earlier field observations (Cranford et al. 1998). This study does not only contribute to the knowledge of the functional relationships between suspension-feeding and environmental settings but it also shows the necessity of further in situ measurements on undisturbed suspension-feeders in order to explain and model large natural spatial variations of suspension-feeding.
ACKNOWLEDGEMENTS This research was supported by the grants of the Estonian Governmental Programme no 0182578s03 and Estonian Science Foundation grant no 5103.
REFERENCES Asmus H Asmus RM 1993 Phytoplankton-mussel bed interactions in intertidal ecosystems. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed), Springer-Verlag, Heidelberg, pp 57-84 Astok V Otsmann M Suursaar Ü 1999 Water exchange as the main physical process in semienclosed marine systems: the Gulf of Riga case. Hydrobiologia 393: 11-18 Barnes RSK Hughes RN 1988 An Introduction to Marine Ecology. Blackwell Scientific Publications, Oxford, 351 p Bayne BL Widdows J Worrall C 1977 Some temperature relationships in the physiology of two ecologically distinct bivalve populations. In: Physiological Responses of Marine Biota to Pollutants, FJ Vernberg A Calabrese FP Thurberg WB Vernberg (Eds). Academic Press, New York. pp 379-400 Clausen I Riisgård HU 1996 Growth, filtration and respiration in the mussel Mytilus edulis: no evidence for physiological regulation of the filter-pump to nutritional needs. Mar Ecol Prog Serr 141: 37-45 Cloern JE 1982 Does the benthos control phytoplankton biomass in South Francisco Bay? Mar Ecol Prog Serr 9: 191-202 Cranford PJ 2001 Evaluating the ‘reliability’ of filtration rate measurements in bivalves. Mar Ecol Prog Serr 215: 303-305 Cranford PJ Emerson CW Hargrave BT Milligan TG 1998 In situ feeding and absorption responses of sea scallops Placopecten magellanicus (Gmelin) to storm-induced changes in the quantity and composition of the seston. J Exp Mar Biol Ecoll 219: 45-70 Cranford PJ Hill PS 1999 Seasonal variation in food utilization by the suspension-feeding bivalve molluscs Mytilus edulis and Placopecten magellanicus. Mar Ecol Prog Ser 190: 223-239 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Doering PH Oviatt CA 1986 Application of filtration rate models to field population of bivalves: an assessment using experimental mesocosms. Mar Ecol Prog Ser 31: 265275 Dolmer P 2000 Feeding activity of mussels Mytilus edulis related to near-bed currents and phytoplankton biomass. J Sea Res 44: 221-231
28 Fréchette M Bourget E 1985 Energy flow between the pelagic and benthic zones: factors controlling particulate organic matter available to an intertidal mussel bed. Can J Fish Aquat Sci 42: 1158-1165 Fréchette M Butman CA Geyer WR 1989 The importance of boundary-layer flows in supplying phytoplankton to the benthic suspension feeder, Mytilus edulis L. Limnol Oceanogrr 34: 19-36 Gili JM Ballesteros E 1992 Structure of cnidarian populations in Mediterranean sublittoral benthic communities as a result of adaptation to different environmental conditions. In: Homage to Ramón Margalef or Why There is Such Pleasure in Studying Nature, JD Ros N Prat (Ed), Publications Universitat de Barcelona. pp 243-254 Hawkins AJS Smith RFM Bayne BL Héral M 1996 Novel observations underlying the fast growth of suspension-feeding shellfish in turbid environments. Mar Ecol Prog Serr 131: 179-190 Hänninen J Vuorinen I Helminen H Kirkkala T Lehtilä K 2000 Trends and gradients in nutrient concentrations and loading in the Archipelago Sea, Northern Baltic, in 19701997. Estuar Coast Shelf Scii 50: 153-171 HELCOM 1996 Third Periodic Assessment of the State of the Marine Environment of the Baltic Sea, 1989-1993; Background document. Balt Sea Environ Proc 64B HELCOM 2002 Environment of the Baltic Sea Area 1994-1998. Balt Sea Environ Proc 82B Herman PMJ Scholten H 1990 Can suspension-feeders stabilize estuarine ecosystems? In: Trophic Relationships in the Marine Environment. Proc 24th Eur Mar Biol Symp, M Barnes RN Gibson (Eds). Aberdeen University Press, Aberdeen, pp 104–116 Jørgensen CB Larsen PS Riisgård HU 1990 Effects of temperature on the mussel pump. Mar Ecol Prog Serr 64: 89-97 Karatayev AY Burlakova LE Padilla DK 1998 Physical factors that limit the distribution and abundance of Dreissena polymorpha (Pall.). J Shellfish Res 17: 1219-1235 Kautsky N Evans S 1987 Role of biodeposition by Mytilus edulis in the circulation of matter and nutrients in a Baltic coastal ecosystem. Mar Ecol Prog Serr 38: 201-212 Kautsky N 1981 On the role of blue mussel Mytilus edulis L. in the Baltic ecosystem. Doctoral dissertation. Stockholm University, Sweden, 22 p Kautsky U 1995 Ecosystem processes in coastal areas of the Baltic Sea. Doctoral dissertation. Stockholm University, Sweden, 25 p Kiørboe T Møhlenberg F 1981 Particle selection in suspension-feeding bivalves. Mar Ecol Prog Serr 5: 291-296 Kiørboe T Møhlenberg F Nøhr O 1980 Feeding, particle selection and carbon absorption in Mytilus edulis in different mixtures of algae and resuspended bottom material. Ophelia 19: 193-205 Kotta J 2000 Impact of eutrophication and biological invasions on the structure and functions of benthic macrofauna. Dissertationes Biologicae Universitatis Tartuensis, 63. Tartu University Press, Tartu, 160 p Kotta J Møhlenberg F 2002 Grazing impact of Mytilus edulis L. and Dreissena polymorpha (Pallas) in the Gulf of Riga, Baltic Sea estimated from biodeposition rates of algal pigments. Ann Zool Fenn 39: 151-160 Lesser MP Witman JD Sebens KP 1994 Effects of flow and seston availability on scope for growth of benthic suspension-feeding invertebrates from the Gulf of Maine. Biol Bull 187: 319-335 Loo LO 1992 Filtration, assimilation, respiration and growth of Mytilus edulis L. at low temperatures. Ophelia 35: 123-131 Møhlenberg F 1995 Regulating mechanisms of phytoplankton growth and biomass in a shallow estuary. Ophelia 42: 239-256 Newell CR Wildish DJ MacDonald BA 2001 The effects of velocity and seston concentration on the exhalant siphon area, valve gape and filtration rate of the mussel Mytilus edulis. J Exp Mar Biol Ecoll 262: 91-111
29 Newell RC Bayne BL 1980 Seasonal changes in the physiology, reproductive condition and carbohydrate content of the cockle Cardium (Cerastoderma) edule (Bivalvia: Cardiidae). Mar Bioll 56: 11-19 Officer CB Smayda TJ Mann R 1982 Benthic filter feeding: a natural eutrophication control. Mar Ecol Prog Serr 9: 203-210 Öst M Kilpi M 1997 A recent change in size distribution of blue mussels (Mytilus edulis) in the western part of the Gulf of Finland. Ann Zool Fenn 34: 31-36 Riisgård HU 1991 Filtration rate and growth in the blue mussel, Mytilus edulis Linnaeus, 1758: dependence on algal concentration. J Shellfish Res 10: 29-35 Riisgård HU 2001 On measurement of filtration rates in bivalves – the stony road to reliable data: review and interpretation. Mar Ecol Prog Serr 211: 275-291 Riisgård HU Randløv A 1981 Energy budgets, growth and filtration rates in Mytilus edulis at different algal concentration. Mar Bioll 61:227-234 Segerstråle SG 1957 Baltic Sea. Mem Geol Soc America 67: 751-800 Strickland JDH Parsons TR 1972 A practical handbook of seawater analysis. Bull Fish Res Bd Can 167: 1-310 Thompson TL Glenn EP 1994 Plaster standards to measure water motion. Limnol Oceanogr 39: 1768-1779 Widdows J 1985 The effects of fluctuating and abrupt changes in salinity on the performance of Mytilus edulis. In: Marine Biology of Polar Regions and Effects of Stress on Marine Organisms, JS Gray ME Christiansen (Eds), J Wiley, Chichester. pp 555–556
CAN BIVALVE SUSPENSION-FEEDERS AFFECT PELAGIC FOOD WEB STRUCTURE?
Theo Prins1 and Vincent Escaravage2 1
National Institute for Coastal and Marine Management/RIKZ, PO Box 8039, 4330 EA Middelburg, The Netherlands 2 Netherlands Institute of Ecology, Yerseke, The Netherlands Abstract: Bivalve suspension-feeders are considered to be keystone herbivores in many estuarine ecosystems. However, bivalves can also feed upon organisms that belong to the microzooplankton and on mesozooplankton. Laboratory experiments showed that nauplii of the copepod Temora longicornis were filtered by mussels at the same rate as algae. Adult T. longicornis were also susceptible to filtration by mussels and oysters, but at a lower rate. Mesocosm experiments compared plankton dynamics in systems with and without mussels. Biomass of diatoms, heterotrophic dinoflagellates and copepods was strongly reduced in the presence of mussels. Some components of the microbial food web, like ciliates and Phaeocystis, did not show a significant effect, due to cascading effects of declining copepod abundance. It is suggested that in the presence of mussels, the pelagic food web may be shifted towards a more dominant microbial food web. Key words: Mytilus edulis, microzooplankton, copepods, grazing, trophic interactions
INTRODUCTION Grazing by bivalve suspension-feeders is considered a major process in many shallow coastal ecosystems, with the potential to control phytoplankton biomass development to a large extent. Observations and model calculations for a wide variety of estuaries and coastal systems, and results from mesocosm and enclosure experiments, support the notion that top-down control of phytoplankton biomass by bivalve suspension-feeding is a widespread phenomenon in bivalve dominated systems (see e.g. Dame 1996). In view of the extensive documentation of bivalve grazing control of phytoplankton biomass, it is rather surprising that there is only limited information on the effects of bivalve grazing on other components of the plankton, particularly zooplankton. Bivalve suspension-feeders may have an indirect effect on herbivorous zooplankton through food competition, but may 31 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 31–51. ©2005 Springer. Printed in the Netherlands.
32 also directly influence zooplankton populations through filtration (Davenport et al. 2000). Carlson et al. (1984) made in situ observations of decreased zooplankton abundance on ebb tides after passage over an intertidal flat with mussels. However, differences were not tested statistically, and the significance of other factors causing the observed decrease in zooplankton abundance (e.g. settlement, other predators) could not be excluded. In recent years, several studies have addressed the effects of bivalve filtration on zooplankton species. There is experimental evidence of filtration of various microzooplankton species by the oyster Crassostrea gigas (Le Gall et al. 1997, Dupuy et al. 1999, Dupuy et al. 2000). It is suggested that feeding by oysters on microzooplankton like protists may constitute a trophic link between the microbial food web and benthic suspension-feeders (Le Gall et al. 1997, Dupuy et al. 1999). Other evidence of a negative impact of bivalves on microzooplankton comes from 3 enclosure experiments, lasting 2-3 weeks, with a mussel biomass in the enclosures high enough to filter the water volume once a day (Riemann et al., 1988). Significantly lower biomass of ciliates and rotifers was observed in enclosures with mussels Mytilus edulis in contrast to enclosures without mussels. However, no effects of the mussels on the abundance of the copepod Acartia tonsa was observed, presumably due to escape responses of the copepods (Horsted et al., 1988). Field observations of declines in copepod abundance, following the introduction of Potamocorbula amurensis in San Francisco Bay, were ascribed to predation by this clam on copepod nauplii (Kimmerer et al. 1994). Experiments demonstrated that bivalve suspension-feeders could capture and ingest copepod nauplii and adults (Kimmerer et al. 1994, Davenport et al. 2000). Escape responses of copepods show interspecific differences (Titelman 2001, Green et al. 2003). Differences in vulnerability of copepod species to bivalve predation could lead to bivalve control of zooplankton species composition, as suggested by Kimmerer et al. (1994). It can be concluded that bivalve filtration can have direct effects through predation on the abundance of zooplankton species. Depending on the zooplankton species involved, bivalve filtration could have an effect on the microbial food web by removing microzooplankton species, or it could have an impact on the classical food chain by predation on copepods. Interspecific differences in susceptibility of zooplankton organisms to bivalve grazing, combined with trophic interactions between different components of the planktonic food web, may result in cascading effects of bivalve grazing on pelagic food web structure. In this study, we use results from various sources to address the hypothesis that bivalve suspension-feeders change pelagic food web structure through predation on zooplankton. Laboratory experiments, aimed at establishing the effects of bivalve filtration on different life-stages of copepods, were carried out to determine the direct, short-term effects of bivalve filtration on copepod survival. Experimental ecosystems (mesocosms)
33 were used for experiments to study the longer-term effects of bivalve filtration on the development of pelagic populations and the composition of the planktonic community.
MATERIALS AND METHODS
Lab Experiments To establish the direct effects of filtration by the blue mussel M. edulis and the pacific oyster C. gigas on naupliar larvae and adult stages of the copepod T. longicornis experiments were carried out between September and December 2002, under controlled conditions at the RIKZ field station. Mussels and oysters had been collected in September 2002 in the Oosterschelde estuary (SW Netherlands). The individual animals were glued on small sticks, and were stored in raceways with flowing natural seawater between experiments. In experiments with mussels (between 15-59 mm shell length), individual animals were placed in 1.15 l bottles, containing a mixture of seawater, algal cells (Rhodomonas ( sp.) and either nauplii or adults collected from a culture of T. longicornis. Mussels were incubated for 30-60 minutes in the bottles that were placed on a rotating wheel to ensure complete mixing of the water in the bottles. Experiments with oysters (67-112 mm shell length) were carried out by placing the oysters in 2.2 l bottles with a mixture of seawater, algae and adult copepods. The water in the bottles was mixed with air bubbles. Initial algal concentrations were kept between 10 and 20 •106 cells l-1, initial copepod concentrations were between 10 and 40 animals l-1. Algal cell concentrations were below levels where the bivalves reduced filtration rates in response to high algal concentrations. Algal cell concentrations at the start and the end of the experiment were counted with a Coulter Counter. Copepods at the start of the experiment were counted while manually adding the copepods to the experimental vessel. At the end of the experiment, the water from the experimental bottles was filtered through a net with a 55 µm mesh size, and copepod abundances were counted under a stereoscope. Each experiment consisted of a series of incubations, with up to 12 bivalve incubations and 2 or more controls (experimental vessels without bivalves). In total, six experiments were carried with nauplii and various sizes of mussels. With adult copepods, two experiments were done with mussels, and four with oysters. Clearance rates CR on algae and on copepods were calculated from the following formula (Coughlan 1969):
34 CR = V/t * ln(C0/Ct) where V = volume of incubation bottle t = time of incubation Co = plankton concentration at the start of the experiment Ct = plankton concentration at the end of the experiment For each experiment, a t-test was used to determine if there was a significant decrease in algal or copepod concentrations in the control incubations.
Design of the Mesocosms The land-based mesocosms are located at the field station of RIKZ near the mouth of the Oosterschelde estuary. The mesocosms consist of black solid polyethylene tanks (height 3 m, diameter 1.2 m, volume 3000 l). The water column in the mesocosms was completely mixed. Daily cleaning prevented fouling of organisms on the walls of the tanks. Inorganic nutrients were continuously added to each of the mesocosms at a rate of 3.7 mmol N m3 , 0.06 mmol P m-3, and 0.8 mmol Si m-3. The mesocosms were flushed with natural seawater at a rate of 100 l day-1, resulting in a residence time of 30 days. Water from the bottom layer of each mesocosm was circulated through a 16 l benthos chamber containing mussels, at a rate of 45 l h-1 by means of a tubing pump. An extensive description of the mesocosms is given in Prins et al. (1995).
Design of the Mesocosm Experiment The period considered was from 30 March 1998 to 20 May 1998, as part of a more extensive mesocosm experiment (see Escaravage and Prins 2002 for details). In this study, we will compare two mesocosm units with a high biomass of mussels M. edulis (“MUS”) with two systems without a benthos and hence a solely pelagic community (“ZOO”). All other experimental conditions were similar in both treatments. In the MUS treatment, 80 mussels were added to each mesocosm. This was equivalent to a mussel biomass of 2.1 g ADW m-3 and a filtration rate of approximately 300 l day-1, or 10% of the mesocosm volume per day. Mussels had been collected a week before the experiment started, from the low tide level at a site close to the field station. At the start of the experiment, natural seawater from the Oosterschelde estuary was added to the mesocosms. Nutrient additions were the same for all mesocosms, and equivalent to loadings to the Dutch coastal zone (Prins et al. 1999). As earlier experiments had indicated that copepod
35 development in the mesocosms could be limited due to low concentrations at the start of the experiment (Prins et al. 1999), copepod biomass was enhanced by addition of zooplankton collected in the field. At the day of the start of the experiment, 12 m3 of water was pumped from approximately 5 m below the surface in the mouth of the Oosterschelde estuary. This water was immediately filtered through a 55 µm plankton net. The collected net plankton was subdivided into 4 subsamples (equivalent to the biomass in 3 m3 of water), and added to each mesocosm within 3 hours after collection. Extensive descriptions of sampling methods and analytical procedures for particulate and dissolved nutrients, chlorophyll-a, phytoplankton and micro- and mesozooplankton biomass, and primary production and bacterial production are given in Escaravage et al. (1995), Prins et al. (1995, 1999), Escaravage and Prins (2002). Samples were collected once a week. Differences between treatments were tested with a two-way ANOVA, with the experimental treatment (presence/absence of mussels) and time as independent factors, and the replicate mesocosms nested within the treatment factor. There was one observation per cell, and the error term for the treatment effect is the nested factor. This design is equivalent to a repeated measures design. Abundance data were log-transformed to reduce heterogeneity of variances.
RESULTS
Lab Experiments Six experiments were done with mussels and nauplii, and two with mussels and adult copepods. In three of these experiments, small but significant changes in concentrations of algae ((Rhodomonas sp.) were observed in the controls, but changes in control concentrations of the copepod T. longicornis were not observed in any of the experiments. Four experiments were done with oysters and adult copepods. In two of these experiments the controls showed small but significant decreases in algal concentrations, but no changes in concentrations of the copepods occurred. In the experiments with significant changes in control algal concentrations, clearance rates were corrected for algal sedimentation in the controls.
36
4
CRcopepods (l h-1)
3
2
1
0 0
1
2
3
-1
CRalgae (l h )
Figure 1. Clearance rates of mussels, calculated from changes in numbers of T. longicornis nauplii (CRcopepods), plotted against clearance rates calculated from Rhodomonas sp. concentration changes (CRalgae).The line y = x is indicated.
3
-1
CRcopepods (l h )
Mussel Oyster
2
1
0 0
1
2
3
4
-1
CRalgae (l h )
Figure 2. Clearance rates of mussels and oysters, calculated from changes in numbers of T. longicornis adults (CRcopepods), plotted against clearance rates calculated from Rhodomonas sp. concentration changes (CRalgae). The line y = x is indicated
37 Individual mussel clearance rates measured in experiments with T. longicornis nauplii are shown in Figure 1. There was no significant difference between clearance rates estimated from changes in algal concentrations (CRalgae), and clearance rates estimated from the decrease in copepod abundance (CRcopepods) (paired t-test, p>0.05). In the experiments with adult copepods, significant decreases in the concentrations of copepods during incubation were observed. However, in contrast to the results with nauplii, there was a lack of correlation between CRcopepods and CRalgae (Figure 2). The clearance rate of mussels on adult copepods (mean ± SD: 0.19 ± 0.21, n=24) was significantly lower (t-test, p<0.01) than the clearance rate of oysters on copepods (0.52 ± 0.46, n=42), even if the observations with high CRalgae of oysters (>2.5 l h-1) were excluded.
Mesocosm Experiment Water temperatures increased from values around 10 °C at the start of the experiment (day 90) to temperatures above 15 °C at day 140. Initial inorganic nutrient concentrations were high (DIN: 178 µM, DIP 4.0 µM, Si 49 µM), and decreased constantly during the course of the experiment. At the end of the experiments inorganic nutrient concentrations (DIN: 114 µM; DIP 0.16 µM; Si 2.0 µM) were still above levels that are considered to be limiting phytoplankton growth (limiting levels: DIN: 2 µM; DIP 0.1 µM; Si 2 µM; Escaravage et al. 1999). No significant differences in nutrient concentrations between treatments occurred. Average biomass of phyto- and zooplankton, and levels of bacterial and primary production are shown in Table 1. Initial phytoplankton biomass was low, with a chlorophyll-a concentration of 1.1 µg l-1. There was a significant treatment effect on chlorophyll concentrations (ANOVA, p<0.01), with levels almost 3 times higher in ZOO than in MUS (Figure 3). An initial diatom bloom between day 100-110, composed mainly of Skeletonema costatum and Rhizosolenia setigera, was followed by a second peak around day 130, dominated by Thalassiosira decipiens. The average diatom biomass (shown as time series in Figure 4) was significantly higher in the ZOO treatment than in the MUS treatment (Table 1). A bloom of the haptophyte Phaeocystis sp. developed at the end of the experiment (Figure 5). Highest biomass levels of Phaeocystis sp. were observed in the ZOO treatment (Figure 5), but variability within treatments was large and differences between treatments were not significant (Table 1). Other small flagellate species (< 5 µm) reached relatively low biomass levels, and average biomass was higher in the ZOO treatment (Table 1). Primary production in each system was strongly correlated with chlorophyll-a concentrations. Phytoplankton primary production and pelagic bacterial production showed a significant treatment effect (Table 1). Average
38 bacterial production was 30-35 % of primary production in the MUS treatment, but only 11-13% in the ZOO treatment. Microzooplankton biomass rapidly increased after the start of the experiment, and more or less followed phytoplankton biomass, culminating in maximum biomass levels between days 110-140 in all mesocosms (Figure 6). Aloricate ciliates (not identified at the species level) were a highly abundant group in the microzooplankton, with peak levels up to 20-35 µg C l-1. There was substantial variation between mesocosms and within Table 1. Biomass of various plankton groups, and levels of bacterial production and primary production. The values represent the averages of replicate mesocosms ± S.D. Significant differences between the two treatments (ANOVA) are indicated (* p<0.050; ** p<0.010; *** p<0.001).
Chlorophyll-a (µg l-1) Diatoms (µg C l-1) Phaeocystis sp. (µg C l-1) Pico- / nanoflagellates (µg C l-1) Primary production (g C m-2 d-1) Bacterial production (g C m-2 d-1) Ciliates (µg C l-1) Dinoflagellates (µg C l-1) Sum of copepod biomass (µg C l-1) Copepod nauplii (µg C l-1) Pseudocalanus minutus elongatus (µg C l-1) Copepodids I-III Copepodids IV-VI Adults Temora longicornis (µg C l-1) Copepodids I-III Copepodids IV-VI Adults
MUS 4.4 ± 0.3
ZOO 10.9 ± 1.3
MUS
49 ±25
177 ± 106
***
55 ± 44
124 ± 94
ns
13.7 ± 0.6
24.3 ± 0.4
**
0.290 ± 0.017
0.863 ± 0.037
**
0.127 ± 0.068
0.157 ± 0.009
*
6.0 ± 2.6
11.0 ± 7.0
ns
3.9 ± 0.4
11.6 ± 7.1
**
5.3 ± 0.3
13.8 ± 4.0
*
1.15 ± 0.33
2.44 ± 0.09
*
0.12 ± 0.03 0.20 ± 0.02 0.50 ± 0.11
0.37 ± 0.03 0.46 ± 0.12 0.89 ± 0.28
** * *
0.22 ± 0.09 0.56 ± 0.21 2.16 ± 0.47
0.61 ± 0.14 1.60 ± 0.48 6.76 ± 2.99
* * *
treatments, and no significant treatment effect was observed (Table 1). Another dominant group in the microzooplankton was formed by heterotrophic dinoflagellates (Figure 7). Protoperidinium bipes and an unidentified Protoperidinium species formed peaks in biomass around day
39 110-120 in ZOO. This was followed by a bloom of Oxyrrhis marina and unidentified thecate dinoflagellate species between day 120-130 in ZOO. Overall, heterotrophic dinoflagellate biomass was significantly higher in treatment ZOO (Table 1). Four different copepod species were observed in the systems. During counting, copepodid stages I-III and stages IV-VI as well as adults were discriminated for all species. The initial copepod biomass was dominated by calanoid nauplii and various stages of Pseudocalanus minutus elongatus and T. longicornis. Copepod biomass peaked between day 120-130, one to two weeks after the diatom bloom started (Figure 8-11). T. longicornis developed highest numbers of copepodids, and also showed good development of adults following the diatom bloom, reaching densities of 1000-2000 m-3. In the last week, numbers of copepodids and adults decreased rapidly, with the exception of adult Acartia clausii that peaked between days 130-140 in the ZOO treatment. Numbers of copepod eggs and naupliar stages reached maximum levels after day 130, hence after the peak levels of adults. The biomass of copepod eggs and of calanoid nauplii was lower in the MUS treatment than in the ZOO treatment (Table 1). A. clausi and Centropages hamatus showed no significant treatment effects, but this was probably due to low numbers. T. longicornis and P. minutus had relatively high numbers of both copepodid groups in treatment ZOO, whereas development in the MUS treatment was significantly lower (p<0.05; Table 1). Adults reached maximum densities around day 130 and then declined. Total copepod biomass was significantly higher in the ZOO treatment (p<0.05; Table 1). The average biomass of heterotrophic dinoflagellates and copepods was higher in the ZOO treatment (Table 1), and biomass levels of these zooplankton groups in the four mesocosms were significantly correlated and proportional to primary production (dinoflagellates: r2=0.99, p<0.001; copepods r2=0.92, p<0.05): as an example, primary production in mesocosm MUS-2 was only 26% of the production in ZOO-2, dinoflagellate biomass was 26% and copepod biomass 30% of that in ZOO-2. Ciliates showed no significant treatment effect and no correlation with primary production. The fraction of ciliate biomass in the total pelagic consumer biomass (ciliates, dinoflagellates and copepods) was, on average, higher in the MUS treatment (37%) than in the ZOO treatment (27%) although differences were not statistically significant due to large variability within the ZOO treatment.
40
35 MUS-1 MUS-2 ZOO-1 ZOO-2
30
Chlorophyll (µg l-1)
25
20
15
10
5
0 90
100
110
120
130
140
Day no
Figure 3. Chlorophyll-a concentrations in the 4 mesocosms
0.6 MUS-1 MUS-2 ZOO-1 ZOO-2
-1
Diatoms (mg C l )
0.5
0.4
0.3
0.2
0.1
0.0 90
100
110
120
Day no
Figure 4. Diatom biomass in the 4 mesocosms
130
140
41
1.4 MUS-1 MUS-2 ZOO-1 ZOO-2
Phaeocystis sp. (mg C l-1)
1.2
1.0
0.8
0.6
0.4
0.2
0.0 90
100
110
120
130
140
Day no
Figure 5. Phaeocystis sp. biomass development in the 4 mesocosms
35 MUS-1 MUS-2 ZOO-1 ZOO-2
30
Ciliates (µg C l-1)
25
20
15
10
5
0 90
100
110
120
130
Day no
Figure 6. Ciliate biomass development in the mesocosms
140
42
35 MUS-1 MUS-2 ZOO-1 ZOO-2
Dinoflagellates (µg C l-1)
30
25
20
15
10
5
0 90
100
110
120
130
140
Day no
Figure 7. Biomass of heterotrophic dinoflagellates in the 4 mesocosms
40 MUS-1 MUS-2 ZOO-1 ZOO-2
Copepods (µg C l-1)
30
20
10
0 90
100
110
120
Day no
Figure 8. Total copepod biomass in the 4 mesocosms
130
140
43
1,2 MUS-1 MUS-2 ZOO-1 ZOO-2
1,0
Nl
-1
0,8
0,6
0,4
0,2
0,0 90
100
110
120
130
140
Day no
Figure 9. Abundance of Pseudocalanus minutus elongatus copepodids
1,2 MUS-1 MUS-2 ZOO-1 ZOO-2
1,0
N l-1
0,8
0,6
0,4
0,2
0,0 90
100
110
120
130
Day no
Figure 10. Abundance of Pseudocalanus minutus elongatus adults
140
44
4 MUS-1 MUS-2 ZOO-1 ZOO-2
Nl
-1
3
2
1
0 90
100
110
120
130
140
Day no
Figure 11. Abundance of Temora. longicornis copepodids
4 MUS-1 MUS-2 ZOO-1 ZOO-2
N l-1
3
2
1
0 90
100
110
120
Day no
Figure 12. Abundance of Temora longicornis adults
130
140
45
DISCUSSION Filtration by bivalves can reduce the biomass of mesozooplankton. The lab experiments gave evidence for direct filtration of copepods by mussels and oysters. In the case of nauplii, mussel clearance rates calculated from algal concentration changes were not different from clearance rates based on changes in nauplii abundance. This indicates that the filtration effect on nauplii is similar to the effect on algae, and apparently nauplii are unable to escape the flow generated by the bivalves. Experiments with adult copepods, on the other hand, showed that clearance rates of oysters and mussels on copepods were much lower than the clearance rates on algae, and there was no correlation between the two clearance rates. In other words, larger bivalves, that have higher clearance rates and thus generate stronger flows, do not have higher chances of capturing adult copepods. This would suggest that the filtration of adult copepods in the experiments was more an accidental process, while the filtration of nauplii was directly related to the bivalve pumping rates. In the experiments with adult copepods, oysters usually showed higher clearance rates on copepods than mussels, at similar CRalgae. Methods differed slightly between mussel and oyster experiments, but the fact that there were no significant changes in copepod concentration in the controls in both methods may suggest that the difference in methods has not affected the results. An explanation for the difference between oysters and mussels might be that the larger size of the oysters, in comparison to the mussels, increased the chance of copepods being captured in the flow field generated by the oysters. This may have determined the higher mortality of the adult copepods in the oyster experiments. The results of the lab experiments indicate that predation on naupliar stages of copepods by bivalves is possible, and this is consistent with earlier experimental evidence of bivalve filtration on copepod nauplii (Kimmerer et al. 1994, Davenport et al. 2000, Green et al. 2003). In addition, our results show that adult copepods are susceptible to filtration as well, although mortality rates are lower than for nauplii, likely as a consequence of stronger escape responses (Green et al. 2003). Finally, mussels and oysters may differ in their impact on adult copepods. The mesocosm study was designed to make a comparison between experimental ecosystems where mussels were dominant consumers, compared to systems without any benthic consumers and hence characterized by an entirely pelagic food web. In the systems with mussels the initial mussel biomass, in relation to the volume of the mesocosm, was 2.1 g ADW m-3. Measured filtration rates of the mussels in the experimental systems indicated that the mussels filtered the entire water volume of the systems every 5-10 days. Similar bivalve densities and grazing pressures have been estimated for
46 estuarine systems with high bivalve biomass and residence times in the order of 10-40 days (Smaal and Prins 1993, Heip et al. 1995). A mortality rate of the planktonic community of 10-20% due to mussel filtration is high enough to limit development (Herman and Scholten 1990, Heip et al. 1995). In the systems without mussels, several species of copepods developed peak numbers. T. longicornis developed highest numbers, comparable to concentrations of natural populations (Fransz et al. 1992, Bakker and Van Rijswijk 1994), showing that the mesocosms were suitable systems for the development of a copepod community. Earlier experiments had already shown that phytoplankton development in the mesocosms largely resemble seasonal trends in Dutch coastal waters (Prins et al. 1999). To establish the effects of the mussels on the development of the pelagic community, a comparison can be made between the MUS treatment and the ZOO treatment. The most straightforward test for mussel filtration effects is a comparison of biomass levels of the various plankton groups. For phytoplankton, this comparison showed that total phytoplankton concentrations, measured as chlorophyll-a, were significantly reduced in the presence of mussels. This is not a surprising observation, as similar effects have been observed in experimental ecosystems (Riemann et al. 1988, Olsson et al. 1992, Granéli et al. 1993, Prins et al. 1995), and have been inferred from in situ observations (see Dame 1993, 1996 and references therein) and model simulations (Officer et al. 1982, Herman and Scholten 1990). Estimates of phytoplankton biomass from microscopic counts showed that not all phytoplankton groups responded similarly, however. There were significant differences between treatments in diatom biomass, but differences in Phaeocystis sp. biomass were not significant. In the microzooplankton, ciliates showed no significant difference between treatments, due to the fact that the replicate mesocosms showed substantial variability. In contrast to ciliates, heterotrophic dinoflagellate biomass was severely reduced in the MUS treatment, illustrating the effect of mussel filtration. Copepod biomass reached much lower levels in the MUS treatment. Significant effects were not equally clear for all species. Whereas the effects on nauplii and on the copepodid stages of the two most abundant species (P. ( minutus and T. longicornis) showed significant reductions in the mesocosms with mussels, differences for adults of these species and for the species with lower abundance ((A. clausi, C. hamatus) were not significant. This is assumed to be a consequence of higher within-treatment variability and of a lower precision in microscopic counts due to lower numbers, instead of an indication of differential effects of mussel filtration. There was no indication that there were differential effects of mussel grazing on the various copepod stages either. It remains an open question whether copepod development was directly limited through filtration by the mussels, or indirectly by food competition between copepods and mussels. In the mesocosms with mussels there was some development of adult copepods in the course of the
47 experiment, which might be an argument against food competition. On the other hand, copepod biomass was proportional to primary production, which could be interpreted as an indication of food limitation. Although it is uncertain whether it is a direct or an indirect effect of mussel grazing, the results clearly indicate a strong effect of mussels on copepod abundance. Within the fifty-day period of the experiment, the various components of the planktonic food web showed substantial temporal dynamics. In the systems without bivalves, two blooms of phytoplankton developed. The initial bloom was dominated by diatom species, and collapsed around day 120, coincident with low DIP and silicate concentrations and a peak in biomass of heterotrophic dinoflagellates and copepods. Copepods are significant consumers of the larger phytoplankton species that are formed by the diatoms but heterotrophic dinoflagellates are also known to ingest diatom chains and these larger protozoans compete with copepods for diatom-prey (Maar et al. 2002). Peak biomasses of heterotrophic dinoflagellates observed in the ZOO treatment may have been high enough to constitute a significant grazing impact. Nutrient limitation and grazing may both have contributed to the decline in diatom biomass after the first peak. The haptophyte Phaeocystis sp. developed several small peaks during the first 30 days of the experiment, but the main bloom of Phaeocystis occurred after day 130, reaching high levels not only in treatment ZOO, but in mesocosm MUS-1 as well. Ciliates are dominant grazers on pico- and nanoflagellates (Brussaard et al. 1995, Granéli and Turner 2002, Escaravage and Prins 2002) and may have exerted a control on Phaeocystis development. The peak levels reached by Phaeocystis in both treatments were inversely related to the biomass levels of ciliates in the preceding weeks (compare Figures 5 and 6). Mussels can filter algae in the size range of Phaeocystis sp., but grazing by mussels seems to have been insufficient to efficiently control the algae, and ciliate grazing determined bloom levels to a large extent. High biomass levels were reached by heterotrophic ciliates (aloricate species), with substantial variability between replicate mesocosms (within treatments). Ciliates are known to be favorite prey for copepods and copepods have been shown to reduce ciliate populations (Sherr et al. 1986, Kivi et al. 1993, Kiørboe 1998, Paffenhöfer 1998, Granéli and Turner 2002). The control of ciliate biomass by copepods is illustrated by biomass levels in ZOO-1 that showed a sharp decline after the peak in copepod biomass, and relatively low ciliate biomass levels in ZOO-2 where copepod densities were high. Thus, ciliate biomass was likely controlled by copepod predation in the ZOO treatment. In the MUS treatment, top-down control on ciliates by copepods was much lower due to the strongly reduced copepod biomass. Mussel filtration may have exerted a top-down control on ciliate development, while at the same time the reduction in copepod abundance had a positive effect on ciliate biomass. Ciliate biomass in the MUS treatment was not reduced as much as other plankton components (e.g. diatoms, dinoflagellates and copepods) and this seems to indicate that mussel filtration was less effective
48 in controlling ciliates. The relatively short generation time of ciliates (circa 12 days, Sherr et al. 1986) is a factor that may have helped ciliates escape from grazing control by the mussels. As a consequence, the relative contribution of ciliates in the total biomass of pelagic herbivores was higher in the mesocosms with mussels. The differences between treatments seem small, but small protozoans like ciliates have a much higher specific ingestion rate than larger mesozooplankton, and a small shift in biomass composition in the zooplankton towards protozoans will result in a much larger shift in grazing intensity, in combination with a shift in prey size towards grazing on smaller phytoplankton. While there have been some observations showing at least the potential for predation on micro- and mesozooplankton species by bivalves (Kimmerer et al. 1994, Le Gall et al. 1997), we are aware of only one other study specifically describing effects of bivalve grazing on a pelagic consumer community. In enclosure experiments, described by Horsted et al. (1988), mussel grazing had significant effects on tintinnid ciliates and rotifers, but not on copepods. The difference between these observations and our results can be explained by differences in the design of the experiments. The enclosure experiments by Horsted et al. (1988) lasted for only 2-3 weeks, which is relatively short compared to the generation time of copepods, and the time scale of the experiment may have prevented a significant response in copepod development. Moreover, mussel filtration rates were estimated to be high enough to filter the entire volume of the enclosures every 1-2 days, which means that the imposed loss rates of the plankton were very high. Unless there is a high water renewal rate, ensuring supply of new food, we believe that a system with such a high grazing pressure of shellfish would not be able to sustain the bivalve biomass present in the systems for a long time. In our study, we used a mussel biomass in the range found in Dutch coastal systems with a major shellfish community, like the Oosterschelde estuary or the western Dutch Wadden Sea. The duration of the experiment was long enough to allow a build-up of micro- and mesozooplankton biomass. Our results show that in the mesocosms with an entirely pelagic food web, copepods and heterotrophic dinoflagellates controlled large phytoplankton. Copepods exerted a top-down control on ciliate biomass by predation, and this resulted in a cascading effect on the biomass of small flagellates (like Phaeocystis sp.) that were controlled through grazing by ciliates. High copepod biomass reduced ciliate biomass resulting in enhanced blooms of small flagellates. This pattern has been observed by various authors (a.o. Hansen et al. 1993, Escaravage and Prins 2002, Granéli and Turner 2002). With a well-developed ‘classical food web’, a large fraction of the primary production may be transferred up the food chain. In the mesocosms with mussels, the classical food chain with large diatoms and mesozooplankton was strongly diminished, as a consequence of the grazing by the mussels. The microbial food web appeared to be less affected. Ciliates partly profited from the reduced copepod predation. In the mesocosms with
49 mussels, bacterial production was relatively high and ciliates formed a relatively large fraction of the herbivore biomass, indicating that the microbial food web was relatively more important. This suggests that in a system dominated by bivalve suspension-feeders; the flow of energy through the classical food web may be low, ultimately resulting in limited production of higher trophic levels like fish (Legendre 1990).
ACKNOWLEDGEMENTS Part of this work was supported by the European Commission under the MAST-III program (contract no MAS3-CT96-0053-PHASE). We thank Arjen Pouwer, Cynthia van der Voorn and Peter de Vries for technical assistance during the mesocosm experiments, and Ainhoa Blanco García and Silvia Mastache Rubio for their assistance in the lab experiments.
REFERENCES Bakker C Van Rijswijk P 1994 Zooplankton biomass in the Oosterschelde (SW Netherlands) before, during and after the construction of a storm-surge barrier. Hydrobiologia 282/283: 127-143 Brussaard CPD Riegman R Noordeloos AAM Cadée GC Witte H Kop AJ Nieuwland G Van Duyl FC Bak RPM 1995 Effects of grazing, sedimentation and phytoplankton cell lysis on the structure of a coastal pelagic food web. Mar Ecol Prog Serr 123: 259-271 Carlson DJ Townsend DW Hilyard AL Eaton JF 1984 Effect of an intertidal mudflat on plankton of the overlying water column. Can J Fish Aquat Scii 41: 1523-1528 Coughlan J 1969 The estimation of filtering rate from the clearance of suspensions. Mar Biol 2: 356-358 Dame RF 1993 Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes. Springer-Verlag, Heidelberg, 579 p Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Davenport J Smith RJJW Packer M 2000 Mussels Mytilus edulis: significant consumers and destroyers of mesozooplankton. Mar Ecol Prog Serr 198: 131-137 Dupuy C Le Gall S Hartmann H Bréret M 1999 Retention of ciliates and flagellates by the oyster Crassostrea gigas in French Atlantic coastal ponds: protists as a trophic link between bacterioplankton and benthic suspension-feeders. Mar Ecol Prog Serr 177: 165175 Dupuy C Vaquer A Lam-Höai T Rougier C Mazouni N Lautier J Collos Y Le Gall S 2000 Feeding rate of the oyster Crassostrea gigas in a natural planktonic community of the Mediterranean Thau lagoon. Mar Ecol Prog Serr 205: 171-184 Escaravage V Peperzak L Prins TC Peeters JCH Joordens JCA 1995 The development of a Phaeocystis bloom in a mesocosm experiment in relation to nutrients, irradiance and coexisting algae. Ophelia 42: 55-74 Escaravage V Prins TC 2002 Silicate availability, vertical mixing and grazing control of phytoplankton blooms in mesocosms. Hydrobiologia 484: 33-48
50 Escaravage V Prins TC Nijdam C Smaal AC Peeters JCH 1999 Response of phytoplankton communities to nitrogen input reduction in mesocosm experiments. Mar Ecol Prog Ser 179: 187-199 Fransz HG Gonzalez SR Cadée GC Hansen FC 1992 Long-term change of Temora longicornis (Copepoda, Calanoida) abundance in a Dutch tidal inlet (Marsdiep) in relation to eutrophication. Neth J Sea Res 30: 23-32 Granéli E Olsson P Carlsson P Granéli W Nylander C 1993 Weak 'top-down' control of dinoflagelate growth in the coastal Skagerrak. J Plankton Res 15: 213-237 Granéli E Turner JT 2002 Top-down regulation in ctenophore-copepod-ciliate-diatomphytoflagellate communities in coastal waters: a mesocosm study. Mar Ecol Prog Ser 239: 57-68 Green S Visser AW Titelman J Kiørboe T 2003 Escape responses of copepod nauplii in the flow field of the blue mussel, Mytilus edulis. Mar Bioll 142: 727-733 Hansen FC Reckermann M Klein Breteler WCM Riegman R 1993 Phaeocystis blooming enhanced by copepod predation on protozoa: evidence from incubation experiments. Mar Ecol Prog Serr 102: 51-57 Heip CHR Goosen NK Herman PMJ Kromkamp J Middelburg JJ Soetaert K 1995 Production and consumption of biological particles in temperate tidal estuaries. Progr Oceanogrr 33: 1-149 Herman PMJ Scholten H 1990 Can suspension-feeders stabilize estuarine ecosystems? In: Trophic Relationships in the Marine Environment, M Barnes R Gibson (Eds) Vol 35. Aberdeen University Press, Aberdeen, pp 104-116 Horsted SJ Nielsen TG Riemann B Pock-Steen J Bjørnsen PK 1988 Regulation of zooplankton by suspension-feeding bivalves and fish in estuarine enclosures. Mar Ecol Prog Serr 48: 217-224 Kimmerer WJ Gartside E Orsi JJ 1994 Predation by an introduced clam as the likely cause of substantial declines in zooplankton of San Francisco Bay. Mar Ecol Prog Serr 113: 8193 Kiørboe T 1998 Population regulation and role of mesozooplankton in shaping marine pelagic food webs. Hydrobiologia 363: 13-27 Kivi K Kaitala S Kuosa H Kuparinen J Leskinen E Lignell R Marcussen B Tamminen T 1993 Nutrient limitation and grazing control of the Baltic plankton community during annual succession. Limnol Oceanogrr 38: 893-905 Le Gall S Bel Hassen M Le Gall P 1997 Ingestion of a bacterivorous ciliate by the oyster Crassostrea gigas: protozoa as a trophic link between picoplankton and benthic suspension-feeders. Mar Ecol Prog Serr 152: 301-306 Legendre L 1990 The significance of microalgal blooms for fisheries and for the export of particulate organic carbon in the oceans. J Plankton Res 12: 681-699 Maar M Nielsen TG Richardson K Christaki U Hansen OS Zervoudaki S Christou ED 2002 Spatial and temporal variability of food web structure during the spring bloom in the Skagerrak. Mar Ecol Prog Serr 239: 11-29 Officer CB Smayda TJ Mann R 1982 Benthic filter feeding: a natural eutrophication control. Mar Ecol Prog Serr 9: 203-210 Olsson P Granéli E Carlsson P Abreu P 1992 Structuring of a postspring phytoplankton community by manipulation of trophic interactions. J Exp Mar Biol Ecoll 158: 249-266 Paffenhöfer G-A 1998 Heterotrophic protozoa and small metazoa: feeding rates and preyconsumer interactions. J Plankton Res 20: 121-133 Prins TC Escaravage V Smaal AC Peeters JCH 1995 Nutrient cycling and phytoplankton dynamics in relation to mussel grazing in a mesocosm experiment. Ophelia 41: 289-315 Prins TC Escaravage V Wetsteyn LPMJ Peeters JCH Smaal AC 1999 Effects of different Nand P-loading on primary and secondary production in an experimental marine ecosystem. Aquat Ecoll 33: 65-81 Riemann B Nielsen TG Horsted SJ Bjørnsen PK Pock-Steen J 1988 Regulation of phytoplankton biomass in estuarine enclosures. Mar Ecol Prog Serr 48: 205-215
51 Sherr EB Sherr BF Paffenhöfer G-A 1986 Phagotrophic protozoa as food for metazoans: a "missing" trophic link in marine pelagic food webs? Mar Microb Food Webs 1: 61-80 Smaal AC Prins TC 1993 The uptake of organic matter and the release of inorganic nutrients by bivalve suspension feeder beds. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, Dame RF (Ed). Springer-Verlag, Heidelberg, pp 271-298 Titelman J 2001 Swimming and escape behaviour of copepod nauplii: implications for predatorprey interactions among copepods. Mar Ecol Prog Serr 213: 203-213
MOTILE SUSPENSION-FEEDERS IN ESTUARINE AND MARINE ECOSYSTEMS
DAVID BUSHEK1 AND DENNIS M. ALLEN2 1
Haskin Shellfish Research Laboratory, Rutgers University, Port Norris, NJ, USA Baruch Marine Field Laboratory, University of South Carolina, Georgetown, SC, USA
2
Abstract: Motile suspension-feeders are defined as macroinvertebrates, fishes, and mammals that strain multiple small particles from the water column during each feeding event. Motile suspension-feeders include some of the most economically and ecologically important species in estuarine, coastal, and open ocean ecosystems. Major fisheries are based on fishes such as anchovies, herrings, and mackerels that feed either mainly or partially on plankton. A variety of mechanisms are used to procure, extract, and select food particles, but particles are not selected and targeted individually. Selectivity for certain-size particles coupled with pulses of high abundance can have major effects on food web and ecosystem structure. Although benthic and motile suspension-feeders share many characteristics, two major distinctions can be identified. Unlike benthic species, water column based forms do not create the kind of stable structures that provide habitat and affect hydrography. A unique character of the motile forms is the ability to migrate, often over long distances, effectively transferring nutrients and energy and otherwise impacting multiple ecosystems. Keywords: motile suspension-feeders, mobile links, nekton, Cnidaria, Arthropoda, fish
INTRODUCTION Traditionally, the term suspension-feeder suggests sessile benthic organisms such as bivalve molluscs, sponges, barnacles, bryozoans, tunicates and a variety of other sedentary invertebrates that feed by filtering plankton from the water column. There are, however, many suspension feeding invertebrates and vertebrates that are not sessile or benthic in nature, instead spending some or all of their lives in the water column. Our intent in this paper is to identify and characterize the roles of large motile organisms that feed on suspended particles in estuarine and marine systems. In this paper, we do not address zooplankton less than about 20 mm, even though we recognize the key role that these widely studied suspension-feeders play in estuarine and marine systems. Instead, we focus on the larger invertebrate and chordate suspension-feeders that have received much less attention. We will broadly characterize the feeding mechanisms and habits of these large motile suspension-feeders, and then identify and contrast their potential or 53 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 53–71. ©2005 Springer. Printed in the Netherlands.
54 known ecological roles with those of benthic suspension-feeders in coastal and marine ecosystems. A universally acceptable definition of the term ‘suspension-feeder’ does not appear to exist. The term is most often applied to those organisms that filter plankton, which are typically a few millimeters or less in size, from the water column. Indeed, a literature search with ‘suspension-feeder’ as a key phrase typically yields an abundance of articles on bivalve molluscs, barnacles, anemones, tunicates and a variety of other predominantly benthic organisms that feed on microscopic plankton. In stark contrast to this robust body of literature are relatively few that address larger motile forms. Sanderson and Wassersug (1990) provide a broad review of suspension feeding vertebrates that covers organisms ranging from tadpoles to bony fish to larger marine mammals and birds. They describe many mechanisms of suspension feeding and make the case that it is not limited to feeding on microscopic particles. In the broadest sense, any organism or object that occurs in the water column is ‘suspended’ regardless of the mechanism that keeps it off the bottom; even particles that rely on currents, differences in water density, or self-propulsion qualify. Thus, any organism feeding on a particle suspended in the water could be classified as a “suspension-feeder”. The shear breadth of this definition, however, makes it virtually trivial. Nobody would associate killer whales attacking a humpback whale with suspension feeding. For that reason, it is critical to recognize that the term suspension feeding unconditionally implies that food particles are consumed en mass from the water column. In fact, the term suspension-feeder is often used synonymously with filter feeder, although a distinction of these terms continues to be debated. The suspension-feeders that we discuss all employ some sort of straining and capture large numbers of small particles during any feeding event. According to our definition, motile suspension-feeders include cnidarians such as the moon jelly (Aurelia ( aurita) that force water through a mesh of tentacles as they pulse through the water, menhaden (the fish, Brevoortia spp.) that strain particles with their gill rakers as they swim with mouths open, and blue whales that engulf enormous parcels of water that they expel through the baleen straining structures in their mouths. Regardless of the size, mobility, or other characteristics of the particles consumed by these different animals, numerous items are consumed at the same time.
General Characteristics of Motile Suspension-feeders Each type of suspension-feeder consumes particles within a specific size range. That range is generally determined by the morphological structures that capture the particles. However, even harvested particles that fall within the range of size may be retained or ejected according to shape, specific gravity or chemical characteristics. Particle size is usually small relative to the
55 size of the consumer, the consumer often being several orders of magnitude larger than the particles it consumes. Young menhaden that measure several centimeters in length consume suspended particles including microbes that are measured in micrometers. Baleen whales that measure several meters in size consume prey that are several millimeters to a few centimeters in length. Suspension feeding is not limited to any particular trophic level or trophic interaction. For example, whales suspension feed on krill that in turn suspension feed on phytoplankton. Hence, suspension-feeders may be herbivores or carnivores or rely mostly on non-living organic materials (detritus). Many, such as menhaden, are likely to be omnivorous, consuming both phytoplankton and zooplankton (Castillo-Rivera et al. 1996). Benthic suspension-feeders are generally considered to be herbivores that feed predominantly on phytoplankton, although many also feed on microzooplankton. Consumption of larger zooplankton is generally thought to be rare among benthic suspension-feeders (Prins and Escaravage 2005). There are a variety of ways that suspended particles may be harvested. Nevertheless, each includes three basic steps. First, water containing suspended particles must be made accessible to the organism (i.e., moved past the particle capturing structure). Next, the particles suspended in a parcel of water must be separated from the water. Finally, the captured particles must be transported to the mouth. Like benthic suspension-feeders, motile suspension-feeders have evolved a variety of mechanisms to feed. Many are analogous or homologous to their benthic counter parts. Some of the organisms that we include in this treatment drift with currents relying on the currents to bring them to their food or their food to them. Others actively swim to areas where their food is concentrated and then they pursue highdensity patches of food particles. As a result, some suspension-feeders in the water column expend significant amounts of energy to feed. Interestingly, perhaps to maximize energy efficiency, some motile species can switch between suspension feeding and selective predation.
Examples of Pelagic Suspension-feeders Large, motile suspension-feeders in estuarine and marine water columns represent a diverse group of invertebrate and vertebrate taxa. Table 1 lists examples that include a variety of jellyfishes, many bony fishes, some cartilaginous fishes, and the large baleen whales. These organisms vary widely in size, behavior, and distributions, but all conform to our general definition of suspension feeding. Unlike most benthic suspension-feeders that consume phytoplankton and microplankton and are referred to as grazers and herbivores, these motile forms consume mostly copepod-sized zooplankton. Accordingly, many of these large motile suspension-feeders are considered predators and carnivores (e.g., Purcell et al. 2001a, Lazzaro 1987). We
56 recognize these kinds of consumers are suspension-feeders because they harvest masses of particles at once rather than target individual food items. Table 1. Major taxonomic groups that include motile suspension-feeders. The bony fishes identified below include some of the world’s most important commercial fishery species. Their relative importance is shown in Figure 1. Juvenile and adult forms of most of these fishes filter phytoplankton and copepod size zooplankton, but some of the anchovies and herrings alternate between suspension feeding and selective predation on larger zooplankton according to conditions. Not all members of these taxa are suspension-feeders. Not all taxa with suspension feeding forms are included. Phylum Cnidaria
Subphylum/Class Hydrozoa Scyphozoa Cubozoa
Examples Rathkea, Obelia Aurelia , Chrysaora Tamoya
Ctenophora
Lobata
Mnemiopsis (comb jelly)
Arthropoda
Crustacea
Euphausiids (krill) Mysids (opossum shrimps) Decapods (shrimps)
Chordata
Urochordata Chondrichthyes
Thaliaceans (salps) Rhincodontidae (carpet sharks) Rhincodon typus (whale shark, Atlantic and Pacific) Cetorhinidae (basking sharks) Cetorhinus maximus (basking shark, Atlantic and Pacific) Clupeidae (herrings) Brevoortia spp. (menhaden, Atlantic) Clupea harengus (Atlantic herring, N Atlantic) Sardinops sagax (pilchard, sardine, sprat, subspecies worldwide) Engraulidae (anchovies) Engraulis ringens (Peruvian anchoveta, SE Pacific) E. mordax (California anchovy, NE Pacific) E. encrasicolus (European anchovy, E Atl./Med.) E. japonicus (Japanese anchovy, W Pacific, So.Africa) Scombridae (mackerels) Scomber japonicus (chub mackerel, circumglobal) Trachurus spp. (jack mackerels, Pacific and Indian O) Myctophidae (lanternfishes) Mobulidae (manta rays ) Baleanidae (right whales) Baleanopteridae (rorqual whales)
Osteichthyes
Mammalia
Most gelatinous zooplankton are suspension-feeders. Jellyfishes of all sizes capture food from suspension as they drift and swim in the water column. Sometimes cnidarians are very abundant, and some hydrozoans occur at densities of hundreds of individuals per cubic meter. In this treatment of suspension-feeders, we include larger species of jellyfishes such as sea nettles (Chrysaora), moon jellies ((Aurelia), and sea wasps (Tamoya), but we
57 do not consider gelatinous forms that prey on small fishes such as the Portuguese man ‘o war’ ((Physalia). Many of the large jellyfishes are sufficiently motile to accomplish considerable lateral and vertical movements in all but the strongest tidal conditions. Shanks and Graham (1987) recorded swimming speeds of 9 m min-1 for Stomolophus meleagris. Ctenophores are another common group of gelatinous suspension-feeders in the water column. The comb jelly Mnemiopsis leidyi is widely distributed in estuaries and coastal seas and has probably received the most attention following its introduction and major expansion in the Black Sea. In oceanic systems, individual salps or colonies ranging in size from a few centimeters to a few meters may dominate the gelatinous zooplankton. Some species of crustaceans can also be important suspension-feeders in the water column. In polar waters, euphausid shrimp or krill often occur in massive swarms and comprise an important trophic link in transfering energy from phytoplankton to fish, birds and whales. Smaller decapod, mysid shrimp and copepod species play a similar role in some temperate coastal systems. Several species of large sharks, rays, and whales are classified as suspension-feeders. They display a wide range of morphological, physiological and behavioral characteristics. Most are from about one meter to tens of meters in length and have large heads and mouths, elongate shapes, and lack teeth. A majority of them are also migratory, moving long distances, to remote feeding grounds. Manta rays and whale sharks are cartilaginous fishes that tend to occur in tropical and sub-tropical habitats. Whale sharks ((Rhincodon spp.) are filter feeders that consume a wide variety of planktonic and nektonic organisms (Coleman 1997). Whale sharks are not dependent on forward motion to feed. Instead they employ a suction filter-feeding mechanism that enables them to capture faster swimming organisms. In contrast, the basking shark, Cetorhinus maximus, feeds more or less continuously as it moves through the water. Table 1 lists some of the most important families of bony fishes that suspension feed. Probably less than 10% of all species of bony fishes are suspension-feeders, but many of these make up important commercial fisheries (Figure 1). Herrings, anchovies, and mackerels are probably the best known and account for a majority of the annual worldwide fishery harvest. These schooling suspension feeding fishes are most prevalent in temperate and boreal waters where upwelling occurs and primary production is high, but some species occur in tropical areas. Lanternfishes and many less familiar deep ocean species may play significant roles in those systems, but almost nothing is known about their feeding habits. Largest among the suspension-feeders are the baleen whales. Baleen whales occur throughout the worlds oceans and migrate great distances to locate high concentrations of energy rich foods. Feeding is accomplished by engulfing large parcels of water that are then strained by the baleen as the water is expelled.
58 Mechanisms of Suspension- feeding in the Water Column Although a variety of mechanisms have evolved for motile organisms to suspension feed, two primary types dominate among vertebrates. The first is through continuous straining or filtering of water caused by forward motion. In this case, water is more or less continuously delivered to a filtering apparatus while the organism swims. This is sometimes called “ram feeding” or “tow net filtering”. Examples include most of the bony fishes; these are the fishes that are often referred to as filter feeders. A second method is often called “intermittent straining” or “gulp filtering”. This method also involves forward motion, but the motion is used to engulf a parcel of water that is subsequently expelled and processed (filtered) to remove suspended food before the next parcel of water is captured. Baleen whales use this technique. Invertebrates, particularly those that are more typically classified as megaloplankton or macrozooplankton may utilize similar methods, but several employ unusual alternative mechanisms including the use of bristled appendages to actively strain particles from water. Salps employ suction feeding in which water is actively drawn into a filtering apparatus via suction created by a pumping motion. Mucous/tentacular entanglement is employed by a wide variety of jellyfish as they swim or drift with currents. In some cases, a mucous net is dragged behind the organism to entangle suspended particles. The net is then consumed to remove the food particles.
Selectivity of Suspended Particles Given the broad range of sizes, composition, and motility among suspended particles in the water column, no single consumer is expected to harvest all particles. Selection for certain subsets of all available particles is the norm and distinguishes the various suspension-feeders from one another. Suspension feeding is not an entirely indiscriminant process. As a result, diets are often not representative of potential food resources or plankton communities (Larson 1991). Rather, consumed prey represent only those food particles that are captured and retained. In cnidarians and ctenophores, prey selection results primarily from the differences in the avoidance abilities of potential prey items before contact or post contact sorting. Most cannot detect, capture and consume specific suspended items from a field of available particles. However, gelatinous pelagic suspension-feeders are known to consume more of some types of particles than others do. This is often accomplished mechanically through the separation of suitably sized particles from smaller ones that pass with ejected water through the straining structures; larger suspended particles do not
59 usually make contact with the primary straining structures (Larson 1991). In a review of how pelagic coelenterates interact with fishes, Purcell and Arai (2001) note a variety of preferences among coelenterate species for eggs and different sizes of larval fishes. The distinction between suspension feeding and selective predation is sometimes blurred in the cnidarians. For instance, some jellyfishes may retract or extend their tentacles in such a way to suggest they were actively seeking particular prey types (Collin et al. 2003). Vertebrate planktivores can be classified into two functional groups, non-visual filter feeders and visual particulate feeders (Leong and O’Connell 1969). Non-visual filter feeders encounter prey and consume them without bias; they tend to consume smaller and less evasive prey than visual particulate feeders in the same water mass. Visual feeders identify and consume prey particles of choice. Recognizing that the distinction between visual particulate feeding and predation is not always clear and that the inclusion of visually oriented consumers falls outside of the interests of most readers interested in suspension feeding, we generally avoid using examples of these kinds of consumers in this paper. However, several of the most abundant fishes in the water column conform to our definition of suspension feeding, at least some of the time. Although most vertebrates are entirely visual particulate or filter feeders, some fishes can switch in response to changes in conditions (O’Connell 1972) or during ontogeny (Drenner et al. 1982). Population or community dynamics such as prey density, prey size or species composition may also result in shifts in feeding mode. Changes in the size distribution and/or densities of suspended particles are known to affect feeding behavior and rates of consumption by planktivorous fishes in a variety of marine systems (O’Connell 1972; Durbin and Durbin 1975; Hunter and Dorr 1982; Gibson and Ezzi 1985). For example, the anchovy Engraulis encrasicolus (Plounevec and Champalbert 2000) and the sea herring Clupea harengus (Batty et al. 1990) can choose to selectively feed on larger zooplankton prey or indiscriminately filter feed on smaller prey. Off the coast of South Africa, the Cape anchovy ((E. capensis) switches between filtering and biting according to densities of prey of a particular size; swimming speed, turning rate, and feeding mode of these fishes also depend on prey size (James and Findlay 1989). Whereas anchovies tend to favor particulate feeding, the co-occurring pilchard, Sardinops sagax, is primarily a filter feeder consuming microzooplankton and, to a lesser extent, phytoplankton (van der Lingen 1994). Selection of different size fractions of the available particles in the water column enables multiple taxa and life stages of suspension-feeders to coexist. Ontogeny is another factor controlling selection of suspended particles of different sizes. Stated simply, juveniles and adults acquire prey of different sizes. Costello and Colin (1994) suggested that selection in the jellyfish A. aurita is determined by whether or not zooplankton have escape speeds slower than the flow velocities at the bell margin (which is related to the diameter of the bell). Difference in size or age may also determine
60 whether the organism uses filter and particulate feeding modes. Increasing size or the development of certain feeding structures may enable larger organisms to suspension feed when smaller ones cannot. There may be an energetic requirement that necessitates the switch. For example, while larval and juvenile fishes may select prey items individually, it may become energetically inefficient to do so as the animal grows larger. Menhaden ( (Brevoortia spp.) smaller than 40 mm (TL) selectively feed on individual prey, but become obligate filter feeders as larger adults (Lazzaro 1987). Environmental conditions may also signal shifts in feeding mode. For example, changes in light (day/night or turbidity) may require that a predator shift between a visual predation and a less selective suspension-feeding mode. Tudela and Palomera (1997) observed that E. encrasicolus filters smaller prey during the day, presumably in deeper water, and sporadically feeds selectively near the surface during the night. In laboratory experiments, Batty et al. (1986) showed that C. harengus filter feeds on brine shrimp larvae in the dark and reverts to selective filter feeding on the same prey by day. Batty et al. (1990) demonstrated that the sea herring uses the filtering mode during the day to capture the same prey as long as densities are above a certain threshold, below which level the fish employs particulate feeding. Even closely related fishes select different subsets of suspended particles. This has been demonstrated in the co-occurring menhaden species Brevoortia gunteri and d B. patronus. Differences in morphological features of their feeding structures result in different diets (Castillo-Rivera et al. 1996). The gill raker system in B. patronus forms a finer mesh compared to B. gunteri. As a result, B. patronus has a diet that is dominated by phytoplankton while B. gunteri has a diet dominated by zooplankton. Similarly, in a comparison of the gill rakers and jaw and branchial dentitions of eight anchovy species in a Venezuelan estuary, Bornsbusch (1988) demonstrated that differences in the food particles selected by the various species were related to differences in morphological structures. The ecological implication of differences in particle size selectivity is that co-occurring species can minimize competition by partitioning the available resources. This occurs in the upwelling ecosystems where anchovies that consume large particles and sardines that filter small particles coexist in high densities (van der Lingen 2002). The morphology and behavior of zooplankton and other motile prey can also affect diet or prey selection (Greene 1985). For example, some prey are more evasive than others. In general, larger more mobile prey are better able to escape both non-selective filter feeders and particulate feeders (Purcell 1985, Bailey and Houde 1989). Larger suspension-feeders are generally better able to capture larger, faster prey. Slow moving or non-motile zooplankters such as fish eggs are typically more susceptible to capture than larvae, and younger larvae are typically more susceptible than older larvae. Some prey may escape gelatinous consumers through chemical or mechanical discharges that affect nematocyst or colloblast responses by the medusa (Purcell 1997).
61 Suspended organic particles (detritus) of a variety of sizes and composition may be an important source of food for some suspension feeding fishes (Peters and Schaaf 1981). In considering the matter of selection, it is also important to point out that not all large particles are retained and not all smaller particles pass through the filtering device. As a result of the various mechanisms of selection, diets of suspension-feeders are often not representative of the community of suspended organisms on which they are feeding. Nevertheless, suspension-feeders can be opportunistic. Lewis and Peters (1994) attribute differences in diet between Atlantic menhaden in estuarine creeks from those in coastal waters to spatial differences in the composition of the three principle food items (amorphous suspended matter, phytoplankton and zooplankton). Knowledge about the selection of different sizes and kinds of particles by different suspension-feeders within a system is necessary to assess ecological impacts of the suspension-feeders.
Roles of Large Suspension-feeders in Marine Fisheries and Ecosystems World fishery harvest statistics indicate that the most important fisheries are comprised largely of suspension feeding fishes. Coastal ocean species such as anchoveta, herring, anchovy, mackerels, and capelin, account for about 75% of the harvests (Figure 1). Among the high seas fisheries, suspension feeding fishes comprise less than 20% of the total harvest. The combined catch of the jellyfish fisheries is much less than the dominant finfish Jellyfish (all)* Blue whiting Capelin Chub mackerel* Largehead hairtail Chilean jack mackerel* Japanese anchovy* Skipjack tuna Atlantic herring* Alaska pollock Anchoveta* 0
1
2
3
4
5
6
7
8
9
10
11
12
Year 2000 production (millions of tonnes)
Figure 1. Annual fishery production for 2000. Except for jellyfishes, only those species with production exceeding one million metric tonnes are shown. Those marked by an asterisk (*) are considered suspension-feeders. Data are from FAO statistics.
62 fisheries, but still significant at around 500,000 metric tons (Figure 1). Jellyfish fisheries are important for China and other Asian countries. Despite their importance as a fishery product, jellyfishes can also be nuisances to other fisheries, negatively impacting them via predation, clogging and damaging nets, contributing to by-catch, and slowing the cooling rates of the catch once harvested (Kingsford et al. 2000). The magnitude of the fishery harvests indicates the high abundances of suspension feeding fishes in some (often small) areas and suggests that they play important roles in estuarine and marine ecosystems. The distribution of suspension feeding finfish and jellyfish fisheries corresponds to areas of highest primary productivity. For example, anchoveta and other anchovies are usually associated with upwelling areas where high phytoplankton and zooplankton densities can support large populations of small schooling fishes. In other nutrient (and thus phytoplankton) enriched estuarine and coastal ocean areas, herrings, mackerels, and a variety of other water column species occur in sufficiently high densities to support fisheries. Jellyfish fisheries are typically located nearshore, often in bays, estuaries, fjords and reef matrices (Kingsford et al 2000). The attraction of other consumers including marine mammals and birds to these areas results in dynamic and complex inter-relations among the constituents of the food webs, and suspension-feeders play key roles. Motile suspension-feeders respond to a variety of physical and biological gradients. Jellyfishes are known to form dense aggregates in many marine areas. Graham et al. (2001) suggested that the passive accumulation of medusae by water circulation alone can not usually explain this phenomenon, leading them to speculate about behavioral responses to various physical gradients (e.g., light, gravity, temperature, salinity, pressure and turbulence). Although stochastic events play important roles in the distribution and impacts of suspension-feeders in estuaries and oceans (Brodeur and Pearcy 1992), directed movements for the purpose of encountering increased densities of food are well documented. In the eastern North Sea, interannual fluctuations in the timing of the departure of sea herring from the traditional fishing grounds was related to the availability of the copepods (Corten 2000). Whale shark migrations appear to coincide with massive coral spawning events off Australia (Taylor 1994), and Heyman et al. (2001) recently documented large whale shark aggregations off the coast of Belize to feed on snapper spawn. Clark and Nelson (1997) documented that young whale sharks ((Rhincodon typus) selectively feed on dense patches of the copepod Acartia clausi. Sims and Quayle (1998) demonstrated that basking sharks actively select areas along thermal fronts to forage on specific and discrete zooplankton assemblages. Sims and Reid (2002) found a coincidental decline between basking sharks and copepods off Ireland over a 27 year period; however, reductions in zooplankton abundance was attributed to climate and water mass changes rather than the impact of shark feeding.
63 Although many factors operating at many temporal and spatial scales influence the structure of food webs and ecosystems, pelagic suspensionfeeders have been shown to exert major influences in some coastal areas. Local impacts of the concentrated effort of large suspension-feeders on small prey populations may be expected. Abundance and size characteristics of jellyfish A. aurita in the Baltic Sea varies considerably among years, but when it reaches peak densities, it can consume more than 60% of the daily production of copepods and other zooplankton (Schneider and Behrends 1994). During some summers, these medusae may be responsible for the decline of mesozooplankton (Schneider and Behrends 1998). A. aurita can affect the population dynamics of a diverse group of zooplankton species (DeLafontaine and Leggett 1988, Bamstedt 1990, Behrends and Schneider 1995). High densities of small hydromedusae were found to considerably reduce copepod numbers in the North Sea, and the presence of various other gelatinous suspension-feeders at different times likely influenced copepod abundances throughout the year (Nicholas and Frid 1999). Purcell (1992) used estimates of medusae densities and feeding rates to conclude that the sea nettle, Chrysaora quinquecirrha, may have caused an observed copepod population decline in tributaries of the Chesapeake Bay; the medusae consumed up to 94% of the copepod standing stock per day. Crustacean suspension-feeders such as euphausiids (krill) can have huge effects on phytoplankton production often consuming about 50% and occasionally removing 100% of the daily primary production in Antarctic waters (Pakhomov et al. 2002). Salps, which can also be abundant in Antarctic waters, are equally effective at removing primary production. Salps and krill usually remain spatially separated in different water masses (Pakhomov et al. 2002), but each of these motile suspension-feeders plays an important role in the south polar food web. Behrends and Schneider (1995) showed that as function of the preference for some copepod species over others, interannual variations in medusae abundance caused shifts in the trophic structure of the zooplankton community. Top down controls on the abundance and composition of mesoplankton may affect primary producers. In the surface waters of a Danish fjord, reductions in zooplankton densities by jellyfish consumption and the subsequent reduction in grazing impact on phytoplankton was thought to be responsible for increases in chlorophyll (Riisgård 1998). Another cascading trophic effect was reported by Feigenbaum and Kelly (1984) when decreases in ctenophore densities by a large medusa lead to an increase in zooplankton and subsequent increase in phytoplankton in a tributary of the Chesapeake Bay. The impacts of ctenophores as agents of change in pelagic communities have been reported from many systems, including the Black Sea (Oguz et al. 2001, Purcell et al. 2001b). In western Canada, ctenophores may largely control the standing stock of zooplankton by early summer (Parsons, et al. 1970; Frank 1986) and have an impact on salmonid fish growth and survival (Parsons and Kessler 1987). In Nova Scotia waters, reduced growth rates of
64 larval cod fish appeared to be due to ctenophore consumption of the same copepod prey (Suthers and Frank 1990). Cowan and Houde (1993) concluded that sea nettle and ctenophore may consume 20-40% daily of the fish eggs and larvae in mid Chesapeake Bay. Diet overlaps between suspension-feeding jellyfishes, ctenophores, and fishes occupying the same area can be as high as 63% (Purcell and Sturdevant 2001). Purcell and Arai (2001) present a review of the interactions of these suspension-feeders and address the ecological and economic consequences of changes in their dominance and roles. Interactions between suspension feeding fishes and other components of marine systems are also complex. Just as motile suspension-feeders can influence certain ecosystem processes, changes in prey availability and environmental conditions can influence the dynamics of suspension-feeder populations. Flinkman et al. (1998) discussed the effects of both bottom-up and top-down processes on interannual variations in zooplankton and herring nutrition in the northern Baltic Sea. Salinity and shifts in mesozooplankton composition among years were both thought to affect herring growth. Largescale changes in resident or migratory suspension-feeder populations may have consequences for local and remote systems at some later time. In some marine systems, suspension-feeders are likely to play important roles in nutrient cycling via excretion and assimilation of different nutrients resulting from metabolic processes. Suspension-feeders in the partially eutrophicated Baltic Sea may influence nutrient dynamics. Hjerne and Hansson (2002) suggested that because fishes accumulate phosphorus in body tissues, suspension-feeding forms might compete with primary producers, and thus play a significant role in ecosystem structure. Tudela and Palomera (1999) found that in a major spawning area of the northwestern Mediterranean Sea, anchovy E. encasicolus excretions accounted for at least 2% of the nitrogen required for sustained phytoplankton production. Even more interesting is the suggestion that nocturnal vertical migrations by E. ensasicolus from deep daytime feeding areas resulted in a net upward flux of nitrogen into the upper photic layer. There is clear evidence of interactions between certain suspensionfeeders and harmful algal blooms (HAB), but many relationships are not yet well understood (Lansberg 2002). Fish kills due to direct toxic effects or oxygen depletion resulting from phytoplankton blooms may have important local consequences for food web structure following the dissipation of bloom conditions. Reductions in grazing pressure could trigger blooms by other phytoplankters and zooplankton. A different interaction has been proposed in association with recent outbreaks of the toxic dinoflagellate Pfiesteria piscicida in North Carolina, USA. Namely, menhaden migrating to feed on algal blooms may have triggered metamorphosis of non-toxic stages of P. piscicida into toxic stages (Burkholder et al. 1995), but this hypothesis is being intensely debated (Dykstra and Kane 2000). Given the potentially significant roles of motile suspension-feeders, it is perhaps surprising that few ecosystem models include them. When they are
65 considered, they are usually either lumped with other suspension feeding components or nekton categories that also include predators. One reason that these distinctions are often lacking is that the feeding mode and diets of many even common consumers is simply not known. Sometimes, jellyfishes are separated from fishes or fishes are separated into suspension and predatory types, but modelers rarely make both distinctions. This is even more surprising considering that spatial patchiness and migrations of schools into and out of systems is expected to dramatically affect system models. Unfortunately, rates of consumption, assimilation, excretion, and energy transfer are difficult to measure for motile organisms. Hence, due to the high spatial and temporal variability associated with both the resources and the processes, quantifying the roles of suspension feeding nekton in dynamic estuarine and open marine systems is difficult. Including suspension-feeding categories in food web and ecosystem models provides opportunities to understand their roles during different seasons and between different locations. In a comparison of trophic structure in three U.S. mid-Atlantic estuaries, Monaco and Ulanowicz (1997) determined that relationships between phytoplankton production, mesozooplankton, and suspension feeding fishes varied considerably among three temperate major U.S. Atlantic coast estuaries even though similar dominant taxa occurred in all systems.
Comparisons Between Benthic and Non-benthic Suspension-feeders
In general, benthic suspension-feeders and their roles in ecosystems have received much more attention than water column-based counterparts, probably because it is logistically easier to measure a wider variety of parameters on sessile organisms than motile organisms. Nevertheless, the two groups share much in common (Table 2). Both consume and impact plankton and other suspended particles, remove biological and non-living contaminants from the water column, circulate nutrients, couple benthic and pelagic subsystems, and are responsible for transferring carbon and energy within ecosystems. One fundamental difference is that benthic organisms can create habitats that, in many cases, themselves comprise major ecological systems (e.g. bivalve, coral, and polychaete reefs). By altering hydrodynamics, such structures can have broad implications for ecosystems. Some pelagic cnidarians, particularly large species that swarm, create structure in the water column that can attract other species much in the same way certain species are attracted to sargassum (Kingsford et al. 2000, Brodeur 1998). But these structures are ephemeral relative to the hard structures formed by many benthic suspension-feeders.
66 Perhaps the single largest functional difference between benthic and pelagic suspension-feeders is the ability of mobile forms to serve as vectors that transfer nutrients and energy among ecosystems that are not linked by circulation patterns. The interconnectivity of coastal and oceanic areas with very different physical, chemical, and biological characteristics enables migratory fishes to play active roles in the dynamics of multiple ecosystems, a function that is not performed by motile freshwater species. Migrations of suspension feeding fishes that occur at scales ranging from a few to thousands of kilometers are known in all of the oceans. For example, American shad on the Atlantic coast of the United States leave their natal rivers in the Southeast region as subadults and migrate to the Gulf of Maine, a distance of more than 1000 km, to over-summer before returning to spawn (Bigelow and Schroeder 1953). Seasonally and life history induced movements between estuaries and oceans (Deegan 1993; Durban and Durbin 1998) and between geographic regions result in transfer of carbon, nutrients, and energy between systems that would otherwise not be linked.
Table 2. Comparison of ecosystem functions between benthic and motile suspension-feeders. X = perform function, O = do not perform function.
Function remove plankton remove other suspended particles remove contaminants excrete nutrients couple bottom and water column transfer carbon/energy within systems create habitat alter water flow dynamics transfer carbon/energy among systems
Benthic
Motile
X X X X X X X X O
X X X X X X O O X
We are not aware of any direct quantitative comparisons of the roles of motile suspension-feeders in freshwater and marine systems, but we can identify some likely differences. Because diversity of taxa and the number of trophic links tend to be lower in most freshwater systems, food webs tend to be simpler and the roles of specific suspension-feeders may be easier to quantify than in marine systems. Indeed, the pioneering work and a large body of research on the impacts of changes in suspension feeding components is from lakes (see Ojaveer, this volume). In confined water bodies, impacts from disturbances including invasive species may be more likely to affect an entire system. We might also expect more rapid responses to disturbances and freshwater systems may be slower to recover than coastal and ocean areas where key species are widely distributed and capable of moving in response
67 to adverse conditions or in search of more favorable conditions. Given the capacity of many marine nekton to migrate long distances, system-wide changes such as the complete failure or extinction of a type of plankton or motile suspension-feeder seems unlikely.
Motile Suspension-feeders in a Changing World The impacts of humans on the world’s ecosystems are increasing rapidly and motile suspension-feeders often have no better refuge than benthic forms. Changes in global climate are likely to change distributions, abundances and roles of suspension feeding invertebrates and fishes in coastal and marine systems. Coincidental with global warming in recent years, changing growth rates, timing of blooms, magnitude of production, and patterns of migrations have been demonstrated for a variety of species and ecosystems. Large-scale climate fluctuations may serve as a major source of interannual variability in jellyfish populations in the Gulf of Mexico (Graham 2001), the Bering Sea (Brodeur et al. 1999, Brodeur et al 2002) and Narragansett Bay, Rhode Island, USA (Sullivan et al. 2001). Temperature changes may be responsible for synchronous multi-decadal shifts in the dominance of anchovies and sardines throughout the Pacific Ocean (Chavez et al. 2003). Alternating warm and cool phases during the 20th century resulted in major biological regime shifts that affected primary production, zooplankton, fishes, and seabird abundance. These multi-decadal changes fluctuations operate at longer scales than more frequently recognized large-scale climatic phenomena such as El Nino and La Nina, which account for interannual fluctuations in the abundance of the prevailing fish species and many other ecosystem features (Chavez et al. 2003). In an analysis of historical records for anchovy, sardine, and herring landings around the world, Schwartzlose et al. (1999) discovered near simultaneous fluctuations in fish stocks occurred in widely separated areas and suggested that these patterns were due to climate effects operating at a global scale. Changes in water quality in coastal systems are also known to affect suspension-feeders in various ways. Mills (2001) concluded that many new jellyfish blooms are a response to the cumulative effects of human activities. It is difficult, however, to separate effects of eutrophication from overfishing, pollution, introductions, and global warming, but in those cases where the link is strongest, most species decline in numbers in response to conditions of eutrophication (Arai 2001). Nonetheless, in some northern marine systems, a few species (e.g., the hydromedusae Aglantha digitale and Rathkea octopunctata and the scyphomedusae A. aurita and Cassiopea spp.) may increase in numbers when water quality is degraded, most likely as a result of adaptations that enable them to tolerate low oxygen levels (Condon et al. 2001) or other environmental conditions (Arai 2001).
68 Introductions of non-native suspension-feeders can have major impacts on food web structure. One of the most intensively studied cases on large-scale introductions in aquatic systems is in the Black Sea where the introduction of the ctenophore, Mnemiopsis leidyi, has had a significant impact on the ecology and economics of the system. When uncontrolled by predation, this ctenophore can affect both zooplankton and fish populations (Purcell and Arai 2001) out-compete other jellyfish for zooplankton (Shusukina et al. 2000). Interestingly, the introduction of the ctenophore Beroe ovata, which preys on M. leidyi, corresponded to a dramatic decline in M. leidyi populations in the Black Sea and a concomitant increase in abundance of its competitor A. aurita (Shusukina et al. 2000). Regardless of the mechanisms of change, fluctuations in the abundance and distributions of large motile suspension-feeders have significant implications for fisheries, related economies and cultures, and ecosystems.
REFERENCES Arai MN 2001 Pelagic coelenterates and eutrophication: a review. Hydrobiologia 451: 69-87 Bailey KM Houde ED 1989 Predation on eggs and larvae of marine fishes and the recruitment problem. Adv Mar Bioll 25: 1-25 Bamstedt U 1990 Trophodynamics of scyphomedusae Aurelia aurita: predation rate in relation to abundance, size and type of prey organism. J Plank Res 12: 215-229 Batty RS Blaxter JHS Libby DA 1986 Herring (Clupea harengus) filter-feeding in the dark. Mar Bioll 91: 371-375 Batty RS Blaxter JHS Richard JM 1990 Light intensity and the feeding behaviour of herring, Clupea harengus. Mar Bioll 107: 383-388 Behrends G Schneider G 1995 Impact of Aurelia aurita medusae (Cnidaria, Scyphozoa) on the standing stock and community compositions of mesozooplankton in the Kiel Bight (western Baltic Sea). Mar Ecol Prog Serr 127: 39-45 Bigelow HB Schroeder WC 1953 Fishes of the Gulf of Maine. US Fish Wildl Serv Fish Bull 74: 1-577 Bornsbusch AH 1988 Gill raker morphologies of anchovies (Teleostei: Engraulidae) from the Rio Orinoco, Venezuela. Copeia 1988: 174-182 Brodeur RD 1998 In situ observations of the association between juvenile fishes and scyphomedusae in the Bering Sea. Mar Ecol Prog Serr 163: 11-20 Brodeur RD Mills CE Overland JE Walters GE Schumacher JD 1999 Evidence for a substantial increase in zooplankton in the Bering Sea, with possible links to climate change. Fish Oceanogrr 8: 296-306 Brodeur RD Pearcy WG 1992 Effects of environmental variability on trophic interactions and food web structure in a pelagic upwelling ecosystem. Mar Ecol Prog Serr 84: 101-119 Brodeur RD Sugisaki H Hunt GL 2002 Increases in jellyfish biomass in the Bering Sea: implications for the ecosystem. Mar Ecol Prog Serr 233: 89-103 Burkholder JM Glasgow HB Jr Hobbs CH 1995 Fish kills linked to a toxic ambush-predator dinoflagellate: distribution and environmental conditions. Mar Ecol Prog Serr 124: 4361 Castillo-Rivera M Kobelkowsky A Zamayoa V 1996 Food resource partitioning and trophic morphology of Brevoortia gunteri and B. patronus. J Fish Bioll 49: 1102-1111
69 Chavez FP Ryan J Lluch-Cota SE Niquen CM 2003 From anchovies to sardines and back: a multidecadal change in the Pacific Ocean. Science 299: 217-221 Clark E Nelson DR 1997 Young whale sharks, Rhincodon typus, feeding on a copepod bloom near La Paz, Mexico. Environ Biol Fishes 50: 63-73 Collin SP Costello JH Klos E 2003 In situ swimming and feeding behavior of eight cooccurring hydromedusae. Mar Ecol Prog Serr 253: 305-309 Coleman JG 1997 A review of the biology and ecology of the whale shark. J Fish Biol 51:1219-1234 Condon RH Decker MB Purcell JE 2001 Effects of low dissolved oxygen on survival and asexual reproduction of scyphozoan polyps (Chrysaora quinquecirrha). Hydrobiologia 451: 89-95 Corten A 2000 A possible adaptation of herring migrations to a change in timing of the Calanus finmarchicus season in the eastern North Sea. ICES J Mar Sci 57: 1261-1270 Costello JH Colin SP 1994 Morphology, fluid motion and predation by the scyphomedusa Aurelia aurita. Mar Bioll 121: 327-334 Cowan JH Jr Houde ED 1993 Relative predation potentials of scyphomedusae, ctenophores and planktivorous fish on ichthyoplankton in Chesapeake Bay. Mar Ecol Prog Ser 95: 55-65 Deegan, LA 1993 Nutrient and energy transport between estuaries and coastal marine ecosystems by fish migration. Can J Fish Aquat Scii 50: 74-79 DeLafountaine Y Legget WC 1988 Predation by jellyfish on larval fish: an experimental evaluation employing in situ enclosures. Can J Fish Aquat Scii 445: 1173-1190 Drenner RW de Noyelles F Jr Kettle D 1982 Selective impact of filter feeding gizzard shad on zooplankton community structure. Limnol Oceanogrr 27: 965-968 Durbin AG Durbin EG 1975 Grazing rates of the Atlantic menhaden Brevoortia tryrannus as a function of particle size and concentration. Mar Bioll 33: 265-277 Durbin AG Durbin EG 1998 Effects of menhaden predation on plankton populations in Narragansett Bay, Rhode Island. Estuaries 21: 449-465 Dykstra MJ Kane AS 2000 Pfiesteria piscicida and ulcerative mycosis of Atlantic menhaden – Current status of understanding. J Aquat Animal Health 12: 18-25 Feigenbaum D Kelly M 1984 Changes in the lower Chesapeake Bay food chain in presence of the sea nettle Chrysaora quiquecirrha (Scyphomedusa). Mar Ecol Prog Serr 19: 39-47 Flinkman J Aro E Vuorinen I Viitasalo M 1998 Changes in northern Baltic zooplankton and herring nutrition from 1980’s to 1990’s: top-down and bottom-up processes at work. Mar Ecol Prog Serr 165: 127-136 Frank KT 1986 Ecological significance of the ctenophore Pleurobrachia pileus off southwestern Nova Scotia. Can J Fish Aquat Scii 43: 211-222 Gibson, RN Ezzi IA 1985 Effect of particle concentration on filter- and particulate-feeding in the herring Clupea harengus. Mar Bioll 88: 109-116 Graham WM 2001 Numerical increases and distributional shifts in Chryaora quiquecirrha (Desor) and Aurelia aurita (Linne’) (Cnidaria: Scyphozoa) in the northern Gulf of Mexico. Hydrobiologia 451: 97-111 Graham WM Pages F Hamner WM 2001 A physical context for gelatinous zooplankton aggregations: a review. Hydrobiologia 451(1-3): 199-212 Greene CH 1985 Planktivore functional groups and patterns of prey selection in pelagic communities. J Plankton Res 7(1): 35-40 Heyman WD Graham RT Kjerfve B Johannes RE 2001 Whale sharks Rhincodon typus aggregate to feed on fish spawn in Belize. Mar Ecol Prog Serr 215: 275-282 Hjerne O Hansson S 2002 The role of fish and fisheries in Baltic Sea nutrient dynamics. Limnol Oceanogrr 47: 1023-1032 Hunter JR Dorr H 1982 Thresholds for filter feeding in northern anchovy, Engraulis mordax. Rep Calif coop oceanic Fish Investt 13: 198-204
70 James AG Findlay KP 1989 Effect of particle size and concentration on feeding behaviour, selectivity and rates of food ingestion by the Cape anchovy Engraulis capensis. Mar Ecol Progr Serr 50: 275-294 Kingsford, MJ Pitt KA Gillanders BM 2000 Management of jellyfish fisheries, with special reference to the Order Rhizostomeae. Oceanogr Mar Biol: Ann Revv 38: 85-156 Lansberg JH 2002 The effects of harmful algal blooms on aquatic species. Rev Fish Sci 10: 113-390 Larson, RJ 1991 Diet, prey selection, and daily ration of Stomolophus meleagrus, a filter– feeding Scyphomedusa from the NE Gulf of Mexico. Est Coastal Shelf Sci 32: 511-525 Lewis VP Peters DS 1994 Diet of juvenile and adult Atlantic menhaden in estuarine and coastal habitats. Trans Am Fish Soc 123: 803-810 Lazzaro X 1987 A review of planktivorous fishes: their evolution, feeding, behaviors, selectivies, and impacts. Hydrobiologia 146: 97-167 Leong RJH O’Connell CP 1969 A laboratory study of particulate and filter feeding of the northern anchovy (Engraulis ( mordax). J Fish Res Board Can 27(3): 557-582 Mills CE 2001 Jellyfish blooms: are populations increasing globally in response to changing climate conditions? Hydrobiologia 451: 55-68 Monaco ME Ulanowicz RE 1997 Comparative ecosystem trophic structure of three midAtlantic estuaries. Mar Ecol Prog Serr 161: 239-254 Nicholas KR Frid CLJ 1999 Occurrence of hydromedusae in the plankton off Northumberland (western central North Sea) and the role of planktonic predators. J Mar Biol Assoc UK K 76: 979-992 O’Connell CP 1972 The interrelation of biting and filtering in the feeding activity of the northern anchovy (Engraulis ( mordax). J Fish Res Bd Canada 29: 285-293 Oguz T Ducklow HW Purcell JE Malonette-Rizzoli P 2001 Modeling the response of topdown control exerted by gelatinous carnivores on the Black Sea pelagic food web. J Geophysical Res Oceans 106: 4543-4564 Pakhomov EA Froneman PW Prissinotto R 2002 Salp/krill interactions in the Southern Ocean: spatial segregation and implications for carbon flux. Deep-Sea Res Pt II – Topical Stud Oceanogrr 49: 1881-1907 Parsons TR Kessler TA 1987 An ecosystem model for the assessment of plankton production in relation to the survival of young fish. J Plank Res 9:125-137 Parsons TR LeBrasseur RJ Barraclough WE 1970 Levels of production in the pelagic environment of the Strait of Georgia, British Columbia: a review. J Fish Res Bd Canada 27: 1251-1264 Peters DS Schaaf WE 1981 Food requirements and sources for juvenile Atlantic menhaden. Trans Am Fish Soc 110: 317-324 Plounevec S Champalbert G 2000 Diet, feeding behaviour and trophic activity of the anchovy (Engraulis encrasicolus L.) in the Gulf of Lions (Mediterranean Sea). Oceanol Acta 23: 175-192 Prins T Escaravage V 2005 Can bivalve suspension-feeders affect pelagic food web structure? In: The Comparative Roles of Suspension-feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 31-52 Purcell JE 1985 Predation on fish eggs and larvae by pelagic cnidarians and ctenophores. Bull Mar Sci 37: 739-755 Purcell JE 1992 Effects of predation by the scyphomedusan Chrysaora quinquecirrha on zooplankton in Chesapeake Bay, USA. Mar Ecol Prog Serr 87: 65-76 Purcell, JE 1997 Pelagic cnidarians and ctenophores as predators: selective predation, feeding rates, and effects on prey populations. Ann I oceanogr Paris 73: 125-137 Purcell JE Arai MN 2001 Interactions of pelagic cnidarians and ctenophores with fish: a review. Hydrobiologia 451: 27-44 Purcell JE Graham WM Dumont HJ (Eds) 2001a Jellyfish Blooms: ecological and societal importance. Developments in hydrobiology 155. Kluwer Academic Publishers, Dordrecht, 334 pp. Reprinted from Hydrobiologia 451.
71 Purcell JE Shiganova TA Decker MB Houde ED 2001b The ctenophore Mnemiopsis in native and exotic habitats: U.S. estuaries versus the Black Sea Basin. Hydrobiologia 451: 145-176 Purcell JE Sturdevant MV 2001 Prey selection and dietary overlap among zooplanktivorous jellyfish and juvenile fishes in Prince William Sound, Alaska. Mar Ecol Prog Serr 210: 67-83 Riisgård HU 1998 Filter feeding and plankton dynamics in a Danish fjord: a review of the importance of flow, mixing and density-driven circulation. J Environ Mgmtt 53: 195207 Sanderson SL Wassersug R 1990 Suspension-feeding vertebrates. Sci Am 262: 96-101 Schneider G Behrends G 1994 Population dynamics and the trophic role of Aurelia aurita medusae in the Kiel Bight and western Baltic. ICES J Mar Sci 51: 359-367 Schneider G Behrends G 1998 Top-down control in a neritic plankton system by Aurelia aurita medusae: a summary. Ophelia 48: 71-82 Schwartzlose RA plus 20 authors 1999 World-wide large-scale fluctuations of sardine and anchovy populations. So Afr J Mar Sci 21: 289-347 Shanks AL Graham WM 1987 Oriented swimming in the jellyfish Stomolophus meleagris L. Agassiz (Schyphozoan: Rhizostomida). J Exp Mar Biol Ecoll 108: 159-169 Shusukina EA Musaeva EI Anokhina LL Lukasheva TA 2000 The role of gelatinous macroplankton jellyfish, Aurelia, and ctenophores Mnemiopsis and Beroe in the planktonic communities of the Black Sea. Oceanoll 40: 809-816 Sims DW Quayle VA 1998 Selective foraging behavior of basking sharks on zooplankton in a small-scale front. Nature 393: 46-464 Sims DW Reid PC 2002 Congruent trends in long-term zooplankton decline in the north-east Atlantic and basking shark (Cetorhinus maximus) fishery catches off west Ireland. Fish Oceanogrr 11: 59-63 Sullivan BK Van Keuren D Clancy M 2001 Timing and size of blooms of the ctenophore Mnemiopsis leidyi in relation to temperature in Narragansett Bay, RI. Hydrobiologia 451: 113-120 Suthers IM Frank KT 1990 Zooplankton biomass gradient off south-western Nova Scotia: nearshore ctenophore predation or hydrographic separation? J Plankton Res 12: 831850 Taylor G 1994 Whale Sharks, the Giants of Ningaloo Reef. f Angus & Robertson, Sydney, 176 p Tudela S Palomera I 1997 Trophic ecology of the European anchovy Engraulis encrasicolus in the Catalan Sea (northwest Mediterranean). Mar Ecol Prog Ser 160: 121-134 Tudela S Palomera I 1999 Potential effect of an anchovy-mediated pump on the vertical availability of nitrogen for primary production in the Catalan Sea (northwest Mediterranean). J Sea Res 42: 83-92 van der Lingen CD 1994 Effect of particle size and concentration on the feeding behaviour of adult pilchard Sardinops sagax Mar Ecol Prog Serr 109: 1-13 van der Lingen CD 2002 Diet of the sardine Sardinops sagax in the southern Benguela upwelling ecosystem. So Afr J Mar Sci 24: 301-31
IMPACT OF SUSPENSION-FEEDING NEKTON IN FRESHWATER ECOSYSTEMS: PATTERNS AND MECHANISMS
Henn Ojaveer Estonian Marine Institute, University of Tartu, Tallinn, Estonia Abstract:
Suspension-feeding fish influence several abiotic and biotic components of freshwater ecosystems. Amongst others, abiotic components include water transparency and retention of essential nutrients. Biotic effects include alterations to bacterioplankton, phytoplankton, zooplankton, phytobenthos and zoobenthos communities. Specifically, species composition, size structure, abundance and biomass of plankton communities can substantially be altered. In addition, growth, reproduction and life history of zooplankton can be changed. Examples of predation with cascading food-web effects or changes in nutrient dynamics largely via excretion are common. More often than not, it is the integrated impacts of the above factors that are responsible for the observed changes.
Key words: lakes, fish, direct and indirect effects, biomanipulation
INTRODUCTION Suspension-feeding is one of the oldest modes of feeding amongst vertebrates, and possibly the most widespread, which evolved further to take individual food particles (Mallatt 1984). In fact, some fish species can switch between the two different feeding modes (e.g., Kajak et al. 1975; Unger et al. 1984), complicating the identification and classification of true suspensionfeeding behaviour. The textbook definition suggests (Gee 1991) that suspension feeders differ in size from their food particles by two orders of magnitude. Thus, it is generally energetically inefficient for suspension feeders to pursue, capture and handle food particles individually. Suspension feeders can roughly be classified as: ram feeders, which rely on forward motion to deliver water into the mouth, and suction feeders, which suck in water while remaining essentially stationary. Both the ram feeders and suction 73 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 73–92. ©2005 Springer. Printed in the Netherlands.
74 feeders are further divided into continuous feeders, which pass water through the mouth continuously, and intermittent feeders which essentially take one gulp at a time and extract the food from it before refilling the mouth (Sanderson and Wassersug 1990). Suspension-feeding fish are important in ecological terms because they occupy a relatively low position in the food chain and consume high amounts of planktonic invertebrates and/or detritus. Economically, suspension-feeding fish account for more than one third of the annual catch of marine and freshwater fish (Sanderson and Wassersug 1990). Although climate has been shown to influence freshwater planktonic communities (e.g., Georg and Harris 1985) and indirectly influence nutrient dynamics in lake ecosystems by altering fish recruitment (Horppila et al. 1998), predation by suspension-feeding fish is the primary factor influencing the structure of lower trophic levels in lakes. Recently, Krivtsov et al. (2001) have shown that incorporation of zooplankton and planktivorous fish into an aquatic ecosystem model dramatically improved the fit of model simulations with observations down to the nutrient level. This result emphasizes the importance of planktivorous fish in terms of structure and function of aquatic ecosystems. Based on the results of eleven different case studies, Lyche (1989) has summarized phyto- and zooplankton responses to reduction of planktivorous fish in lakes with different trophic status and discussed possible ‘top-down’ but also ‘bottom-up’ mechanisms governing the observed changes. The present review is substantially wider in scope and attempts to: (1) reveal the magnitude and direction of responses of selected abiotic (water transparency and nutrient dynamics) and biotic (bacterio- phyto- and zooplankton and benthos) components of freshwater ecosystems to suspension-feeding fish, (2) summarize the proposed mechanisms responsible for such changes, (3) illustrate the complex, multiple and cascading nature of impacts of suspension-feeding fish in aquatic ecosystems, and (4) discuss the application of the gathered knowledge to biomanipulation of waterbodies.
IMPACTS OF SUSPENSION-FEEDING FISH TO ABIOTIC ENVIRONMENT Water Transparency Changes in water transparency, observed in both enclosure experiments and field studies, is mainly related to indirect effects of fish via food consumption. By removing large bodied zooplankton, planktivorous fish reduce phytoplankton grazing. Reduced grazing pressure allow phytoplankton to increase in abundance and biomass that decreases water transparency. For example, water transparency decreased significantly in experimental enclosures with planktivorous young-of-the-year (YOY) perch Perca fluviatilis (Bertolo et al. 2000) or bluegill Lepomis macrochirus (Drenner et al.
75 1990) compared to controls without fish (Table 1). Similarly, a significant increase in water transparency was observed in fishless enclosures in a shallow hypereutrophic lake compared to the enclosures containing an obligate filter-feeding planktivore, the silver carp Hypophthalmichthys molitrix (Tang et al. 2002). Finally, at a much larger scale, Scavia et al. (1986) documented a significant increase in water clarity following a decline of the major open water planktivorous alewife Alosa pseudoharengus in Lake Michigan from 1975-1984. Nutrient dynamics In oligotrophic systems, planktivorous fish can be significant sources of phosphorus. Based on their own experiments with zooplanktivorous fish (the cyprinid Phoxinus eos) in an unproductive lake and results of earlier studies, Vanni et al. (1997) concluded that zooplanktivorous fish may have strong effects on the dynamics of and fluxes of the essential nutrients. It appears that fish consistently increase water column total phosphorus (TP) and particulate phosphorus concentrations, consistently decrease the mean particle size of particulate phosphorus and have variable effects on total nitrogen (TN) concentration. The high treatment of planktivorous fish Lepomis auritus (ca 25 m3 mesocosm with 18 1-year old fish) added more than 2.5 times as much phosphorus as the zooplankton in the control treatment and the same fish treatment had the highest phosphorus excretion rates (PerezFuentetaja et al. 1996). Fish also influence nutrient ratios in the zooplankton fraction of the water column. When large cladocerans ((Daphnia and Holopedium) increased upon exclusion of fish, C:P and N:P ratios of the zooplankton fraction showed distinct declines, corresponding to the relatively high body contents of these taxa. Measured loss of P from the water column through sedimentation and wall growth increased with fish biomass suggesting that the presence of fish increases the relative retention of P in the water column (Vanni et al. 1997). Fish impacts to nutrient dynamics seems to be irrespective to trophic status of a waterbody. This is evident from an experimental study in an eutrophic humic lake carried out by Attayde and Hansson (2001b) who observed postive effects of fish to both to TP and TN concentrations. Predation induced shifts in plankton size structure (by favouring the dominance of smaller plankton, which contribute more to total biomass and less to sedimentation) affects epilimnetic total phosphorus, through changes in sedimentation rates. That is, fish reduce the fraction of epilimnetic TP sedimentation and in this way inhibit the seasonal decline in TP (Mazmuder et al. 1989). Despite these studies, nutrients do not always respond to fish manipulations. For instance, different enclosure experiments with various silver carp H. H molitrix concentrations (0-316 g m-2) did not result in any significant nutrient (TN and TP) changes in a shallow eutrophic lake (Tang et al. 2002). Likewise, in a study with four different planktivorous fish,
76 no significant effects on nutrients were detected in a tropical eutrophic water of Paranoa Reservoir, Brazil, even in cases in which there was fish death and decomposition that caused short-term increases of nutrient concentrations (Starling and Rocha 1990). However, in the two last cases, it could be that the response is non-linear and the fish density threshold level was not reached.
IMPACTS OF SUSPENSION-FEEDING FISH TO BIOTA Bacteria The impact of suspension-feeding fish on bacterioplankton is poorly known and existing evidence is to some extent controversial. Nonetheless, results of the limited number of studies that have been conducted indicate that the influence from suspension feeders to aquatic bacteria may be substantial. Additions of planktivorous fish to experimental enclosures of eutrophic lake water increased bacterial production (up to 4-fold) although the bacterial cell number did not change significantly (Riemann and Sondergaard 1986). In another pond study, the presence of the silver carp H. molitrix increased and stabilized bacterial abundance to 18-25 million cells/ml by constant consumption of the secondary bacterial production (up to 97%) by the fish (Kuznetsov 1980). Fish stock alterations may markedly affect the grazing rate of zooplankton on bacterioplankton. Enclosure experiments in eutrophic lakes show that zooplankton grazing rate of bacterial production can increase 10 times (from 4.6 to 48-51%) when planktivorous fish are removed (Riemann 1985). Results of in vitro grazing experiments in a shallow hypertrophic lake show that decreasing planktivorous fish densities can increase grazing on bacterioplankton production by more than 15 times (Jeppesen et al. 1996). In contrast, Elliott et al. (1983) did not find any treatment effects of fish manipulations on free-living planktonic bacterial numbers in the water column. They concluded that planktonic bacteria are not a significant part of epilimentic systems. In another study in a shallow eutrophic lake, bacteria were generally insensitive to the gradient of silver carp H. H molitrix biomass -3 (initial biomass range 0-54.1 g m ; Fukushima et al. 1999). However, it cannot be excluded that in the last two studies, the fish density gradient did not exceed the threshold value.
Phytoplankton The impact of suspension-feeding fish upon phytoplankton abundance, biomass and community composition has been a major focus of
77 both laboratory enclosure and field studies. Fish-induced changes have been observed for a variety of species and in waterbodies with different trophic status and food-web structures. Changes in algal abundance, biomass, biovolume, size structure and species composition have been observed (Table 1). In a study using large enclosures with multiple trophic levels and water from Tuesday Lake, Michigan, Vanni and Layne (1997) found that, in addition to total phytoplankton biovolume, abundances of 13 phytoplankton taxa (belonging to dinoflagellates, chrysophytes, green algae, diatoms and cryptomonads) and total phytoplankton biomass increased in the presence of fish compared to enclosures without fish. In another study, Attayde and Hansson (2001a) observed a significant increase in total algal biovolume in the presence of the planktivorous perch P. fluviatilis in an eutrophic humic Swedish lake. The increase was mainly due to an increase in the biovolumes of cyanobacteria, diatoms and cryptomodads while a decrease in the total algal biovolume in the control was mainly due to the disappearance of cyanobacteria. In a study by Drenner et al. (1984), a comparison of ponds containing an omnivorous pump filter feeder, the gizzard shad Dorosoma cepedianum, with control ponds showed that the fish suppressed the dinoflagellate Ceratium, the only phytoplankton species large enough to be ingested at a maximum rate. In contrast, the presence of D. cepedianum had no effect on populations of Synedra, Peridinium, Navicula, Kirchneriella, Cyclotella and Chlamydomonas, but abundances of Ankstrodesmus, Cryptomonas, Cosmarium and Rhodomonas were enchanced (Drenner et al. 1984). The presence of bluegill L. macrochirus in an outdoor mesocosm experiment significantly increased diatoms, unicellular green algae, colonial blue-green algae and filamentous blue-green algae (Drenner et al. 1990). A field mesocosm study showed that presence of sockeye salmon (Oncorhynchus nerka) fry significantly increased chlorophyll a content not only by increasing the biomass of the grazable component of the phytoplankton community (Chlamydomonas spp. and Cryptomonas spp.), but also stimulating blooms of the ungrazable Dinobryon relative to fishless controls (Schindler 1992). The remaining ungrazable components of the phytoplankton community exhibited no significant treatment response. Similarly, McDonald (1985) recorded significant enchancement of grazed algal communities by the filter-feeding blue tilapia Tilapia aurea in a laboratory experiment: cell densities of the green algae Ankistrodesmus falcatus increased 2-7 times above the ungrazed cell densities when grazed by the tilapia. Finally, Perin et al. (1996) found that the presence of hybrid redbelly dace ((Phoxinus sp.) in either shallow or deep enclosures of a stratified lake resulted in a 2-fold increase in total phytoplankton productivity. The fish had no effect on pikoplankton but increased productivity of nanoplankton by 1.5 times (Perin et al. 1996). Based on studies in which planktivorous Phoxinus sp. were added to enclosures installed at different depths in a dimictic temperate lake, Proulx et al. (1996) concluded that
78 changes in phytoplankton community structure may be a more general outcome of fish manipulations (regardless of zooplankton community structure) than changes in total biomass and chlorophyll a. In another study performed with the same fish species by Fukushima et al. (1999), suppression of cyanobacterial blooms and small nanoplankton, together with domination of picocyanobacteria in the phytoplankton community in the fish enclosures was observed. However, in difference of results from many other studies, some large filamentous phytoplankton still persisted in relatively high abundances in the fish enclosures. The above study also demonstrated, in difference of many other studies, that the silver carp H. molitrix is able to reduce the total algal biomass in a specific conditions – in a simple system with no large herbivorous zooplankton (Fukushima 1999). Not all the suspension-feeding fish have similar effects on phytoplankton. This was demonstrated in a study that compared the effects of several facultative planktivores: Congo tilapia Tilapia rendalli, bluegill L. macrochirus, tambaqui Colossoma macropomum and the silver carp H. molitrix. While presence of the first three species was associated with increases of the dominant blue-green algae, Cylindrospermopsis reciborskii, silver carp clearly reduced the algal abundance (Starling and Rocha 1990). Zooplankton Numerous short-term experimental studies have been carried out to describe and explain the impact of planktivorous fish on zooplankton community structure and biomass. In general, the presence of fish substantially reduces zooplankton biomass and shifts species dominance from larger (e.g., Daphnia and Holopedium) to smaller taxa (Bosmina ( , copepod nauplii). For instance, such results have been obtained in eutrophic lakes with the silver carp (Tang et al. 2002) and enclosure experiments using eutrophic lake water containing yellow perch (Post and McQueen 1987, Mazumder et al. 1990). However, there exist several examples of deviations from the general pattern. For instance, Attayde and Hansson (2001a) observed a doubling of zooplankton biomass, due mainly to an increase in rotifer, Bosmina, Ceriodaphnia and Cyclops biomass, when planktivorous perch P. fluviatilis were present in enclosures with eutrophic and humic lake water. In a study with bluegill L. macrochirus, only rotifers increased in abundance whereas cladocerans, cyclopoid copepodids and copepod nauplii decreased (Drenner et al. 1990). Unfortunately, the latter study gives only the abundance-based results and therefore it remains unclear whether the total zooplankton biomass incerased or decreased. In treatments with the pump filter-feeding Sacramento blackfish Orthodon microlepidotus by Byers and Vinyard (1990) zooplankton densities were lower and significant reductions relative to control tanks was observed for all cladocera and copepods and most rotifer taxa whereas the
79 only zooplankter showing significant increase was Kellicottia. The addition of mosquitofish Gambusia holbrooki to experimental tank communities containing Daphnia further reduced the abundance of some small zooplankton (Moina, Keratella) but, in contrast to the general theory for temperate lakes, increased the numbers of large zooplankters - Daphnia and adult Boeckella. The addition of juveniles of Australian golden perch Macquaria ambigua to the same communities reduced the abundance of small zooplankton (Moina), increased the abundance of large zooplankton (adult Boeckella) and had no significant effect on Daphnia (Matveev et al. 2000). In a long-term study carried out in an eutrophic Chinese lake, Lake Donghu in 1957-1996 it was demonstrated that continuous increase in the planktivorous fish abundance, silver carp H. molitrix and bighead carp Aristichthys nobilis, resulted finally in a density decrease of an invertebrate predator Leptodora kindti although both fish and L. kindti exhibited a density increase during the first 25 years of study. This initial increase was due to considerable increase in the densities of their zooplankton prey. The latter abundance decline of L. kindti was accompanied by remarkable decline of their body length, presumably due to increased fish predation (Xie et al. 2000). In a long-term (1979-1995) biomanipulation experiment in mesotrophic lakes where all planktivorous fish were remoced from an experimental lake, the immediate response was characterised by a strong increase in both Daphnia biovolume and size and an increase in the abundance of the invertebrate predator Chaoborus flavicans. After 12 years, the larger and darker C. obscuripes successfully colonised and displaced the small C. flavicans. Subsequently, zooplankton biovolume and size decreased dramatically due to heavy predation from small rotifers to large daphnids whereas zooplankton biovolume was substantially lower than in adjacent reference lake dominated by planktivorous fish (Wissel and Benndorf 1998). Fish may also exert major impact to important life history parameters and fecundity of zooplankton. Planktivorous YOY roach Rutilus rutilus influenced indirectly energy allocation pattern in the postembryonic development of Daphnia galeata: both the size of primiparae and the average egg volume were smaller in individuals reared from neonata in water inhabited by fish compared to the control. The smaller size of experimental primiparae resulted from smaller increments in some instars and from earlier maturation. There were also larger clutches in experimental individuals compared with the controls. Thus, in presence of predators, part of energy allocated to growth is reduced in favor of energy allocated to reproduction (Machacek 1991). Similarly, in another study, the dominating herbivorous Ceriodaphnia was observed to bear smaller eggs, have higher fecundity and reach earlier maturity in presence of planktivorous fish (YOY perch P. fluviatilis, Bertolo et al. 2000). In treatments containing the Sacramento blackfish O. microlepidotus the number of eggs per female Daphnia schodleri was higher in enclosures with fish and the numbers of females carrying
80 ephippia was significantly lower in the presence of fish (Byers and Vinyard 1990). Fish also have marked influence on some dynamic parameters of the zooplankton community: filtering rates of zooplankton were 2-5 times higher in the enclosures without fish than in manipulations with fish. This was largely due to the higher individual filtering rates of plankton. Microzooplankton had higher community filtering rates than mesozooplankton in the enclosures with fish, and sometimes were as great as mesozooplankton community filtering rates even in the enclosures without fish. Biomass-specific filtering and ingestion rates of both micro- and mesozooplankton were higher in the enclosures without fish (Mazumder et al. 1990).
Benthos The main scientific effort in studies with suspension-feeding fish has been directed towards changes in pelagic ecosystem. However, at least several of the recorded changes (e.g., water transparency, nutrient dynamics, plankton abundance, and sedimentation) should most likely have also an indirect effect to benthic life. Only a few direct evidences are available for changes in benthic communities. For instance, submerged macrophytes Potamogeton crispus and Elodea canadensis increased in abundance after reduction of planktivorous fish (Sondergaard et al. 1990). Presence of the silver carp H. molitrix at high densities, 450 and 1350 kg ha-1, resulted in increased total benthic biomass, but especially that of Oligochaeta by 2.6 and 5.6 times, respectively and decrease of the share of invertebrate predators in benthos over 2 times (Kajak et al. 1975). The flathead gray mullet Mugil cephalus strongly modified the benthic community (e.g., increase in mud snail densities was observed), presumably as a result of direct predation (Torras et al. 2000). These available results confirm that in suspension-feeding fish may have strong impact to benthic communities and this should reach much more attention in future when managing ecosystems. Table 1. Summary overview of multiple ecosystem effects of suspension-feeding fish in freshwater systems Fish species
System
P. fluviatilis H. molitrix P. flavescens
Eutrophic lake Hypereutrophic lake Eutrophic lake
D. cepedianum L. macrochirus
Pond Eutrophic pond
Impact Water Decrease Decrease
Parameter
Reference
Transparency Transparency
Bertolo et al. 2000 Tang et al. 2002
Decrease
Transparency
Decrease Decrease
Transparency Transparency
Post and McQueen 1987 Drenner et al. 1984 Drenner et al. 1990
81 Several fish
Eutrophic lake
Decrease
Transparency
H. molitrix
Eutrophic lake
Increase
Transparency
Nutrients O. microlepidotus Lake Manzanita Increase water P. fluviatilis Eutrophic humic Increase lake L. auritus Oligotrophic lake Increase L. macrochirus P. flavescens
Eutrophic pond Increase Lake St. George Increase
Ph. eos
Unproductive lake Eutrophic lake
Increase Variable Insignificant
Hypereutrophic lake
Insignificant
Several fish H. molitrix
H. molitrix Pimepheles promelas H. molitrix
Bacteria Pond Increase Eutrophic humic Insignificant lake Eutrophic lake Insignificant
Several fish
Eutrophic lake
G. holbrooki M. ambigua D. cepedianum
Eutrophic water Eutrophic water Pond
Several fish
Eutrophic lake
L. macrochirus
Eutrophic pond
H. molitrix
Eutrophic lake
H. molitrix
Eutrophic lake
H. molitrix
Pond
Ph. eos
Unproductive lake Hypereutrophic lake Unproductive lake Small softwater lake Pond Lake St. George
H. molitrix Ph. eos Phoxinus sp. D. cepedianum P. flavescens
Increase
Starling and Rocha 1990 Fukushima et al. 1999
DIP, NH4, N
Byers and Vinyard 1990 TN, TP Attayde and Hansson 2001b TP Perez-Fuentetaja et al. 1996 TP Drenner et al. 1990 Epilimnic TP Mazumder et al. 1989 TP, particulate P Vanni et al. 1997 TN Several nutrients Starling and Rocha 1990 TN, TP Tang et al. 2002
Abundance Abundance
Kuznetsov 1980 Elliott 1983
Abundance
Fukushima et al. 1999 Riemann and Sondergaard 1986
Production
Phytoplankton Increase Abundance Insignificant Abundance Variable Abundance of various species Increase Abundance of blue-green algae Increase Abundance of blue-green algae Decrease Abundance of blue-green algae Decrease Abundance of blue-green algae Increase Abundance of nanoplankton Increase Biomass
Matveev et al. 2000 Matveev et al. 2000 Drenner et al. 1984 Starling and Rocha 1990 Drenner et al. 1990 Fukushima et al. 1999 Starling and Rocha 1990 Milstein et al. 1988
Increase
Biomass
Vanni and Layne 1997 Tang et al. 2002
Increase
Biomass
Vanni et al. 1997
Variable
Biomass
Proulx et al. 1996
Decrease Increase
Biomass Biomass of nanoplankton
Drenner et al. 1984 Mazumder et al. 1990
82 P. promelas O. nerka Several fish
Experimental conditions Shallow pond Eutrophic lake
Phoxinus spp. Phoxinus spp.
Oligotrophic lake Increase Oligotrophic lake Increase
P. fluviatilis Ph. eos
Eutrophic lake Unproductive lake Eutrophic humic lake Small softwater lake
P. fluviatilis Phoxinus sp.
P. promelas P. fluviatilis H. molitrix
Experimental conditions Eutrophic lake
Increase
Production
Elliott 1983
Increase Increase
Production Production
Schindler 1992 Riemann and Sondergaard 1986 Perin et al. 1996 Perin et al. 1996
Increase Increase
Production Production of nanoplankton Chl a Biovolume
Increase
Biovolume
Decrease Zooplankton Decrease Decrease
Attayde and Hansson 2001a Size distribution Proulx et al. 1996
Abundance, size distribution Biomass, size distribution Biomass, size distribution Biomass, size distribution Biomass, size distribution Size distribution
P. flavescens
Hypereutrophic lake Eutrophic lake
P. flavescens
Lake St. George Decrease
L. auritus
Oligotrophic lake Decrease
Ph. eos
Unproductive lake
P. fluviatilis
Eutrophic humic Increase lake Eutrophic water Increase Eutrophic water Increase
Biomass
Eutrophic lake
Egg size Clutch size Maturity Primiparae size Egg size Clutch size Maturity Clutch size
G. holbrooki G. holbrooki M. ambigua P. fluviatilis
R. rutilus
Lab conditions
Decrease Decrease
Decrease
Decrease Increase Earlier Decrease Decrease Increase Earlier Increase
Bertolo et al. 2000 Vanni et al. 1997
Elliott 1983 Bertolo et al. 2000 Tang et al. 2002 Post and McQueen 1987 Mazumder et al. 1990 Perez-Fuentetaja et al. 1996
Size distribution Vanni et al. 1997
Attayde and Hansson 2001a Biomass Matveev et al. 2000 Size distribution Matveev et al. 2000 Bertolo et al. 2000
Machacek 1991
O. microlepidotus Lake Manzanita Byers and Vinyard water 1990 DIP – dissolved inorganic phosphorus, NH4 – ammonium, TN – total nitrogen, TP – total phosphorus
83 EXAMPLES OF COMPLEX ECOSYSTEM EFFECTS In most systems, suspension-feeding fishes exert multiple effects that create complex interactions and alter the structure and function of the ecosystem. To illustrate the multiple and complex cascading nature of ecosystem impacts attributable to suspension-feeding fish, four case studies are summarized below. Each of them is based on a single summary paper, referred to at the end of each case-study.
Lake Michigan Dramatic alterations to the ecosystem of Lake Michigan occurred during the 1970s and 1980s. Concurrently with reductions in both phosphorus load and concentration there was a sharp decline in abundance of the dominant zooplantkivore, the alewife A. pseudoharengus. Food-web model estimates indicated that increases in alewife abundance would be accompanied by: (1) decreases in total zooplankton biomass and a zooplankton composition shift from dominance of Daphnia to dominance off Diaptomus, (2) increases in total phytoplankton biomass with a change in composition from a dominace of flagellates to a mixture of flagellates and blue-green algae, (3) decreases in primary production and zooplantkon ingestion rates, (4) more dramatic decreases in phytoplankton growth rates than production rates, (5) increases in sedimentation rate, (6) decreases in P regeneration by zooplankton and available P concentrations, and (7) increases in zooplankton secondary production and growth rates with Diaptomus growth rates always below Daphnia growth rates. These simulated changes are similar to those observed for Lake Michigan during changing abundances of alewife, suggesting that shift in zooplanktivory was the main cause for the observed cascading effects (Scavia et al. 1988).
Lake Mendota A massive mortality of the planktivorous fish cisco, Coregonus artedii, in 1988 had a dramatic impact on the plankton community dynamics of lake Mendota, Wisconsin, USA. After the fish mortality event, the larger Daphnia pulicaria replaced the smaller Daphnia galea mendotae that dominated when cisco was abundant. Spring phytoplankton biomass and the duration of the spring clear-water period appear to be closely related to the presence or absence of cisco. When cisco are absent, spring phytoplankton biomass is relatively low and the duration of the clear-water period is long. When cisco is abundant, spring phytoplankton biomass is relativly high and the duration of the spring clear-water period short. Summer Daphnia biomass appears to be inversely correlated with cisco abundance. The effects of cisco
84 on summer phytoplankton appears to be more complex than the effects on spring phytoplankton, potentially as a result of nutrient and trophic cascade interactions (Vanni et al. 1990).
Lake Vaeng After removal of about 50% of the planktivorous/benthic fish (bream Abramis brama and roach Rutilis rutilus) in the shallow eutrophic lake Vaeng, Denmark, the zooplankton community changed from a dominance of rotifers to large cladocerans and zooplankton biomass increased 3-6 fold. Phytoplankton biomass decreased three-fold within three years. Qualitatively, the phytoplankton community shifted from a dominance of cyanobacteria and small diatoms to a dominance of larger diatoms, larger green algae and cryptophytes. Secci depth increased by a factor of two. A concurrent reduction of phosphorus concentration was solely ascribed to a lower net phosphorus release from the sediment, which was, in turn, attributed to changes caused by the fish manipulation. The reduced net release of phosphorus was attributed to several factors, especially to enhanced redox conditions in the sediment due to reduced sedimentation and increase in microphytobenthic primary production that increased phosphorus uptake. Submerged macrophytes increased in abundance due to improved light climate at the lake bottom (Sondergaard et al. 1990).
Lake Warniak Enclosure experiments that introduced the silver carp H. H molitrix into the eutrophic Warniak Lake, Poland, changed the structure and function of planktonic and benthic communities in the lake. First, phytoplankton and zooplankton biomass decreased. The composition of the phytoplankton community shifted towards more nanoplankton and dinoflagellates with a smaller proportion of blue-green algae. The numbers of hetertrophic bacteria decreased, but production increased. There were also increases in benthic biomass, especially Oligochaeta, and increases in individual biomass of Chironomidae, but the proportion of invertebrate predators in the total benthic biomass declined (Kajak et al. 1975).
SYNTHESIS OF MECHANISMS RESPONSIBLE Planktivorous fish can affect the structure and dynamics of planktonic communities through a variety of pathways that involve grazing and nutrient
85 recycling. An overview of these mechanisms is depicted in Figure 1 and summarized below. Vanni and Layne (1997) offered three mechanisms that may be responsible for the effects of fish on primary producers: (1) decreased herbivory by zooplankton when biomass of zooplanktivourous fish is high, (2) modification of nutrient recycling rates and ratios at which zooplankton release nutrients, and (3) nutrient recycling by fish. Research on the relative importance of these mechanisms is relatively scarce. The available data suggest that relative importance of the mechanisms involved varies among waterbodies (incl. trophic status, depth of a water column and lake thermal structure). For instance, the same authors (Vanni and Layne 1997) have suggested that nutrient recycling by fish can have stronger effects on phytoplankton communities than nutrient recycling by zooplankton in lownutrient systems, whereas data from Attayde and Hansson (2001a) indicates that increased nutrient excretion by zooplankton might account for most of the fish-mediated nutrient cycling effects in eutrophic systems. Schindler (1992) and Perez-Fuentetaja et al. (1996) have noted that fish and zooplankton induced nutrient increases appear to be more important for algal communities than grazing. In a similar trophic system (oligotrophic lake), Proulx et al. 1996 found that the effects of planktivorous fish were more evident with community structure variables (size or taxonomic distribution) than with coarser variables like algal biomass. The same study suggested that the effects of fish on phytoplankton appeared to be due to herbivore grazing and nutrient regeneration by fish and/or zooplankton would be relatively significant in more oligotrophic waterbodies. By controlling zooplankton species composition fish may indirectly control the N:P ratio supplied to phytoplankton. For instance, some large zooplankton taxa that dominate when zooplanktivorous fish are scarce or absent (e.g., Daphnia), have a low N:P ratio. These plankters excrete nutrients at a relatively high N:P ratio. The opposite is true for some small-bodied zooplankton taxa, e.g., calanoid copepods and Bosmina, that dominate when the abundance of planktivorous vertebrate predators is high (e.g, Sterner et al 1992, Vanni et al. 1997). Because the N:P ratio has important effects on phytoplankton structure and dynamics, shifts in this ratio induced by fish may have pronounced effects on phytoplankton. In addition, zooplankton communities dominated by small taxa will generally recycle nutrients to phytoplankton at higher rates than communities dominated by large plankters (Attayde and Hansson 2001a and references therein). Planktivorous fish can also affect phytoplankton communities directly by predation and release of nutrients through excretion. Direct predation effects (primarily the reduction of abundance of large taxa) have been observed only for a few species (see above). Since fish release nutrients at low N:P ratio (e.g., Attayde and Hansson 1999), this might favor cyanobacteria. In addition, the nutrient supply to phytoplankton occurs in more distinct pulses when fish are present compared to the situation when fish
86 are absent and large zooplankters are abundant (Reinertsen et al. 1986). Faeces produced by different fish species may selectively influence phytoplankton production. For example, blue tilapia T. aurea faecal material suppressed algal chlorophyll a production dramatically whereas gizzard shad Dorosoma cepedianum faeces increased production tenfold (Fernandes et al. 1994). By controlling zooplankton fish can also affect bacterioplankton. Changes in bacterial secondary production is controlled by the release of extracellular organic carbon by phytoplankton, especially when grazing pressure from zooplankton is reduced and algal growth is high. In addition, other important mechanisms for regulating the bacterial production are release of dissolved organic carbon from zooplankton (high when abundance of planktivorous fish is low) and algal senescence, induced by phosphate depletion (Riemann and Sondergaard 1986). Jeppesen et al. (1996) argued that Daphnia, which dominates when planktivorous fish abundance is low, is a more efficient bacterioplanktivore than small cladocerans ((Bosmina), cyclopoid copepods and rotifers that dominate during high predation pressure. Less DOC
Pulsed release, low N:P ratio, selective impact by feces
Higher TP, TN and particulate P, increased P retention, less epilimnic sedimentation
Water clarity Worsened light conditions
FISH BACTERIA More substrate? Less large plankters, reduced grazing
PHYTOBENTHOS
Suppression of large taxa
Changed habitat, altered communities?
ZOOPLANKTON Lower N:P, higher nutrient recycling
Selective feeding on nanoplankton
PHYTOPLANKTON
Release of EOC, algal senescence
?
Changed & less food?
ZOOBENTHOS
Fig. 1. Direct (grazing, grey straight arrows) and indirect (black curvy arrows) effects of suspension-feeding fish in freshwater ecosystems. The graph illustrates the situation when fish are present or their abundance is higher compared to the situation when it is lower. The mechanisms responsible are given in boxes. The thickness of grey lines denotes the grazing intensity (low or high). Direct black arrows indicate biotic effects on some selected important abiotic variables (nutrient concentrations, water transparency).
87 However, direct predation by fish may stabilize the bacterial population abundance and fish excrements may serve as important sources of organic matter and energy for bacteria (Kuznetsov 1980). An important mechanism for shaping plankton communities (and affecting water transparency) is selective feeding by microzooplankton (rotifers, Bosmina, nauplii and cyclopoids) on nanoplankton. This enables nanoplankton to become abundant in the enclosures without fish. It also explains how high community filtering rates of microzooplankton in the enclosures with fish can be associated with high biomass of nanoplankton. Finally, it helps explain how changes in the biomass of mesozooplankton, but not in microzooplankton, produce the fish manipulation effects of reduced biomass of nanoplankton and improved water quality (Mazumder et al. 1990). In fact, increases in the pico- and nanoplankton biomass in the microzooplankton dominated enclosures resulted from a combination of three factors (as summarised by Mazumder 1990): (1) the absolute loss of pico- and nanoplankton to zooplankton via grazing was low, (2) under low grazing pressure, regenerated nutrients favor small cells, and (3) these small cells contribute less to sedimentation. One potential mechanism for counteracting high grazing pressure is shifting the algal community towards forms with grazer resistant adaptations such as large size or grazer avoidance spines. This has been suggested from a study in a simple system of sub-Antarctic lakes (Hansson and Tranvik 1996). As pointed in the above-cited study, in species-rich and complex systems such adaptations may be more important and the observed effects of trophic interactions from top predators on lower trophic levels may be obscured. It is generally assumed that the majority if not all fish impacts on zooplankton structure and production are mainly due to direct predation. However, the food quality and competition for dietary resources should be taken into account (Lyche 1989). For instance, the often dominant algae before planktivore reduction, the filamentous cyanocbacteria, are normally poorly assimilated by herbivorous zooplankton and interfere with food collection. The cryptomonads that often increase in relative abundance after planktivore reduction are excellent food for Daphnia and the same is probably true for small chrysomonads. Unchanged zooplankton biomass may be the result of a combination of reduced food resource (primary production) and higher relative abundance of edible algae with better quality (Lyche 1989 and references therein). Changes in the benthic vegetation as a result of fish manipulations could be directly related to water transparency: less fish ĺ less phytoplankton ĺ higher transparency. This facilitates development of more rich benthic vegetation and expansion of distribution area of bentic phytobenthos communities. However, response of overall benthic communities to manipulations of suspension-feeding fish might not be so straight forward as other factors are likely to be involved (e.g., resuspension of sediments,
88 changes in sedimentation and accumulation of organic matter, intraspecific relations etc.). This issue should be addressed in future studies. Based on the variable response of zooplankton and phytoplankton communities to additions of fish in tank experiments, Matveev et al. (2000) concluded that the relative roles of zooplankton grazing and fish nutrient recycling may vary with different patterns of community structure. The effects of fish on algal growth may also vary with fish species, being largely dependent on their behaviour. Thus, the ultimate response of algal biomass in a given lake is more likely to depend on a combination of factors and their relative weights rather than on either ‘top-down’ or ‘bottom-up’ contol alone. Several other papers have concluded similarly: the observed effects of planktivorous fish is not be explained by one single factor but rather by complex set of interdependent factors (e.g., Starling and Rocha 1990; Rhew et al. 1999; Attayde and Hansson 2001).
RELEVANCE TO BIOMANIPULATION WITH FISH While the qualitative response of plankton to manipulations with planktivorous fish seems often (but always not easily) predictable, the magnitude of the response may differ. These differences may be related to perturbation intensity, food-chain complexity and/or trophic state (Lyche et al. 1990). Different responses of phytoplankton communities to fish manipulations have also been recorded in relation to lake thermal structure and depth of the water column (Proulx et al. 1996). Clearly, fish have strong negative impact on zooplankton total biomass and size structure but the negative impact is not uniform across species. However, the relationship at the next trophic level is much weaker and the predictability of alterations of phytoplankton and water clarity, as a result of fish manipulations is much lower. Therefore, it has been suggested (Post and McQueen 1987) that the success of biomanipulation of waterbodies by using suspension-feeding fish as a technique may be limited and nutrient reduction may be more effective than biomanipulation efforts by fish (Drenner et al. 1990 and references therein). When comparing effects of an invertebrate predator and planktivorous fish on plankton communities in a long term biomaipulation experiment, Wissel and Benndorf (1998) concluded that the positive effect of invertebrate predators for biomanipulation seems to be valuable only for short term experiments. In the long-term a suitable invertebrate predator that is able to prey on large zooplankton may immigrate in lakes without planktivorous fish. Therefore, the above study suggests that in order to get long-lasting positive results in biomanipulated systems it is necessary to have at least a minimum planktivorous fish biomass to control invertebrate predators (Wissel and Benndorf 1998).
89 In an effort to control small suspended solids and nutrients in wastewater stabilization ponds, Smith (1993) examined several treatment options where fish with different feeding habits were allowed to coexistence with zooplankton. The best result was achieved when combination of filterfeeders (the silver carp) with zooplankton was applied. This combination reduced algal biomass by 99% compared to controls and the suspended solids and nutrients were presumably likewise controlled. Therefore, this method was suggested as a successful tool for biomanipulation (Smith 1993). A few suspension-feeding fish species are of special interest in biomanipulation, partly due to their commercial value. One of them, silver carp, can also feed on bottom sediments – a fact that is important when considering the use of suspension-feeding fish to control algal blooms. Namely, the ability to utilize other food resources allows the fish to survive periods of low plankton biomass and be ready to consume the developing plankton blooms (Kajak et al. 1975). Silver carp has been suggested as a good candidate species to control phytoplankton development, and especially the blue-green algae (Starling and Rocha 1990). A few available studies of this subject highlight the complexity of such solutions. In one of them, Fukushima et al. (1999) showed that a shift from fish biomass to presence/absence of the fish, as the main factor responsible for changes in phytoplankton community, occurs during the course of seasonal development of the blue-green algae. The another study, carried out by Milstein et al. (1988), suggested that more numerical density than individual weight of the fish can be related to observed changes in phytoplankton. Successful management of aquatic ecosystems also depends on a variety of other factors. It was briefly mentioned earlier that climate may affect freshwater ecosystems. Unusual weather conditions (e.g., prolonged ice cover) may also cause deviations in several abiotic and biotic parameters (Scavia et al. 1986). Several other studies have indicated that trophic cascades in lakes are likely to be weaker when trophic status of the lake is higher (e.g., Vanni et al. 1990). Finally, in manipulation efforts with fish, it should be kept in mind that food web effects resulting from the presence of planktivorous fish can affect the trophic status of planktonic systems and the efficacy of external nutrient load is dependent on the food-web (Mazmuder et al. 1989).
ACKNOWLEDGEMENTS Comments of two anonymous reviewers on the manuscript are highly appreciated. These improved substantially quality of the paper.
REFERENCES Attayde JL Hansson L-A 2001a Fish-mediated nutrient recycling and the trophic cascade in lakes. Can J Fish Aquat Scii 58: 1924-1931
90 Attayde JL Hansson L-A 2001b The relative importance of fish predation and excretion effects on planktonic communities. Limnol Oceanogrr 46: 1001-1012 Attayde JL Hansson L-A 1999 Effects of nutrient recycling by zooplankton and fish on phytoplankton communities. Oecologia 121: 47-54 Bertolo A Lacroix G Lescher-Moutoue F Cardinal-Legrand C 2000 Plankton dynamics in planktivore- and piscivore-dominated mesocosms. Arch Hydrobioll 147: 327-349 Byears S Vinyard GL 1990 The effects on the plankton community of filter-feeding Sacramento blackfish, Orthodon microlepidotus. Oecologia 83: 352-357 Drenner RW Mummert JR deNoyelles F Jr Kettle D 1984 Selective particle ingestion by a filter-feeding fish and its impact on phytoplankton community structure. Limnol Oceanogr 29: 941-948 Drenner RW Smith JD Mummert JR Lancaster HF 1990 Responses of a eutrophic pond community to separate and combined effects of N:P supply and planktivorous fish: a mesocosm experiment. Hydrobiologia 208: 161-167 Elliott ET Castanares LG Porter KG 1983 Trophic-level control of production and nutrient dynamics in an experimental planktonic community. Oikos 41: 7-16 Fernandes CA Philips EJ Crisman TL Hopson MS 1994 Relative importance of blue tilapia and gizzard shad to lake sediment and water column nutrient concentration - analysis of fish feces. Lake Reserv Manage 9: 73 Fukushima M Takamura N Sun L Nagawa M Matsuhige K Xie P 1999 Changes in the plankton community following introductions of filter-feeding planktivorous fish. Freshwat Bioll 42: 719-735 Gee JHR 1991 Specialist Aquatic Feeding Mechanisms. In: Fundamentals of Aquatic Ecology, RSK Barnes and KH Mann (Eds). Blackwell Scientific Publications, Oxford. pp 186212. George DG Harris GP 1985 The effect of climate on long-term changes in the crustacean zooplankton biomass of Lake Windermere, UK. Nature 316: 536-539 Hansson L-A Tranvik LJ 1996 Quantification of invertebrate predation and herbivory in food chains of low complexity. Oecologia 108: 542-551 Horppila J Peltonen H Malinen T Luokkanen E Kairesalo T 1998 Top-down or bottom-up effects by fish: issues of concern in biomanipulation of lakes. Restor Ecoll 6: 20-28 Jeppesen E Sondergaard M Jensen JP Mortensen E Sortkjaer O 1996 Fish-induced changes in zooplankton grazing on phytoplankton and bacterioplankton: A long-term study in shallow hypertrophic Lake Sobygaard. J Plankton Res 18: 1605-1625 Kajak Z Rybak JI Spodniewska I Godlewska-Lipowa WA 1975 Influence of the planktonivorous fish Hypophthalmichthys molitrixx (Val.) on the plankton and benthos of eutrophic lake. Pol Arch Hydrobioll 22: 301-310 Krivtsov V Goldspink C Sigee DC Bellinger EG 2001 Expansion of the model 'Rostherne' for fish and zooplankton: role of top-down effects in modifying the prevailing pattern of ecosystem functioning. Ecol Modell 138: 153-171 Kuznetsov YeA 1980 The Effect of Fish With Different Feeding Habits on the Development of Planktonic Bacteria. J Ichthyoll 20: 87-98 Lyche A 1989 Plankton community response to reduction of planktivorous fish populations. A review of 11 case studies. Aqua Fenn 19: 59-66 Machacek J 1991 Indirect effect of planktivorous fish on the growth and reproduction of Daphnia galeata. Hydrobiologia 225: 193-197 Mallatt J 1984 Feeding ecology of the earliest vertebrates. Zool J Linn Soc 82: 261-272 Matveev V Matveeva L J Jones GJ 2000 Relative impacts of Daphnia grazing and direct stimulation by fish on phytoplankton abundance in mesocosm communities. Freshwat Bioll 44: 375-385 Mazumder A McQueen DJ Taylor WD Lean DRS Dickman MD 1990 Micro- and mesozooplankton grazing on natural pico- and nanoplankton in contrasting plankton communities produced by planktivore manipulation and fertilization. Arch Hydrobiol 118: 257-282
91 Mazumder A Taylor WD McQueen DJ Lean DRS 1989 Effects of fertilization and planktivorous fish on epilimnetic phosphorus and phosphorus sedimentation in large enclosures. Can J Fish Aquat Scii 46: 1735-1742 Mazumder A Taylor WD McQueen DJ Lean DRS 1990 Effects of fish and plankton on lake temperature and mixing depth. Science (Wash) 247: 312-315 McDonald ME 1985 Growth of a grazing phytoplanktivorous fish and growth enhancement of the grazed alga. Oecologia 67: 132-136 Milstein A Hepher B Teltch B 1988 The effect of fish species combination in fish ponds on plankton composition. Aquacult Fish Manage 19: 127-137 Perez-Fuentetaja A McQueen DJ Ramcharan CW 1996 Predator-induced bottom-up effects in oligotrophic systems. Hydrobiologia 317: 163-176 Perin S Pick FRA Lean DRS Mazumder A 1996 Effects of planktivorous fish and nutrient additions on primary production of shallow versus deep (stratified) lake enclosures. Can J Fish Aquat Scii 53: 1125-1132 Persson A 1997 Effects of fish predation and excretion on the configuration of aquatic food webs. Oikos 79: 137-146 Post JR McQueen DJ 1987 The impact of planktivorous fish on the structure of a plankton community. Freshwat Bioll 17: 79-89 Proulx M Pick FR Hamilton PB Lean DRS 1996 Effects of nutrients and planktivorous fish on the phytoplankton of shallow and deep aquatic systems. Ecology 77: 1556-1572 Reinertsen H Jensen A Langeland A Olson Y 1986 Algal competition for phosphorus: The influence of zooplankton and fish. Can J Fish Aquat Scii 43: 1135-1141 Rhew K Baca RM Ochs CA Therkeld ST 1999 Interaction effects of fish, nutrients, mixing and sediments on autotrophic picoplankton and algal composition. Freshwat Bioll 42: 99-109 Riemann B 1985 Potential importance of fish predation and zooplankton grazing on natural populations of freshwater bacteria. Appl Envir Microbioll 50: 187-193 Riemann B Sondergaard M 1986 Regulation of bacterial secondary production in two eutrophic lakes and in experimental enclosures. J Plankton Res 8: 519-536 Rudstam LG Lathrop RC Carpenter SR 1993 The rise and fall of a dominant planktivore: Direct and indirect effects on zooplankton. Ecology 74: 303-319 Sanderson SL Wassersug R 1990 Suspension-feeding vertebrates. Sci Am 262: 96-10 Scavia D Fahnenstiel GL Evans MS Jude DJ Lehman JT 1986 Influence of salmonine predation and weather on long-term water quality trends in Lake Michigan. Can J Fish Aquat Sci 43: 435-443 Scavia D Lang GA Kitchell JF 1988 Dynamics of Lake Michigan plankton: A model evaluation of nutrient loading, competition, and predation. Can J Fish Aquat Sci 45: 165-177 Schindler DE 1992 Nutrient regeneration by sockeye salmon (Oncorhynchus nerka ) fry and subsequent effects on zooplankton and phytoplankton. Can J Fish Aquat Sci 49: 24982506 Smith DW 1993 Wastewater treatment with complementary filter feeders: A new method to control excessive suspended solids and nutrients in stabilization ponds. Water Environ Res 65: 650-654 Sondergaard M Jeppesen E Mortensen E Dall E Kristensen P Sortkjaer O 1990 Phytoplankton biomass reduction after planktivorous fish reduction in a shallow, eutrophic lake: a combined effect of reduced internal P-loading and increased zooplankton grazing. Hydrobiologia 200-201: 229-240 Starling FLRM Rocha AJA 1989 Experimental study of the impacts of planktivorous fishes on plankton community and eutrophication of a tropical Brazilian reservoir. Hydrobiologia 200-201: 581-591 Sterner RW Elser JJ Hessen DO 1992 Stoichiometric relationships among producers, consumers and nutrient cycling pelagic ecosystems. Biogeochemistry 17: 49-67
92 Tang H Xie P Lu M Xie L Wang J 2002 Studies on the effects of silver carp ((Hypophthalmichthys molitrix) on the phytoplankton in a shallow hypereutrophic lake through an enclosure experiment. Int Rev Hydrobioll 87: 107-119 Torras X Cardona L Gisbert E 2000 Cascading effects of the flathead grey mullet Mugil cephalus in freshwater eutrophic microcosms. Hydrobiologia 429: 49-57 Unger PA Lewis WM McClean DH 1984 Nonvisual feeding in a visual planktivore Xenomelaniris venezuelae. Oecologia 64: 280-283 Vanni MJ Layne CD 1997 Nutrient recycling and herbivory as mechanisms in the "TopDown" effect of fish on algae in lakes. Ecology 78: 21-40 Vanni MJ Layne CD Arnott SE 1997 Top-down" trophic interactions in lakes: effects of fish on nutrient dynamics. Ecology 78: 1-20 Vanni MJ Luecke C Kitchell JF Magnuson JJ 1990 Effects of planktivorous fish mass mortality on the plankton community of Lake Mendota, Wisconsin: Implications for biomanipulation. Hydrobiologia 200-201: 329-336 Wissel B Berndorf J 1998 Contrasting effects of the invertebrate predator Chaoborus obscuripes and planktivorous fish on plankton communities of a long term biomanipulation experiment. Arch Hydrobioll 143: 129-146 Xie P Huang X Takamura N 2000 Changes of Leptodora kindti abundance (1957-1996) in a planktivorous fishes-dominated subtropical Chinese lake (Lake Donghu). Arch Hydrobiol 147: 351-372
INFLUENCE OF EASTERN OYSTERS ON NITROGEN AND PHOSPHORUS REGENERATION IN CHESAPEAKE BAY, USA
RIE Newell, TR Fisher, RR Holyoke and JC Cornwell Horn Point Laboratory, University of Maryland Center for Environmental Science, PO Box 775, Cambridge, MD 21613, USA.
[email protected] Abstract: Suspension-feeding bivalves couple pelagic and benthic processes because they consume seston from the water column, and their biodeposits (feces and pseudofeces) settle on the sediment surface. Abundant stocks of bivalves can exert grazer control on phytoplankton, and this results in some nitrogen and phosphorus being regenerated to the water column as excreta and via microbial decomposition of biodeposits. Bivalve biodeposition, however, enhances net ecosystem losses of N and P via sediment burial and bacterially mediated, coupled nitrification-denitrification. Bivalve feeding also reduces turbidity and thereby increases light available for microphytobenthos. Although microphytobenthos may compete with nitrifying bacteria for N, potentially reducing coupled nitrification-denitrification, they retain N and P within sediments, further reducing net regeneration to the water column. Keywords: benthic-pelagic coupling, bivalves, Chesapeake Bay, nitrogen, denitrification, nutrient burial, oyster, phosphorus
INTRODUCTION Many estuaries and coastal water bodies worldwide are anthropogenically enriched with Nitrogen (N) and Phosphorus (P). These fertilizing elements emanate from point sources (e.g., sewage treatment plant effluent), non-point sources (e.g., agricultural run-off and septic-tank discharge), and atmospheric deposition. These anthropogenically enhanced sources of N and P cause fundamental changes in the magnitude and distribution of phytoplankton biomass and primary production (Fisher et al. 1988, Malone 1992, Conley 1999), and the resulting enhanced phytoplankton production and blooms of both toxic and nontoxic microalgae frequently have deleterious effects on the structure and function of coastal ecosystems (Cloern 2001). In some locations, such as Chesapeake Bay and Long Island Sound, USA, microbial decomposition of the excess phytoplankton biomass supported by nutrient enrichment causes bottom water and sediments to become anoxic during warmer months (Boicourt 1992). In shallow eutrophic water bodies, consumption of phytoplankton by 93 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 93–120. ©2005 Springer. Printed in the Netherlands.
94 abundant stocks of bivalve suspension feeders directly reduces the amount of particulate organic matter (POM) available to be remineralized by pelagic consumers and bacterioplankton (Cloern 1982, Officer et al. 1982). Hence, observed phytoplankton concentrations are not only the result of anthropogenic nutrient inputs - the so called “bottom-up” influence - but also are affected by changes in the abundance of grazer organisms that exert "topdown" control (Newell 1988, Dame 1996, Strayer et al. 1999, Newell 2004). Newell (1988) postulated that present-day high phytoplankton biomass in Chesapeake Bay is caused, in part, by the reduction of “top-down” grazing control. In particular, there has been a large decline in the abundance of suspension feeding eastern oysters, Crassostrea virginica, superimposed on the well-recognized effects of nutrient enrichment stimulating phytoplankton production (Kemp and Boynton 1992, Cloern 2001, Harding et al. 2002). Intensive harvesting of oysters in the 19th century led to a severe decline in oyster stocks and their reef habitat (Kennedy and Breisch 1981, Rothschild et al. 1994). This loss of material, in combination with shell becoming covered in sediment from erosion and resuspension (Smith et al. 2001), led to a steady decline in the clean shell necessary for oyster larval settlement. In addition, since the 1950's, two major parasitic diseases, MSX (Haplosporidium nelsoni) and Dermo ((Perkinsus marinus), exacerbated the harvest-related decline in oysters in Chesapeake Bay (Ford and Tripp 1996). Newell (1988) estimated that eastern oyster populations in the early 19th Century filtered 80% of the <9 m water volume in Chesapeake Bay per day during summertime; the precipitous decline in oysters, however, reduced this to <1% filtration of the shallows per day by 1988. This collapse of the eastern oyster population in Chesapeake Bay has also occurred in other estuaries (e.g., Delaware Bay and Long Island Sound) along the Atlantic coast of North America (Mackenzie 1996). This suggests that these other estuaries have also lost the benthicpelagic coupling function once provided by abundant stocks of eastern oysters. Bivalve suspension-feeders serve an important biogeochemical role in coastal ecosystems because N and P from the water column are transferred to the sediments in their biodeposits. Many investigators have focused on the nutrient regeneration aspects of bivalve grazing (Prins and Smaal 1990, Asmus and Asmus 1991, Yamamuro and Koike 1993, Pietros and Rice 2003); here we emphasize the importance of bivalve biodeposition in enhancing the processes leading to nutrient burial and denitrification (Fig. 1, Newell et al. 2002, Newell 2004). For simplicity in Fig. 1 we only include processes related to bivalve feeding, and ignore gravitational settling of senescent phytoplankton, etc. If all of the phytoplankton N and P that is removed from the water column by suspension-feeding bivalves is regenerated either in excreta (urine) or microbially regenerated from bivalve biodeposits, then bivalve populations simply serve to recycle N and P (T = R in Fig. 1). Under these conditions, recycling of N and P by bivalves maintains primary production at a maximum level determined by inputs (I) less physical losses
95 (net Exchange in Fig. 1). To the extent that N and P from biodeposits either
Fig. 1. Conceptual diagram of how phytoplankton production (PP) in the Choptank River estuary is supported by inputs (I) and sediment regeneration (R) of N and P. The amount of N and P flushed from the estuary, buried in the accumulating sediments, and denitrified controls the maximum phytoplankton biomass supported. Illustrated in the three boxes for locations A, B, and C are how the percentage of surface photosynthetically active radiation (PAR) and degree of oxygen saturation varies with water column depth from the shallows flanks (< 2 m) of the estuary (A), to slightly deeper areas (2 to 8 m) that form the majority of the estuary bottom (B), to deep water sites (C) located in the deepest (> 8 m) main channel of the estuary.
accumulate in sediments (A > 0 in Fig. 1), are denitrified (D > 0 in Fig. 1), or are trapped by microphytobenthos (Fig. 2A), then bivalve feeding enhances nutrient loss from the ecosystem (R R < T). Due to this removal mechanism within the sediments, phytoplankton production is set at a level lower than that determined by the nutrient inputs from the watershed, airshed, and exchanges with the next downstream ecosystem. In addition to altering nutrient regeneration processes, bivalve grazing, by removing phytoplankton and inorganic particles from the water column, can reduce turbidity. The resulting increased light penetration (Fig. 1 Box A) can potentially enhance the area of sediments with sufficient light intensity to support photosynthesis of benthic plants (Newell and Koch 2004). Other deeper or more turbid areas may either have insufficient light (Fig. 1 Box B) or insufficient light and low dissolved oxygen (Fig. 1 Box C). It is well documented that there have been severe economic
96
Fig. 2. Role of eastern oysters in removing phytoplankton from the water column and transferring undigested particulate material as biodeposits to the sediment surface. Illustrated are benthic-pelagic scenarios for three locations (Boxes A, B, and C; Fig. 1) with different levels of photosynthetically active radiation (PAR) and dissolved oxygen concentrations. From surficial aerobic sediments (middle panel; Location B) N and P are released to the water column. The microbially mediated process of nitrification in the aerobic surface sediments coupled to denitrification within the underlying anaerobic sediments causes N to be lost from biodeposits as N2 gas. N not regenerated is buried in accumulating sediments and P is immobilized in the aerobic sediments. In contrast, little nutrient regeneration into the water column takes place in locations with sufficient light to support active microphytobenthos that absorb regenerated N and P at the sediment surface (upper panel; Location A). Coupled nitrification-denitrification is also reduced because the microphytobenthos out-compete bacteria for NO2-, NO3-, and NH4+. In locations where the sediments are anoxic (lower panel; Location C) nitrification is inhibited and all N and P is regenerated from the sediments as NH4+ and PO43-. Some burial of N occurs but P sorption is precluded. Solid lines indicate transfer of materials; dashed lines indicate diffusion of materials; dotted lines indicate microbially mediated reactions.
97 consequences to the fishing communities in Chesapeake Bay (Kennedy and Breisch 1981, Rothschild et al. 1994) stemming from the decline in eastern oyster harvests from the peak of ~60,000 metric tons wet tissue weight annually in the late 19th Century to ~ 300 metric tons in 2002 (National Marine Fisheries Service; www.st.nmfs.gov). Less easy to assess are the adverse ecological consequences and the economic costs to coastal ecosystems of essentially the complete loss of eastern oysters. Ecological effects associated with eastern oyster populations can be divided into 1) the influence of their suspension-feeding activity on water quality, and 2) the role of the physical oyster reef structure itself in providing habitat for other organisms. We focus here on the primary effects associated with oyster populations filtering particles from the water column and transferring material to the sediment surface. We do not consider the secondary effects of oyster reefs providing habitat for many invertebrate and vertebrate species (Coen et al. 1999). The objective of this paper is to review available data to determine whether suspension-feeding bivalves are simply rapidly regenerating nutrients, thereby maintaining phytoplankton production at levels set by inputs; alternatively, if by enhancing denitrification and burial, can bivalve feeding be considered a “sink” for N and P? We then apply this information to estimate the possible effects of stocks of sub-tidal eastern oysters on the watershedlevel nitrogen and phosphorus budgets for the Choptank River, a mesohaline tributary of Chesapeake Bay, USA. EASTERN OYSTER FEEDING AND PARTICLE REMOVAL Crassostrea virginica is an active suspension-feeder that exhibits a complex feeding response when exposed to seasonal variations of temperature and seston concentration (reviewed by Newell and Langdon 1996). Captured particles are sorted, and the less nutritious ones or those in excess of gut capacity are immediately rejected as pseudofeces (Newell and Jordan 1983, Ward et al.1994, Newell and Langdon 1996). Nutritious particles are ingested and subject to extracellular and intracellular digestion and the remains are defecated within ~ 24 h (Langdon and Newell 1996). The number of captured particles that can be ingested by bivalves is dependent on gut residence time and gut volume, which is a function of body size (Bayne et al. 1984). Once maximum ingestion rates are attained, any further increase in seston concentration results in increasing amounts of material being rejected as pseudofeces (Haven and Morales-Alamo 1966, Newell and Jordan 1983). Thus, at low seston concentrations eastern oyster pseudofeces are composed primarily of inorganic material, but as seston concentrations increase the proportion of POM in their pseudofeces also increases. Eastern oysters maintain high clearance rates even when the volume of particles captured on their gills exceed their limited gut capacity. Under such circumstances excess particles are rejected as pseudofeces prior to
98 ingestion. By maximizing the number of particles captured and subjected to efficient pre-ingestive selection, eastern oysters maximize their ingestion of nutritious particles (Ward et al. 1994, Newell and Langdon 1996). Today, throughout the majority of the eastern oyster’s range, anthropogenic nutrient enrichment has increased phytoplankton biomass; consequently, once the oyster’s nutritional needs are satisfied, even phytoplankton cells are rejected in pseudofeces, in addition to less nutritious detrital and silt particles. This response of eastern oysters to increasing seston concentrations is quite different from other suspension-feeding bivalves, such as infaunal cockles and clams, which mainly regulate their ingestion rates by constraining their clearance rates rather than rejecting excess particles as pseudofeces (Hawkins et al. 1998, Grizzle et al. 2001). Consequently, the species of bivalves that can exert the greatest influence on benthic-pelagic coupling are those, such as oysters and mussels, which maintain high clearance rates and reject relatively large amounts of POM as pseudofeces. One of the more important effects of bivalve feeding is the repackaging of small seston particles into large aggregates. Particles in eastern oyster feces are tightly bound in a mucoid matrix and voided as pelleted strings that can be as long as several millimeters. Pseudofeces are less tightly bound in mucus and may be subject to some disaggregation when voided from the oyster. As a consequence of the aggregation of both feces and pseudofeces, biodeposits have a faster sinking velocity that is up to 40 times that of non-aggregated particles (Kautsky and Evans 1987, Widdows et al. 1998). Even though epibenthic bivalves like eastern oysters live on reefs relatively close to the sediment surface, their biodeposits may be widely distributed by wave action or tidal currents before final deposition (Haven and Morales-Alamo 1968, Dame et al. 1991a, Silvert and Cromey 2001). Biodeposits can only settle where the friction velocity (u*), which is a function of current velocity and bed roughness, is below a critical velocity required to suspend particles of that particular mass. If biodeposits become fragmented into smaller particles, they will sink more slowly and are resuspended at lower friction velocities (Sanford and Chang 1997). Oyster shells and reefs add appreciably to bed roughness and hence increase friction velocity (Wildish and Kristmanson 1997), thereby enhancing the transport of biodeposits to sediments away from the reef structure. In locations and at times where u* is below the critical velocity, biodeposits undergo a consolidation process and gradually become incorporated into the sediments (Haven and Morales-Alamo 1966, 1968, Kaspar et al. 1985, Jaramillo et al. 1992, Widdows et al. 1998). Once consolidated into the sediments, a much greater bottom shear stress is required to resuspend the cohesive material (Sanford and Chang 1997). Despite the clearly recognized importance of bivalve biodeposition in benthic pelagic coupling (Fig. 1), exactly how much of the material is transferred to the sediment surface, how much is resuspended, and how much accumulates in the surficial sediment remains poorly characterized. Further research is required to quantify biodeposit
99 sinking rates and physical conditions leading to their resuspension before we can fully quantify the contribution of bivalves to benthic-pelagic coupling. EASTERN OYSTER DIGESTION Bivalves digest and assimilate N from different sources of POM with efficiencies from ~20 to 90%. These efficiencies also vary seasonally (Bayne and Newell 1983, Kreeger and Newell 2001), leading to substantial amounts of undigested particulate organic nitrogen (PON) being transferred to the sediment surface in feces in addition to the PON rejected in pseudofeces. Newell and Jordan (1983) reported that eastern oysters feeding on natural seston at concentrations of 5 to 20 mg L-1 assimilated ~ 50% of the PON filtered, and the remainder was voided as biodeposits. For eastern oysters feeding during summer months on natural seston in the mesohaline Chesapeake Bay, Jordan (1987) found that biodeposits contained ~ 2 to 3 times as much C, N, and P per unit weight as particles settling-out naturally from the water column (Table 1). In numerous studies of sediments near large aggregations of bivalves, an increase in sediment N content has been confirmed (Tenore et al. 1982, Kaspar et al. 1985, Kautsky and Evans 1987, Deslous-Paoli et al. 1992, Hatcher et al. 1994). Therefore, oyster populations remove substantial amounts of planktonic N and P from the water column and enrich the underlying sediment. The N and P absorbed by bivalves from the ingested food undergo internal metabolic processing. The majority is used for tissue growth, and some is excreted in urine (Bayne and Hawkins 1992). Magni et al. (2000) reviewed the extensive literature on weight specific N excretion for 11 species Table 1. Comparison of mean (SE; n = 20 to 25) concentrations (mg g -1) of carbon, nitrogen, and phosphorus in dry eastern oyster biodeposits and in natural seston that settled from the water. Biodeposits were collected by holding oysters in the Choptank River for 2 to 14 d in May, June, July, August, and November 1983 and in May, June, and July 1984. Seston material that settled from the water due to gravity was collected concurrently in an apparatus identical to that used to hold oysters and collect their biodeposits. Data from Jordan (1987).
Biodeposits
Seston Material
Carbon (mg C g -1)
34.8 + 3.15
14.6 + 1.19
Nitrogen (mg N g -1)
4.8 + 0.44
2.1 + 1.19
Phosphorus (mg P g -1)
0.58 + 0.085
0.32 + 0.028
C:N:P ratio (molar)
154:18:1
117:14:1
of bivalve mollusc, and average ammonium excretion rates were ~ 6.0 micromol NH4+ g-1 DW h-1, (DW = dry tissue weight), in agreement with data for Crassostrea virginica (Srna and Baggaley 1976). In comparison to the
100 extensive literature on N excretion by bivalves, the rate of P excretion has received little attention. Magni et al. (2000) directly measured P excretion in different size classes of Ruditapes philippinarum and Musculista senhousia, and their average measured rates were 1.9 and 1.4 micro-mol PO4 g-1 DW h-1, respectively. Magni et al. (2000) only found literature data for rates of P excretion in 3 other species of bivalves; when combined with their data, these indicate average excretion rates of ~ 1.2 micro-mol P g-1 DW h-1. OYSTERS AS AGENTS OF BENTHIC-PELAGIC COUPLING Natural sediments have well-developed microbial communities inhabiting distinct zones of oxygen content (Fig. 2; Henriksen and Kemp 1988). Therefore, bivalve biodeposits that settle on sediments with an oxic surface layer are subject to initial decomposition by aerobic bacteria (Fig. 2B). Organic materials are oxidized to CO2, PO43-, and NH4+, and other aerobic bacteria further oxidize NH4+ to NO2- and NO3-. Some of the NO2- and NO3diffuses down into underlying anaerobic sediments, and some diffuses out of the sediment and enters the water-column nutrient pool. In the underlying anaerobic sediments, denitrifying bacteria use the oxidized forms of N as terminal electron acceptors, reducing the NO2- and NO3- to N2 gas (Henriksen and Kemp 1988, Seitzinger 1988, Risgaard-Petersen et al.1994). Absent Nfixation, N2 is unavailable to plants and passes to the atmosphere. Denitrification can only occur where there is a close juxtaposition between oxygenated sediments that support nitrifying bacteria and anaerobic sediments that support denitrifying bacteria (Kristensen 1988). Bacterial degradation of particulate organic N and P from bivalve biodeposits that settle to anoxic sediments is solely via anaerobic pathways (Fig. 2C). Because the initial nitrification step is precluded, all regenerated N remains as NH4+, and there is negligible sorption of PO4 to iron complexes (Krom and Berner 1981). The microbial communities associated with sediments are a crucial element mediating nutrient regeneration processes from biodeposits. Consequently, studies of nutrient regeneration from bivalve communities held in conditions without underlying sediments containing a well-developed microbial community are missing a crucial biological element. For instance, rates of N and P regeneration from oyster biodeposits incubated in small chambers (Jordan 1987) and from eastern oysters maintained in MERL mesocosms without sediments (Pietros and Rice 2003), should not be extrapolated to actual effects of bivalves on coastal ecosystems, where nutrient regeneration processes are much more complex. Rates of net N flux from natural bivalve communities (direct excretion by the animals plus regeneration from biodeposits in the sediments) can be substantial, ranging from ~ 1 to 5 mmol N m-2 h-1 (Dame et al. 1989, 1991a, Asmus and Asmus 1991, Magni et al. 2000), with rates being greater in summer than in winter months (Dame et al. 1992). Dame et al. (1989)
101 estimated that a South Carolina coastal intertidal eastern oyster reef transferred in biodeposits ~189 g N m-2 y-1 from the water to the sediments, with the majority subsequently being regenerated as NH4+ (125 g N m-2 y-1). The difference between these two values, about 33% of the net N removal, represents N incorporated either into biomass of reef fauna and flora, denitrified, or buried in the sediments. The nitrogen released from bivalve populations comes not only from ingested phytoplankton but also nonphytoplankton material, such as N-rich bacteria and flagellates (Asmus and Asmus 1991), and these heterotrophic organisms are readily captured by bivalves (Bayne and Hawkins 1992, Kreeger and Newell 2001). The lack of simple techniques to measure denitrification prior to the development of the membrane inlet mass spectrometric method (Kana et al.1998, Cornwell et al. 1999, Newell et al. 2002), meant that earlier studies on the ecological role of bivalves in altering patterns of inorganic nutrient regeneration could not measure enhanced N-loss via denitrification (e.g., Jordan 1987, Dame et al. 1989, 1991a, 1992, Asmus and Asmus 1991, Hatcher et al. 1994, Prins and Smaal 1994). Kaspar et al. (1985) used the indirect acetylene block technique to measure denitrification and reported appreciably higher denitrification potential in sediments underlying ropecultured mussels than in nearby reference sites with similar sediments. More recently, Pelegri and Blackburn (1995), using the 15N isotope pairing technique, observed stimulation of coupled nitrification-denitrification associated with bivalves. Although bivalves are clearly important mediators of N cycling, their role in P cycling is equivocal. Primary reasons for this ambiguity are the fundamental differences in the chemistry of these two elements and their relative concentration in biodeposits. Not only are biodeposits more enriched in N than P (N:P atomic ratio of 18 in oyster biodeposits compared to 14 in seston; Table 1) but the remineralized N is predominately in the form of NH4+ that can readily diffuse out from the sediment. Conversely the balance between binding and release of P from oyster biodeposits is highly dependent on sediment oxygenation and the development of a redox gradient within the sediments (Fig 2B). In estuarine sediments, P fluxes are controlled by y interfacial adsorption and desorption processes, often involving oxidized Fe3+ iron and sulfur cycling (Krom and Berner 1981). Iron oxides at the sedimentwater interface are a diffusive barrier to P fluxes across the sediment-water interface. Under fully oxygenated conditions most of the P regenerated from biodeposits will be buried in the accumulating sediments. If the depth of the oxygenated zone in the surficial sediments decreases, as it does seasonally in Chesapeake Bay sediments (Fig 2C), the P-adsorbing iron oxide can be reduced to ferrous monosulfides, thereby allowing the release of sedimentary inorganic P (Boynton and Kemp 1985). Studies on natural bivalve populations indicate that they do not increase P regeneration above background levels (e.g., Dame et al. 1991a) or increase regeneration only slightly (Asmus and Asmus 1991, Souchu et al.
102 2001). Similarly, Doering and Oviatt (1986) and Doering et al. (1987) reported no appreciable phosphate fluxes from the sediment in MERL macrocosm tanks containing Mercenaria mercenaria, but ammonium fluxes were 60 times higher in tanks containing these clams. Dame et al. (1989) estimated that a South Carolina coastal reef of eastern oysters and associated fauna and flora affected a net P transfer to the sediments of 98 g P m-2 yr-1, presumably either incorporated into fauna and flora or buried in the sediments, with little P release. Magni et al. (2000) calculated the upward diffusive flux of N (0.2 to 1.5 mmol NH4+-N m-2 d-1) and P (0.01 to 0.05 mmol PO43--P m-2 d-1) based on pore water nutrient concentrations in sediments collected from within dense natural field populations of the bivalves R. philippinarum and M. senhousia. At the measured field densities of these two species of bivalve, they estimated direct excretory activity of up to 35.2 mmol NH4+-N m-2 d-1 and 5.8 mmol PO43--P m-2 d-1, based on laboratory measured excretion rates. The direct excretion of N and P were ~23 and 116 times greater, respectively, than sediment regeneration rates. The results from their study may be somewhat atypical, however, in that the rates of P excretion they reported for the smallest size class of R. philippinarum that were highly abundant in the field populations were ~ 3 times greater than rates they measured for other size classes.
OYSTERS: PROMOTING OR REDUCING N AND P CYCLING? Data reviewed above indicates that rates of N release from bivalve excretion and biodeposit regeneration can be substantial although P regeneration appears to be relatively low. Many investigators have surmised, either from these rates of recycling or by direct measurements of higher primary production and phytoplankton biomass in the immediate vicinity of the bivalves, that bivalve feeding activity serves to enhance primary production (e.g., Doering et al. 1986, 1987, Dame and Dankers 1988, Dame and Libes 1993, Dame et al. 1989, 1991b, Prins and Smaal 1990,1994, Asmus and Asmus 1991, Yamamuro and Koike 1993, Magni et al. 2000, Pietros and Rice 2003). What is ignored in these studies, however, is that bivalves, by virtue of their high clearance rates, filter phytoplankton from large volumes of water. This has the localized effect of focusing nutrients that are then regenerated with a potential increase in the area-specific rates of nutrient recycling. Nonetheless, at the ecosystem level (e.g., estuary or embayment) the maximum phytoplankton standing stock supported by the nutrients regenerated through bivalve populations cannot exceed the level that can be sustained by nutrient inputs from the watershed or from adjoining water bodies (Fig. 1); i.e., regeneration only maintains levels of primary production set by external inputs. Furthermore, these earlier investigations typically neglect the role that bivalve biodeposition has in enhancing N and P burial
103 and sediment denitrification (Fig. 2B). These processes remove N and P from the water column, thereby reducing phytoplankton biomass and primary production. In situations where bivalves are either at very high population densities or living in locations with low water circulation, biodeposition can stimulate microbial metabolism sufficiently to cause the sediments to become anaerobic (Newell 2004). In such situations (Fig. 2C), nutrients are regenerated primarily as NH4+ and PO43, with little or no loss due to burial and denitrification. Newell et al. (2002) measured changes in nitrogen fluxes and denitrification in laboratory incubations of sediment cores subject to loading by pelletized phytoplankton cells, an experimental analog for oyster biodeposits. When organics were regenerated under aerobic conditions (Fig. 2B), typical of those associated with oyster habitat, coupled nitrificationdenitrification was promoted, resulting in denitrification of ~20% of the added N. In contrast, under anoxic conditions typically found in summer beneath the pycnocline in main-stem Chesapeake Bay (Kemp and Boynton 1992), nitrogen from the added organics was released solely as ammonium (Fig. 2C). Newell et al. (2002) postulated that denitrification of PON remaining in oyster biodeposits may enhance nitrogen removal from estuaries compared to locations without oysters. Furthermore, in aerobic incubations with sufficient light (~70 mol photons-2 s-1), Newell et al. (2002) found that microphytobenthos absorbed the inorganic nitrogen released from the added organics (Fig. 2A). These results suggest that an ecosystem dominated by benthic rather than planktonic primary production may develop in shallow waters when reduced turbidity associated with bivalve feeding increases light penetration to a level that can sustain benthic microalgal production (Fig. 1A). In summary, it is well-established that bivalve feeding activity reduces ecosystem levels of phytoplankton biomass. This reduction in biomass and partial regeneration of N and P will tend to make these nutrients less limiting locally to the remaining phytoplankton. This may well alter phytoplankton physiological condition and thereby enhance growth rates, (i.e., production is the product of phytoplankton biomass times cell specific growth rate.). This accounts for the observation from many field and mesocosm studies of increasing rates of phytoplankton production in the vicinity of the bivalve consumers (e.g., Prins et al. 1998, Pietros and Rice 2003, reviewed by Dame 1996). We contend, however, that because sediment N and P regeneration are less than 100% efficient due to burial and denitrification (i.e. T > R; see Fig. 1), bivalve feeding serves to reduce recycling and ultimately will reduce system-level phytoplankton production and biomass.
INFLUENCE OF BIVALVE FEEDING ON TURBIDITY Bivalve filtration of suspended particles from the water column reduces turbidity and thereby increases photosynthetically active radiation
104 (PAR) penetration (Box A Fig. 1). Newell and Koch (2004) developed a numerical model to simulate the interaction between wave-induced sediment resuspension, bivalve filtration, and seagrass growth. Their model predicted that the presence of subtidal oysters at 25 g DW m-2 (25 oysters of 7.5 cm shell length) reduced suspended sediment concentrations in shallow waters by nearly an order of magnitude when water temperatures were ~25oC and oyster clearance rates were high, compared to situations where oysters are absent. This reduction in suspended particles produced a significant increase in water clarity and hence the depth to which submerged aquatic plants, such as seagrasses and microphytobenthos (MacIntyre et al.1996) were predicted to grow. In eutrophic systems where there has been a substantial increase in bivalves through the rapid colonization by a non-native species, there has often been a corresponding increase in abundance of submerged aquatic macrophytes (Cohen et al. 1984, Strayer et al. 1999). In open and wellflushed estuaries, the exchange of suspended particles from adjacent waters means that the localized enhancement of bivalve stocks may not reduce turbidities sufficiently to permit seagrasses to grow. Microphytobenthos, however, require less PAR than seagrasses to sustain positive growth (MacIntyre et al.1996), and therefore may be able to colonize deeper sediments in the immediate vicinity of an enhanced bivalve population. Actively growing microphytobenthos can directly intercept much of the inorganic nitrogen regenerated from the sediments by bacterial decomposition, thereby limiting N release to the water column (Fig. 2A; Krom 1991, Rysgaard et al. 1995, Sundbäck et al. 2000, Newell et al. 2002). The actual magnitude of these N fluxes varies depending on the nutrient concentrations in the overlying water, microphytobenthos biomass and species composition, and light conditions. In addition to the role of microphytobenthos in intercepting N regenerated from the sediments, the oxygen produced from their photosynthesis can alter depths of oxygen penetration into the sediments (Fig. 2A). This enhanced oxygen supply can by used by bacteria at the sediment water interface to maintain nitrification (Risgaard-Petersen et al. 1994, Rysgaard et al.1995, An and Joye 2001). This is important in situations where the POM remaining in bivalve biodeposits stimulates microbial metabolic oxygen demand to levels that exceed the oxygen resupply from the water column. When growth of the microphytobenthos is high, their uptake of NO3- competes with the bacteria, resulting in a decline or cessation of denitrification while still preventing nutrient regeneration to the water column (Henriksen and Kemp1988, Risgaard-Petersen et al. 1994, Rysgaard et al. 1995, An and Joye 2001). Chesapeake Bay management activities are designed to reduce turbidities, thereby increasing PAR at the sediment surface in order to enhance seagrass growth. This strategy does not consider that some species of macroalgae (e.g., Ulva spp., Enteromorpha spp., and Cladophora spp.) flourish in locations that have elevated levels of inorganic nutrients and relatively low irradiances (from 18 to 175 micro mol photons-2 s-1). In some
105 shallow and enclosed embayments these nuisance species out-compete other macroalgae (e.g., Taylor et al. 2001) and grow so profusely that they restrict water flow and cause sediment hypoxia when they decay (Peckol and Rivers 1995, Rafaelli et al.1998). In locations where bivalve grazing reduces phytoplankton stocks, thereby concurrently decreasing inorganic nutrients stored in phytoplankton biomass and increasing PAR at the sediment surface, nuisance macroalgae may become established, rather than a more balanced flora of seagrasses and microphytobenthos.
MODELING OYSTER INFLUENCE ON N AND P REMOVAL Several modeling studies have been undertaken to evaluate the importance of eastern oysters in the ecology of Chesapeake Bay. Bartleson and Kemp (1990) manipulated an ecosystem model of C and N flow to assess the improvements in water quality associated with a tenfold increase in benthic suspension feeders in water <10 m deep. This increase in benthic suspension-feeding activity reduced the organic material available to be microbially decomposed in the deep channel, hence leading to a marked increase in bottom oxygen concentrations. Ulanowicz and Tuttle (1992) used an ecosystem model of carbon flow through Chesapeake Bay food webs to explore the trophic consequences of an increase in oyster stocks. Their results suggest that there would be substantial changes in the food web, including lower phytoplankton biomass as a consequence of higher grazing rates by greater oyster stocks. Similarly, predictions from a model that includes the interaction of water flow and bivalve feeding (Gerritsen et al. 1994) suggest substantial reductions in phytoplankton biomass could be achieved by increasing oyster stocks through off-bottom oyster culture. None of these previous modeling efforts estimated the possible ecological benefits associated with suspension-feeding bivalves altering N and P cycling as a result of removing particles from the water column and transferring undigested remains to the sediment surface. Consequently, because of a lack of quantitative information on these putative benefits it has been difficult to incorporate the value of these ecosystem services in watershed management plans that are attempting to reduce anthropogenic nutrient inputs. We develop here an elementary “spread-sheet” model to assess the influence of eastern oysters on removal of N and P inputs to the Choptank River estuary, a mesohaline Maryland tributary to Chesapeake Bay. We estimated the monthly amount of P buried and N removed due to burial and coupled nitrification-denitrification resulting from the biodeposition activity of adult eastern oysters (shell height 7.6 cm; ~1 g dry tissue weight). These estimates are applicable to natural oyster reefs where the majority of biomass is in adult oysters. They include neither the reduced levels of N and P removal during the time it takes oysters to grow from juvenile to adult size
106 nor the enhanced levels of N and P removal as oysters grow larger than 7.6 cm. Monthly environmental data (average water temperature, chlorophyll a, and seston concentrations; Table 2) were taken from the EPA Chesapeake Bay monitoring program for station ET5.2 in the Choptank River (www.chesapeakebay.net). Biodeposition rates were estimated from data obtained by Jordan (1987) for eastern oysters held for 4 to 36 h in a flume supplied with flowing (0.8 to 10 cm s-1) Choptank River water at ambient temperatures and seston concentrations over an annual cycle. We calculated the clearance rate of a 1 g dry tissue weight adult eastern oyster at each monthly water temperature and seston load (Table 2) from Jordan’s (1987) data (reproduced as Fig. 15 in Newell and Langdon 1996) by dividing the weight of biodeposits voided per unit time by the seston concentration. This method provides the most accurate possible laboratory estimates of bivalve feeding rates (Cranford 2001). We then calculated the amount of chlorophyll a removed from the water column and converted this to N removed using a Chlorophyll a :Nitrogen ratio of 1 g Chl a :14 g N (Parsons et al. 1984). We applied an average N assimilation efficiency of 50% derived from physiological data for eastern oysters (Newell and Jordan 1983) to estimate the total amount of undigested PON that was egested as biodeposits by the oysters per month. We estimated the amount of PON from biodeposits that becomes buried in the sediments by applying a burial rate of 10%, from field data collected in the Choptank River by Boynton et al.(1995), to these monthly biodeposition data (Table 2). We estimated the amount of particulate P transferred to the sediments by applying an average N:P molar ratio of 18:1 (Table 1) to the monthly PON values calculated for oyster biodeposition and then assumed that 90% of P becomes buried in the sediments (Table 2). This 90% value was estimated from the data reviewed above that indicates negligible P release from bivalve biodeposits in aerobic sediments. We used a laboratory derived estimate that 20% of the PON in biodeposits is denitrified (Newell et al. 2002) to calculate the monthly amount of N removed from biodeposits associated with coupled nitrification-denitrification (Table 2). This 20% denitrification value is conservative as it is at the low end of the range calculated from literature data by Seitzinger (1988), who estimated that 20 to 70% of the total N flux from coastal marine sediments is in the form of N2 gas. Many seasonal and physical factors, such as temperature, sediment porosity, water flow, abundance of bioturbators, and sediment oxygen content, can alter the rate of coupled nitrification-denitrification. The combined nitrogen loss rate of 30% we used in our calculations [burial rate (10%) and denitrification (20%)] is close to the 30% annual loss estimated for
107 Table 2. Monthly average water temperature (oC), seston concentrations (mg L-1), and phytoplankton chlorophyll a (micro g L-1) in the Choptank River (EPA Chesapeake Bay Program monitoring station ET 5.2). Clearance rates (L h-1g-1 dry tissue weight) calculated from Jordan (1987) were used to estimate the monthly amount of phytoplankton N filtered from the water column. We then estimated the monthly amounts of N and P biodeposition that were buried (mg month-1g-1 dry tissue weight) and N denitrified (mg month-1g-1 dry tissue weight). See text for complete details of these calculations.
Water Temp oC Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
3 3 6 11 17 23 27 27 25 19 11 6 Annual Total
Seston (mg L-1)
Chl a (microg L-1)
Clearance Rate (L h-1g-1 )
11.4 14.3 13.2 16.7 14.5 10.7 13.0 13.0 13.4 12.8 9.4 11.4
5.5 8.7 8.9 9.6 12.2 12.3 15.4 16.0 11.9 7.3 6.0 5.7
0 0 0.45 0.90 1.72 3.74 9.62 9.62 7.46 2.34 1.38 0.44
monthly nutrient removal g-1 dry tissue weight mg N denitrified 0 0 4.08 8.69 21.20 46.35 149.26 155.08 89.52 17.25 8.36 2.52 502.31
mg N buried 0 0 2.04 4.35 10.60 23.17 74.63 77.54 44.76 8.62 4.18 1.26 251.15
mg P buried 0 0 2.21 4.71 11.50 25.13 80.92 84.08 48.53 9.35 4.53 1.37 272.34
natural intertidal oyster reefs by Dame et al. (1989). They estimated this loss term from field studies in which they found that the difference between the seasonal average fluxes into an oyster reef (T in Fig. 1) and regeneration (R in Fig.1) was about 33%. Dame et al. (1989) suggested that the difference between these two fluxes was associated with N incorporation into oysters and other organisms, denitrification, and burial in the sediments. In order to place our estimates of biomass-specific N and P removal into an ecosystem perspective we calculated the size of the eastern oyster population currently present in the upper Choptank River. The total standing stock of oysters > 7.6 cm shell height in Maryland’s portion of Chesapeake Bay in 2002 was estimated to be 342.3 x 106 g DW (SJ Jordan, pers. comm., MD Dept. Natural Resources). Assuming that these oysters are uniformly distributed over the 800 km2 of oyster habitat in Maryland (Smith et al. 2001), adult 1 g dry tissue weight oysters are currently at a population density of ~0.43 oysters m-2. Because the lower salinities in the upper portion of the Choptank River reduces the virulence of MSX and Dermo epizootics (Ford and Tripp 1996), we assumed that the extant oyster stocks were at a density of 1 oyster m-2. We used acoustic survey data (Smith et al. 2001) on the aerial extent of oyster bars to calculate that there are 1,736 ha of oyster bottom in the upper Choptank River. At the current oyster abundance of 1 m-2 this area of oyster bottom supports a total of ~ 17 x 106 oysters (>7.6 cm shell length).
108 We estimated monthly N and P inputs into the drainage basin of the Choptank River with the hydrochemical model, Generalized Watershed Loading Functions (Table 3). The GWLF model was calibrated with data from the USGS gauging station at Greensboro, MD., that receives inputs from 17% of the land area of the Choptank watershed (Fisher et al. 1998, Lee et al. 2000). Diffuse source inputs of N and P from ungauged areas were modeled using local data on human populations, soils, land-use patterns, atmospheric deposition, and measured waste water treatment plant discharges were added as point sources (Lee et al. 2001). Inputs from all three source types raise N concentrations in the estuary during winter and spring (Fig. 3) and stimulate phytoplankton production later in the year (Fig. 3) when temperatures are higher (Fisher et al. 1988, 1998, Berndt 1999). Inputs of nutrients from the watershed and airshed are out of phase with the period of maximum phytoplankton biomass. Inputs of DIN were highest in January to May (Table 3) due to high river discharge in these months transporting terrestrial nutrients. High DIN concentrations (predominately NO3-) along the length of the estuary at this time of the year (Fig. 3) indicate that nutrients are effectively being stored in the estuary. In contrast to the high delivery of nutrients at low temperatures, in summer nutrients decline because removal processes (e.g., accumulation if N in phytoplankton biomass) exceed inputs (Fig. 3). We estimated the monthly amounts of N buried and denitrified and P buried by eastern oysters at their current population density of 1 m-2 on 1,736 ha of oyster bottom in the upper portion of the Choptank River (Table 3 ). We then expressed these monthly removal rates associated with 1 oyster/m2 as a percentage of monthly inputs (Fig. 4). Seasonal N and P removal associated with biodeposition at current oyster densities were greatest (~5% and ~34%), respectively, from July to September due to highest oyster feeding activity at this time of greatest phytoplankton abundance. We also estimated monthly nutrient removal for oysters at densities of 10 m-2 (Fig. 4) that corresponds to the Chesapeake Bay 2000 Agreement oyster restoration goal of a ten fold higher oyster abundance by 2010 (www.chesapeakebay.net). At this modest oyster density of 10 m-2, ~50% and ~350% of the monthly summer N and P inputs, respectively, were estimated to be removed: i.e., P would be removed faster than inputs and half of the N inputs would be removed. Even this oyster restoration goal is possibly a factor of 10 below the ecosystem carrying capacity. Newell (1988) estimated 1988 oyster stocks to be ~1% of historical densities present in Chesapeake Bay prior to commercial exploitation in the 19th Century. Furthermore, adult oyster abundances of 10 m-2 (10 g DW m-2) are low for eastern oysters living in natural reefs. For example, Dame (1976) reported that intertidal reefs in North Inlet, South Carolina had an oyster biomass of between 300 to 500 g DW m-2.
109 Table 3. Table 3. Total monthly N and P (kg) inputs into the Choptank watershed and airshed estimated by Lee et al. (2001). Values presented are averages for 1980 to 1996. The total amounts (kg month-1) and % of the monthly N and P inputs that are buried and denitrified associated with biodeposition from oysters at a density of 1 g DW m-2 on 1,736 ha of restorable oyster bottom in the Choptank River were calculated as described in the text.
Total-N inputs (kg) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total
281,450 261,970 312,350 292,500 243,930 148,250 75,480 80,810 99,140 97,940 114,500 254,270 2,262,580
Total-P Monthly nutrient removal for oysters at a density of inputs 1 m-2 on 1,736 ha oyster bottom (kg) N (kg) P (kg) % N inputs % P inputs 5,245 4,837 5,351 5,338 6,022 4,641 4,059 4,274 4,587 4,015 4,344 6,373 59,085
0 0 106 226 552 1,207 3,887 4,038 2,331 449 218 66 13,080
DIN Concentration
0 0 38 82 200 436 1,405 1,460 843 162 79 24 4,728
0.0 0.0 0.0 0.1 0.2 0.8 5.1 5.0 2.4 0.5 0.2 0.0 0.6
0.0 0.0 0.7 1.5 3.3 9.4 34.6 34.2 18.4 4.0 1.8 0.4 8.0
Chlorophyll a concentration
Fig. 3. Temporal and Spatial variability in dissolved inorganic nitrogen inputs (micro M) and Chlorophyll a (micro g L-1) over an annual cycle (July 1997 to June 1998) from the mouth towards the headwaters of the Choptank River Estuary, MD (Berndt 1999).
110 We summed the total annual amount of N and P removed for oysters at population densities of 1, 10, and 100 m-2 on 1,737 ha of oyster bottom and expressed this as a % of total annual N and P inputs to estimate their potential contribution at various abundances (Fig. 4). At their current population density of 1 m-2, oysters can only remove 0.6 and 8% of annual N and P inputs respectively. Obviously, if oyster stocks are increased tenfold as a consequence of management activities then this will increase to 6% and 80% of N and P inputs, respectively. The factors that govern N and P removal via bivalve biodeposition are complex and variable; consequently, in our model we have made a number of simplifying assumptions to predict the magnitude of nutrient removal. The magnitude of our estimates are sensitive to the phytoplankton Chl a concentrations (Table 2). Phytoplankton blooms are highly variable spatially and temporally (Fig. 3; Fisher et al 1988, Harding et al 2002) and, therefore, our calculations can only be considered an approximation of the magnitude of possible N and P removal associated with eastern oyster feeding. Field studies can be designed to measure the actual N and P fluxes and burial
Fig. 4. The monthly amounts of N and P buried and denitrified (Table 3) expressed as % of monthly inputs into the Choptank River MD. N and P were removed via biodeposition by 1 g oysters at a density of 1 and 10 m-2 on 1,737 ha of oyster bottom in the Choptank River. The % removal of total annual N and P inputs at a density of 1, 10, and 100 m-2 oysters are tabulated.
measurements in sediment cores collected seasonally from around natural stocks of bivalves. Information on how much of the biodeposits from the bivalve population are incorporated into the sediment is also necessary in order to assess the incremental changes in sediment nutrient regeneration and burial associated with a certain amount of biodeposition. Unfortunately, this is a major technical limitation as we do not currently have a usable method to
111 determine how widely the biodeposits are being distributed across receiving sediments.
DISCUSSION Prevailing eutrophic conditions in Chesapeake Bay are commonly ascribed to anthropogenic nutrient enrichment from the watershed and airshed (D’Elia et al.1992). Undoubtedly this is a major causal factor in stimulating excess primary production that ultimately leads to oxygen deficient bottom waters (Boicourt 1992, Kemp and Boynton 1992). In addition, the demise of the once abundant suspension-feeding eastern oyster has drastically reduced benthic consumption of phytoplankton (Newell 1988), and this has likely caused major disruptions to the Chesapeake Bay food web (Ulanowicz and Tuttle 1992). Today, pelagic metazoan and microbial organisms are the major consumers of phytoplankton (Baird and Ulanowicz 1989), and N and P regenerated by these organisms are released directly back to the water column where they can sustain primary production. In the mesohaline area of Chesapeake Bay, large amounts of phytoplankton that are unconsumed in the upper water column accumulate in the oxygen-poor waters beneath the pycnocline where N and P are regenerated as NH4+ and PO43- from anaerobic sediments (Kemp and Boynton 1992). Current management efforts to reduce levels of phytoplankton production in Chesapeake Bay are based on a policy of reducing point and non-point source nutrient inputs (D’Elia et al.1992). While anthropogenic nutrient loadings are indeed the most critical component of curbing excess phytoplankton production, we suggest that the effect of eastern oysters in reducing phytoplankton concentrations and altering patterns of nutrient regeneration may also be crucial to achieving the long-term goal of improving water quality. Our analysis of the interactions between eastern oyster feeding and nutrient inputs to the Choptank estuary indicates that the highest N and P inputs occur at times of lowest benthic-pelagic coupling associated with oyster feeding (winter /spring); whereas times of maximum oyster feeding correspond to times of maximum primary production (Fig. 3; Malone 1992). Despite these apparent imbalances, eastern oysters can exert a direct influence on reducing N and P recycling by enhanced feeding on the accumulated phytoplankton biomass and removing some of this biomass to the sediments. Our proposition that eastern oysters can be extremely important in governing the response of Chesapeake Bay to nutrient enrichment fits the new conceptual model of coastal eutrophication developed by Cloern (2001). Cloern (2001) proposed that coastal systems do not respond either simply or uniformly to nutrient enrichment and that “system-specific attributes act as a filter to modulate the (system) response to enrichment”. He suggests that a complete understanding of the biological responses to nutrient enrichment within the physical regime of a particular ecosystem is a prerequisite to
112 developing strategies necessary for ecosystem restoration and rehabilitation. We agree with Cloern’s (2001) assessment, and we further suggest that three factors largely determine estuarine response to nutrient inputs: 1) Sufficient water residence time to allow consumption of nutrient inputs (typically > 30 d), 2) sufficient water clarity (or a sufficiently shallow upper mixed layer; Fisher et al. 2003) to permit net phytoplankton growth, and 3) the presence of active and abundant suspension-feeding bivalves in the shallow waters. The first two factors concern the ability of phytoplankton to intercept the incoming nutrients, whereas the last concerns the interception of phytoplankton by the benthos. Increasing oyster populations in Chesapeake Bay will not be without cost. For a commercially harvested species, such as the eastern oyster, there is an obvious trade-off between harvesting oysters and leaving them on the bottom to provide ecosystem services. One way to compare these divergent uses is to denominate their respective benefits in monetary terms. The dockside harvest value of eastern oysters is ~$25 per bushel of whole live oysters [a bushel is the customary harvest unit (1 MD bushel 0.046 m-33 300 oysters of 7.6 cm shell length)]. At the current assumed oyster abundance of 1 m-2, the 1,736 ha of oyster bottom in the upper Choptank estuary supports a total of 17.36 x 106 oysters (>7.6 cm shell length). If this entire stock of oysters were harvested they would yield ~57.9 x 103 Maryland bushels (2,663 m-3) with a dockside value of ~$1.5 million. While consumers may value this harvest at $1.5 million, it must be noted that its value to harvesters is significantly less. Factoring in costs of harvesting (fuel, labor, and capital), it is likely that the net value to harvesters of harvesting the entire stock of oysters would be less than one half the total dockside value (<$750,000). Our simple model indicates that present-day oyster stocks of 1 m-2 in the upper Choptank River are responsible for burying and denitrifying 13,080 kg N and 4,728 kg P annually (Table 3). The value of this nutrient reduction in dollar terms is the sum of all commercial and non-commercial benefits generated by its impact on water quality. A direct estimate of these benefits is beyond the scope of the present study. However, a much simpler way to estimate a value for these reductions is to determine their opportunity cost. That is, what is the cost of alternative ways of obtaining these same nutrient reductions? To the extent that reductions of any given amount of nutrients by oysters obviate the need to incur those costs, this is their value in terms of nutrient reduction. The EPA Chesapeake Bay Program has recently undertaken a “Use Attainability Analysis” that provides estimates of the cost of reducing nutrient inputs necessary for meeting the water quality goals in the Chesapeake Bay (http://www.chesapeakebay.net/uaasupport.htm). / The annual Baywide nutrient reduction requirement (relative to year 2000 loads) is 47.17 x 106 kg N and it is estimated that it will cost $1,138 million annually (operating and amortized capital costs) to achieve this level of N reduction.
113 Given these available figures, the average cost of removing N from Chesapeake Bay loads is $24.07 kg-1. This average masks large variations, however, with costs ranging from $4.6 kg-1 for planting cover crops to $1,250 kg-1 (www.chesapeakebay.net/ecoanalyses.htm) to implement erosion and sediment controls. However, under the assumption that reducing nutrient loads in order to achieve desired water quality outcomes is justified on a cost-benefit basis, this average unit reduction cost provides a useful measure of the nutrient reduction value of oysters. With respect to the modeled nutrient reduction estimates, the calculated value of the entire stock of oysters in the upper Choptank estuary in removing 13,080 kg N annually is $314,836. While this is less than the one-time benefit of harvesting these oysters, oysters remaining in the system will continue to generate nutrient reduction benefits over their 10+ y life span. Taking this into account, the ecosystem value of current Choptank River oyster stocks increases to $3.1 million, or over twice the value of harvesting them. This analysis ignores the economic value of the hard reef substrate as habitat for many other animal species that have value for commercial and recreational fisheries (Coen et al.1999). It should be noted, however, that some nutrients are removed when oysters are harvested because oyster tissue and shell contains phosphorus (~0.8% and ~0.1% dry weight respectively) and nitrogen (~7% and ~0.3% dry weight respectively (Galtsoff 1964, Newell, unpublished data). For a 1 g dry tissue weight oyster of shell length 7.6 cm the shell weighs ~150 g. When this size oyster is harvested 0.52 g N and 0.16 g P are removed in flesh and shell, which is ~2/3 of the amounts estimated to be removed annually as a consequence of the oysters normal feeding and deposition processes (Table 2) Newell (1988) calculated that the oyster stocks that existed in Chesapeake Bay prior to heavy exploitation in the 19th Century required between 23% and 41% of present-day phytoplankton carbon production to sustain their metabolic carbon demands. How is it possible that these oyster stocks required such a large proportion of primary production that is currently sustained by high levels of anthropogenic nutrient enrichment? We speculate that these highly abundant oysters were food limited and slow growing and therefore had high food digestion and assimilation efficiencies. Consequently, denitrification and N and P burial may have been minimal because little POM was voided in their biodeposits. It is likely that excreted NH4+ promoted tight N coupling in what used to be an oligotrophic estuary. How did the system change when that tight coupling between phytoplankton production and benthic consumption was changed? Were these changes necessarily linear? It is plausible that phytoplankton biomass in Chesapeake Bay was not appreciably enhanced by the initial loss of eastern oyster grazing until a critical threshold of lower oyster abundance and higher nutrient inputs was reached.After that point the ecosystem switched from one where benthic processes dominated to the current situation where phytoplankton mainly
114 flows through pelagic consumers with substantial nutrient regeneration through the microbial food web (Baird and Ulanowicz 1989). Ecological theory suggests that when a dominant species in an ecosystem is lost, the vacant niche is filled by another species. What has happened to the niche once occupied by eastern oysters in Chesapeake Bay? Pelagic suspension feeders, such as zooplankton and various planktivorous fish species may have increased in abundance due to reduced competition for phytoplankton; however, they are not sufficiently abundant to consume all of present-day phytoplankton production (Baird and Ulanowicz 1989). Certainly in some parts of Chesapeake Bay, benthic consumers, such as polychaetes (Thompson and Schaffner 2001) have probably increased to the point that today they are potentially filling part of the ecological niche once occupied by oysters. In the tidal freshwater and oligohaline reaches of some tributaries the asiatic clam Corbicula fluminea and the wedge clam, Rangia cuneata, can attain extremely high densities (Gerritsen et al.1994). These authors estimated that >50% of the annual phytoplankton production in the oligohaline regions is consumed by these abundant benthic suspension feeders. But it is in the mesohaline portions of the Bay, the region where eastern oysters were once highly abundant Baywide, which no benthic consumer, perhaps with the exception of the tunicate, Molgula manhattensis, seems to have filled the niche once occupied by oysters. Yet it is in this mesohaline region of the Bay that nutrient enrichment generates the highest level of primary production (Harding et al. 2002). Consequently, it is this unconsumed autochthonous carbon production from this region that settles to the sediment surface beneath the pycnocline and is responsible for generating bottom water hypoxia and anoxia (Boicourt 1992, Kemp and Boynton 1992). It is this anoxic bottom water that is one of the obvious signs of ecosystem degradation that ongoing management activities are trying to correct. In summary, eastern oysters, in common with other suspensionfeeding bivalves, serve to regenerate large quantities of N to the water column. It is also apparent that bivalve biodeposition can enhance the permanent removal of N and P from the water column via burial and denitrification, thereby reducing the amounts of these two nutrients available to maintain phytoplankton production. The relative balance between nutrient regeneration and permanent removal depends on the amount of biodeposition by the particular species of bivalve and the environmental conditions (Fig. 1). We suggest that a possible management strategy for improving water quality in Chesapeake Bay is to increase top-down control on phytoplankton through the enhancement of natural eastern oyster populations. This restoration strategy would be most beneficial if it promoted enhancement of bivalves in well-mixed waters because this leads to enhanced nitrogen loss through coupled nitrification-denitrification and the burial of both N and P in the accumulating sediments. In such well-oxygenated locations, the further
115 decomposition of POM-rich biodeposits by both metazoan and microbial decomposers will not lead to the development of hypoxic bottom waters and sediments. Unfortunately, due to ongoing epizootics of Dermo and MSX disease, the long term survival of restored eastern oyster beds in regions of Chesapeake Bay with salinities >12 must be questioned (Jordan 1995, Ford and Tripp 1996), and hence reliance on these populations to achieve longterm water quality goals is not recommended. Despite the uncertainty about the survival of oysters, we believe that they can be a useful supplement to management activities that are centered on controlling nutrient inputs. The use of oysters to help attain water quality goals is important because it offers one of the few opportunities to reduce nutrients once they have entered a receiving body of water.
ACKNOWLEDGMENTS This review was supported with funding to Newell and Cornwell from Maryland Sea Grant (SA07528051-F) through NOAA award NA16RG2207and to Fisher from NASA’s LCLUC program and Maryland Department of Natural Resources. We are grateful to Robert Wieland for his encouragement and helping with the economic analyses. We thank Dr. G. Smith for the oyster bar survey data and Greg Radcliffe for drawing some of the figures. The US government is authorized to publish reprints of this work and the authors reserve the right to post a copy on their academic websites for the private and non-commercial use of individuals, notwithstanding any copyright notations hereon.
REFERENCES An S
Joye SB 2001 Enhancement of coupled nitrification-denitrification by benthic photosynthesis in shallow estuarine sediments. Limnol Oceanogrr 46: 62-74 Asmus RM Asmus H 1991 Mussel beds, limiting or promoting phytoplankton? J Exp Mar Biol Ecoll 148: 215-232 Baird D Ulanowicz RE 1989 The seasonal dynamics of the Chesapeake Bay ecosystem. Ecol Monogr 59: 329-364 Bartleson RD Kemp WM 1990 Preliminary ecosystem simulations of estuarine planktonicbenthic interactions: The planktonic submodel. In: New Perspectives in the Chesapeake System: A Research and Management Partnership. Proceedings of a Conference. JA Mihursky and A Chaney (Eds). Chesapeake Research Consortium Publication No.137 Solomons MD USA pp 243-252. (Available from www.vims.edu/GreyLit/CRC/crc137) Bayne BL Hawkins AJS 1992 Ecological and physiological aspects of herbivory in benthic suspension-feeding molluscs. In: Plant-Animal Interactions in the Marine Benthos. Systematics Association Special Volume No. 46. DM John SJ Hawkins and JH Price (Eds). Clarendon Press, Oxford, UK, pp 265-288 Bayne BL Klumpp DW Clarke KR 1984 Aspects of feeding, including estimates of gut residence time, in three mytilid species (Bivalvia, Mollusca) at two contrasting sites in the Cape Peninsula, South Africa. Oecologia 64: 26-33 Bayne BL Newell RC 1983 Physiological energetics of marine molluscs. In: The Mollusca,
116 Volume 4. Physiology, Part 1, ASM Saleuddin and KM Wilbur (Eds), Academic Press, New York, USA, pp 407-515 Boicourt WC 1992 Influences of circulation processes on dissolved oxygen in the Chesapeake Bay. In: Oxygen Dynamics in the Chesapeake Bay, DE Smith M Leffler and G Mackiernan (Eds), Maryland Sea Grant Publication, College Park, MD, USA, pp 7-59 Boynton WR Garber JH Summers R Kemp WM 1995 Input, transformations, and transport of nitrogen and phosphorus in Chesapeake Bay and selected tributaries. Estuaries 18: 285-314 Boynton WR Kemp WM 1985 Nutrient regeneration and oxygen consumption by sediments along an estuarine salinity gradient. Mar Ecol Prog Serr 23: 45-55 Berndt HL 1999 Effects of nutrients and turbidity from point and non-point source inputs on phytoplankton in the Choptank estuary. MS thesis, Univ. of MD, College Park, MD, 168 pp Coen LD Luckenbach MW Breitburg DL 1999 The role of oyster reefs as essential fish habitat: A review of current knowledge and some new perspectives. American Fisheries Society, Symposium 22, Bethesda, MD, 438-454 Cloern JE 1982 Does the benthos control phytoplankton biomass in South San Francisco Bay? Mar Ecol Prog Serr 9: 191-202 Cloern JE 2001 Our evolving conceptual model of the coastal eutrophication problem. Mar Ecol Prog Serr 210: 223-253 Cohen RRH Dresler PV Philips EJP Cory RL 1984 The effect of the asiatic clam Corbicula fluminea, on phytoplankton of the Potomac River, Maryland. Limnol Oceanogrr 29: 170-180 Conley DJ 1999 Biogeochemical nutrient cycles and nutrient management strategies Hydrobiologia 410: 87-96 Cornwell JC Kemp WM Kana TM 1999 Denitrification in coastal ecosystems: methods, environmental controls, and ecosystem level controls, a review. Aquat Ecol 33: 41-54 Cranford PJ 2001 Evaluating the 'reliability' of filtration rate measurements in bivalves. Mar Ecol Prog Serr 215: 303-305 Dame RF 1976 Energy flow in an intertidal oyster population. Est Coast Mar Sci 4: 243-253 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Dame RF Dankers N 1988 Uptake and release of materials by a Wadden Sea mussel bed. J Exp Mar Biol Ecoll 118: 207-216 Dame RF Dankers N Prins TC Jongsma H Smaal AC 1991a The influence of mussel beds on nutrients in the western Wadden Sea and eastern Scheldt estuaries. Estuaries 14: 130-138 Dame RF Libes S 1993 Oyster reefs and nutrient retention in tidal creeks. J Exp Mar Biol Ecoll 171: 251-258 Dame RF Spurrier JD Williams TM Kjerfve B Zingmark RG Wolaver TG Chrzanowski T H McKellar HN Vernberg FJ 1991b Annual material processing by a salt marshestuarine basin in South Carolina, USA. Mar Ecol Prog Serr 72: 153-166 Dame RF Spurrier JD Wolaver TG 1989 Carbon, nitrogen and phosphorus processing by an oyster reef. Mar Ecol Prog Serr 54: 249-256 Dame RF Spurrier JD Zingmark RG 1992 In situ metabolism of an oyster reef. J Exp Mar Biol Ecol 164: 147-159 D’Elia CF Harding Jr. LW Leffler M Mackiernan GB 1992 The role and control of nutrients in Chesapeake Bay. Water Sci Technoll 26: 2635-2644 Deslous-Paoli J-M Lannou A-M Geairon P Bougrier S Raillard O Heral M 1992 Effects of the feeding behaviour of Crassostrea gigas (Bivalve Mollusc) on biosedimentation of natural particulate matter. Hydrobiologia 231: 85-91 Doering PH Kelly JR Oviatt CA Sowers T 1987 Effect of the hard clam Mercenaria
117 mercenaria on benthic fluxes of inorganic nutrients and gases. Mar Bioll 94: 377-383 Doering P H Oviatt CA 1986 Application of filtration rate models to field populations of bivalves: as assessment using experimental mesocosms. Mar Ecol Prog Serr 31: 265275 Doering PH Oviatt CA Kelly JR 1986 The effects of the filter-feeding clam Mercenaria mercenaria on carbon cycling in experimental marine mesocosms. J Mar Res 44: 839861 Fisher TR Gustafson AB Radcliffe GR Sundberg KL Stevenson JC 2003 A long-term record of photosynthetically active radiation (PAR) and total solar energy at 38.6DEG N,78.2DEG W. Estuaries 26: 1450-1460 Fisher TR Harding LW Stanley DW Ward LG 1988 Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and Hudson River Estuaries. Est Coastal Shelf Sci 27: 61-93 Fisher TR Lee K-Y Berndt H Benitez JA Norton MM 1998 Hydrology and chemistry of the Choptank River basin in the Chesapeake Bay drainage. Water Air Soil Polll 105: 387-397 Ford SE Tripp MR 1996 Diseases and Defense Mechanisms. In The Eastern Oyster, Crassostrea virginica, VS Kennedy RIE Newell and A Eble (Eds), Maryland Sea Grant Publication, College Park, MD, USA, pp 581-660. Galtsoff PS 1964 The American Oyster. Fish Bulll 64: 1-480 Gerritsen JA Holland AF Irvine DE 1994 Suspension-feeding bivalves and the fate of primary production. An estuarine model applied to Chesapeake Bay. Estuaries 17: 403-416 Grizzle RE Shumway SE Bricelj VM 2001 Physiology and Bioenergetics of Mercenaria mercenaria. In: The Hard clam, Mercenaria mercenaria. JN Kraeuter and M Castagna (Eds). Elsevier, Amsterdam, NL, pp 305-382 Harding LW Mallonee ME Perry ES 2002 Toward a predictive understanding of primary productivity in a temperate, partially stratified estuary. Estuar Coastal Shelf Sci 55: 437-463 Hatcher A Grant J Schofield B 1994 Effects of suspended mussel culture (Mytilus spp.) on sedimentation, benthic respiration and sediment nutrient dynamics in a coastal bay. Mar Ecol Prog Serr 115: 219-235 Haven DS Morales-Alamo R 1966 Aspects of biodeposition by oysters and other invertebrate filter feeders. Limnol Oceanogrr 11: 487-498 Haven DS Morales-Alamo R 1968 Occurrence and transport of faecal pellets in suspension in a tidal estuary. Sed Geoll 2: 141-151 Hawkins AJS Bayne BL Bougrier S Heral M Iglesias JIP Navarro E Smith RFM Urrutia MB 1998 Some general relationships in comparing the feeding physiology of suspension-feeding bivalve molluscs. J Exp Mar Biol Ecoll 219: 87-103 Henriksen K Kemp WM 1988 Nitrification in estuarine and coastal marine sediments. In: Nitrogen Cycling in Coastal Marine Environments. TH Blackburn J Sorensen (Eds). Wiley and Sons, Chichester, England, pp 205-249 Jaramillo E Bertran C Bravo A 1992 Mussel biodeposition in an estuary in southern Chile. Mar Ecol Prog Serr 82: 85-94 Jordan SJ 1987 Sedimentation and remineralization associated with biodeposition by the American oyster Crassostrea virginica (Gmelin). Doctoral dissertation. University of Maryland, College Park, MD, USA, 200 p Jordan SJ 1995 Population and disease dynamics of Maryland oyster bars: a multivariate classification analysis. J Shellfish Res 14: 459-468 Kana TM Darkangelo C Hunt MD Oldham JB Bennett GE Cornwell JC 1994 A membrane inlet mass spectrometer for rapid high precision determination of N2 , O2 , and Ar in environmental water samples. Anal Chem 66: 4166-4170 Kaspar HF Gillespie PA Boyer IC MacKenzie AL 1985 Effects of mussel aquaculture on the nitrogen cycle and benthic communities in Kenepuru Sound, Marlborough Sounds, New Zealand. Mar Bioll 85: 127-136
118 Kautsky N Evans S 1987 Role of biodeposition by Mytilus edulis in the circulation of matter and nutrients in a Baltic coastal ecosystem. Mar Ecol Prog Serr 38: 201-212 Kemp WM Boynton WR 1992 Benthic-Pelagic interactions: Nutrient and oxygen dynamics. In: Oxygen Dynamics in the Chesapeake Bay, DE Smith M Leffler and G Mackiernan (Eds). Maryland Sea Grant Publication, College Park, M.D, USA, pp 149-221 Kennedy VS Breisch LL 1981 Maryland's Oysters, Research and Management, Maryland Sea Grant, University of Maryland, College Park, MD, USA, 286 p Kreeger DA Newell RIE 2001 Seasonal utilization of different seston carbon sources by the ribbed mussel, Geukensia demissa (Dillwyn) in a mid-Atlantic salt marsh. J Exp Mar Biol Ecoll 260: 71-91 Kristensen E 1988 Benthic fauna and biogeochemical processes in marine sediments, microbial activities and fluxes. In: Nitrogen Cycling in Coastal Marine Environments. TH Blackburn and J Sorensen (Eds). Wiley and Sons, Chichester, England, p 275-299 Krom MD 1991 Importance of benthic productivity in controlling the flux of dissolved inorganic nitrogen through the sediment-water interface in a hypertrophic marine ecosystem. Mar Ecol Prog Serr 78: 163-172 Krom MD Berner RA 1981 The diagenesis of phosphorus in a nearshore marine sediment. Geochim Cosmo Acta 45: 207-216 Lee K-Y Fisher TR Jordan TE Correll DL Weller DE 2000 Modeling the hydrochemistry of the Choptank River basin using GWLF and Arc/Info: 1. Model calibration and validation. Biogeochem 49: 143-173 Lee K-Y Fisher T R Rochelle-Newall E 2001 Modeling the hydrochemistry of the Choptank River basin using GWLF and Arc/Info: 2. Model Application. Biogeochem 56: 311-348 MacIntyre HL Geider RJ Miller DC 1996 Microphytobenthos: The ecological role of the "secret garden" of unvegetated, shallow-water marine habitats. I. Distribution, abundance and primary production. Estuaries 19: 186-201 MacKenzie CL Jr. 1996 Management of natural populations In The Eastern Oyster, Crassostrea virginica, VS Kennedy RIE Newell and A Eble (Eds). Maryland Sea Grant Publication, College Park, MD, USA, pp 707-721 Magni P Montani S Takada C Tsutsumi H 2000 Temporal scaling and relevance of bivalve nutrient excretion on a tidal flat of the Seto Inland Sea, Japan. Mar Ecol Prog Ser 198: 139-155 Malone TC 1992 Effects of water column processes on dissolved oxygen, nutrients, phytoplankton and zooplankton. In: Oxygen Dynamics in the Chesapeake Bay, DE Smith M Leffler and G Mackiernan (Eds), Maryland Sea Grant, College Park, MD, USA, pp 61-112 Newell RIE 1988 Ecological Changes in Chesapeake Bay, Are they the result of overharvesting the Eastern oyster (Crassostrea virginica)? In: Understanding the Estuary, MP Lynch EC Krome (Eds). Chesapeake Research Consortium Publication 129 (CBP/TRS 24/88), Gloucester Point, VA, USA. pp 536-546. (Available at www.vims.edu/GreyLit/crc129.pdf) Newell RIE 2004 Ecosystem influences of natural and cultivated populations of suspensionfeeding bivalve molluscs: A review. J Shellfish Res 23: 51-61 Newell RIE Cornwell JC Owens MS 2002 Influence of simulated bivalve biodeposition and microphytobenthos on sediment nitrogen dynamics, a laboratory study. Limnol Oceanogrr 47: 1367-1379 Newell RIE Jordan SJ 1983 Preferential ingestion of organic material by the American Oyster Crassostrea virginica Mar Ecol Prog Serr 13: 47-537 Newell RIE Koch EW 2004 Modeling seagrass density and distribution in response to changes in turbidity stemming from bivalve filtration and seagrass sediment stabilization. Estuaries. In Press Newell RIE CJ Langdon 1996 Mechanisms and Physiology of Larval and Adult Feeding. In
119 The Eastern Oyster, Crassostrea virginica, VS Kennedy RIE Newell A Eble (Eds). Maryland Sea Grant Publication, College Park, MD, USA, pp 185-230 Officer CB Smayda TJ Mann R 1982 Benthic filter feeding, A natural eutrophication control. Mar Ecol Prog Serr 9: 203-210 Parsons TR Takahashi M Hargrave B 1984 Biological Oceanographic Processes. Pergamon Press. 330 p Peckol PJ Rivers JS 1995 Physiological responses of the opportunistic macroalgae Cladophora vagabunda (L.) van den Hoek and Gracilaria tikvahiae (McLachlan) to environmental disturbances associated with eutrophication. J Exp Mar Biol Ecoll 190: 1-16 Pelegri SP Blackburn TH 1995 Effect of bioturbation by Nereis sp., Mya arenaria, and Cerastoderma sp. on nitrification and denitrification in estuarine sediment. Ophelia 42: 289-299 Pietros JM Rice MA 2003 The impacts of aquacultured oysters Crassostrea virginica (Gmelin, 1791) on water column nitrogen and sedimentation: results of a mesocosm study. Aquaculture 220: 407-422 Prins TC Smaal AC 1990 Benthic-pelagic coupling: The release of inorganic nutrients by an intertidal bed of Mytilus edulis, In: Trophic Relationships in the Marine Environment. M Barnes and RN Gibson (Eds). Proc. 24th European Marine Biology, Symposium, Aberdeen University Press, Scotland, pp 89-103 Prins TC Smaal AC 1994 The role of the blue mussel Mytilus edulis in the cycling of nutrients in the Oosterschelde estuary (The Netherlands). Hydrobiologia 283: 413429 Prins TC Smaal AC Dame RF 1998 A review of the feedbacks between bivalve grazing and ecosystem processes. Aquat Ecol 31: 349-359 Rafaelli DG Raven JA Poole LJ 1998 Ecological impact of green macroalgal blooms. Oceanogr Mar Biol Ann Revv 36: 97-125 Risgaard-Petersen N Rysgaard S Nielsen LP Revsbech NP 1994 Diurnal variation of denitrification and nitrification in sediments colonized by benthic microphytes. Limnol Oceanogrr 39: 573-579 Rothschild BJ Ault JS Goulletquer P Heral M 1994 Decline of the Chesapeake Bay oyster population: a century of habitat destruction and overfishing. Mar Ecol Prog Serr 111: 29-39 Rysgaard S Christensen PB Nielsen LP 1995 Seasonal variation in nitrification and denitrification in estuarine sediment colonized by benthic microalgae and bioturbating infauna. Mar Ecol Prog Serr 126: 111-121 Sanford LP Chang M-L 1997 The bottom boundary condition for suspended sediment deposition. J Coastal Res Special Issue 25: 3-17 Seitzinger SP 1988 Denitrification in freshwater and coastal marine ecosystems, ecological and geochemical significance. Limnol Oceanogrr 33: 702-724 Silvert W Cromey CJ 2001 Modeling impacts In: Environmental Impacts of Aquaculture, KD Black (ed), CRC Press Boca Raton FL, pp 154-181 Smith GF Greenhawk KN Bruce DG Roach EB Jordan SJ 2001 A digital presentation of the Maryland oyster habitat and associated bottom types in the Chesapeake Bay (1974-1983). J Shellfish Res 20: 192-206 Souchu P Vaquer A Collos Y Landrein S Deslous-Paoli J-M Bibent B 2001 Influence of shellfish farming activities on the biogeochemical composition of the water column in Thau lagoon. Mar Ecol Prog Serr 218: 141-152 Srna RF Baggaley A 1976 Rate of excretion of ammonia by the hard clam Mercenaria mercenaria and the American oyster Crassostrea virginica. Mar Bioll 36: 251-258 Strayer DL Caraco NF Cole JJ Findlay S Pace ML 1999 Transformation of freshwater ecosystems by bivalves, a case study of zebra mussels in the Hudson River. BioScience 49: 19-27 Sundbäck K Miles A Goransson E 2000 Nitrogen fluxes, denitrification and the role of microphytobenthos in microtidal shallow-water sediments, an annual study. Mar Ecol Prog Serr 200: 59-76
120 Taylor R Fletcher RL Raven JA 2001 Preliminary studies on the growth of selected “green tide” algae in laboratory culture: effects of irradiance, temperature, salinity and nutrients on growth rate. Bot Marina 4: 327-336 Tenore KR and 19 others 1982 Coastal upwelling in the Rias Bajas, NW Spain. Contrasting the benthic regimes of the Rias de Arosa and de Muros. J Mar Res 40: 701-768 Thompson ML Schaffner LC 2001 Population biology and secondary production of the suspension feeding polychaete Chaetopterus cf. variopedatus: Implications for benthic-pelagic coupling in lower Chesapeake Bay. Limnol Oceanogrr 46: 1899-1907 Ulanowicz RE Tuttle JH 1992 The trophic consequences of oyster stock rehabilitation in Chesapeake Bay. Estuaries 15: 298-306 Ward JE Newell RIE Thompson RJ MacDonald BA 1994 In vivo studies of suspensionfeeding processes in the eastern oyster, Crassostrea virginica Gmelin. Biol Bulll 186: 221-240 Widdows J Brinsley MD Salkeld PN Elliott M 1998 Use of annular flumes to determine the influence of current velocity and bivalves on material flux at the sediment-water interface. Estuaries 21: 552-559 Wildish DJ Kristmanson DD 1997 Benthic Suspension Feeders and Flow. Cambridge University Press, Cambridge, UK. 409 p Yamamuro M Koike I 1993 Nitrogen metabolism of the filter-feeding bivalve Corbicula japonica and its significance in primary production of a brackish lake in Japan. Limnol Oceanogrr 38: 997-1007
HOW DOES ESTIMATION OF ENVIRONMENTAL CARRYING CAPACITY FOR BIVALVE CULTURE DEPEND UPON SPATIAL AND TEMPORAL SCALES?
Pedro Duarte1, Anthony J. S. Hawkins 2 and António Pereira1 1
CEMAS – University Fernando Pessoa, Praça 9 de Abril, 349, 4249-004 Porto, Portugal, Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth PL1 3DH, United Kingdom
2
Abstract: The simplest computational approach for estimating environmental carrying capacity (CC) for bivalve suspension-feeders is to compare the combined rate of filtration with rates of processes that contribute to food renewal. More realistic approaches are based on mathematical models that take into account complex sets of feedbacks, both positive and negative, whereby cultured organisms interact with ecosystem processes. Each of these methods requires spatial and temporal integrations. Yet densities of cultured animals and rates of ecological processes vary in space and time. We illustrate strong dependencies of estimated CC on the spatial and temporal scales chosen for associated integrations. Where food availability is the primary limitation upon CC, low resolution models may lead to overestimates of CC, when the potential for error increases in positive relation with the spatial scale resolved by a model. Considering both spatial and temporal integrations, we recommend a procedure to help evaluate the maximum appropriate scale for the situation at hand, thereby avoiding bias in estimates of CC stemming from any “dilution” of bivalve densities. Keywords: Spatial scales, carrying capacity, modelling
INTRODUCTION Suspension-feeding bivalves have a remarkable capacity to filter the water column. Nevertheless, growth may be limited both by the quantity and composition of available food items, which may include bacteria, phytoplankton and detritus (Bayne 1992, Grant et al. 1992, Hawkins et al. 1998). Food availability is therefore a critical element in estimating environmental carrying capacity for bivalve culture (CC), when particle supply rate is primarily dependent upon both water residence time (RT) and primary production rate, as reflected by cell doubling time (PT) (Dame and Prins 1998). 121 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 121–135. ©2005 Springer. Printed in the Netherlands.
122 The rate at which food is supplied to suspension-feeding bivalves can potentially limit CC at different geographic scales. Among others, these include the scale of the cultivation unit as blocks of ropes or lantern nets, as seston-depleted water flows from one cultivation unit to another, and the ecosystem scale. CC also depends on temporal variations in food availability, which may vary over both short (i.e. tidal) and long (i.e. seasonal) time scales (Pilditch et al. 2001). Estimates of CC must therefore integrate over different spatial and temporal scales. CC has been defined as the maximum standing stock that may be kept within a particular ecosystem to maximize production without negatively affecting growth rate (Carver and Mallet 1990). More recently, CC has been described as the standing stock at which the annual production of the marketable cohort is maximized (Bacher et al. 1998), or the total bivalve biomass supported by a given ecosystem as a function of particle supply rate and the time taken by bivalves to clear water of all available particles (CT) (Dame and Prins 1998). These and other definitions are generally based upon food limitation alone, despite variable and significant interrelated effects of food availability, water quality and other environmental factors upon bivalve survival and population dynamics. Indeed, considering that ecosystems have multiple functions, with a need for integrated management, ecologists are increasingly challenged to model the various interactions between and among species, including with their environment, on a large scale. These interactions underlie more holistic definitions of CC, such as “the amount of change that a process or variable may suffer within a particular ecosystem, without driving the structure and function of the ecosystem beyond certain acceptable limits” (Duarte et al. 2003). Certainly, there are examples where CCs for bivalve cultivation have been exceeded by unsustainable practices. These include the bay of Marénnes-Óleron, France, where growth in the oyster Crassostrea gigas has reduced significantly with increased stock densities over the years (Raillard and Ménesguen 1994). In addition, growth in the mussel Mytilus edulis has been diminished by increased standing stocks in the Oosterschelde estuary, Netherlands (Smaal et al. 2001). On the above basis, approaches used to define CC may be divided into two main categories: traditional budget calculations and ecological models. Budget calculations consider the time taken for phytoplankton renewal, calculated from PT, the time for water renewal, expressed as RT, and the time taken for bivalves to filter all water within a particular area (CT) (Dame and Prins 1998). These comparisons may be performed over daily, seasonal or other intervals to determine the biomass of bivalves that can be sustained in a given ecosystem. Ecological models, on the other hand, may be divided in box models, local depletion models and integrated physicalbiogeochemical models, all based on established interrelations between environmental variables, biogeochemical processes and animal or plant
123 physiology (e.g. Bacher 1989, Jørgensen et al. 1991, Bacher et al. 1998, Hawkins et al. 2002; Duarte et al. 2003). Such models divide ecosystems into distinct state variables (e.g. bivalve biomass, phytoplankton biomass). Flows of energy or material between state variables are quantified as biological fluxes (e.g. grazing), mediated by external forcing functions (e.g. light intensity). Fluxes are normally represented by a series of differential equations that define internal processes. To account for spatial heterogeneity, the ecosystem may be divided in boxes. The size of each box determines the spatial resolution of the model. Typically, box size in coastal ecosystem models has a scale of hundreds to thousands of meters. For a description of the general structure of an ecosystem box model with bivalve suspension-feeders, see Herman (1993) and Dowd (1997). The main difference between box models and coupled physicalbiogeochemical models is that in the latter, physical and biogeochemical processes are computed simultaneously over the same temporal and spatial frameworks. Coupled physical-biogeochemical models calculate the velocity field with the equations of motion and the equation of continuity (Knauss 1997), solving the transport equation for all pelagic variables as follows: dS w dt x
w vSS y
w
z
Ax
w 2S wx2
A
w 2S y2
Az
w 2S Sources Sinks z2
(1)
where u, v and w = current speeds in x, y and z directions (m s-1); A = the coefficient of eddy diffusivity (m2 s-1); S = a conservative (Sources and Sinks are null) or a non-conservative variable in the respective concentration units. Whilst coupled physical-biogeochemical models are a more accurate representation of natural systems than box models, their main drawback is the required computing time, which especially complicates calibration and validation. Local depletion models are applied to reduced spatial scales, usually being restricted to a single cultivation unit, towards being a practical tool for local farmers or managers. This unit may be divided in several cells, but with no feedback to the ecosystem. Instead, models are forced solely by seston availability and current velocities at the boundaries, solving the transport equation using those boundary conditions for local sources and sinks. By these means, outputs may be used to predict effects of bivalve filtration, including the geometry of cultivation structures, on seston supply and bivalve growth downstream (e.g. Pilditch et al. 2001, Bacher et al. 2003). Calculated for the whole system, or in models with low spatial resolution, average bivalve density may be lower than at the scale of cultivation units; because bivalves are not normally spread evenly throughout the ecosystem. This distribution factor would not matter in estimations of CC, if seston renewal was sufficiently fast that bivalve growth was not limited in densely cultivated areas. Otherwise, irrespective of whether using box or
124 coupled physical-biogeochemical models, unless appropriate scales are chosen to account for variable bivalve densities, predicted CC may exceed the true potential. Our objectives are to help answer the following questions: 1) How does CC depend on the spatial resolution of the computing method? 2) How should one evaluate the right spatial and temporal scales to compute CC?
METHODS AND CONCEPTS In the present work, some theoretical considerations are developed in order to estimate appropriate spatial and temporal scales in CC estimates. Model simulations and comparisons of results obtained with different distributions and densities are also carried out to obtain some empirical insight into the dependence of CC estimates on the spatial resolution of the computing methods. Both approaches are applied to Sungo Bay (People’s Republic of China), based upon our previous modelling there (Duarte et al. 2003).
Spatial and Temporal Scales – Theoretical Considerations We suggest that one possible way to evaluate appropriate spatial and temporal scales to compute CC, thereby avoiding potential errors described above, is to compare CT with PT at different scales ranging from the whole system to cultivation units. The larger the scale chosen, the more likely it is that PT will be lower than CT, given “dilution” of high bivalve densities. This, we define here as Condition 1. PT < CT
(1)
At lower scales, focussing upon cultivation sites with high bivalve densities, the opposite may happen. If there is a geographic scale ('X) below which CT is smaller than PT, then bivalves below 'X remove phytoplankton cells faster than they divide, and therefore depend upon food input from adjacent waters. This, we define here as Condition 2. PT > CT
(2)
At this stage, CT should be compared with water residence time ((RT), which may be estimated as 'X/v, where v = resultant current speed (m s-1). If
125 RT is smaller than CT, then 'X may be an appropriate scale to use in further computations. If not, then 'X must be reduced to ensure that water properties do not change much across the scale considered. This, we define here as Condition 3. RT < CT
(3)
If Condition 2 is true and Condition 3 does not hold even for very small 'X, predicted bivalve biomass will exceed system CC. If Condition 2 never holds, predicted bivalve biomass will be below CC. If Condition 2 holds and Condition 3 is always attained, predicted bivalve biomass may be above or below CC, depending on whether water renewal decreases or increases the available seston. After assessing and choosing an appropriate spatial scale, temporal scale may be assessed using the numerical Courant condition (Condition 4) in such a way to avoid all water in any box being cleared within each model time step for numerical stability:
(4) where CR R = bivalve clearance rate (m3 d-1), Phyt = phytoplankton biomass (mg m-3) and V = box volume (m3). Model Simulations We assess the above a priori expectations on the basis of empirical evidence for Sungo Bay, an area of 180 km2 of intensive multi-species aquaculture of kelp ((Laminaria japonica), oyster (C. gigas) and scallop (Chlamys farreri) in the People’s Republic of China (Fig. 1). Firstly, we calculate values of CT, PT and RT at different spatial scales, ranging from 500 m to 15000 m, to assess how estimated carrying capacity may depend on the spatial and temporal scales chosen for associated integrations. Secondly, we use the two-dimensional vertically integrated, coupled hydrodynamic-biogeochemical model developed and described by Duarte et al. (2003) for Sungo Bay. The model has a land and an ocean boundary, and is based on a finite difference bathymetric staggered grid (Vreugdenhil 1989) with 1120 cells (32 columns X 35 lines) and a spatial resolution of 500 m. The model time step is 18 seconds. It is forced by tidal height at the sea boundary, light intensity, air temperature, wind speed, cloud cover and
126 boundary conditions for some of the simulated state variables. It solves the general 2D transport equation (Equation 1) (Neves 1985, Knauss 1997). The hydrodynamic sub-model solves the speed components, whereas biogeochemical processes such as primary productivity and grazing, as well as physical processes such as sediment deposition and resuspension provide the sources and sinks terms of Equation 1. Duarte et al. (2003) used this model to estimate CC for oysters (C. gigas) and scallops (C. farreri) in Sungo Bay. The model resolved bivalve density in aquaculture and non-aquaculture areas, with no bay-averaged values. Scenarios modeled by Duarte et al. (2003) represented past (Scenario I) and current culture practice (Scenario IIa), with initial stocks of 2850, 0.6 and 1860 tons DW of kelps, oysters and scallops, respectively; including hypothetical changes of x 0.5, x 2 and x 3 in seeding densities of scallop and oyster (Scenarios IIb to d,
16 km N
17.5 km
Sungo Bay
Kelp culture Scallop culture Oyster culture
Fig. 1 – Areas cultivated in Sungo Bay since 1999, including part of the model grid (upper left corner), for which the spatial step is 500 m (refer Methods and Concepts, Model simulations).
However, considering the variability of phytoplankton standing stock and net primary production obtained in the simulations described in Duarte et al. (2003) it can be estimated that PT would have varied over a range of 3 to 9
127 days, with an average of 5 days. Therefore, Conditions 2 and 3 were true at the spatial scale of 500 m x 500 m cells used by Duarte et al. (2003). In addition, the Courant Condition 4, with a model time step of 18 s, chosen to ensure hydrodynamic model stability, gives a value of 8 x 10-12 (cf.Methods and Concepts, Spatial and temporal scales – theoretical considerations, above), which is considerably less than one. Therefore, both the spatial scale of 500 m and the model time step of 18 s used by Duarte et al. (2003) seem appropriate for the scallop cultivation site. Table 1 – Aquaculture scenarios simulated with the model (refer Methods and Concepts) (Fig. 1). Culture densities
Past simulations
Current simulations
(Duarte et al. 2003) Scenarios
Oysters -2
(indiv. m ) IIa
55.0
Scallops
Scenarios
-2
Oysters -2
(indiv. m ) 56
IIIa
Scallops
(indiv. m )
(indiv. m-2)
18.2
19.8
IIb
27.3
28
IIIb
9.1
9.9
IIc
110.0
112
IIIc
36.5
39.6
IId
165.0
168
IIId
54.7
59.3
respectively), whilst maintaining seeding periods and spatial distributions as in Scenario IIa (Fig. 1 and Table 1). Scenarios analyzed in the present study (IIIa to d) maintain the same total numbers of oysters and scallops as under Duarte et al’s (2003) Scenarios IIa to d, respectively. However, whereas our past Scenarios IIa to d restricted the spatial distributions of each cultured species to their normal areas of culture, present Scenarios III a to d distribute each cultured bivalve species both through their own normal area of culture and that area used for kelp culture (Fig. 1). Densities were therefore reduced, whilst maintaining the same biomass throughout the bay (Table 1). Comparing CCs predicted for Scenarios IIIa to d with those reported by Duarte et al. (2003) for Scenarios IIa to d, we can assess any effects of local food limitation on CC estimates. A priori, one would expect that spreading the bivalves over a larger area should increase estimates of bay-scale CC, thereby providing empirical evidence for the importance of spatial resolution.
128
RESULTS AND DISCUSSION In Sungo Bay, spatially and temporally averaged CT (10 days) is smaller than RT (20 days) but larger than PT (5 days), suggesting an unutilized environmental capacity for increased culture of filter-feeding bivalves (Table 2) (Duarte et al. 2003). Considering issues of scale, current velocities were in fact slowest within the nearshore regions of
Table 2 – Physical and biological characteristics of Sungo Bay (refer Results and Discussion). Characteristics 2
Area (km )
Values 179.5
Depth (m)
10.0 6
3
Volume (10 m )
1800.0
Residence time (RT; d) Average annual Chl a
20.0 1.5
-3
concentration (mg m ) Primary production (106 g C d-1) Cell doubling time (PT; d) Total biomass (106 g) Bivalve clearance time (CT; d)
26.5 5.0 44000.0 10.1
Sungo Bay (Duarte et al. 2003). At the scallop culture site in northwestern part of the bay, average current velocities were as low as 0.025 m s-1. On the basis of a spatial resolution of 500 m as used in the ecosystem model of Duarte et al. (2003), it is possible to estimate the maximum RT as roughly 0.2 days in one cell of the model grid. Assuming an average clearance rate for a commercial-sized scallop of 5 to 6 cm shell length as 2.5 l h-1 (Hawkins et al. 2002), a cultivated density of 59 ind. m-2 (Duarte et al. 2003), and an average depth of 6 m, it is possible to estimate CT for one 500 m x 500 m cell of the model as 1.7 days. It was impractical within the context of present study to determine effects of scale on PT, as this would have required revision of hydrodynamic component for input to our coupled model at each chosen scale.
129
Fig. 2 – Relationships between water residence time (RT), mean, maximum and minimum production time (mean PT, max PT and min PT), clearance time (CT, for two ranges of bivalve densities) and spatial scale for (a) scallops and (b) oysters. Arrows on lower left corners of both graphs depict spatial scales appropriate for the Sungo model (refer to Results and Discussion).
130 Repeating the above calculations of CT and RT for different spatial scales ranging from 500 m to 15000 m, it is important to note that bivalve density decreased as ǻX increases (Fig. 2). For each ǻX, density was recalculated by a weighted average of density within the cultivated areas and density of 0 ind. m-2 outside those areas. This produced a linear decrease in density from 56 ind. m-2 in culture areas to 5.5 ind. m-2 when integrated for the whole bay. To assess consequences of increased culture densities, similar calculations were also carried out with a starting density of 112 ind m-2. Resulting CT, RT and PT values are illustrated in Fig. 2a, showing that ǻX should be lower than 1800 m for Conditions 2 and 3 to hold. Certainly, only by working at this lower spatial resolution within areas of existing culture, is the potential afforded to establish areas with greater potential for bivalve production. Similar calculations were carried out for oysters, and results presented in Fig. 2b. Given that the average clearance rate of a commercialsized oyster of 7 cm shell length is 4 l h-1 (Barillé et al. 1997), ǻX in areas of the highest bivalve densities should be less than 900 m. Minimum length scales in areas of bivalve cultivation in Sungo Bay all exceed a minimum of 1000 m (Fig. 1), and which therefore exceeded appropriate ǻXs estimated here for the culture of both scallops and oysters. The results presented in Table 3 summarize scallop and oyster productions predicted under scenarios IIa, IIb, IIc, IId, IIIa, IIIb, IIIc and IIId (cf. – Methods and Concepts). Bivalve growth isolines prior to harvesting (cf. – Fig. 1) for simulations IIa and IIIa are shown in Fig. 3. The model predicts that under scenarios IIIa, IIIc and IIId, there is a considerable increase in scallop production (between one and two orders of magnitude). It also predicts that compared with a 2x increase in bivalve density, a 3x increase in density results in a decreased scallop yield. Oysters, on the other hand, show a decrease in production for scenario IIIa relative to scenario IIa. This result may be explained by the poorer performance of oysters that are near the sea boundary (Fig. 3), as a result of lower seston organic contents (model results not shown). However, as bivalve density increases in scenarios IIIc and IIId the model predicts an important increase in production compared with scenarios IIc and IId. This result reflects both intra and interspecific competitions, when oysters out-compete scallops (Duarte et al. 2003), over-riding effects of lower food quality near the sea
131
Fig. 3 - Scenario IIa (on the left) and IIIa (on the right). Growth isolines (cm shell length) predicted by the model for scallops and for oysters. For the former, results are shown for May, just before harvest (top two figures). For the latter, results are shown for October and February, just before the first and second harvest periods, respectively. (refer to text and Table 3).
132
Table 3 – Harvest (103 T FW) predicted for aquaculture scenarios IIa to IId by Duarte et al. (2003) and in the present study for aquaculture scenarios IIIa to IIId. Scenarios and results are given for normal, decreased (1/2) and increased (2 and 3 fold) bivalve loads (refer Methods and Concepts).
Scallops
Oysters
Past simulation s (Duarte et al. 2003)
Current simulation s
Past simulation s (Duarte et al. 2003)
Current simulation s
Normal
9 (IIa)
18(IIIa)
42 (IIa)
37(IIIa)
1/2X
8 (IIb)
9(IIIb)
26 (IIb)
20(IIIb)
2X
0.6 (IIc)
31(IIIc)
58 (IIc)
61(IIIc)
3X
0.9 (IId)
20(IIId)
25 (IId)
75(IIId)
Scenarios
boundary. Even following a 3x increase in bivalve density, the model still predicts increased oyster production, including an increased total production of both scallops and oysters (Table 3). Average concentrations of chlorophyll, total particulate matter and particulate organic matter over the whole bay and simulation period (Jan 99 – August 2000) are compared for simulations IIa – IId and IIIa – IIId in Table 4. Concentrations for former simulations (IIa to d) are all lower than those for current simulations (IIIa to d). This result indicates that spreading bivalve biomass over a larger area, towards the sea boundary, where RT is lower, results in reduced impacts of bivalve filter feeding on ecosystem properties, consistent with the higher bivalve production predicted by simulations IIIa, IIIc and IIId. These results confirm the a priori hypothesis that spreading bivalves over a larger area should increase estimates of bayscale CC (cf. – Methods and Concepts). In fact, the results shown in Table 3 suggest a significant potential for increasing scallop and oyster production under scenarios III, whereas the opposite was predicted under scenarios II, mostly for the scallops.
133 Table 4 – Annual mean concentrations of chlorophyll, total particulate matter (TPM) and particulate organic matter (POM) predicted by the model under the 99-00 aquaculture simulations IIa, IIb, IIc and IId (Duarte et al., 2003) and the current simulations IIIa, IIIb, IIIc and IIId (cf. – Methods and Concepts, Table 1 and Fig. 1).
IIa
IIIa IIb
IIIb
IIc
IIIc
IId
IIId
Chl. ( g l-1)
1.5
2.1
2.5
1.0
1.6
0.8
1.2
TPM (mg l-1)
14.9 15.1 15.1 15.1 14.6
POM (mg l-1)
2.4
2.6
2.0
2.6
2.7
2.1
14.9 14.5 14.6 2.4
2.0
2.2
The local depletion model of Bacher et al. (2003) was used to evaluate bivalve growth and food depletion at different parts of Sungo Bay. Results obtained suggest that scallop growth under current densities (c.a. 50 ind m-3) is below optimal at inner parts of the bay, where current velocities and bay-sea exchanges are reduced. The negative effects of density on scallop growth ranged from 5% in the eastern part of the bay to more than 30% in the southwestern part of the bay. This is consistent with the model results of Duarte et al. (2003), and with the larger RT at the inner parts of the bay predicted by the same authors.
CONCLUSIONS A general conclusion from the above results is that in estimating CC, 'X should be smaller than the length scale of the areas currently used for suspended bivalve cultivation within Sungo Bay. This would not be the case if food supply was greater, perhaps associated with faster RT or PT. Whatever, as high bivalve densities at sites of cultivation are “diluted” upon integration over larger areas, CT increases geometrically with 'X. This is because there is a similar decrease in bivalve density. Assuming that CT may define an upper threshold for CC, it follows that low resolution models may lead to overestimates of CC, when the potential for error increases in positive relation with the spatial scale resolved by a model. We have demonstrated a strong dependence of estimated CC on the spatial scales chosen for associated integrations. To evaluate appropriate scales under different culture situations, we recommend following the procedures described above, comparing RT, PT and CT at scales ranging from less than the smallest cultivation units to the scale of whole system, for an indication of maximum appropriate scale that must be resolved to avoid bias in estimates of CC stemming from any “dilution” of bivalve densities (refer Methods and Concepts).
134
REFERENCES Bayne BL 1992 Feeding physiology of bivalves: time-dependence and compensation for changes in food availability. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed). Springer-Verlag, Heidelberg, pp 1-24 Bacher C 1989 Capacité trophique du bassin de Marennes-Oléron: couplage d’un modèle de transport particulaire et d’un modèle de croi ssance de l’huître Crassostrea gigas. Aquat Living Resourr 2: 199-214 Bacher C Duarte P Ferreira JG Héral M Raillard O 1998 Assessment and comparison of the Marennes-Oléron Bay (France) and Carlingford Lough (Ireland) carrying capacity with ecosystem models. Aquat Ecoll 31: 379–394 Bacher C Grant J Hawkins AJS Fang J Zhu M Besnard M 2003 Modelling the effect of food depletion on scallop growth in Sungo Bay (China). Aquat Living Resourr 16: 10– 24 Barillé L Héral M Barillé-Boyer A 1997 Modélisation de l’ecophysiologie de l’huître Crassostrea gigas dans un environment estuarine. Aquat Living Resourr 10: 31-48 Carver CEA Mallet AL 1990 Estimating carrying capacity of a coastal inlet for mussel culture. Aquaculture 88: 39-53 Dame RF Prins TC 1998 Bivalve carrying capacity in coastal ecosystems. Aquat Ecoll 31: 409–421 Dowd M 1997 On predicting the growth of cultured bivalves. Ecol Modell 104: 113–131 Duarte P Hawkins A Meneses R Fang J Zhu M 2003 Mathematical modelling to access the carrying capacity for multi-species culture within coastal waters. Ecol Modell 168: 109-143 Grant J Dowd M Thompson K Emerson G Hatcher A 1992 Perspectives on field studies and related biological models of bivalve growth and carrying capacity. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed), Springer-Verlag, Heidelberg, pp 371-420. Hawkins AJS Bayne BL Bougrier S Héral M Iglesias JIP Navarro E Smith RFM Urrutia MB 1998 Some general relationships in comparing the feeding physiology of suspension-feeding bivalve molluscs. J Exp Mar Biol Ecoll 219: 87-103 Hawkins AJS Duarte P Fang JG Pascoe PL Zhang JH Zhang XL Zhu M 2002 A functional model of responsive suspension-feeding and growth in bivalve, configured and validated for the scallop Chlamys farreri during culture in China. J Exp Mar Biol Ecoll 281: 13-40 Herman MJ 1992 A set of models to investigate the role of benthic suspension feeders in estuarine ecosystems. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed), Springer-Verlag, Heidelberg, pp 421-454 Jørgensen SE Nielsen S Jørgensen L 1991 Handbook of Ecological Parameters and Ecotoxicology. Elsevier, Amsterdam Knauss JA 1997 Introduction to Physical Oceanography. Prentice-Hall, New York Neves RJJ 1985 Étude Expérimentale et Modélisation Mathématique des Circulations Transitoire et Rédiduelle dans L’estuaire du Sado. Ph.D. Thesis, Université de Liège Pilditch CA Grant J Bryan KR 2001 Seston supply to sea scallops (Pacopecten ( magellanicus) in suspended culture. Can J Fish Aquat Scii 58: 241-253 Raillard O Ménesguen A 1994 An ecosystem model for the estimating the carrying capacity of a macrotidal shellfish system. Mar Ecol Prog Serr 115: 117–130 Smaal A Stralen M Schuiling E 2001 The interaction between shellfish culture and ecosystem processes. Can J Fish Aquat Sci 58: 991–1002
135 Vreugdenhil CB 1989 Computational Hydraulics, An Introduction. Springer-Verlag, Berlin. 183 p
IMPACT OF INCREASED MINERAL PARTICLE CONCENTRATION ON THE BEHAVIOR, SUSPENSION-FEEDING AND REPRODUCTION OF ACARTIA CLAUSII (COPEPODA)
N Shadrin1, L Litvinchuk2 1. 2.
Institute of Biology of the Southern Seas, Sevastopol, Ukraine Zoological Institute, Saint Petersburg, Russia
Abstract: Erosion and dumping lead to increased concentrations of mineral particles in coastal waters. The muddy waters contained high concentrations of hydrophylic limestone particles 2-50 microns in diameter. Accumulations of these particles on the sea floor can lead to the mortality of benthic macro-organisms. Using laboratory experiments, we studied the impact of dumping from mines on the mobility, feeding and reproduction of the pelagic suspension-feeder Acartia. Increased particle load changes copepod locomotive and digestive processes. Nauplii production per female decreased about 7 times in waters polluted with mineral particles. A conceptual model of mineral particle concentration impact on suspension-feeders is presented. Key words: Copepoda, mineral particles, suspension-feeding, reproduction, Black Sea.
INTRODUCTION Present changes in many productive coastal ecosystems are the result of anthropogenic factors and long-term climate variability. Some causes of these changes are well studied while others are not. One of the least studied processes is the increasing influx of mineral particles (MP) into the sea as a result of coastal erosion and dumping. The growth in MP concentrations leads to increased turbidity of sea water with the concurrent reduction in light penetration and primary production (Shadrin et al. 1991). MPs also adhere to microalgae, which causes autolysis of the cells and drastically decreases phytoplankton primary production. Suspension-feeders play an important role in energy and matter transfer from primary producers to higher trophic levels, however the impact of MP concentration on their functioning has received little attention. The authors examine how Balaklava mine mudwaters influenced the planktonic 137 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 137–146. ©2005 Springer. Printed in the Netherlands.
138 suspension-feeder Acartia clausi (Copepoda). Limestone for gravel production is extracted at the Balaklava mine situated near Sevastopol (Crimea, Black Sea). About 1,500,000 m3 of effluent containing 2-50 µm limestone particles with concentrations ranging between 1-29 g/l are annually discharged into the sea. This effluent is associated with a 0.1 km2 lifeless area on the sea bottom. However, in the same area, zooplankton may still be found at abundances and species compositions similar to those in unpolluted sea water. Nevertheless, all Copepoda and Cladocera in the area have the MP lumps in their intestines. Presumably, the strong along coastal currents carry fresh zooplankton from unpolluted areas into the lifeless zone and projects the false impression that there is little negative impact on the zooplankton. This article describes the results of experiments designed to show the influence of MP on the planktonic suspension-feeder A. clausi. These results may be applied not only to this particular copepod species but also to a variety of other pelagic suspension-feeders.
MATERIALS AND METHODS The copepod Acartia clausi, which is the prevailing zooplankton species in coastal Black Sea, was used in laboratory experiments. The copepods were collected with a Juday net in the vicinity of Sevastopol. Only intact female copepods of standard habit were used in the experiments. Waste water which contained limestone particles 2-50 µm in size was taken directly where the outlet discharging the effluent from Balaklava mines was situated. Then the sample was diluted with filtered sea water until the final, 0.1-0.2 g/l, concentration of MP in experimental containers was reached. Each experiment was done in triplicate using 0.5 L laboratory containers each with 10 female copepods. Controls used filtered sea water (10 micron) and experimentals used water with MPs as described above. We filtered the sea water through a filter with the pore size 10 µm. The same protocol was used for subsequent experiments with and without vibratory water motion (jars were placed on a shaker table). The mechanical agitation prevented MPs from settling, thus simulating the marine environment receiving the effluent. Acartia exhibits two main moving patterns leaps (jumps) and glides (Petipa 1981, Kovaleva and Shadrin 1984, Tiselius and Jonsson 1990, Leising and Franks 2002, etc.). In our study, a stopwatch was used to measure time parameters.Leap and glide length were measured under a binocular dissecting microscope with 8-X magnification and using an ocular eyepiece micrometer.
139 Behavioral characteristics were recorded for 20-30 min. For each characteristic motion, a mean was calculated from 20-40 observations. SINKING DEPTH Acartia sink during a pause between leaps or glides (Kovaleva and Shadrin 1984, Hansen et al. 1991). The sinking distance during the pause was determined in a graduated cylinder with imprinted scale at 1 mm resolution. 100 observations were made in the experiment and 100 in control. Egg Laying Capacity. Experiments to test the effects of MPs on the egg-laying capacity of the copepod were also conducted. As before, 10 females were placed into the laboratory containers and fed on the cultured dinoflagellate, Peridinium trochoideum (3 mg/l). Three days after feeding the number of eggs laid and nauplii hatched were counted. STARVATION Six experiments were conducted to determine the time to death due to starvation. There were 10-20 females in each 0.5 L container. Altogether, 185 copepods were observed. Every day we counted the number of alive and dead females. One experiment was made without vibratory motion, and five with shaking. The Gut Clearance Rate Gut clearance rate was determined for fed copepods ((P. trochoideum, 3 mg/l) in containers with and without MPs. As soon as the intestines were full, usually after 20-40 min of visual observation, the copepods were captured and put individually onto small glass plates with the filtered sea water. The gut clearance rate, i.e. the time required for fecal pellet evacuation, was assessed using stopwatch under a binocular dissecting microscope at 12X magnification. MPs adhesion onto the copepods was observed visually using a binocular dissecting microscope. Statistical Analysis Statistical analysis of the data was performed after Sokal and Rohlf (1995). Mean and standard deviations were calculated. Significant differences were with the Student t-test.
140
RESULTS MPs Adhesion Onto the Copepods. Figure 1. Acartia clausi from experimental vessel
141 In the experiments, MPs adhered to the copepods (Fig. 1) within 2030 min. The particles might be seen covering different spots on the copepod’s body; however, small particles were found predominantly on the legs and the frontal part of cephalothorax. Larger particles usually stuck to the abdomen, furca and furcal setulae. Movement of buccal appendages covered with MPs was impaired. The second antennae were often spread apart and motionless. Without the interference of water motion, after the MPs had settled onto the bottom, the particles covered the copepod’s bodies for 2-3 hours, until they were washed off by locomotion with agitation in pure water. Total MP volume adherence to Acartia was approximated through visual observation as 10-30% of a copepod’s body volume. Knowing density of limestone it was possible to approximately assess total MP weight on a female copepod as about 40-60% body weight on the average and occasionally values even larger than the body mass were observed. Influence of MPs on Locomotion A. clausi encrusted with MPs moved only in leaps, while in the control group copepods moved with hops, leaps and glides. Table 1 shows results obtained from the observation of the copepod mobility in the experiment versus control. The means of behavioral characteristics measured in the experiment and in the control were significantly different (P<0.05 ÷ 0.001) and the same was true for the distribution patterns. The length of leaps of copepods from the experimental and the control jars also considerably differed. Copepods covered with MPs usually moved in short leaps: 63.5% of them were 0.2-1.2 mm long leaps. In the control the portion of such leaps was only 22.5% of the total. Intermediate (1.2-2.4 mm long) leaps constituted 24% in the experimental and 61.5 in the control groups, and long (>2.4 mm) leaps composed 12.5% and 16%, correspondingly. The resting pause was shorter (5.5 sec) in the MP encrusted copepods as compared to the controls: the registered maximums were 5.5 and more than 7 sec, correspondingly. The average number of the pauses per minute measured in the experimental group and in the control groups was 6 and 12.5 respectively. The downward distance the copepods moved during a resting pause was larger in the experimental containers. Short (0.2-1.0 mm) distances were not observed in the control group while in the experiment, the percentage of descent for 1-2 mm was evaluated as 18% and 40% in the control group; the maximum sinking distances were 8.2 and 2.6 mm; correspondingly.
142 Table 1. Behavioral characteristics of Acartia clausii in the experiment and in the control (average ± Standard deviation) Ex p. ---------Ob s.
Leap
Glide
mm
mm
1 n 2 n 3 n
0.98r0.24 100 1.20r0.38 90 1.11r0.47 30
0
1 n 2 n 3 n
1.74r0.24 100 1.87r0.43 90 2.06r0.50 30
0.23r0.08 40 0.25r0.05 20 0.27r0.06 40
0 0
Number of Leaps Per min
Experiment 36.7r1.1 30 37.3r3.2 30 31.5r1.6 30 Control 26.8r1.0 30 30.2r1.8 30 28.3r2.1 30
Number of Glides Per min
Pause min
Downward Motion During Pause mm
0
0.92r0.12 60 1.26r0.21 50 1.53r0.23 30
3.32r1.84 60 4.23r1.46 50 4.03r1.59 30
1.41r0.21 60 1.60r0.27 50 2.04r0.35 30
1.04r0.47 60 1.36r0.41 50 1.61r0.53 30
0 0
9.01r1.24 30 11.00r2.0 30 9.70r1.85 30
In the test group, 65% of descents were greater than 2.6 mm. Experiments with mechanical motion yielded results, which differed slightly (1st day) and drastically during the following days. Resting pauses got shorter, while the distance of downward motion during the pause got larger. Generally, differences between copepods in the experimental and in the control vials became more pronounced. The following tendencies were revealed: in the control the number of leaps was 21-23 per minute during the first three days, while in the trial group it decreased from 23 to 20 to 16 in the 1st, 2nd and 3d days, correspondingly. In the controls, the number of pauses per minute (6-7) was almost invariable, and in the experiment steadily decreased from 10.5 during the 1st day to 6.3 during the 3d day. The duration of resting pause decreased slightly from 4.2 to 3 sec in the control group and increased from 2.2 to 4.6 and to 8.6 sec (1st, 2nd and 3d days, correspondingly) in the experimental group. We estimated average locomotion rate using equation 1 (values were taken from Table 1). V = L1˹n1 + L2˹n2
(1)
where V is average locomotion rate, L1 – leap length (mm), N1 – the number of leaps a minute, L2 – gliding distance (mm), N2 – the number of glides a
143 minute, it is possible to compute average forward motion rate of the copepods, that is 56.2 mm/sec in the control and 38.6 mm/sec, i.e., 1.5 times as less, in the experimental group. Influence of MPs on Survival Rate The survival rate of A. clausi with and without vibratory motion showed no significant difference (only 1.09 times) between the controls and the treatments with MPs. However, when the water was agitated the results were different: in the experimental group all copepods died during the 3d day and in the control – during the 9th day. In the experimental containers with the food and MPs, copepods died by the fourth day (on a shake table). We can conclude that in the water with suspended MPs and food the copepods are factually starving.
Influence of MPs on the Nutrition In the presence of MPs, copepods swallow the phytoplankton cells and simultaneously a large amount of MPs. This was easily seen with the unaided eye (Fig. 1). After the anterior and the posterior sections of the intestines were filled (Fig. 1), two food pellets formed, one in the anterior and one in the posterior sections. Digested food is first evacuated from the posterior section. In the experiment it took 87-96 min for the first fecal pellet to be excreted, while in the control it took 99-106 min. The anterior food pellet had been excreted within 105-108 and 115-125 min, correspondingly in experiment and in control. Visual examination showed that in the experimental copepods the faecal pellets had not been digested as fully as in the control: nearly 85-90% of the food pellets were MPs (Fig. 1). Influence of MPs on Copepod Reproduction Females in the experimental containers usually laid 2.4±0.92 eggs/day, which was about 6.5 times less than the control females (15.6±0.92 eggs/day), the differences were significant (P<0.001).
DISCUSSION Our results support earlier reports that non-toxic limestone particles have a negative impact on the copepods. This effect can be explained through energy balance approach though our study is not directly linked to bioenergetics of copepods. The bioenergetics of Acartia has been studied thoroughly (Petipa 1981, Shadrin1990b), therefore we can apply the resulting
144 deductions and rules to the present investigation. In brief, the main equations describing energy balance approach are: P=A–R
(2)
where P is production generated by organism (population), A – assimilation food energy, R – total energy expenditure A = Į·C
(3)
where C is the ingested ration and Į is the assimilated part of ration. Ivlev (1961) found that a fasting organism dies because of a critical body weight loss due to metabolism. In our experiments, the period during which the fasting copepods remained alive in the sea water contaminated with MPs was 3 times less than in the control. In the control, the survival time under fasting was similar to that known for Acartia from earlier experiments (Dagg 1977, Kovaleva and Shadrin 1984). Therefore, the same portion of copepod’s body weight was used for the metabolism 3 times faster in the experimental containers than in the control. These data are supported by our observations of copepod behaviour and literature data (Busky 1998, Svetlichny et al. 2000). A. clausi “armored” with MPs would consume by 40-50% more energy (proportional to the body plus MPs total weight) for active moving than copepods without MPs. MPs adhered to the body also caused additional hydrodynamical drag on movement. Increase in the sinking rate during the pause would correspondingly require that more energy to be consumed to keep the copepod submerged at a definite depth in the water column. In the experimental chambers, copepods sank down to the bottom of the jars, in nature A. clausi covered with MPs may similarly sink beneath the productive water layer rich in phytoplankton and doom themselves to starvation and death. Earlier experiments showed that the concentration of phytoplankton cells in the A. clausi ration is proportional to the rate of their locomotion (Kovaleva 1987, Shadrin 1990a). The ration estimate also depends on the intestinal capacity (working volume). Our results show that clogging with MPs (Fig. 1) reduces working capacity of the intestine approximately by 80%. The portion of food assimilated from the ration (Į, equation (3)) is inversely proportional to the rate at which the food passes the gut (Petipa 1981, Kioerboe and Tiselius 1987). Therefore, accelerated pellet evacuation may decrease assimilation rate by 10-20% in the experimentals. These results can explain why in the water with MPs and food the copepods are starving.
145 Our results suggest that decrease of total production would be 5-7 times as large. The adult female copepods do not grow, their production is entirely generative. Generative production in the experiment was 6.5 times less than in the control. Based on relevant data of different independent investigations, this estimate can be considered reliable. In experimental containers egg production is probably the product of the energy and substance the females accumulated before the experiment and not to the food consumed during the experiments. White and Dagg (1989) determined that concentrations of MP suspension from 100 to 1000 ppms had a negative impact on the generative production of A. tonsa. However, the harmful effect can be reduced through increased food (phytoplankton) concentrations. In another experiment with Acartia, the effect of non-toxic latex beads (diameter 45 µm) was shown on increased sinking rate and short leap frequency as well as a lower rate of phytoplankton consumption (Hansen et al. 1991). Not only do adhered MPs make the copepods heavier and impair their movement, but so do epibiont algae covering their bodies. T. Kovaleva (1982) studied A. clausi females covered with unicellular algae and intact (uncovered) to compare between their rations, movement and survival rate under fasting. In copepods bearing the epibiontic algae, the ration was 3.2 times less, egg production was 2.6 times less and the time they staved to death in the sea water without food 1.7 times faster than the control. In brief, we can conclude that increased concentration of MPs, especially hydrophilic ones, will always lead to drastic decrease of pelagic suspension-feeder productivity, even to extinction of the population because the energy costs are always greater in the presence of MPs.
CONCLUSIONS Increased concentrations of non-toxic natural mineral particles are the major factor, which endangers benthic and planktonic suspension-feeders. MPs should be taken into account in environmental impact assessments of waste water discharge, damping and erosion of coastal zone. It is essential to further develop relevant bioindication and biotesting methods as well as computation techniques allowing assessing such environmental damage.
ACKNOWLEDGEMENTS We would like to express our thanks to two anonymous referees for the friendly constructive reviews of this paper. We also thank Ms. O. Klimentova for her help with translation into English. Mr. O. Eryomin provided technical assistance.
146 REFERENCES Dagg M. 1977 Some effects of patchy food environment on copepods. Limnol Oceanogr 22: 99-107 Hansen B Hansen PJ Nilsen TG 1991 Effects of large nongrazable particles on clearence and swimming behaviour of zooplankton. J Exp Mar Biol Ecol 152: 257-269 Ivlev VS 1961 Experimental Ecology of the Feeding of Fishes. New Haven, Yale Univ. Press, 292 p Kiørboe T Tiselius PJ 1987 Gut clearence and pigment destruction in herbivorous copepod, Acartia tonsa, and the determination of in situ grazing rates. J Plank Res 9: 525-534 Kovaleva TM 1982 The influence of algae fouling of copepods on their vital activity. Ecol morja 11: 29-36 (in Russian) Kovaleva TM 1984 Effect of algae size and concentration on the consumption rate of two species of marine copepods. Ecol morja 31: 20-35 (in Russian) Kovaleva TM Shadrin NV 1983 Changes in motor activity and fat consumption in Acartia clausi Giesbr. during long fasting. Ecol morja 14: 44-50 (in Russian) Kovaleva TM Shadrin NV 1984 Changes in motor activity and fat consumption in Acartia clausi Giesbr. during long fasting. Can Trans Fish Aqua Sci 5047 20 p. (Trans from RU: Ecol morja, 1983, v. 14) Leising A.M Franks PJ 2002 Does Acartia clausii (Copepoda; Calanoida) use an area restricted food? Hydrobiologia 480: 193-207 Petipa TS 1981 Trophodynamics of Copepoda in Marine Planktonic Ecosystems. Naurova Dumka, Kiev, 242 p (in Russian) Shadrin NV 1990 Influence of biotic factors on energy balance of hydrobionts. In: Bioenergetics of Hydrobionts. GE Shulman and GA. Finenko (Eds), Naukova Dumka, Kiev, pp 109-118 (in Russian) Shadrin NV 1990b Population mechanisms of regulation of copepods energetic budget and generation abundance. In: Bioenergetics of Hydrobionts. GE Shulman GA Finenko (Eds.) Naukova Dumka, Kiev, pp 110-118 (in Russian) Shadrin NV Kovaleva TM Fedorova L 1991 Mudwater impact upon shelf plankton. 26 Eur Mar Biol Symp, Abstract, Middelburg, 142 p Sokal RR Rohlf FJ 1995 Biometry, WH Freeman, New York, 450 p Tiselius P Jonsson PR 1990 Foraging behaviour of six calanoid copepods: observations and hydrodynamic analysis. Mar Ecol Progr Serr 66: 23-33 White JR Dagg MJ 1989 Effect of suspended sediments on egg production of the calanoid copepod Acartia tonsa. Mar Bioll 102: 315-319
SUSPENSION-FEEDERS AS FACTORS INFLUENCING WATER QUALITY IN AQUATIC ECOSYSTEMS
SA Ostroumov Department of Hydrobiology, Faculty of Biology, Moscow State University, Moscow 119992, Russia Abstract: Suspension-feeders are found in both pelagic and benthic systems. They function as an important part of an ecosystem’s biomachinery that maintains water quality in aquatic systems. They remove suspended matter and excrete faeces, pseudofaeces and dissolved inorganic materials that contribute to nutrient cycling between the water column and the benthic habitats. Suspension-feeders are a key part of many natural aquatic remediation systems and they can decrease some negative anthropogenic impacts. Recent experiments are reported that demonstrate new effects of pollutants on the filtration rates of suspension-feeders. Keywords: suspension-feeders, water purification, bivalves, pollution, filtration rate, surfactants, detergents
The river Rhine, it is well known, Doth wash your city of Cologne; But tell me, Nymphs, what power divine Shall henceforth wash the river Rhine?
Samuel Taylor Coleridge 1772-1834
INTRODUCTION Water is a key of the world economy and is essential for the existence of life. "It is essential that humans understand the fundamentals of water science for its responsible use and the effective management of water resources for both hydrological availability and acceptable water quality" (Wetzel 2001). Water quality in aquatic ecosystems is the result of a number of intertwined processes (Wetzel 2001, Ostroumov et al. 2002). Organisms of various trophic levels are involved in complex webs of interactions within aquatic systems. Among those organisms, suspension-feeders play a 147 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 147–164. ©2005 Springer. Printed in the Netherlands.
148 significant role in both pelagic (Sushchenya 1975) and benthic (Dame 1972, 1976, 1996) trophic webs. The filtering activity and ecology of suspension-feeders has been examined from various perspectives by many authors who studied both pelagic (e.g., Sushchenya 1975, Zankai 1979, Starkweather and Bogdan 1980) and benthic systems (e.g., Alimov 1981, Gutelmaher 1986, Newell 1988, Monakov 1998, Dame et al. 1989, 2001, 2002). New evidence of the vulnerability of that activity of suspension-feeders to certain chemicals is presented and synthesized with previous findings (Ostroumov 1998, 2002 a, b, 2003a, 2004). The role and patterns of the filtration rates of suspensionfeeders is examined in order to better understand the implications of the effects of these chemicals. Specifically, this paper addresses: why maintaining water quality is a necessity in natural ecosystems; the role of suspension-feeders in some of the key processes contributing to water quality maintenance; clearance time in some water bodies; the variety of taxa of suspension-feeders; the effects of the concentrations of suspended matter on filtration rates; and the ecological role of the production of pellets by suspension-feeders. Why Natural Processes that Improve Water Quality Are a Necessity in Natural Ecosystems In natural waters, there are at least four main factors that practically decrease water quality: dissolved organic matter, nutrients, suspended matter, xenobiotics and pollutants resulting from anthropogenic activities (Table 1). Table 1. Some factors decreasing water quality in aquatic ecosystems Factors decreasing water quality Organic matter
Nutrients
Suspended matter Xenobiotics and pollutants
Autochthonous sources of contamination Excretion of organic molecules and decay of organisms Nitrogen fixers; sediments and reentry of materials from sediments to water Plankton, sediment resuspension Secondary pollution from sediments
Allochthonous sources of contamination Other ecosystems, man-made activities
References
Other ecosystems, air
Wetzel (2001)
Land, streams, manmade activities Man-made activities
Wotton (1994)
Wetzel (2001)
Rand (1995); Ostroumov (2001c)
149 Table 2. Water filtration by suspension-feeders may influence other biotic and abiotic processes that are involved in water purification. Water purification Chemical oxidation by oxygen
Biotic/ Abiotic Abiotic
Photodegradation of organic matter and pollutants
Abiotic
Pollutant sorption by sediments
Abiotic
Pollutant sorption by pellets
Abiotic
Sedimentation of particles of seston
Abiotic
Material accumulation by aquatic organisms
Biotic
Oxidative biodegradation of organic molecules by bacteria and fungi
Biotic
Recycling nutrients making them available to organisms involved in water purification
Biotic
Decreasing sediment erodibility
Biotic/ Abiotic
Suspension-feeders influence Suspension-feeders remove some suspended matter from water column; by increasing light penetration, they help benthic algae to carry out photosynthesis and to produce oxygen; availability of oxygen is especially important at the bottom where the organic matter is accumulated Suspension-feeders via filtration increase the water transparency and light penetration into water The sorption of pollutants by sediments depends on the percentage of organic matter in the sediments; suspension-feeders produce pellets rich in organics and by doing so increase the percentage and amount of organic matter in the sediments; as a result, the capacity of sediments to adsorb pollutants increases. Suspension-feeders produce pellets (faeces, pseudofaeces) which are the additional centers for adsorption of pollutants Suspension-feeders remove small particles from water and aggregate them into bigger particles of pellets; the latter sediment faster than the small particles Suspension-feeders produce shells, which contain C and other elements that stay on the bottom for long time. It is important in terms of biomineralization and cycling of carbonates bound in shells. By removing bacteria and fungi from water suspension-feeders participate in regulating the abundances of many species in the ecosystem. As a result, suspension-feeders contribute to the control of the rate of processes performed by bacteria and fungi (e.g., Ostroumov, 2000 e, 2001d) Suspension-feeders actively participate in recycling nutrients. E.g., they excrete N and P. By doing so they contribute to recycling. Sediment erodibility depends on the epifaunal bivalves (Widdows et al. 2000)
Some natural processes may lead to decreasing water quality if other natural processes do not properly counterbalance them. Thus, the excretion of
150 large amount of soluble organic matter by algae and cyanobacteria during algal blooms may decrease water quality. The decay of the biomass of cyanobacteria in eutrophic water bodies may decrease the quality of drinking water. The Role of Suspension-feeders in Improving Water Quality There are both anthropogenic and natural processes that decrease water quality and there are a number of processes contributing to improving water quality. Taken together, the relevant processes form a complex mechanism that unifies the individual processes into the mechanism ("biomachinery") of water self-purification in aquatic ecosystems (Ostroumov 2004). Those processes can be categorized into three groups: (1) physical, (2) chemical, and (3) biotic processes. The focus of this paper is on the biotic processes (Table 2).
Estimates of Individual Filtration Rates and System Clearance Times A compilation of published individual filtration rates of a diversity of macroinvertebrates is presented in Table 3. The filtration rate can be as -1 -1 high as 10.2 l g h (dry body mass; see Dame et al. 2001). The filtration -1 rate FR (l h ) generally increases with increasing body mass W (dry mass of the soft parts) according to relationship (1) FR = a W b
(1)
where a and b are species-specific constants that are empirically determined (e.g., Sushchenya 1975, Alimov 1981, Kryger and Riisgård 1988). For example, the a value for Dreissena polymorpha feeding on Chlorella vulgaris is 6.82 (Kryger and Riisgård 1988). The values of b range from 0.46 to 0.92 (Kryger and Riisgård 1988). The clearance time for a given water body as a result of the suspension-feeders in that system can be calculated by multiplying the species specific individual filtration rates by the number of individuals per species and summing the various products to determine the system filtration rate. The inverse of the system filtration rate is the clearance time. Table 4 presents the clearance times of a number of different systems.
151 Table 3. Individual filtration rates. Unless another reference is indicated, the data are from Dame et al. (2001), Shuntov (2001), and Krylova (1997). AFDW (ash-free dry weight). Organisms
Crustacea
Rotifers
Corals Brachiopoda
Polychaeta
Spongia (Porifera)
Ascidia Bryozoa Cirripedia Mollusca Decapoda Echinodermata
-1
-1
Weight specific filtration rates l g h ( unless the other units are indicated); comments; Plankton Eudiaptomus gracilis (copepod) in Lake Balaton: in winter 0.11 ml -1 -1 -1 -1 animal day and in summer 1.44 ml animal day , their maximal -1 -1 activity was as high as 3.27 ml animal day (Zankai 1979). -1 -1 Keratella cochlearis : 5-6 µl animal h (Starkweather and Bogdan 1980). Brachionus calyciflorus (when feeding in the laboratory on -1 -1 ciliates Coleps sp.) filtered water at a rate of 5-30 µl animal h (Mohr and Adrian, 2000; cit. in Ostroumov et al. 2003b). Immobile Benthos -
Alcyonium digitatum : the removal rate of phytoplankton 0.16 mg C g 1 -1 h (Migne and Davoult, 2002) some examples of brachiopods, which are part of water-filtering community:Laqueus californianus ; depth: 2-1600 m; up to 3000 -2 animals m -2 Diestothyris frontalis at 0-435-m depth, up to 900 animals m Tythothyris rosimarginata, at 5-25 –m depth Pelagodiscus atlanticus at 366-5530 – m depth Multitentacula (Dentaria) amoenaa at 5060 – m depth Bathyneaera hadalis at 2970-8430 – m depth (Zezina, 1997) -1 -1
Sabellidae; sabellids remove particles 3-8 µm; 0.5-1.8 l g h 1 (Sabella spallanzanii; AFDW) -1 -1 Serpulidae: remove particles 2-12 µm; 4.7-10.2 l g h 3 -3 -1 0.04 – 0.06 cm water cm sponge sec ((Baikalospongia bacillifera) (Savarese et al. 1997) 1.2 ((Ficulina ficus; AFDW); -2 up to 12 (Thenea abyssorum; AFDW); biomass up to 1524 mg m (Norwegian-Greenland Sea, depth 2020-2630 m) (Witte et al. 1997: -1 -1 2 ml g sec (Spongilla lacustris dry weight) (Frost 1980); -2 (Thenea abyssorum AFDW; the biomass up to 1524 mg AFDW m ) (Witte et al. 1997) 2.7 ((Ascidiella aspersa; AFDW) 2.2 (Plumatella ( fungosa; dry weight) (Monakov 1998); other data: 0.42 per 1 zooid; 1 (Balanus ( crenatus AFDW); 8.8 ml per 1 animal ((Balanus perforatus) up to 8.8 (Ostrea edulis AFDW) Mobile Benthos Porcellana longicornis 0.1-0.27 l crab-1 h-1(Achituv and Pedrotti 1999) Echinarachnius parma (Echinoidea, Clypeastroida) some of Ophiuroidea 10.4 (Ophiothrix fragilis; AFDW)
152
Table 4. Estimates of suspension-feeders filtering the entire water column: clearance time (amount of the days for the water column to be filtered).
System
Organism
% vol 100
Time (d) 1.2
Lake Baikal, Russia
Lake Punnusjarvi, Russia
Baikalospongia bacillifera, B. intermeda, Lubomirskia baikalensis Zooplankton (various)
Lake Tuakitoto, New Zealand
Sphagnum bogpond; Wisconsin, USA Laholm Bay in the Kattegat, Denmark
References; comments
100
3
Hyridella menziesi (bivalve)
100
1.4
Spongilla lacustris (sponge)
130
1
Frost (1978); areas of high biomass 31.8 g m-2
Cardium edule, Mya arenaria
50100
3
Loo and Rosenberg (1989); area 60 km2
Northern San Francisco Bay, USA South San Francisco Bay, USA North Inlet (South Carolina, USA)
M. arenaria
100
ca. 1
Potamocorbula amurensis
100
0.6
Crassostrea virginica
100
0.86.1
Dame et al. (1980); volume 22 ·106 m3
Carlingford Lough, Ireland
C. gigas, Tapes semidiscussata, Mytilus edulis
100
87.5
Narragansett Bay (Rhode Island, USA)
Mercenaria mercenaria
100
32.1
Ball et al. (1997); cit. in Dame et al. (2001); volume of water 196·106 m3 Pilson (1985); volume of water 2724·106 m3
Oosterschelde, The Netherlands
Mytilus edulis, Cerastoderma edule
100
3.7
Western Wadden
M. edulis, C. edule
100
5.8
Savarese et al., 1997; littoral zone, 12-m depth
Andronikova (1976, 1978) – cit. in Gutelmaher (1986) Ogilvie and Mitchell, (1995)
Nichols (1985)
Cloern (1982); water volume 2500·106 m3
Smaal et al. (1986) ; volume of water 2740·106 m3 Dame et al. (1991a); Van
153 Sea, The Netherlands
Stralen (1995); cit. in Dame et al. (2001); volume of water 4020·106 m3 Tenore et al.(1982); cit. in Dame et al. (2001); volume of water 4335·106 m3 Newell (1988) ; volume27 300·106 m3
Ria de Arosa, Spain
M. edulis
100
12.4
Chesapeake Bay, USA
C. virginica
100
87.5
Marina da Gama
several species
100
1.1
Davies et al. (1989) volume 0.025·106 m3
Kertinge Nor, Denmark
Ciona intestinalis (ascidian)
100
0.8-5
Bay of Brest, France
multiple species
100
2.8-6
Petersen and Riisgård (1992) ; volume 11 000·106 m3 Hily (1991); volume 1480·106 m3
MarennesOléron, France
C. gigas, M. edulis
100
2.9
Königshafen, Germany
M. edulis, C. gigas
100
0.92.8
Héral et al. (1988); Bacher (1989) - cit. in Dame et al. (2001); volume 675·106 m3 Asmus and Asmus (1991); volume 7.2 ·106 m3
The Effects of Material Concentration of Suspended Matter on Filtration Rates Studies of filtering by suspension-feeders at various concentrations of particles of suspended matter have found non-linear dependence of the filtering rate on the concentration of particles (e.g., Schulman and Finenko 1990, Monakov 1998). In some experimental systems the higher the concentration of suspended particles (e.g., algal cells), the lower the filtration activity of suspension-feeders and the lower the grazing pressure on seston. This pattern has been seen in the benthic bivalve Mytilus galloprovincialis (feeding on Gymnodinium kowalevskii), and the zooplankter Brachionus calyciflorus (feeding on Langerheimia ciliata and Scenedesmus acuminatus), Daphnia magna, D. rosea, Diaphanosoma brachyurum as well as other suspension-feeders (Sushchenya 1975, Shulman and Finenko 1990, Monakov 1998). That pattern is of ecological interest as the grazing pressure on suspended particles (including phytoplankton) is an important part of control of phytoplankton. The non-linear dependence of the filtering rate on the concentration of algal cells means that when for some reason the concentration of phytoplankton increases, the grazing pressure on it - at least
154 in some cases - might not increase proportionately, which indicates that some relative decrease in the control of the phytoplankton takes place. It might open the way for further growth of the phytoplankton. That pattern of cause and effect is an exemplification of positive feedbacks. A detailed discussion of the role of positive feedbacks was given in (DeAngelis et al. 1986, Ulanowicz 1997– cit. in Dame et al. 2001; Herman et al. 1999). The Ecological Role of the Production of Pellets by Suspension-Feeders: Biosediment Table 5. Biosediments formation by suspension-feeders System Laholm Bay
Organisms Cardium edule M. arenaria C. edule + M. arenaria
Rocky shores
Mytilus edulis
Norwegia nGreenland Sea, depth 2020-2630 m
sponge Thenea abyssorum, biomass up to 1524 mg AFDW m-2 T. abyssorum
Baltic coastal ecosystem
M. edulis
Amount N: 199 t y-1 C: 1449 t y-1 N: 36 t·y-1 C: 262 t·y-1 N: 235 t y-1 C: 1711 t·y-1 C: 29 g·m -2 y-1 11.9 kg ·m-2 y-1 (dry weight), 0.6-2.2 mg C·m-2 ·d-1
up to 0.7 g C m-2 y-1 -1
1092 g ·m-2 y dry weight, including C 80.7 -1 g ·m-2 ·y
N 10.4 g· m-2 ·y-1 P 1.6 g ·m-2 ·y-1 Marine/ estuarine NETHER LANDS Marine/ estuarine
of 11.9 kg, faeces 9.2 kg, pseudofaeces 2.7 The poriferan community possibly adds up to 10% to the vertical particle flux -
Tsuchiya (1980)
Total sedimentation (total amount of all types of sedimenting material) was -1 3521 g ·m-2 ·y dry weight
-
25 g ·m-2 ·h
-1
bivalves
-
60 g ·m-2 ·h
-1
18 g·m-2 ·h
References Loo and Rosenberg (1989)
-1
M. edulis
M. chilensis
Comments per 60 km2; 0-10 m depth
-
Witte et al. (1997)
author's estimate based on the data (Witte et al. 1997) Kautsky and Evans, 1987;
Widdows et al., 1998 Smaal et al. 1986 Jaramillo et al., 1992; cit. in Ostroumov et al., 2001
155 As a result of the production of pellets by suspension-feeders, some amount of organic matter is deposited on the bottom. The material deposited as biosediment is important part of biogeochemistry of aquatic ecosystems. The creation of biosediment has been quantified in a number of studies (Table 5).
Polyfunctional Role of Suspension-feeders Regulating Ecosystem Processes The entire community of aquatic species is important in controlling and regulating ecosystem properties and functions related to filtering rates. Suspension-feeders have a double role: they are both actors who themselves perform the function of removing particles from water and regulators who control other actors – e.g., other biological species who are involved in water purification. Thus, the regulatory role of suspension-feeders in ecosystems is poly-functional as they regulate the abundances of many key players in the ecosystem including not only algae but also bacteria (Ostroumov 2002a) and possibly fungi. As a result, suspension-feeders may control the key process of oxidation of organic matter in the ecosystem by the bacteria (plus fungi) who are often the most active factor in performing the oxidation of organic matter in aquatic ecosystems (Wetzel 2001). Often bacteria are responsible for over 55% of the total respiration of a community involved in the oxidation of organic matter (Sorokin et al. 1997). The impact of suspension-feeders on bacteria and fungi is dualistic. The negative effects: suspension-feeders graze on bacteria and fungi, thereby removing them from water. Some positive effects are also of significance: the same suspension-feeders may help bacteria and fungi, e.g., by producing extra-cellular organic metabolites and recycling nutrients (via excretion of various forms of nitrogen and phosphorus) that are available for bacteria. The regulatory role of suspension-feeders might be important towards optimization of the rates of oxidation of organic molecules by bacteria and some other heterotrophic organisms. Assessing the role of suspensionfeeders in regulating the abundance of bacteria and therefore their oxidation activity, we should underline existence of several aspects of the influence of suspension-feeders on heterotrophic bacteria. Those aspects include the production by suspension-feeders of the pellets rich in organic matter that provide the organic substrate for heterotrophic bacteria, and also production of some other substances, e.g. production of the organic molecules that function as electron acceptors for anaerobic bacteria in sediments (Wetzel 2001). The Effect of Chemical Pollutants on Filtration Rates
The traditional studies of the impacts of pollutants on aquatic organisms are focused on the lethal and semi-lethal effects, and usually
156 measured by the concentrations required to kill half of a population, LC50 (Rand 1995). It is also important to understand the sublethal effects of pollutants on suspension-feeders and specifically on their functional activities such as removal of particles from the water (Ostroumov 1998, 2000d, 2001c). Tests of the effects of surfactants and some other chemicals on the filtering rate of invertebrates have found inhibition of filtering (Ostroumov et al. 1997, Kartasheva and Ostroumov 1998, Ostroumov 1998, 2000a,b, c, 2002a,b). The anionic surfactant sodium dodecyl sulphate (SDS) inhibited the filtration activity of Mytilus edulis (Ostroumov et al. 1997) and M. galloprovincialis (Ostroumov 2000d, 2001c). Surfactants tetradecyltrimethylammoinium bromide (TDTMA) and SDS (both at 0.5 mg l-1) inhibited the filtration activity of C. gigas (Ostroumov 2003a). Triton X100, a non-ionic surfactant, inhibited the filtration rate of M. edulis (Ostroumov et al 1998), M. galloprovincialis, and the freshwater bivalve, Unio tumidus (Ostroumov 2000d, 2001c). TDTMA inhibited water filtration by rotifers (Kartasheva and Ostroumov 1998, Ostroumov et al. 2003b). New data on the effects of some chemicals on suspension-feeders are presented in Table 6. Those data are in accord with the previous studies by Table 6. Some chemicals that decrease the filtration rates of the suspension-feeders (Ostroumov 2003a, with additions). (SD - synthetic laundry detergent (powder); LD - liquid detergent) Chemical (see text) Organism Reference oil hydrocarbons Mytilus galloprovincialis New data pesticides M. edulis Donkin et al. (1997) tributyltin, dibutyltin M. edulis Widdows and Page (1993) SDS, TDTMA Crassostrea gigas New data SDS M. edulis Ostroumov et al. (1997) Triton X-100 M. edulis Ostroumov et al. (1998) Triton X-100 Unio tumidus Ostroumov (2002a) TDTMA M. edulis/M .galloprovincialis New data TDTMA Unio pictorum Ostroumov (2002a) SD (Lanza) M. galloprovincialis, C. gigas New data LD (E) M. galloprovincialis, C. gigas New data LD (Fairy) M. galloprovincialis Ostroumov (2001b) LD (Fairy) C. gigas New data SD (IXI) M. galloprovincialis Ostroumov (2002c) SD (Deni) C. gigas Ostroumov (2002c) SDS M. galloprovincialis Ostroumov (2000c) Cu M. galloprovincialis New data Cu Dreissena polymorpha Kraak et al. (1993) Cd M. galloprovincialis New data Cd Dreissena polymorpha Kraak et al. (1993) SD OMO Unio tumidus Ostroumov (2001a) TDTMA Brachionus angularis Kartasheva and Ostroumov (1998) TDTMA Brachionus calyciflorus Ostroumov et al. (2003 b) PCBs, PAHs M. edulis Olsson et al. (2004)
157 other authors who have studied other chemicals (pesticides, inorganic salts of metals) which also inhibited the filtration rates of bivalves (e.g., Mitin 1984, Stuijfzand et al. 1995, Donkin et al. 1997). Filtering can be inhibited not only by man-made chemicals, but also by some natural substances, i.e., kairomones. Thus, the filtration activity of Ceriodaphnia was inhibited by kairomones exuded by eastern rainbowfish Melanotaenia duboulayi at fish densities 0.125 fish l-1 (Rose et al. 2003). Suspension-feeders as Bio-filters Some of the facts discussed above are relevant to many taxa of suspension-feeders, those in the plankton and benthos, and those in freshwater and marine systems. The material above leads to the formulation of some conclusions about the fundamental features of suspension-feeders. This is presented in Table 7. Each line of the table starts with a summary of factual data in the left cell of the line. The factual data are commented in the middle cell of each line, which pave the way for the formulation of a more general principle in the right cell of each line. Table 7. Some key facts and principles that characterise suspension-feeders as part of waterfiltering biomachinery maintaining water quality and some features of aquatic ecosystem. Key facts
Comment / Consequences
The amount of water filtered (per unit of biomass of animals or per unit of area or per unit of time) is very significant There are several taxa of suspension-feeders which filter water
Significant contribution to the removal of particles (seston) from water; contribution to water purification
The higher the concentration of particles, the lower the filtration rate and relative grazing pressure
Positive feedbacks that in turn may lead to the increase in heterogeneity of parts of the water column
The amount of suspension that is being filtered out of water is usually more than needed for metabolism
A significant amount of the formerly suspended matter is finally packed, ejected and/or excreted as pellets
Suspension-feeders produce pellets
The pellets gravitate towards the bottom or the lower layers
Increase in reliability of the biomachinery of water filtration
Fundamental principles concerning the role of suspension-feeders Large-scale repair of water quality
Contribution to the stability of water quality in ecosystem; maintaining stability of habitats of many aquatic species Suspension-feeders have a potential to contribute to creating habitat heterogeneity (in terms of patchiness of concentrations of suspended matter in water) Suspension-feeders provide some ecological services to the system (by upgrading water quality); ecological taxation: suspension-feeders pay ecological tax to the community (ecosystem) Acceleration of migration of elements through the water
158 of the water column
Suspension-feeders remove bacteria and fungi
Regulatory effect (control) on planktonic bacteria and fungi; regulatory impact on benthos benthic community
column of the ecosystem; pellets-mediated acceleration of the removal of particles (seston) from water column Contribution to the regulation of ecosystem metabolism
The material of this paper together with that previously presented (Yablokov and Ostroumov 1991, Ostroumov 2000d, 2001c, 2003a,b) provides a basis for a new perspective of how to analyze and classify manmade disturbances in ecosystems (Table 8). According to the approach presented in Table 8 and discussed in more detail in (Ostroumov 2003b), man-made inhibition of the water filtration rate can be a significant disturbance to the ecosystem. The inhibition of the filtration activity of suspension-feeders may lead to the situation previously described as that of ecological bomb of the second type (Ostroumov 1999). Table 8. The level–block approach to the analysis of ecological hazards of anthropogenic effects on the biota. After Ostroumov (2001c, 2003b) with some modifications.
Disturbance level
Examples of disturbances and their consequences (some of them may be assigned to different levels)
1
Individual responses
Toxic effects on individual organisms (increased mortality, decreased fertility, ontogenetic disturbances, diseases, etc.), changes in morphological and physiological variability, and behavioral changes
2
Aggregated responses of groups of organisms
Changes in primary and secondary productivity, biomass of groups of organisms, water chlorophyll, and dissolved O2 concentrations
3
Ecosystem stability and integrity (which is an attribute that depends on the sum of functional links among parts of the ecosystem)
Weakening and/or rearrangements of plankton–benthos connections; rearrangements and/or weakening of links in the food web; changes in the level of bacterial and fungal biodegradation; decrease in the filtration rates and elimination of suspended particles (seston) from water; decrease in water self-purification. Decrease in some regulatory effects because of the loss, migration, or trophic inertness of organisms belonging to higher trophic levels
159 4
Ecosystem contribution to biospheric processes: biogeochemical flows of chemical elements
Changes in C flows (e.g., sedimentation of pellets formed by suspension-feeding organisms) and N flows (e.g., nitrogen fixation), as well as in fluxes and cycles of other elements, including S, Si, P, and others (e.g., Ostroumov, Kolesnikov, 2001, 2003)
Ecosystem contribution to biospheric processes: flows of energy
Changes in energy (heat, light, UV etc.) flows
CONCLUSIONS 1. Suspension-feeders are key components in an ecosystem's water purification activity. 2. Their role is powerful, labile, and subject to subtle adjustment and regulation. 3. The functioning of suspension-feeders is vulnerable to manmade stressors: sub-lethal concentrations of manmade chemicals (surfactants and others) can inhibit the filtration activity of suspension-feeders. 4. The analysis presented shows the important role of suspension-feeders in aquatic ecosystem. As the group of suspension-feeders includes many taxa that are a part of biodiversity, the analysis leads to better appreciation of the value of biodiversity of aquatic ecosystems. 5. The data and analysis presented show the environmental hazard of some sub-lethal concentrations of chemicals exemplified by their negative effects on the filtration rates of filter-feeders. New priorities in assessing potential environmental hazard of chemicals can now be formulated. Our data underline the hazards sublethal concentrations of pollutants have on aquatic organisms. From that perspective, the current set of criteria to assess these hazards is to be adjusted so that more attention is given to sublethal effects of pollutants including those on suspension-feeders. The contribution of a diversity of suspension-feeders to water purification is another important argument that emphasizes the necessity of biodiversity protection, for the role of quality of water is immense: "Water is life's materr and matrix, mother and medium. There is no life without water." (Albert von Szent-Györgyi)
ACKNOWLEDGEMENTS The author thanks many colleagues at Moscow State University, the Institute of Biology of Southern Seas (Sevastopol), and the Plymouth Marine Laboratory for assistance, to Prof. R. Dame and the other participants of the
160 workshop on suspension-feeders in Nida (2003), the participants of the series of conferences 'Aquatic Ecosystems and Organisms', Prof. AF Alimov, Prof. M.E. Vinogradov, Prof. V.V. Malakhov, Prof. E.A. Kriksunov, Prof. V. D. Fedorov, Prof. A.O. Kasumyan, Prof. G. E. Shulman, Prof. N. Walz, Prof. P. Wangersky, and Prof. R. Krueger for stimulating discussions, to Dr. N. N. Kolotilova, N.E. Zourabova, L.V. Podshchekoldina and L.I. Shpitonova for help, to Dr. R. Dame, Dr. D. Bushek, and Dr. J. Thompson for sending scientific materials. Special thanks to the reviewers of the paper. The grants from the MacArthur Foundation and the Open Society Foundation supported some part of the research. The work with M. edulis was done together with Dr. P. Donkin and Prof. J. Widdows.
REFERENCES Achituv Y Pedrotti ML 1999 Costs and gains of porcelain crab suspension feeding in different flow conditions. Mar Ecol Prog Ser 184: 161-169 Alimov AF 1981 Functional Ecology of Freshwater Bivalves (Funktzionalnaja Ekologija Presnovodnykh Dvustvorchatykh Molluskov). Nauka Press, Leningrad, 248 p Asmus RM Asmus H 1991 Mussel beds: limiting or promoting phytoplankton? J Exp Mar Biol Ecoll 148: 215-232 Bacher C 1989 Capacité trophique du bassin de Marennes-Oléron: couplage d'un modele de transport particulaire et d'un modele de croissance de l'huitre Crassosrea gigas. Aqua Living Resources 48: 199-214 Ball B Raine R Douglas D 1997 Phytoplankton and particulate matter in Carlingford Lough, Ireland: an assessment of food availability and the impact of bivalve culture. Estuaries 20: 430-440 Bern L 1987 Zooplankton grazing on [methyl-3H]thymidine-labelled natural particle assemblages: determination of filtering rates and food selectivity. Freshwater Biol 17: 151-159 Cloern J 1982 Does the benthos control phytoplankton biomass in South San Francisco Bay? Mar Ecol Prog Ser 9: 191-202 Dame RF 1972 The ecological energies of growth, respiration and assimilation in the intertidal American oyster, Crassostrea virginica. Mar Biol 17: 243-250 Dame RF 1976 Energy flow in an intertidal oyster population. Est Coast Mar Sci 4: 243253 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Dame RF Bushek D Edwards D Koepfler ET Gregory L 2002 Ecosystem response to bivalve density reduction. Aqua Ecoll 36: 51-65 Dame RF Bushek D Prins T 2001 Benthic suspension feeders as determinants of ecosystem structure and function in shallow coastal waters. In: Ecological Comparisons of Sedimentary Shores, K Reise (Ed), Springer-Verlag, Berlin, pp 11-37 Dame RF Dankers N Prins T Jongsma H Smaal A 1991 The influence of mussel beds on nutrients in the Western Wadden Sea and Eastern Scheldt estuaries. Estuaries 14: 130-138. Dame RF Spurrier JD Wolaver TG 1989 Carbon, nitrogen and phosphorus processing by an oyster reef. Mar Ecol Prog Ser 54: 249-256 Dame RF Zingmark R Stevenson H Nelson D 1980 Filter feeding coupling between the water column and benthic systems. In: Estuarine Perspectives, V Kennedy (Ed). Academic Press, NY, pp 521-526
161 Davies B Stuart V de Villiers M 1989 The filtration activity of a serpulid polychaete population ((Ficopomatus enigmaticus (Fauvel) and its effects on water quality in a coastal marina. Estu Coast Shelf Sci 29: 613-620 Donkin P Widdows J Evans SV Staff F Yan T 1997 Effects of neurotoxic pesticides on the feeding rate of marine mussels Mytilus edulis. Pesticide Sciencee 49: 196-209. Frost T 1978 Impact of the freshwater sponge Spongilla lacustris on a Sphagnum bog-pond. Verhandlungen Internationale Vereinigung für Theoretische und Angewandte Limnologie 20: 2368-2371 Frost T 1980 Clearance rate determination for the freshwater sponge Spongilla lacustris: effects of temperature, particle type and concentration, and sponge size. Archiv für Hydrobioll 90: 330-356 Gutelmaher BL 1986 Metabolism of Plankton as the Whole (Metabolizm Planktona Kak Edinogo Tzelogo) Nauka, Leningrad. 156 p Herman P Middelburg J Van de Koppel J Heip C 1999 Ecology of estuarine macrobenthos Adv Ecol Res 29:195-240 Hily C 1991 Is the activity of benthic suspension feeders a factor controlling water quality in the Bay of Brest? Mar Ecol Prog Ser 69: 179-188 Kartasheva NV Ostroumov SA 1998 Tetradecyltrimethylammonium bromide (tetradetziltrimetilammonij bromid). Toxicol Bull (Toksikologicheskii Vestnik) k 5: 30-32 Kautsky N Evans S 1987 Role of biodeposition by Mytilus edulis in the circulation of matter and nutrients in a Baltic coastal ecosystem. Mar Ecol Prog Serr 38: 201-212 Kraak M Lavy D Toussaint M Schoon H Peeters W Davids C 1993 Toxicity of heavy metals to the zebra mussel ((Dreissena polymorpha). In: Zebra Mussels: Biology, Impact, and Control. TN Nalepa D Schloesser (Eds), Lewis Publishers, Boca Raton, pp 491-502 Krylova EM 1997 Abyssal bivalves of the superfamily Cuspidarioidea (Septibranchia) from the Far-Eastern seas and the North-Western Pacific. In: Composition and Distribution of Bottom Invertebrate Animals in the Seas of Russia and Adjacent Waters. AP Kuznetzov ON Zezina (Eds), Institute of Oceanology, Moscow, pp 80-84 Kryger J Riisgård HU 1988 Filtration rate capacities in 6 species of European freshwater bivalves. Oecologia 77: 34-38 Kryuchkova NM Rybak VH 1989 Patterns of feeding of Moina rectirostris Leydig on phytoplankton of a hypertrophic water body (osobennosti pitaniya Moina rectirostris Leydig fitoplanktonom gipertrofnogo vodoema. Hydrobiol J (Gidrobiologicheskiy Zhurnal) 25 (3): 28-31 Loo L Rosenberg R 1989 Bivalve suspension-feeding dynamics and benthic – pelagic coupling in a eutrophicated marine bay. J Exp Mar Biol Ecoll 130: 253-276 Migne A Davoult D 2002 Experimental nutrition in the soft coral Alcyonium digitatum (Cnidaria: Octocorallia): removal rate of phytoplankton and zooplankton. Cah de Biol Marr 43 (1): 9-16 Mitin A 1984 Effect of some environmental factors on the water-clearing activity of bivalves. Summary of Ph.D. Thesis, Moscow. 22 p Monakov AV 1998 Feeding of Freshwater Invertebrates (Pitanie Presnovodnykh Bespozonochnykh). Institute of Ecological and Evolutionary Problems, Moscow. 320 p Newell RIE 1988 Ecological changes in Chesapeake Bay: are they the result of overharvesting the American oyster, Crassostrea virginica? In: Understanding the Estuary: Advances in Chesapeake Bay Research , MP Lynch EC Krome (Eds), Chesapeake Bay Research Consortium, Solomon’s Maryland, pp 536-546 Nichols F 1985 Increased benthic grazing: an alternative explanation for low phytoplankton biomass in northern San Francisco Bay during the 1976-1977 drought. Estu Coast Shelf Sci 21: 379-388 Ogilvie S Mitchell S 1995 A model of mussel filtration in a shallow New Zealand lake, with reference to eutrophication control. Arch Hydrobioll 133: 471-482
162 Olsson B Bradley B Gilek M Reimer O Shepard J Tedengren M 2004 Physiological and proteomic responses In Mytilus edulis exposed to PCBs and PAHs extracted from Baltic Sea sediments. Hydrobiologia 514: 15-27 Ostroumov SA 1998 Biological filtering and ecological machinery for self-purification and bioremediation in aquatic ecosystems: towards a holistic view. Rivista di Biologia/ Biology Forum 91: 247-258 Ostroumov SA 1999 Integrity-oriented approach to ecological biomachinery for selfpurification and bioremediation in aquatic ecosystems: stopping an ecological time bomb. Limnology and Oceanography: Navigating into the Next Century. February 15, 1999, Santa Fe, New Mexico. ASLO, Waco, TX. Page 134 Ostroumov SA 2000a Criteria of ecological hazards due to anthropogenic effects on the biota: searching for a system (Kriterii ekologhicheskoj opastnosti antropoghennykh vozdejstvij na biotu: poiski sistemy). Doklady Biological Sciences 371: 204-206 (the Russian edition: Doklady Akademii Naukk 371: 844-846) Ostroumov SA 2000b The concept of aquatic biota as a labile and vulnerable component of the water self-purification system (Kontzeptzija vodnoi bioty kak labil'nogo i ujazvimogo zvena sistemy samoochishchenija vody). Doklady Biological Sciences 372: 286-289 (the Russian edition: Doklady Akademii Naukk 372: 279-282) Ostroumov SA 2000c Aquatic ecosystem: a large-scale diversified bioreactor with a water self-purification function (Vodnaja ekosistema: krupnorazmernyj diversifitzirovannyj bioreaktor s funktzijej samoochishchenija vody). Doklady Biological Sciences 374: 514-516 (the Russian edition: Doklady Akademii Naukk 374: 427-429) Ostroumov SA 2000d Biological effects of surfactants in connection with the anthropogenic impact on the biosphere. MAX Press, Moscow. 116 p Ostroumov SA 2000e Inhibitory analysis of the regulatory interactions in food webs. Doklady Akademii Naukk 375: 847-849 Ostroumov SA 2001a Responses of Unio tumidus to a mixture of chemicals and the hazard of synecological summation of anthropogenic effects (Reagirovanie Unio tumidus pri vozdeistvii smesevogo himicheskogo preparata i opasnost sinekologicheskogo summirovaniya antropogennyh vozdeistviy). Doklady Akademii Naukk 380: 714-717 Ostroumov SA 2001b Synecological foundation for the solution of the problem of eutrophication (Sinekologicheskie osnovy resheniya problemy evtrofirovaniya). Doklady Akademii Naukk 381: 709-712 Ostroumov SA 2001c Biological Effects of Surfactants on Organisms. MAX Press, Moscow. 334 p Ostroumov SA 2001d Imbalance of factors which control the abundance of unicellular plankton under the anthropogenic impact. Doklady Akademii Naukk 379: 136-138 Ostroumov SA 2002a Inhibitory analysis of top-down control: new keys to studying eutrophication, algal blooms, and water self-purification. Hydrobiologia 469: 117129 Ostroumov SA 2002b Polyfunctional role of biodiversity in processes leading to water purification: current conceptualizations and concluding remarks. Hydrobiologia 469: 203-204 Ostroumov SA 2002c Biodiversity and water quality: the role of feed-backs (Sokhranenie bioraznoobrziya i kachestvo vody: rol obrantykh svyazej v ekosistemakh) Doklady Akademii Naukk 382: 138-141 Ostroumov SA 2003a Studying effects of some surfactants and detergents on suspension feeding bivalves. Hydrobiology 500: 341-344 Ostroumov SA 2003b Anthropogenic effects on the biota: towards a new system of principles and criteria for analysis of ecological hazards. Rivista di Bilogia / Biology Forum 96: 159-169
163 Ostroumov SA 2004 On biotic self-purification of aquatic ecosystems. Elements of theory (O bioticheskom samoochishchenii vodnykh ekoistem. Elementy teorii) Doklady Akademii Naukk 396: 136-141. Ostroumov SA Kolesnikov MP 2001 Pellets of molluscs in biogeochemical flows of C, N, P, Si, and Al. Doklady Akademii Naukk 379: 426-429 Ostroumov SA Kolesnikov MP 2003 Molluscs in biogeochemical flows (C, N, P, Si, Al) and water self-purification: effects of surfactants. Vestnik MGU (Bulletin of Moscow State University). Series 16. Biologyy 1: 15-24 Ostroumov SA Donkin P Staff F 1997 Inhibition by the anionic surfactant, sodium dodecyl sulphate, of the ability of mussels Mytilus edulis to filter and purify the sea water (Anionnoje poverkhnostno-aktivnoje veshchestvo inghibirujet sposobnost' midij Mytilus edulis filtrovat' i ochishchat' morskuju vodu). Vestnik Moskovskogo Universiteta. Seria 16. Biologija. (Bulletin of Moscow University. Series 16. Biology) 3: 30-36 Ostroumov SA Donkin P Staff F 1998 Filtration inhibition induced by two classes of synthetic surfactants in the bivalve mollusc (Narushenije filtracii dvustvorchatymi molluskami pod vozdejstvijem poverkhnostno-aktivnykh veshchestv dvukh klassov). Doklady Akademii Naukk 362: 574-576 Ostroumov SA McCutcheon S Steinberg C (Eds) 2002 Ecological Processes and Ecosystems: Functioning towards Water Purification. (Hydrobiologi ( a, vol. 469). Kluwer Academic Publishers, Dordrecht, 204 p Ostroumov SA Walz N Widdows J 2001 Measuring activity of suspension feeders: linking organismal and superorganismal levels of hydrobiology and biological oceanography. hy y Ecological Studies, Hazards, Solutions 5: 26-27 Ostroumov SA Dodson S Hamilton D Peterson S Wetzel R 2003a Medium-term and longterm priorities in ecological studies. Rivista di Biologia/ Biology Forum 96: 327-332 Ostroumov SA Walz N Rusche R 2003b Effects of a cationic amphiphilic chemical on rotifers. Doklady Akademii Naukk 390: 423-426 Petersen J Riisgård HU 1992 Filtration capacity of the ascidian Ciona intestinalis and its grazing impact in a shallow fjord. Mar Ecol Prog Ser 88: 9-17 Pile A Patterson M Savarese M Chernykh VI Fialkov VA 1997 Trophic effects of sponge feeding within Lake Baikal's littoral zone. 2. Sponge abundance, diet, feeding efficiency, and carbon flux. Limnol Oceanogrr 42: 178-184 Pilson M 1985 On the residence time of water in Narragansett Bay. Estuaries 8: 2-14 Rand G 1995 Fundamentals of Aquatic Toxicology, Taylor & Francis, Philadelphia, 1125 p Reeders H bij de Vaate A Noordhuis R 1993 Potential of the zebra mussel (Dreissena ( polymorpha) for water quality management. In: Zebra Mussels: Biology, Impact, and Control. TN Nalepa D Schloesser (Eds), Lewis Publishers, Boca Raton, FL, pp 439451 Rose R Warne M Lim L 2003 Exposure to chemicals exuded by fish reduces the filtration and ingestion rates of Ceriodaphnia cf. dubia. Hydrobiologia 501: 215-217 Savarese M Patterson M Chernykh VI Fialkov VA 1997 Trophic effects of sponge feeding within Lake Baikal's littoral zone. 1. In situ pumping rates. Limnol Oceanogr 42: 171178 Smaal AC Verhagen J Coosen J Haas H 1986 Interactions between seston quantity and quality and benthic feeders in the Oosterschelde, The Netherlands. Ophelia 26: 385-399 Sorokin YA Sorokin PY Mamaeva TI Sorokina OV 1997 The Sea of Okhotsk bacterioplankton and planktonic ciliates. In Complex Studies of Ecosystem of the Sea of Okhotsk. VV Sapozhnikov (Ed), VNIRO Press, Moscow, pp 210-216 Shulman GE Finenko GA 1990 Bioenergetics of Aquatic Organisms. Naukova Dumka Press, Kiev, 246 p Shuntov VP 2001 Biology of Far-Eastern Seas of Russia. Vol.1. TINRO-Centre, Vladivostok. 580 p
164 Starkweather P Bogdan K 1980 Detrital feeding in natural zooplankton communities: discrimination between live and dead algal foods. Hydrobiologia 73: 83-85 Stuijfzand SC Kraak MHS Wink YA Davids C 1995 Short-term effects of nickel on the filtration rate of the zebra mussel Dreissena polymorpha. Bulletin of Environmental Contamination and Toxicology 54: 376-381 Sushchenya LM 1975 Quantitative Trends in the Feeding of Crustaceans. Nauka i Tekhnika press, Minsk, 208 p Tsuchiya M 1980 Biodeposit production by the mussel Mytilus edulis L. on rocky shores. J Exp Mar Biol Ecoll 47: 203-222 Vinogradov ME Shushkina EA 1987 Functioning of Plankton Communities of the Epipelagial of the Ocean. Nauka Press, Moscow, 240 p Wetzel B 2001 Limnology. Academic Press, San Diego. 1006 p Widdows J Page D 1993 Effects of tributyltin and dibutyltin on the physiological energetics of the mussel, Mytilus edulis. Mar Env Res 35: 233-249 Widdows J Brinsley M Salkeld P Elliott M 1998 Use of annular flumes to determine the influence of current velocity and bivalves on material flux at the sediment-water interface. Estuaries 21 (4A): 552-559 Widdows J Brown S Brinsley M Salkeld P Elliott M 2000 Temporal changes in intertidal sediment erodibility: influence of biological and climatic factors. Cont Shelf Res 20: 1275-1289 Witte U Brattegard T Graf G Springer B 1997 Particle capture and deposition by deep-sea sponges from the Norwegian-Greenland Sea. Mar Ecol Prog Serr 154: 241-252. Wotton R 1994 The Biology of Particles in Aquatic Systems. Lewis Publishers, Boca Raton, FL, 325 p Yablokov AV Ostroumov SA 1991 Conservation of Living Nature and Resources: Problems, Trends, Prospects. Springer-Verlag, Berlin, 271 p Zankai N 1979 Jahreszeitliche Änderung der Filtrierrate der Copepoden Eudiaptomus gracilis (G.O.Sars) im Balaton-See. Verhandlungen Internationale Vereinigung für Theoretische und Angewandte Limnologie 20: 2551-2555 Zezina ON 1997 Brachiopods in a natural bottom biofilter at the shelves and slopes in the Far-Eastern Seas of Russia. In: Composition and Distribution of Bottom Invertebrate Animals in the Seas of Russia and Adjacent Waters, AP Kuznetzov ON Zezina (Eds), Institute of Oceanology, Moscow, pp 52-60
NEOPLASIA IN ESTUARINE BIVALVES: EFFECT OF FEEDING BEHAVIOUR AND POLLUTION IN THE GULF OF GDANSK (BALTIC SEA, POLAND)
Maciej Woáowicz, Katarzyna Smolarz, Adam Sokoáowski Laboratory of Estuarine Ecology, Institute of Oceanography, University of GdaĔsk, Al. Marszaáka á J.Piáásudskiego 46, 81-378 Gdynia, Poland Abstract:
The incidence of tumors in bivalve molluscs is receiving increased attention due to possible detrimental effects on harvested stocks. Although the etiology or causes of neoplasias remains unclear, pollution by carcinogenic agents is implicated in the heavily exploited littoral zones of coastal waters. In the Gulf of GdaĔsk, southern Baltic Sea, a higher prevalence of the disorder was observed in infaunal facultative (deposit/suspension) feeders compared to epifaunal obligate suspension-feeders, providing a new behavioural aspect of the tumor. Recent studies also reveal a potential cause-and-effect relationship between sediment factors and the incidence of neoplasia across a range of environmental properties.
Keywords: neoplasia, bivalves, feeding behaviour, sediment toxicity, Baltic Sea
INTRODUCTION Over the last several decades estuarine and coastal ecosystems have been subject to considerable change in terms of ecosystem functioning. Intense human (economic) activity in the coastal zone and catchments areas have brought about an increase in pollution and eutrophication, inducing alterations of both physical and chemical properties of the water column and bottom sediments. Increased suspended matter and high loadings of nutrients and pollutants discharged to the sea have had direct effects on benthic communities. Increased sedimentation and oxygen depletion have degraded natural habitats, resulting in changes of spatial distribution and biodiversity of benthic species. For example, the elevated contents of suspended matter and of silt fractions in surficial sediments have enabled suspension/deposit feeder and detritivorous populations to expand. Among them suspension-feeding bivalves play an important role in the transfer of organic matter from the water column to bottom sediments (through filtration and biodeposition) and thus contribute significantly to energy fluxes between pelagic and benthic 165 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 165–182. ©2005 Springer. Printed in the Netherlands.
166 systems (Kautsky and Evans 1987). Suspension-feeders act as a natural biological filter and remove large amounts of suspended particles from the water column, together with particle-bound chemical contaminants. Thus, harmful toxic substances might be removed from the water column and deposited as faeces and pseudo-faeces to the surface sediments or be accumulated in tissues. The direct consequences of these suspension-feeder mediated processes are a potential improvement of water quality, an increased risk of cumulative pollutant transfer along the food chain to higher consumers (including humans) or food chain magnification and a general threat to the ecological and economic environments. In contrast to suspension-feeders, deposit feeders take up organic material that has already been transported to the bottom by natural sedimentation processes or through biodeposition. If harmful pollutants that are usually linked to organic matter are enriched in the sediments, then deposit feeders may be exposed to a higher degree of harmful substances than suspension-feeders. Deleterious effects can be expected in stagnated deep-water regions where low water flow often leads to oxygen depletion and increased hydrogen sulphide concentrations (Janas and Szaniawska 1996) that may account for massive mortalities in resident bivalve populations (Cederwall and Elmgren 1990, Baden et al. 1990). Synergistic interactions between various environmental factors, including different pollutants are thought to induce pathological changes in the ecosystem such as toxic algal blooms, teratological forms of diatoms, increased invertebrate parasitism, greater incidences of fish illness, in addition to the morphological deformation of shell and the appearance of neoplasia in bivalves. In the case of sedentary benthic invertebrates, shell deformation and neoplasia may indicate a decrease in the natural (immunological) resistance of these organisms to harmful substances and the ecotoxicological risk in the ambient environment. In the Gulf of GdaĔsk part of the southern Baltic Sea, neoplastic cells have been found in two common species of infaunal facultative (deposit/suspension) feeders, namely Macoma balthica and Mya arenaria, but not in epifaunal obligate suspension-feeders such as Mytilus edulis trossulus andd Cerastoderma glaucum, highlighting a new biological (behavioural) aspect of the tumor. In this review, the hypothesis comparing the risk of carcinogenic changes of facultative feeders versus suspension-feeders is discussed based on the records of neoplasia in southern Baltic Sea. By relating the prevalence of gill neoplasia and the data from the concurrent monitoring of basic environmental parameters to the toxicological data of the surface sediments, a new approach is offered on the etiologic basis of the cancer.
167 NEOPLASIA IN BIVALVE MOLLUSCS One of the most deleterious phenomena observed recently in bivalves is tumor disorder. Neoplasia usually starts with local and multifocal microscopic lesions in respiratory system and the presence of abnormal cells freely circulating within hemolymph. In the middle phase of the disease, bivalves exhibit emaciation, cellular degeneration, and poor cellular differentiation. A terminal phase is manifested by reduction in mitotic activity and further degeneration of tissue leading to the death of the organism (e.g. Farley 1969a, Villalba et al. 1995, Villalba et al. 2001). In disturbed coastal and estuarine systems, marine invertebrates, specifically bivalves, often develop this fatal tumor in their hemolymph and tissues. An increase in both the numbers and types of tumors found in shellfish has been noted over the past several decades (Elston et al. 1992). Neoplasms or the proliferation of cellular disorders were first described in oysters (Farley 1969a), mussels (Farley 1969b, Pauley 1969) and clams (Christensen et al. 1974) from North American coastal waters. Since that time, neoplasia has spread widely and been reported in at least 22 bivalve species (Table 1) from more than 20 locations around the world (Fig. 1; Elston et al. 1992, Peters et al. 1994, Villalba et al. 2001 among others). Table 1. Survey of records of the occurrence of neoplasia in different bivalve species from various geographical regions. Species
Location
Ostreidae Crassostrea commercials East Australia Crassostrea gigas East and West cost of U.S., Japan Breton Coast, France Crassostrea virginica Apalachicola Bay, FL, U.S. East and West cost of U.S. Virginia, U.S. New Haven, U.S. Rhode Island, U.S. Crassostrea rhizophorae Brazil Ostrea chilensis Chiloe, Chile Ostrea lurida Yaquina Bay, OR, U.S. Ostreola conchaphila Yaquina Bay, OR, U.S. Ostrea edulis
Mytilidae Mytilus edulis trossulus
Reference
Wolf (1977) Farley (1969a)
Galicia, Spain Mali-Ston, Yugoslavia Breton Coast, France
Balouet et al. (1986) Couch (1969) Farley (1969a) Frierman (1976) Newman (1972) Peters et al. (1994) Nascimento et al. (1986) Mix and Breese (1980) Mix (1976) Farley and Sparks (1970) Mix et al. (1977) Alderman et al. (1977) Alderman et al. (1977) Balouet et al. (1982, 1986)
Yaquina Bay, OR, U.S. Departure Bay, B.C., Canada Vancouver Island, Canada Penn Cove, WA, U.S. Sequim Bay, WA, U.S.
Farley (1969b), Mix (1983) Cosson-Mannevy et al. (1984), Bower (1989) Emmett (1984) Elston et al. (1988a)
168 Species
Location Denmark British Columbia, Canada
Mytilus edulis complexx
Puget Sound, WA, U.S. Plymouth, U.K.
Mytilus edulis galloprovincialis
Adula californiensis Myidae Mya arenaria
Humboldt Bay, CA, U.S. East and West Cost, U.S. Rias of Galicia, NW Spain Ria of Vigo, NW Spain West Cost, U.S. Long Island Sound, CT, U.S. Maine New Jersey, U.S. Maine Rhode Island, U.S. Rhode Island, U.S. Chesapeake Bay, U.S. Maine, U.S.
Barnstable Harbour, Massachusetts, U.S. Alsea Bay, OR, U.S. Atlantic coast, Canada Gulf of GdaĔsk, Baltic Sea, Poland Tellinidae Macoma truncata Macoma nasuta Macoma inquinata Macoma balthica
Cardiidae Cerastoderma edule
Mercenaria compechiensis Mercenaria mercenaria
Green and Alderman (1983) Lowe and Moore (1978) Elston et al. (1988b) Hillman (1993) Villalba et al. (1997) Alonso et al. (2001) Mix (1975a, b) Brousseau (1987) Brown et al. (1976) Reno et al. (1994) Brown et al. (1977) Cooper and Chang (1982) Farley et al. (1986) Dungan et al. (2002) Leavitt et al. (1990) Yevich and Barszcz (1977) Barber et al. (2002) Reno et al. (1994) House et al. (1998) Barber et al. (2002) Woáowicz et al. (2000)
Baffin Island, Canada Yaquina Bay, OR, U.S. Yaquina Bay, OR, U.S. Chesapeake Bay, U.S. Finnish Cost Gulf of Gdansk, Baltic Sea, Poland Tred Avon River, Maryland, U.S.
Neff et al. (1987) Farley (1976a) Farley (1976a) Christensen et al. (1974) Pekkarinen (1993) Thiriot-Quiévreux and Woáowicz (1996, 2001) Farley (1976b)
Cork Harbour, Ireland
Twomey and Mulcahy (1988) Villalba et al. (2001)
Galicia (NW Spain) Veneridae Ruditapes decussatus
Reference Pauley (1969) Rasmussen (1986) Bower (1989) Krishnakumar et al. (1999)
Ria de Arousa, Galicia, NW Spain Coastal Florida Lagoon, U.S.
Villalba et al. (1995)
Narragansett Bay, RI, U.S. South Carolina, U.S.
Yevich and Barry (1969) Evarsole and Heffernan (1995)
Bert et al. (1993)
169
Fig. 1. Locations with recorded neoplasia in bivalves (for references see Table 1).
Most neoplasias have been described as a growth pattern of hemolymph cells, characterized primarily by excessive cell proliferation. Three main cytological characteristics are associated with neoplastic cells, namely nuclear polymorphism, nuclear hyperchromatism and cell polymorphism. Other features include the presence of more than one nucleolus in the nucleus, a high frequency of mitotic division and a high nucleus to cytoplasm ratio (Mix 1983, Krishnakumar et al. 1999). In molluscs, most identified neoplasms have been reported as sarcomas of haematopoietic origin (Bower et al. 1994, Rodriguez et al. 1997) due to proliferation of enlarged cells with a large lobate nucleus. Gonad neoplasms have also been noted (Bert et al. 1993, Peters et al. 1994, Alonso et al. 2001). To date, the neoplasia of bivalves is the most well-defined and widespread tumor in marine and brackish water invertebrates (Elston et al. 1992). Over the last several decades, shellfish culture has developed considerably around the world and, in many cases neoplasia has been found in cultured bivalves (Ford et al. 1997, Alonso et al. 2001) or in wild populations that occur in regions of intense aquaculture activity (Rodriguez et al. 1997, Villalba et al. 2001). Due to potential deleterious effects that epizootics may have on monospecific communities, the prevalence and etiology of neoplastic conditions have received considerable attention. Initially, viruses, retroviruses and infectious agents were most commonly hypothesised as the main cause of neoplastic changes in marine bivalves (Farley et al. 1972, Oprandy et al. 1981, Oprandy and Chang 1983,
170 Ford et al. 1997). More recently, environmental carcinogens such as hydrocarbons (Yevich and Barszcz 1977, Naes et al. 1995), herbicides (Van Beneden et al. 1993) and so-called biotoxins (Landsberg 1996, Roy et al. 1998) have been suggested as risk factors in the development of mutations that may provoke neoplastic features. In many cases, while the increase in tumor incidence can be correlated with the increase of aquatic toxicant levels, causality cannot be definitively proven. Bivalves have been proposed as ideal indicator organisms for neoplasias (Van Beneden 1994). The development of techniques that allow the effects of harmful compounds on bivalve biology and physiology to be studied may lead to the establishment of programmes for monitoring the distribution and transfer of pollution in coastal areas similar to the Mussel Watch Program.
THE GULF OF GDANSK - A CASE STUDY The Gulf of Gdansk is considered a seriously polluted water basin. Over several decades, a number of adverse alterations in the ecosystem of the Gulf have been recorded (Glasby and Szefer 1998). Most threatening was the increasing contamination by nutrients (Nowacki et al. 1993, Andrulewicz 1996), heavy metals, radionuclides (Szefer et al. 1996, Szefer 2002), organic compounds (e.g. pesticides, cyclic hydrocarbons, organotin; Falandysz et al. 1997 1998, Konat and Kowalewska 2001, Albata et al. 2002), and military wastes that were deposited on the sea bottom after the Second World War (Korzeniewski 1999). Potential effects to the system may be also induced by invasive and non-indigenous estuarine and marine organisms (NEMO), associated parasites and pathogenic microorganisms to which local species are not specifically resistant (Leppakoski and Olenin 2000). These result in an integrated and functional response at population, community and ecosystem levels including morphological changes in diatoms e.g. increased contribution of teratological forms (BogaczewiczAdamczak et al. 2001), worse physiological condition and a higher degree of parasite infection in marine invertebrates and vertebrates (Sokoáowski et al. 1999, Chibani et al. 2001), gill neoplasia in bivalves (Thiriot-Quiévreux and Woáowicz 1996, 2001, Smolarz et al. 2003) and the appearance of toxic algal blooms (PliĔski et al. 1998). This pathology is most likely attributed to indirect effects of pollution. Less well known yet are interactions between the various constituents of pollutant mixtures that may mutually enhance, or alternatively, inhibit responses of the system. Because of the presence of serious pollution and the considerable asymmetry in the distribution of such contamination, the ecosystem of the
171 Gulf of Gdansk provides an ideal environment in which to test potential cause-and-effect relationships between pollutants and their biological effects.
Prevalence of Neoplasia and Cytogenetic Approach In the Baltic clam M. balthica, neoplastic cells were first seen in the respiratory system. A gill-originating neoplasia has been reported from only two sites in the Baltic Sea, namely the coastal waters of Finland (Pekkarinen 1993) and the Gulf of Gdansk, Poland (Thiriot-Quiévreux and Woáowicz 1996). In the Gulf of Gdansk, M. balthica (12 mm to 22 mm shell length) were collected by dredging at four regular sites: SW60, SW40, PB30 and Hel45 (the number at each station designates its depth in meters, Fig. 2) between 1995 and 2002. Sampling sites were selected to represent different hydrological and environmental conditions in order to better elucidate etiology of neoplasia. Recent studies have confirmed a relatively high prevalence of the tumors in respiratory, digestive and reproductive systems of the clam in the time studied (Table 2), and the appearance of neoplastic cells in another bivalve species, the soft-shell clam M. arenaria, at the rate of 16.4 % (Woáowicz et al. 2000). Table 2. Average contribution of diseased Macoma balthica in the Gulf of Gdansk over the period of 1995-2002 (Thiriot-Quiévreux and Woáowicz 1996, 2001).
Year 1995 1998 1999 2000 2001 2002 Total
No of animals studied 47 152 648 701 696 234 2478
No of abnormal animals 13 51 162 302 281 37 846
% of abnormal mitosis 27.7 33.0 25.0 43.1 40.4 16.1 Mean 34.1
Chromosome analyses based on Giemsa-stained metaphases of the M. balthica gill tissue from different regions of the Gulf, revealed two categories of animals, a normal one with a diploid chromosome number of 2n = 38 and an abnormal one with hyperploidy and chromosomal abnormalities related to neoplasia. According to the histological representation of the cancer, affected cells were large and actively proliferative with pleomorphic nuclei. Neoplastic cells were first found in gills, and then they often invaded surrounding connective tissues, leaving gills lesions severely changed. Heavily diseased individuals showed infiltration of neoplastic cells in all tissues and numerous mitotic figures.
172 The prevalence of affected animals varied from 0 to 94% at the different sites studied (Thiriot-Quiévreux and Woáowicz 1996, 2001). The highest prevalence of the disease occurred at Hel45 and the lowest prevalence was noted at SW40 (Vistula River plume) and PB30 (inner part of the Gulf).
10 m 40 m 80 m
Clear seasonal tendencies were observed in the contribution of animals with neoplastic features, with generally higher intensities of neoplasia during spring and summer than in winter. At the site with the highest average prevalence of the tumor maximal number of affected clams was noted in the spring (63.3% in May) and summer (65.6% in July). In late summer/autumn the occurrence of animals with neoplastic cells decreased but still remained high (ca. 60%).
Etiology Although the etiology of neoplasia in bivalves remains essentially undefined, many authors suggest a cause-and-effect relationship with environmental pollution. However, none of the studies was successful in the
173 clear demonstration of a single (or multifactorial) agent that can account for disseminated changes. Recently, the focus has been on benthic sediments that are both a direct habitat and the principal food source of deposit and facultative (deposit/suspension) feeders such as M. balthica. Surficial sediments were therefore sampled at the same sites in the Gulf of Gdansk and on the same occasions as the clams, and analysed for various basic environmental parameters including granulometry, organic matter content and the content of fine-grained fraction (< 63 µm), and potentially causal (toxic) factors such as the heavy metals (Cu, Zn, Cd, Ni, Pb, Mn and Fe) and PCBs. Sediments were also tested for toxicity using the Bioassay "ToxAlert" test with bacteria Vibrio fischeri. ENVIRONMENTAL PARAMETERS The content of the < 63 µm fraction in the surficial sediments and the contribution of organic components in the fraction showed spatial differences. The highest average contribution of the <63 µm fraction (ca 29%) was measured at SW60 and PB30 at depths of 60 m and 30 m. This fraction was also relatively rich in organic matter, the annual percentage of which averaged 5.7% and 7.9%, respectively. At Hel45 and SW40 the finegrained fraction constituted on average 12.0-18.1% of the sediments, and the organic matter content 5.0-6.5%. Grain-size analysis revealed a similar contribution of sand faces in surficial sediments at Hel45 and SW40 (82.9 % and 88.0 %, respectively) and at SW60 and PB30 (71.5% and 71.3%, respectively). Although a profile of sediment parameters could be drawn across the Gulf of Gdansk, no clear pattern of relationship with the neoplasia in the Baltic clam was apparent. HEAVY METALS Top 5 cm sediment samples from all sites were collected in triplicate as this depth represents the fraction of the sediments that M. balthica can penetrate and is the biologically relevant portion of the sediment. Dry sediments were sieved through a 63 µm mesh to obtain the fine-grained fraction. Separation of the < 63 µm fraction is a commonly used standardised procedure to compensate for grain size effects on element variability in sediments (Szefer et al. 1995). The residue was subject to extraction with 1 M HCl. This method determines the fraction of elements in the sediments associated with Fe- and Mn-oxides and estimates the labile and presumably bioavailable phase of the metal (Bryan and Langston 1992). Concentration of seven metals, namely Cu, Zn, Cd, Ni, Pb, Mn and Fe were measured in the extract by AAS. The sites differed substantially in the degree of sediment metal bioavailability, but no evident cause-and-effect relationship could be deduced
174 when the frequency of the neoplasia was included in a regression analysis. More general estimates explaining possible variation in the incidence of gill neoplasia in M. balthica in terms of metal bioavailability was obtained by Multiple Regression Analysis, which showed some effects of iron compounds. However, no clearly conclusive statements could be made from the results of the studies.
HYDROCARBONS (PCB) Polychlorinated biphenyls (PCB) are considered carcinogenic and highly toxic to fish and aquatic invertebrates by the Environmental Protection Agency (EPA) and International Agency for Research on Cancer (IARC). Due to very low solubility in water, PCBs can be expected to transfer rapidly from water or food to the fatty tissues of animals where they are very resistant to metabolic breakdown and thus are present in biota in relatively high concentrations. Sediment samples from the sites with the highest prevalence of gill neoplasia in M. balthica were analysed for organic chlorine with fluorescent X-ray spectrometry. In all cases the signal was below a detection limit of 5 ppm Cl. Furthermore, in three additional samples the total concentration of organic chlorine was determined in a coulometer after extraction with xylene and combustion in a pipe furnace. The results demonstrated that none of the samples analysed contained amounts of polychlorinated biphenyls or other non-volatile and volatile chloro-organic substances that can provoke carcinogenic changes in clams.
SEDIMENT TOXICITY Sediment toxicity tests were performed with a ToxAlert 100 kit (Merck®), which measures the fluorescent activity of bacteria V. fischeri. This method allows a direct estimation of the cumulative toxicity of all toxicants present in the sediments. The ecotoxicological test is based on the definition of an inhibitory effect or EC50 (concentration which affects 50% of the organisms tested) or the concentration with no noticeable effect (NOEC-No Observed Effect Concentration). In the case of the V. fischeri test the value of EC20 refers to the dilution of extractant that induces no statistically significant decrease of luminescence in acute toxicity tests i.e. a sample is not toxic. Sediments from nearly all sites demonstrated the effect of acute toxicity to disappear only at dilutions below 10 % allowing the classification of the sediments as toxic according to the Recommendations of the Helsinki
175 Commission (HELCOM Recommendation 23/11 adopted March, 6 2002). Additionally, in many cases (specifically at sites SW60 and Hel45) the toxicity was very high (Table 3). A strong relationship of sediment toxicity to the prevalence of gill neoplasia in M. balthica was found (p<0.02), highlighting the importance of surficial sediments as a source of the causal factors of the tumor. Table 3. Average sediment toxicity (3
SITE Hel45 PB30 SW60 SW40
Dilution 1:1 75.5 66.8 66.9 69.3
1:2 50.5 40.1 43.8 41.4
1:6 22.0 14.0 18.5 9.2
1:8 17.7 10.0 13.3 7.3
EC20% 8.0 12.6 9.8 12.2
CONCLUSIONS AND PROSPECTS A considerable number of recent research results seem to show that the prevalence and the rate of progression of disseminated neoplasia in bivalves is elevated in seriously polluted environments under increasing anthropogenic pressure from increasing population growth, urbanisation, industrialisation and intensification of agriculture e.g. in the U.S., Europe and Japan (Table 1 and Fig. 1). This might suggest the presence of toxicants and other anthropogenic compounds as carcinogenic agents, although they may act solely as an additional stress. So far a clear-cut etiologic basis has been shown only in one species Unio pictorum facultative feeder (Khudoley and Syrenko 1978), and uniform initialising casual factors applicable to other bivalve species have not yet been identified from the variety of potentially harmful substances present in coastal and estuarine waters in elevated concentrations. A synergistic effect of various factors cannot be excluded either. Health disorders in bivalves are becoming a widespread problem of current world concern, given the elevated pollutant status and sustainable development of coastal waters. This problem has particular relevance since the exploitation and development of coastal systems is intimately related to the socio-economic health of many countries. In this context, the Gulf of Gdansk (southern Baltic Sea) with its specific hydrological conditions (inland and brackish character, restricted connection with the North Sea) and geographical position (dense human population in drainage area, highly developed industry and agriculture that results in serious pollution) may
176 provide a good environment to develop hypotheses on the disease process in bivalves. In the Gulf neoplastic cells were recorded in two of four bivalve species commonly distributed in the Gulf, namely M. balthica and M. arenaria. The two species are infaunal facultative (deposit/suspension) feeders that are able to take up particles from the water column and surficial sediments (Lin and Hines 1994, Strasser 1999). The cancer was not detected in epifaunal obligate suspension-feeders, the mussel M. edulis trossulus and the common cockle C. glaucum. Although the distribution ranges of the bivalve species often overlap, particularly those of M. arenaria and C. glaucum, the cancer cells were found only in the first one, possibly indicating a different risk of carcinogenic changes of facultative and suspension-feeders. Due to cancer-induced disrupted reproduction (Barber 2002), potential genetic heredity of the disease and an increased mortality of the facultative and deposit feeders present an increased danger to populations of facultative feeders, leading in extreme cases to elimination of species. The significance of bivalve-disseminated neoplasia may therefore occur in terms of loss of marine resources, redistribution of gene pools and perturbations in functioning of coastal ecosystems (Elston et al. 1992). The interspecific incidence of the disorder (suspension versus suspension/deposit feeders) might be also indicative of the source of potential causal factors in the environment. Chemical compounds present on the sea bottom, available to deposit feeders but not to filter feeders, appear therefore to act as a carcinogenic agent. The risk of induction of neoplasia in bivalves should then increase with depth, particularly in water basins with a trough geomorphological structure and an enhanced accumulation/sedimentation of pollutant contaminants in deep areas. Indeed, the highest prevalence and most advanced stage of neoplasia progression were observed in the Baltic clam M. balthica, which is the only bivalve species that can occupy habitats of a depth of up to 90 m (Janas and Szaniawska 1996). Deeper regions of the Gulf of Gdansk represent the ultimate sink for effluents e.g. nutrients, metals and organic compounds associated with particles, discharged from the land via run-off and rivers, so the clams are therefore subject to adverse conditions that may induce an increase in sensitivity to stress and make the etiological agent more virulent. Differences in the bivalve’s physiological performance with changes in oxygen conditions and the appearance of contaminants have been well documented in eutrophic stagnant deep-water layers in various marine and estuarine systems (Jahn and Theede 1997, de Zwaan and Babarro 2001). An increased prevalence of neoplasia in the clams was recorded at deeper sites in agreement with the results of the sediment toxicity test (Table 3). The highest percentage contribution of diseased animals occurred at Hel45, which is located on the steep slope of the Hel Peninsula where deepsea waters from an open part of the Baltic encounter water inflow ultimately
177 from the North Sea. The water inflow modifies substantially environmental conditions by increasing salinity and oxygenation and reducing water temperature, so leading to strong thermohaline stratification. The hydrological situation thus changes rapidly over a short time and is unstable. However, when the inflow of more saline and denser waters from the North Sea is restricted, as has been observed at the end of `90s, hypoxia or even anoxia appears in stagnated bottom waters and often hydrogen sulphide occurs in the deeper part of the Gulf. During the recent years a strong inflow of water from the North Sea has been observed. This leads to the exchange of both seas' water masses and an improvement of the environmental conditions, which appears to be a reason for the observed changes in the number of neoplasia. Although no single etiology factor was demonstrated in this study and the cause of the neoplasms remains unknown, the relationship of prevalence of neoplasia in M. balthica to sediment toxicity indicates clearly the existence of carcinogenic agents in the sediments. Elucidation of the specific causes of the tumors requires the quantification of a broader range of potential causal factors in the environment. This statement generates hypotheses for further studies, which can help to elucidate the environmental significance of bivalve disseminated neoplasia and to assess its biological risk at the population and ecosystem levels. Questions relating to causes and mechanisms of disease transmission in bivalves represent a priority challenge, thus further studies should be done in order to distinct the illness degree between mollusc species M. arenaria and M. balthica. Important clues to the etiological basis of the cancer can be revealed through laboratory experiments in which healthy clams will be exposed to controlled doses of specific toxicants (e.g. TBT, cyanobacterial toxins, atrazine) over a limited period of time. A key area is also to understand the effect of disseminated neoplasia on the ecophysiological performance of the bivalves, such as M. balthica, measured as respiration rate, condition index or population mortality estimates. Long-term experiments across generations encompassing a full gonad development cycle and spawning, larvae formation, settlement, and maturation of young individuals under laboratory controlled conditions can address another relevant concern of genetic heredity of the disease. These three topics represent high priorities for future research on bivalve neoplasia.
ACKNOWLEDGEMENTS Department of Analytical Chemistry (headed by Prof. J NamieĞnik), Chemical Faculty, Gdansk University of Technology, Gdansk, Poland is sincerely acknowledged for help in performing sediment toxicity tests and hydrocarbon analyses. We thank Prof. P Rainbow from Darwin Center, The Natural History Museum, London for English editing and two unknown referees for adorable and helpful comments.
178 REFERENCES Albata A Potrykus J Pempkowiak J Porte C 2002 Assessment of organotin pollution along the Polish coast (Baltic Sea) by using mussels and fish as sentinel organisms. Chemosphere 47: 165-171 Alderman DJ van Banning P Perez-Colomer A 1977 Two European oyster (Ostrea edulis) mortalities associated with abnormal haemocytic condition. Aquaculture 10: 335-340 Alonso A Suarez P Alvarez C San Juan F Molist P 2001 Structural study of a possible neoplasia detected in Mytilus galloprovincialis collected from the Ria of Vigo (NW Spain). Dis Aquat Orgg 47: 73-79 Andrulewicz E 1996 The regional characteristic of the GdaĔ a sk Basin. Oceanol Studd 1-2: 1116 Baden SP Loo LO Phil L Rosenberg R 1990 Effect of eutrophication on benthic communities including fish: Swedish west coast. AMBIO 19: 113-122 Balouet G Poder M Cahour A 1982 Hyaline cell proliferation in Ostrea edulis L.: histogenetic and oncologic approach. In: Third International Colloquium on Invertebrate Pathology, UK. pp 268-270 Balouet G Poder M Cahour A Auffret M 1986 Proliferative hemocytic condition in European flat oysters: a 6-years survey. J Invert Patholl 48: 208-215 Barber BJ MacCallum GS Robinson SMC McGladdery S 2002 Occurrence and lack of transmissibility of gonadal neoplasia in softshell clams, Mya arenaria, in Maine (USA) and Atlantic Canada. Aquat Living Resourr 15: 319-326 Bert TM Hesselman DM Arnold WS Moore WS Cruz-Lopez H Marelli DC 1993 High frequency of gonadal neoplasia in a hard clam (Mercenaria spp.) hybrid zone. Mar Bioll 117: 97-104 Bogaczewicz-Adamczak B Klosinska D Zgrundo A 2001 Diatoms as indicators of water pollution in the coastal zone of the Gulf of GdaĔ a sk (southern Baltic Sea). Oceanol Studd 30(3-4): 59-75 Bower SM 1989 The summer mortality syndrome and haemocytic neoplasia in blue mussels (Mytilus edulis) from British Columbia. Canadian Technical Report of Fisheries and Aquatic Sciences No 1703 Bower SM McGladdery SE Price IM 1994 Synopsis of infectious diseases and parasites of commercially exploited shellfish. Annu Rev Fish Dis 4: 1-199 Brousseau DJ 1987 Seasonal aspects of sarcomatous neoplasia in Mya arenaria (soft shell clam) from Long Island Sound. J Invert Pathol 50: 269-276 Brown RS Wolke RE Saila SB 1976 A preliminary report on neoplasia in feral populations of the soft-shell clam, Mya arenaria, prevalence, histopathology and diagnosis. In: Proceedings of the First International Colloquium on Invertebrate Pathology, Kingston, Ontario, Canada. pp 356-361 Brown RS Wolke RE Saila SB Brown CW 1977 Prevalence of neoplasia in 10 New England populations of the soft shell clam (Mya arenaria). Ann NY Acad Sci 298: 522-534 Bryan GW Langston WJ 1992 Bioavailability, accumulation and effects of heavy metals in sediments with special references to United Kingdom estuaries: a review. Environ Pollutt 76: 227-240 Cederwall H Elmgren R 1990 Biological effects of eutrophication in the Baltic Sea, particularly the coastal zone. AMBIO 19: 109-112 Chibani M Zióákowska á M Kijewska A Rokicki J 2001 Pomphorhynchus laevis parasite of flounder Platichthys flesus as a biological indicator of pollution in the Baltic Sea. J Mar Biol Ass UK K 81: 165-166
179 Christensen DJ Farley CA Kern FG 1974 Epizootic neoplasm in the clam Macoma balthica (L.) from Chesapeake Bay. J Natl Cancer Instt 52: 1739-1749. Cooper KR Chang PW 1982 A review of the evidence supporting a viral agent causing a hematopoietic neoplasm in the soft shell clam, m Mya arenaria. Third International Colloquium on Invertebrate Pathology, UK. pp 271-272 Cosson-Mannevy MA Wong CS Cretney WJ 1984 Putative neoplastic disorders in mussels (Mytilus edulis) from southern Vancouver Island waters, British Columbia. J Invert Patholl 44: 151-160 Couch JA 1969 An unusual lesion in the mantle of the American oyster Crassostrea virginica. Natl Cancer Inst Monogrr 31: 557-562 Dungan CF Hamilton RM Hudson KL McCollough CB Reece KS 2002 Two epizootic diseases in Chesapeake Bay commercial clams, Mya arenaria and Tagelus plebeius. Dis Aquat Orgg 50: 67-78 Elston RA Kent ML Drum AS 1988a Progression, lethality and remission of hemic neoplasia in the bay mussel Mytilus edulis. Dis Aquat Orgg 4: 135-142 Elston RA Kent ML Drum AS 1988b Transmission of hemic neoplasia in the bay mussel, Mytilus edulis, using whole cells and cell homogenate. Dev Comp Immunol 12: 719727 Elston RA Moore JD Brooks K 1992 Disseminated neoplasia of bivalve molluscs. Rev Aquat Orgg 6: 405-466 Emmett B 1984 The analysis of summer mortality in cultured mussels, Mytilus edulis, final report. British Columbia Science Council Grant No 58 (RC-7), 74 p Evarsole AG Heffernan PB 1995 Gonadal neoplasia in northern Mercenaria mercenaria (Linneus, 1758) and southern M. compechiensis (Gmelin, 1791) quahogs and their hybrids cultured in South Carolina. J Shellfish Res 14: 33-39 Falandysz J Strandberg L Bergqvist P-A Strandberg B Rappe C 1997 Spatial distribution and bioaccumulation of polychlorinated naphthalenes (PCBs) in mussel and fishes from the Gulf of Gdansk, Baltic Sea. Sci Total Environ 203: 93-104 Falandysz J Strandberg L Strandberg B Rappe C 1998 Polychlorinated naphthalenes in three-spined stickleback Gasterosteus aculeatus from the Gulf of Gdansk. Chemosphere 37: 2473-2487 Farley CA 1969a Probable neoplastic disease of the hematopoietic system in oysters, Crassostrea virginica and C. gigas. Natl Cancer Inst Monogr 31: 541-555 Farley CA 1969b Sarcomatoid proliferative disease in a wild population of the blue mussels (Mytilus edulis). J Nat Canc Instt 43: 509-516 Farley CA 1976a Ultrastructural observations on epizootic neoplasia and lytic virus infection in bivalve molluscs. Prog Exp Tumor Res 20: 283-294 Farley CA 1976b Proliferative disorders in bivalve molluscs. Mar Fish Rev 38: 30-33 Farley CA Banfield WG Kasnic G Jr Foster WS 1972 Oyster herpes-type virus. Science 178: 759-760 Farley CA Otto SV Reinish CL 1986 New occurrence of epizootic sarcoma in Chesapeake Bay soft clams Mya arenaria. Fish Bull USS 84: 851-857 Farley connective tissue cells in molluscs. Bibl Haemato 36: 610-617 Ford SE Barber RD Marks E 1997 Disseminated neoplasia in juvenile oyster Crassostrea virginica and its relationship to the reproductive cycle. Dis Aquat Orgg 28: 73-77 Frierman EM 1976 Occurrence of hematopoietic neoplasms in Virginia oysters (Crassostrea virginica). Mar Fish Revv 38: 34-36 Glasby G P Szefer P 1998 Marine pollution in Gdansk Bay, Puck Bay and the Vistula Lagoon, Poland: an overview. Appl Geochem 212: 49-57 Green M Alderman DJ 1983 Neoplasia in Mytilus edulis L. from United Kingdom waters. Aquaculture 30: 1-10 Hillman RE 1993 Relationship of environmental contaminants to occurrence of neoplasia in Mytilus edulis populations from east and west coast mussel-watch sites. J Shellfish Res 12: 109
180 House ML Kim CH Reno PW 1998 Soft shell clam Mya arenaria with disseminated neoplasia demonstrates reverse transcriptase activity. Dis Aquat Orgg 34: 187-192 Jahn A Theede H 1997 Different degrees of tolerance to hydrogen sulphide in populations of Macoma balthica (Bivalvia, Tellinidae). Mar Ecol Prog Serr 154: 185-196 Janas U Szaniawska A 1996 The influence of hydrogen sulphide on macrofaunal biodiversity in the Gulf of GdaĔsk. Oceanologia 38 (1): 127-142 Kautsky N Evans S 1987 Role of biodeposition by Mytilus edulis in the circulation of matters in a Baltic coastal ecosystem. Mar Ecol Prog Ser 38: 201-212 Khudoley VV Syrenko OA 1978 Tumor induction by N-nitroso compounds in bivalve mollusks Unio pictorum. Cancer Lettt 4: 349-354 Konat J Kowalewska G 2001 Polychlorinated biphenyls (PCBs) in sediments of the southern Baltic Sea-trends and fate. Sci Tot Environ 280: 1-15 Korzeniewski K 1999 Chemical warfare agents dumped in the Baltic Sea. An overview. Oceanol Studd 28: 23-34 Krishnakumar PK Casillas E Snider RG Kagley AN Varanasi U 1999 Environmental contaminants and the prevalence of hemic neoplasia (leukemia) in the common mussel (Mytilus edulis complex) from Puget Sound, Washington, U.S.A. J Invert Patholl 73: 135-146 Landsberg JH 1996 Neoplasia and biotoxin in bivalves: is there a connection? J Shellfish Res 15: 203-230 Leavitt DF Capuzzo J Smolowitz RM Miosky DL Lancaster BA Reinish CL 1990 Hematopoietic neoplasia in Mya arenaria: prevalence and indices of physiological condition. Mar Biol 105: 313-321 Leppakoski E Olenin S 2000 Non-native species and rates of spread: lessons from the brackish Baltic Sea. Biol Invas 2: 151-163 Lin J Hines AH 1994 Effects of suspended food availability on the feeding mode and burial depth of the Baltic clam, Macoma balthica. Oikos 69: 28-36 Lowe DM Moore MN 1978 Cytology and quantitative cytochemistry of a proliferative atypical hemocytic condition in Mytilus edulis (Bivalvia, Mollusca). J Natl Cancer Instt 60: 1455-1459 Mix MC 1975a Neoplastic disease in Yaquina Bay bivalve mollusks. Proc 13th Ann Hanford Biol Symp pp 369-386 Mix MC 1975b Proliferative characteristics of atypical cells in native oysters (Ostrea lurida) from Yaquina Bay, Oregon. J Invert Patholl 26: 289-298 Mix MC 1976 A general model for leucocyte cell renewal in bivalve molluscs. Mar Fish Rev 1212: 37-49 Mix MC 1983 Haemic neoplasms of bay mussels, Mytilus edulis, from Oregon: Occurrence, prevalence, seasonality and histopathological progression. J Fish Dis 6: 239-248 Mix MC Breese WP 1980 A cellular proliferative disorder in oysters (Ostrea chilenis) from Chiloe, Chile, South America. J Invert Patholl 36: 123-124 Mix MC Pribble HJ Riley RT Tomasovic SP 1977 Neoplastic disease in bivalve molluscs from Oregon estuaries with emphasis on research on proliferative disorders in Yaquina Bay oysters. Ann NY Acad Sci 298: 356-373 Naes K Knutzen J Berglind L 1995 Occurrence of PAH in marine organisms and sediments from smelter discharge in Norway. Sci Tot Environ 163: 93-106 Nascimento IA Smith DH Kern F III Pereira SA 1986 Pathological findings in Crassostrea rhizophorae from Todos os Santos Bay, Bahia, Brazil. J Invert Path 47: 340-349 Neff JM Hillman RE Carr RS Buhl RL Lahej JI 1987 Histopathologic and biochemical responses in Arctic marine bivalve molluscs exposed to experimentally spilled oil. Arctic 40 (Suppl 1): 220-229 Newman MS 1972 An oyster neoplasm of apparent mesenchymal origin. J Natl Cancer Inst 48: 237-243
181 Nowacki J Kowalewski M Krezel A 1993 Long-term trends of concentration of nitrates, phosphates and oxygen in the Gulf of Gdansk. Stud Mat Oceanoll 64: 123-130 Oprandy JJ Chang PW 1983 5-Bromodeoxyuridine induction of hematopoietic neoplasia and retrovirus activation in the soft shell clam, Mya arenaria. J Invert Patholl 42: 196-206 Oprandy JJ Chang PW Pronovost AD Cooper KR Brown RS Yates VJ 1981 Isolation of viral agent causing hematopoietic neoplasia in the soft shell clam, Mya arenaria. J Invert Patholl 38: 45-51 Pauley GB 1969 A critical review of neoplasia and tumor-like lesions in mollusks. Nat Canc Inst Monogrr 31: 509-539 Pekkarinen M 1993 Neoplastic disease in the Baltic Macoma balthica (Bivalvia) off the Finnish coast. J Invert Patholl 61: 138-146 Peters EC Yevich JC Harshbarger JC Zaroogian GE 1994 Comparative histopathology of gonadal neoplasms in marine bivalve molluscs. Dis Aquat Orgg 20: 59-76 PliĔski M Musiaá A Ostrowski B 1998 Blue-green algae blooms in the Gulf of GdaĔsk and surrounding area. Oceanol Studd 1: 55-60 Rasmussen L 1986 Cellular reactions in mollusks with special reference to chemical carcinogens and tumors in natural populations of bivalve mollusks. In: Proceedings, Fourth International Colloquium on Invertebrate Pathology, UK, pp 441-443 Reno PW House M Illingworth A 1994 Flow cytometric and chromosome analysis of softshell clams, Mya arenaria, with disseminated neoplasia. J Invert Path 64: 163172 Rodriguez H Soto M Arias C Estevez J 1997 A case of disseminated haemic sarcoma in Cerastoderma (=Cardium) glaucum Poiret 1789 collected from Ria de Vigo (NW Spain). Bull Eur Assoc Fish Pathol 17: 88-90 Roy D Colerangle JB Kamleshwar PS 1998 Is exposure to environmental or industrial endocrine disrupting estrogen–like chemicals able to cause genomic instability? Front Biosci 3: 913-921 Smolarz K Woáowicz M Thiriot-Quievreux C 2003 Argylophylic nuclear organiser regions (AgNORs) in interphases and metaphases of normal and neoplastic gill cells of Macoma balthica (Bivalvia, Tellinidae) from the Gulf of Gdansk (Baltic Sea). Dis Aquat Org 56: 269-274 Sokoáowski A Woáowicz M Hummel H Bogaards R 1999 Physiological responses of Macoma balthica to copper pollution in the Baltic. Oceanol Acta 22(3): 431-439 Strasser M 1999 Mya arenaria-an ancient invader of the North Sea. Helgoländer Meeresunters 52: 309-324 Szefer P 2002 Metals, Metalloids and Radionuclides in the Baltic Ecosystem. Elsevier, Amsterdam, 764 p Szefer P Glasby GP Pempkowiak J Kaliszan R 1995 Extraction studies of heavy-metal pollutants in surficial sediments from the southern Baltic Sea off Poland. Chem Geol 120:111-126 Szefer P Szefer K Glasby GP Pempkowiak J Kaliszan R 1996 Heavy metal pollution in surficial marine sediments from the Gulf of GdaĔ a sk and the Southern Baltic Sea off Poland. J Environ Sci Health 31A: 2723-2754 Thiriot-Quiévreux C 2002 Review of the literature on bivalve cytogenetics in the last ten years. Cah Biol Marr 43: 17-26 Thiriot-Quiévreux C Woáowicz M 1996 Etude caryologique d'une néoplasie branchiale chez Macoma balthica (Mollusca, Bivalvia). C R Acad Sci 319: 887-892 Thiriot-Quiévreux C Woáowicz M 2001 Chromosomal study of spatial variation of the prevalence of a gill neoplasia in Macoma balthica from the Gulf of GdaĔsk (Baltic Sea). Ophelia 54: 75-81 Twomey E Mulcahy MF 1988 Transmission of a sarcoma in the cockle Cerastoderma edule (Bivalvia, Mollusca) using transplants. Dev Comp Immunol 12: 195-200 Van Beneden RJ 1994 Molecular analysis of bivalves tumors: models for environmental /genetic interactions. Environl Health Persp 102 (12): 81-83
182 Van Beneden RJ Gardner GR Blake NJ Blair DG 1993 Implication of the presence of transforming genes in gonad tumors in two bivalve mollusc species. J Cancer Res 53: 2976-2979 Villalba A Carballal MJ Lopez C 2001 Disseminated neoplasia and large foci indicating heavy haemocytic infiltrations in cockles Cerastoderma edule from Galicia (NW Spain). Dis Aquat Orgg 46: 213-216 Villalba A Mourelle SG Carballal MJ Lopez C 1997 Symbionts and diseases of farmed mussels Mytilus galloprovincialis throughout the culture process in the Rias of Galicia (NW Spain). Dis Aquat Orgg 31: 127-139 Villalba A Peters CE Lopez MC Caraballal MJ 1995 Disseminated sarcoma in the clam Ruditapes decussatus in Galicia (NW Spain). J Invert Patholl 65: 76-78 Wolf PH 1977 Diseases and parasites in Australian commercial shellfish. Haliotis 8: 75-83 Woáowicz M Thiriot-Quievreux C Smolarz K Sokoáowski A 2000 Baltic bivalve neoplasia as an effect of the environmental pollution? Proc of 10th Symp on Environmental Toxicology and Risk Assessment: Science, Policy and Standardization-Implications for Environmental Decisions. Toronto, Canada Yevich PP Barry MM 1969 Ovarian tumors in the quahog Mercenaria mercenaria. J Invertebr Patholl 14: 267 Yevich PP Barszcz CA 1977 Neoplasia in soft-shell clams (Mya arenaria), collected from oil-impacted sites. Ann NY Acad Sci 298: 409-426 de Zwaan A Babarro JMF 2001 Studies on the causes of mortality of the estuarine bivalve Macoma balthica under conditions of (near) anoxia. Mar Bioll 138: 1021-1028
BIVALVES AS BIOFILTERS AND VALUABLE BYPRODUCTS IN LAND-BASED AQUACULTURE SYSTEMS
Muki Shpigel The National Center for Mariculture, Israel Oceanographic and Limnological Research, P.O. Box 1212, Eilat 88112, Israel Abstract: Effluents from land-based mariculture have a deleterious environmental impact by enriching littoral waters with particulate and dissolved nutrients. The use of suspension-feeders (e.g. bivalves) in biofiltration systems provides an inexpensive option for the biological removal of these nutrients from effluent water. Land-based facilities are safer because the quality of the incoming water can be controlled, and integrated polyculture systems can save resources, diversify the product, allow intensification, and are environmentally friendly, thus offering a valid alternative to open sea monoculture. Several approaches have been developed to reach this goal. These approaches are based on the use of algae and invertebrates as biofilters, which remove the dissolved and particulate nutrients from fishpond effluents. Sunlightdependent assimilation turns excess nutrients into microalgae biomass, which is then consumed with other organic particulate matter by the invertebrates. These systems consist of three units: a) a fish or shrimp culture unit, b) an earthen pond that serves as both a microalgal production and a bivalve culture unit, and c) a seaweed biofilter unit. Microalgae, especially benthic diatoms, are responsible for most of the nutrient uptake. The main cultured herbivores are filterfeeding bivalves, which grow in the earthen pond. The seaweed biofilter strips the remaining dissolved nutrients from the effluents. The harvested yields (fish, bivalve and seaweed) contain 63% of the nitrogen introduced into the system, particulate matter (faeces and uneaten feed) contain 33%, and suspended and dissolved outflow, approximately 4% of the nitrogen budget. More complex systems require highly skilled operators and more sophisticated technology, but they provide increased product diversity. Depending on practical factors such as market prices, quality and cost of labor, seed and fingerlings supply, environmental conditions, market size and property cost, the operational protocol can be shifted between the three product types (fish, algae and invertebrates) to maximize profits. Keywords: land-based, mariculture, effluent, integrated system, suspension-feeders, biofiltration
183 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 183–197. ©2005 Springer. Printed in the Netherlands.
184 INTRODUCTION World production of filter-feeding bivalves has increased dramatically in the last decade. According to FAO, fisheries and aquaculture production increased fourfold over the last 30 years, reaching production of almost 12 million MT in year 2000 (Figure 1). This growth is primarily due to aquaculture (Anderson 2000). As the global population continues to grow, demand and production of bivalves will continue to grow as well. The main bivalves from the fishery and aquaculture industries are mussels, oysters, scallops and clams (Table 1). Oysters and clams are the leading products, showing the most significant rise in recent years. Bivalve culture is attractive to growers, probably by virtue of its low production cost (no feed input).
Figure 1: World bivalve production (FAO data).
China is the major producer, contributing 78% of the world bivalve production (Figure 2). Only mussel production rates remain relatively unaffected by the input from China. Molluscan bivalve aquaculture is a “green” industry. Shellfish feeds on the base of food chain. They are highly efficient filter feeders that directly remove particulate matter, reduce turbidity, and remove nitrogen from the water. In an area with heavy loading of nitrogen to the environment by agriculture or aquaculture, shellfish activity can recirculate this nitrogen excess into a harvestable commodity. For example: 10,000 oysters contain 13.6 kg of nitrogen and 1.4 kg of phosphate. By harvesting these valuable oysters we are able to remove 100 kg of nitrogen per year from the
185 environment (Shumway et al. 2003). Shellfish culture is considered by the US Organic Standards Board’s (NOSB) as an “organic aquaculture” which calls for “an ecological production management system that promotes and enhances biodiversity, biological cycles and soil biological activities” (Shumway et al. 2003). Table 1. Molluscs cultured in the world Common name Scientific name Family: Mytilidae Mediterranean mussel Mytilus galloprovincialis Blue mussel Mytilus edulis California sea-mussel Mytilus californianus Green mussel Perna canaliculus; P .viridis Family: Ostreidaea Mangrove oyster Sydney rock oyster American oyster Pacific oyster European flat oyster Chilean flat oyster
Crassostrea rhizophorae Saccostrea commercialis Crassostrea virginica Crassostrea gigas Ostrea edulis Ostrea chilensis
Common name Scientific name Family: Pectinidae Japanese scallop Patinopecten yessoensis Bay scallop King scallop
Argopecten irradians Pecten maximus
Sea scallop
Placopecten magellanicus
Family: Veneridae Manila clam
Ruditapes philippinarum
Hard clam
Mercenaria mercenaria
Butter clam
Saxidomus gigantea Other Families
Ark shell Cockle Ocean quahog Surf clam Geoduck Soft shell clam
Anadara spp. Cerastoderma edule Arctica islandica Spisula solidissima Panopea abrupta Mya arenaria
Metric tonnes (x 1000)
4,500 4,000
Including China Excluding China
3,500 3,000 2,500 2,000 1,500 1,000 500 0 Clams
Oysters
Mussels
Scallops
Figure 2. World aquaculture bivalve production with and without including China (for year 2000, FAO data).
186 The decision to grow a particular bivalve species in a commercial aquaculture system requires careful evaluation of a wide range of variables biological, technological, marketing-related and economical. An attractive species for culture should have good biological traits - fast growth rate, low food conversion ratio (FCR), resistance to pests, and tolerance to a wide range of environmental conditions; the technology for its reproduction and culture should be straightforward and user-friendly for the growers; it should be in demand on the market with respect to appearance, taste, smell, texture, processing considerations and market behaviour; profitability depends on yield per unit of area, cost effective seafood, grow-out time, harvest frequency, farm-gate price, price trends and the cost of waste treatment. Bivalves can be cultured by sea ranching or in land-based facilities. Sea ranching is done in the open sea, natural tidal ponds, bays and lagoons. In the sea, bivalves utilise natural phytoplankton and their excretions (dissolved N and faeces) are removed by the currents.Culture can be done on the sea bottom (intertidal and subtidal culture) or suspended in the water. Bottom culture methods include rack, tray, string, or stick culture. Suspended culture methods include raft, long-line, string, floating string, or tray culture (Quayle and Newkirk 1989). These methodologies are generally applied to extensive or semi-intensive cultures. The seed animals can be obtained from the wild or from hatcheries. In the open sea the bivalves are vulnerable to weather conditions, predation, red tide and poaching. Since bivalves are generally cultured close to shore, they are additionally subjected to urban pollution, from microbes to chemicals (e.g. heavy metals), sometimes wiping out the entire local bivalve industry. In British Columbia, Canada, for example, habitat closures doubled in less than 30 years due to human-induced pollution (BC Ministry of Agriculture 2003). As filterfeeders, bivalves can accumulate high concentrations of toxic and pathogenic material. Such sanitary problems can affect the economics of bivalve culture. In land-based systems, bivalves can be cultured in artificial tidal pools, ponds, tanks or indoor hatcheries and nurseries. Land-based facilities are safer, since the quality of the incoming water can be controlled. However, because of the high costs of construction, trained technicians, water pumping, food (microalgae) supply and waste control, the profitability of bivalve monoculture in such systems is doubtful. Land-based integrated polyculture systems with fish or shrimp hold more economic promise because they save resources, diversify the product, allow intensification and optimisation, and are environmentally friendly. This paper summarises the state of the art research and development in the use of bivalves as biofilters and as a safe, valuable byproduct in land-based integrated polyculture systems, making them a valid, efficient alternative to open sea monoculture.
187 BIVALVES IN LAND-BASED POLYCULTURE SYSTEMS Protein is the most expensive component in fish feed. Breakdown of protein by the cultured animals and microbes in the water generates a high level of nitrogen waste in the effluents. Fish and shrimp assimilate only 2030% of the nitrogen in their prepared feed. The rest is excreted into the water, mainly as dissolved ammonia and faeces. Uneaten feed also becomes a significant source of nitrogen. To be sustainable, expansion of aquaculture necessitates an economical solution to this waste problem. Because of the relatively low nutrient concentration in aquaculture effluents, biofiltration approaches, rather than standard industrial wastewater treatment, have been the treatments of choice in aquaculture (van Rijn 1996). Bacteria dissimilate the waste carbon and nitrogen into gasses, while plants assimilate them into biomass. Bacterial biofilter technologies can be technically effective and allow significant water recirculation, but cost and complexity have prevented them from becoming widespread. Bacterial biofilter technologies are suitable for relatively small intensive land-based culture of lucrative organisms. Wastewater biofiltration by photosynthetic nutrient assimilation into vegetation, seaweed and microalgae is a more promising method of water treatment. The added cost and complexity involved with the assimilative biofiltration approach can be profitably recovered by the sale of the biofilter organisms. In this way the waste nutrients are turned into a resource. Additional benefits of plant biofilters include higher market yield per feed use, diversification of the farm products, additional jobs, and, most importantly, reduced environmental pollution. While the market for seaweeds is relatively large, there is little use for mixed populations of microalgae except in shellfish polyculture. Polyculture can be done in a single-pond system or in a system of linked ponds (integrated system). Single-pond polyculture is not suitable for intensification because of the different cultured organisms’ conflicting requirements. In a typical integrated system, the different species are cultured in separate spatial entities, permitting intensification and optimisation of production. Seawater is pumped from the sea to ponds containing fish or shrimp fed a pelleted diet. The effluent water from these ponds, rich in organic matter, can be stripped of nutrients in ponds containing microalgae or macroalgae. Microalgae can be utilized by filterfeeders such as the brine shrimp Artemia or bivalves, while macroalgae can be sold for human consumption, the phyco-colloid industry and other uses, or utilised on site to grow abalone and sea urchins. Detritus can be utilised by detritivores such as mullet or sea cucumbers (Figure 3).
188 Food Filter feeders, Bivalves
Shrimps
Artemia
Microalgae
Detritivores
Fish
Macroalgae Abalone
Herbivores
Sea urchins
Figure 3. Schematic pathway of flow options in integrated system.
Table 2: Marine integrated systems in the world.
Oysters, shrimp Fish, bivalves, seaweed Oysters, seaweed Fish, abalone, seaweed Fish, oysters Fish, oysters, sea urchins, seaweed Shrimp, oysters, seaweed
China Israel China Israel France Chile Australia
Wang 1990 Shpigel et al. 1993b Qian et al. 1996 Shpigel et al. 1996 Lefebvre et al. 2000 Chow et al. 2001 Jones et al. 2001
In recent years, several research facilities and enterprises have developed integrated culture systems, most of them still in experimental stages. They have usually included two or three species. In most of the studies, seaweed and microalgae are used as biofilters for the dissolved nutrients (Table 2).
189 The first practical and quantitative integrated land-based cultures of marine fish and shellfish, with phytoplankton as biofilter and shellfish food, was described in Israel by Hughes-Games (1977) and Gordin et al. (1981). A semi-intensive seabream and grey mullet pond system with silicate-rich green water, located on the coast of the Gulf of Eilat (Red Sea), supported dense populations of diatoms, excellent food for oysters (Krom et al. 1989, Erez et al. 1990). Later on, the development of a practical intensive culture of bivalves in phytoplankton-rich effluents was described in a series of articles (Shpigel and Friedman 1990, Shpigel and Blaylock 1991, Shpigel et al. 1993a 1993b, Neori and Shpigel 1999, Neori et al. 2001). Lefebvre et al. (2000) found that detritus waste from intensive fish farming can contribute to the growth of bivalves and reduce particulate matter in the water. Jones et al. (2001) found that the Sydney rock oyster Saccostrea commercialis reduced the concentration of suspended particulates including algae, bacteria and inorganic particles in an integrated system of fish, oysters and the seaweed Gracilaria edulis. Today, the integrated mariculture of finfish-phytoplankton-shellfish is a developed discipline, with a significant amount of quantitative information regarding performance, facility size, yields of the different organisms, expected degree of pollution, and anticipated income. In our fish-microalgae-bivalves-seaweed integrated system, the dissolved nutrients from fish effluents coupled with high solar radiation result in high phytoplankton production, which supports high rates of growth and production of the oyster Crassostrea gigas (0.7% d-1; stocking densities of 20-40 kg m-3, in tanks) and the clam Tapes philippinarum (0.6% d-1; stocking densities of 5-9 kg m-2, on the bottom of the sedimentation pond) (Figure 4, Shpigel et al. 1993b, Shpigel and Neori 1996). In the total nitrogen budget of this system, less than 5% of the fish feed-N input is discharged into the sea. The fish assimilate 21%, the bivalves 15% and the seaweed 22%. About 32% of the N-budget is faeces, pseudofaeces and uneaten feed that sink to the bottom. As a result, 60% of the nitrogen budget of the farm is expected to reach commercial products, about three times more than in present day fish cage farms (Figure 5). In such a system, microalgae are able to assimilate 1-3 g N m-2d-1, and bivalves assimilate 0.5-1g N kg-1d-1. A daily ratio of 10 tonnes of feed supports a standing stock of 1000 tonnes of fish. This amount of food is equivalent to 640 kg of nitrogen. The fish assimilate 160 kg (25%) of nitrogen. Ninety-six kg (15%) of the nitrogen is comprised of particulate nitrogen, and 380 kg (60%) is comprised of dissolved nitrogen. Twenty hectares of microalgae are required to remove the ammonia from 1000 tonnes of fish. Annual production will be 1600 tonnes of fish and 1000 tonnes of oysters. The successful running of such an integrated system requires maintaining a careful balance between nutrient production by the main
190 organism, and nutrient uptake capacity of algae and shellfish. We found that a minimum of three fishponds is required.
14
18
13
16
12
14
11
12
10
10
9
8
8
6
7
4
6
2
5
0
Day:
Condition Index
Whole Weight (g)
Tapes philippinarum 20
4
0
30
60
90 120 150 180 210 240 270 300 330 360 390 420 450 480
Crassostrea gigas 10 9
60
8 50
7 6
40
5 30
4 3
20
2 10
Day:
1
0
Condition Index
Whole Weight (g)
70
0 0
30
60
90 120 150 180 210 240 270 300 330 360 390 420 450 480
Figure 4: Average whole weight (dark line) per individual and condition index (pale line) of Crassostrea gigas and Tapes philippinarum measured biweekly for more than a year (after Shpigel et al. 1993b).
Three units of the integrated system were compared (Figure 6). The first unit recirculated water between a single PVC-lined fishpond and an oyster tank. In the second unit, the outflow water of all three PVC-lined fishponds drained to a common reservoir, and from there passed to an oyster tank. In the third unit, the oysters were grown in a tank using water from a sandy-bottom sedimentation pond into which water from the three fishponds flowed with a 24-hour retention time. Oysters using sedimentation pond water grew significantly faster than those grown in water from the PVC-lined ponds and the C unit (Figure 7). When comparing the abiotic and biotic parameters in the three units (Table 3), most of the parameters were similar with the exception of algal concentration, particulate organic matter (POM), and particulate organic nitrogen (PON), which were higher in the PVC fishpond, and specific growth rate (SGR) which was higher in the sedimentation pond.
191 FISH FOOD
BIODEPOSITS
SEDIMENTATION PONDS
PN 2.0% DN 25% TOTAL 32% PN 1.8% DN 2.3%
BIVALVE FILTRATION UNIT
BIODEPOSITS DN
6%
FISH
21%
BIVALVES
11%
BIVALVE POLISHING
5%
SEAWEED
22%
TOTAL
63%
BIODEPOSITS DN
PN 18.5% DN 18.5%
16%
(% NITROGEN)
FISH PONDS PN 54% DN 10%
10%
YIELD
SEA WATER
(100% NITROGEN)
BIODEPOSITS DN
SEAWEED UNIT
SEDIMENTATION DENITRIFICATION
SEA WATER
Figure 5. Flow diagram of nitrogen in the proposed integrated system. All the numbers are percentages of the N introduced in the fish feed (PN = particulate nitrogen; DN = dissolved nitrogen; Biodeposits = faeces and uneaten food) (after Shpigel et al. 1993b).
SEAW ATER INFLOW FISH POND
FISH POND
FISH POND
OYSTER TANK
C UNIT
OYSTER TANK
SEDIMENTATION POND
OYSTER TANK
OYSTER REACTORS MACROALGAE UNIT
TO SEA
Figure 6. Schematic diagram of the fish-bivalve integrated system (after Shpigel et al. 1993a).
192
Sed pond
PVC pond
14
C unit
WET WEIGHT (g)
12 10 8 6 4 2 0 0
10
20
30
40
50
60
DAYS
Figure 7. Growth rates of Crassostrea gigas in the sedimentation pond, PVC-lined pond, and C Unit. (after Shpigel et al. 1993a).
Table 3. Summary of the abiotic and biotic parameters in three units of the integrated system during the experiment (+SD). PIM = particulate inorganic matter, POM = particulate organic matter, PON = particulate organic nitrogen, C/N = carbon/nitrogen, SGR = specific growth rate (after Shpigel et al. 1993). Parameter
PVC Pond
Temperature (oC) PH Oxygen (mg/L) Ammonia (P mole) Algae conc. (cells/ml) PIM (mg/L) POM (mg/L) PON (mg/L) C/N ratio (Algae) SFG growth rate (%/d) Condition index
12.5-18.2 7.8-8.7 6.8-10.1 5.95 0.2-60 x 105 59r15 134r50 2.78r0.4 6.11r0.84 0.32 9.14r1.1
Sedimentation Pond 13.0-`7.0 7.9-8.4 7.5-10.1 5.60 2.0-4.2 x 105 65r18 61r11 1.01r0.2 7.05r0.65 1.56 12.34r1.8
Unit C 12.0-18.2 7.8-8.4 6.8-9.9 5.87 0.15-4.1 x 105 59r12 70r7 1.06r0.4 7.04r0.1 7.24 11.21r2.1
Algal population in the PVC system suffered periodic blooms and crashes, usually dominated by planktonic microalgae such as Chlorella, Tetraselmis, Chaetoceros and Piramimonas. The periodic phytoplankton crashes involved vigorous grazing by heterotrophic microflagellates and microciliates. Algal populations in the sedimentation pond were relatively constant
193 and consisted of two major groups: species that developed in the pond, mostly benthic diatoms such as Navicula, Amphora a and Nichiaa which are an excellent diet for the bivalves; and mixed species, mainly planktonic microalgae that developed in the fishponds. It was found that planktonic algae comprised 67% of the microplankton in the PVC-lined pond, and the composition of the oysters' stomach contents reflected these proportions. In the sedimentation pond, 67% of the microplankton consisted of benthic diatoms and these comprised 80% of the oysters’ stomach contents (Figure 8).
SEDIMENTATION POND
PVC-LINED POND 16%
12%
67%
67%
21%
SEDIMENTATION POND OYSTERS’ STOMACH CONTENTS 10% % 10%
PVC-LINED POND OYSTERS’ STOMACH CONTENTS
80%
63%
21% 16%
Microflagellates
Benthic diatoms
Planktonic algae
Figure 8: Proportion of benthic diatoms, planktonic algae and microflagellates in the sedimentation and PVC-lined ponds and stomach contents of Crassostrea gigas (after Shpigel et al. 1993a).
In conclusion, three factors are responsible for the superior growth in the earthen pond: (1) highly diverse mixed diet, (2) additional nutritious food from benthic diatoms, and (3) a stable algal concentration. BIVALVES IN POLISHING REACTORS
194 The integrated mariculture concept developed at NCM calls for the sedimentation pond water to drain into polishing reactors with bivalves to remove particulate matter and seaweed to removed dissolved nutrients on its way back into the sea. When we compare bivalve performance, we can see a positive correlation between alga flux and oyster growth, and a negative correlation between alga flux and filtration efficiency. This means that we need two management procedures: one to culture the bivalves in the sedimentation pond water, and the other to function as a polishing reactor. Shpigel et al. (1997) studied the concept of mechanical biofiltration for removing particulate matter (PM) from fishpond effluents prior to discharge. The treatment process was based on the different and complementary biofiltration by two species of marine bivalves (C. gigas and T. philippinarum), as well as by mechanical sedimentation. The filtration efficiency of the bivalves was assessed in two reactor designs with different flow patterns, a plug flow reactor (PFR) and a continuously stirred flow reactor (CSFR; Figure 9). In the PFR, water flow is in plug form. The filtration activity of the bivalves is expected to vary as a result of the change in PM concentration along the longitudinal axis of the reactor. In the CSFR the water is thoroughly mixed. In an ideal CSFR, the PM is continuously redistributed and its concentration is uniform. These reactor types have been successfully used for treating large volumes of wastewater (Metcalf and Eddy 1983). The rationale for testing these types of reactors is based on the fact that the filtration rates and efficiencies of bivalves are species-specific and vary with flow rate and particle concentration. Removal efficiency was tested in two ways: by turbidity measurements for all the PM in the water, and chlorophyll-a reduction for the phytoplankton in the water. The control represents the physical sedimentation of the particles in both reactor types, in the presence of bivalve shells. No significant differences were observed between the two types of reactors. PM was reduced by 25% and chlorophyll-a by 11-15% in the presence of bivalve shells (Table 4). The PFR was significantly more efficient for both species of bivalves when these were exposed to high levels of PM and chlorophyll. When both species were exposed to low concentrations of PM and chlorophyll, the PFR still showed better removal efficiency than the CFSR (Table 4). Best results can be anticipated when the integrated system includes two bivalve units: a highly productive earthen sedimentation pond and a number of polishing reactors. Two different units are necessary, as the conditions that maximize bivalve yield differ from that necessary for high filtration efficiency. Although considerable information is already available for bivalve production in land- based integrated systems; there are still some bottlenecks that need to be worked through. Key issues for further studies are:
195 minimizing heat loss or gain downstream the integrated system by using greenhouses and working in modular systems; regeneration of the biodeposit materials; disease control of the cultured organisms; sanitary aspects of the bivalves for the consumer; and maintenance of a constant balance between nutrient production by the main organism and nutrient uptake capacity of algae and shellfish.
OUTLET
INLET
bivalves
bivalves
bivalves
bivalves
View from above
CFSR INLET bivalves
bivalves
bivalves
bivalves
bivalves bivalves bivalves Cross section view
Figure 9. Schematic diagram of the Plug Flow Reactor (PFR) and Continuous Stirred Flow Reactor (CSFR) containing stacks of trays with bivalves. Arrows indicate direction of flow (after Shpigel et al. 1997).
The challenge for the future is to produce a large quantity of aquaculture products that are cost-effective for producers, at a reasonable price for the consumers, and ecologically sustainable. Bivalve culture in land-based integrated systems can provide a valuable, high quality sustainable product while helping reduce pollution returning to the sea. Such mariculture can further help provide jobs and socio-economic development for the community.
196 Table 4. Comparison of PM removal (nephelometric turbidly units = NTU) and chlorophyll-a; mean+SD) by T. philippinarum and C. gigas (after Shpigel et al. 1997).
Inlet
Outlet
Initial concentration % removal Turbidity Chl-a Turbidity Chl-a (µg l-1) (NTU) (µg l-1) (NTU) PFR CSFR
61 + 2.4 61 + 2.4
27 + 2.5 27 + 2.5
25 + 2.5 11 + 1.5 23 + 1.9 15 + 1.3
T. philippinarum
PFR CSFR PFR C. gigas CSFR
48 + 2.3 48 + 2.3 48 + 2.3 48 + 2.3
35 + 2.5 35 + 2.5 35 + 2.5 35 + 2.5
97 + 2.7 84 + 1.4 91 + 3.7 79 + 4.6
97 + 3.8 87 + 1.4 93 + 4.5 69 + 5.6
T. philippinarum
20.2 + 1.8 20.2 + 1.8 15.1 + 1.3 15.1 + 1.3
11 + 1.7 11 + 1.7 11 + 1.7 11 + 1.7
88 + 3.7 69 + 4.1 83 + 3.7 75 + 3.1
83 + 3.3 57 + 4.1 86 + 4.2 79 + 3.5
Control (bivalves)
PFR CSFR PFR C. gigas CSFR
REFERENCES BC Ministry of Agriculture, Food & Fisheries 2003 The 2002 British Columbia Seafood Industry Year in Review. Ministry of Agriculture, Food and Fisheries, Victoria, BC. June 2002. Chow F Macchiavello J Santa Cruz S Fonck E Olivares J 1994 Utilization of Gracilaria chilensis (Rhodophpyta: Gracilariaceae) as a biofilter in the depuration of effluents from tank cultures of fish, oysters, and sea urchins. World Aquacult Soc 32(2): 215220 Erez J Krom MD Neuwirth T 1990 Daily oxygen variations in marine fish ponds, Elat, Israel. Aquaculture 84: 289-305 Goldman JC Tenore RK Ryther HJ Corwin N 1974 Inorganic nitrogen removal in a combined tertiary treatment-marine aquaculture system. I. Removal efficiencies. Water Res 8: 45-54 Gordin H Motzkin F Hughes-Games WL Porter C 1981 Seawater mariculture pond - an integrated system. Euro Maricult Soc (special publication) 6: 1-13 Hughes-Games WL 1977 Growing the Japanese oyster (Crassostrea gigas) in subtropical seawater fishpond: 1. growth rate, survival and quality index. Aquaculture 7: 225229 Huguenin JH 1976 An examination of problems and potentials for future large-scale intensive seaweed culture systems. Aquaculture 9: 313-342 Jones AB Dennison WC Preston NP 2001 Integrated mariculture of shrimp effluent by sedimentation, oyster filtration and macroalgal absorption: a laboratory scale study.
197 Aquaculture 193: 155-178 Krom MD Erez J Porter CB Ellner S 1989 Phytoplankton nutrient uptake dynamics in earthen marine fishponds under winter and summer conditions. Aquaculture 76: 237253 Lefebvre S Barille L Clerc M 2000 Pacific oyster (Crassostrea gigas) feeding responses to a fish-farm effluent. Aquaculture 187: 185-198 Metcalf L Eddy HP 1983 Wastewater Engineering: Treatment, Disposal, Re-use. Tata McGraw-Hill, New Delhi, 919 pp. Neori A Shpigel M 1999 Algae treat effluents and feed invertebrates in sustainable integrated mariculture. World Aquaculture 30: 46-49, 51 Neori A Shpigel M Scharfstein B 2001 Land-based low-pollution integrated mariculture of fish, seaweed and herbivores: principles of development, design, operation and economics. Aquaculture Europe 2001 Book of Abstracts. Eur Aquac Soc Spec Publ 29: 190-191 Qian P Wu CY Wu M Xie YK 1996 Integrated cultivation of the red alga Kappaphycus alvarezii and the pearl oyster Pinctada martensi. Aquaculture 147: 21-35 Quayle DB and Newkirk GF 1989 Farming Bivalve Molluscs: Methods for Study and Development. World Aquaculture Society, International Development Research Centre, Canada. 294 pp. Ryther JH Goldman JC Gifford JE Huguenin JE Wing AS Clarner JP Williams LD Lapointe BE 1975 Physical models of integrated waste- recycling marine polyculture systems. Aquaculture 5: 163-177 Shpigel M Blaylock RA 1991 The Pacific oyster, Crassostrea gigas, as an biological filter for a marine fish aquaculture pond. Aquaculture 92: 187-197 Shpigel M Fridman R 1990 Propagation of the Manila clam Tapes semidecussatus in the effluent of fish aquaculture ponds in Eilat, Israel. Aquaculture 90(2): 113-122 Shpigel M Gasith A Kimmel E 1997 A bio-mechanical filter for treating fish-pond effluents. Aquaculture 152: 103-117 Shpigel M Lee J Soohoo B Fridman R Gordin H 1993a Use of effluent water from fishponds as a food source for the Pacific oyster, Crassostrea gigas Thunberg. Aquaculture and Fisheries Managementt 24: 529-543 Shpigel M Neori A 1996 The Integrated Culture of Seaweed, Abalone, Fish and Clams in Modular Intensive Land-based Systems: I. Proportions of Size and Projected Revenues. Aquacult Engg 15(5): 313-326 Shpigel M Neori A Popper DM Gordin H 1993b A proposed model for "environmentally clean" land-based culture of fish, bivalves and seaweeds. Aquaculture 117: 115-128 Shumway SE Davis C Downey R Karney R Kraeuter J Parsons J Rheault R Wikfors G 2003 Shellfish aquaculture – In praise of sustainable economies and environments. World Aquaculture 34(4): 15-18. Tenore KR 1976 Food chain dynamics of abalone in a polyculture system. Aquaculture 8: 23-27 van Rijn J 1996 The potential for integrated biological treatment systems in recirculating fish culture - a review. Aquaculture 139: 181-201 Wang JK Lam CY Jacob GS 1990 Preliminary investigation of an oyster-shrimp joint production system. Trans Am Soc Ag Engg 33(3): 975-980
SIGNIFICANCE OF SUSPENSION-FEEDER SYSTEMS ON DIFFERENT SPATIAL SCALES Harald Asmus and Ragnhild M. Asmus Alfred-Wegener-Institut für Polar- und Meeresforschung, Wattenmeerstation Sylt, 25992 List, Sylt, Germany Abstract: The role of suspension–feeders varies according to the spatial and temporal scales at which assessments are made. In this paper, biodiversity, productivity, and filtration capacity are used to assess the influences of suspension-feeders at scales ranging from individual mussels to biogeographic regions. Quantitative comparisons based on the `catchment areas´ of single mussels, mussel beds and entire bays are used to illustrate how the role of these organisms varies as a function of the unit of measurement. One key factor influencing the relative importance of suspension-feeders at different scales is the rate of water movement and, thus, the volume of water available to the consumers. The biodiversity within suspension-feeder guilds is important because of the way it can affect the amount and sizes of particles removed from the spectrum of available food items. Variations in the role of these animals are also observed at time scales. The complexity of the scaling problem is illustrated using examples from the suspension-feeder guild in a tidal basin of the Wadden Sea (North Sea) where experiments and field measurements have provided insights into processes and mechanisms accounting for spatial and temporal variations. Keywords: suspension-feeders, biodiversity, productivity, food webs, scale
INTRODUCTION Feeding on suspended material and organisms is a widespread process in aquatic ecosystems and occurs over a large range of spatial and temporal scales (Dame 1996, Dame et al. 2001, Riisgård 1998). Suspension-feeders may drift in the water column or are fixed to the bottom. Some suspension-feeders move actively when feeding to increase the yield of food. Depending on the size of the suspension-feeders, from ciliates to whales, the size range of their prey varies from bacteria to krill, prawns and small fishes. A common feature of the food source of suspension-feeders is the three dimensional distribution in a cloud-like pattern which is either moving with the water masses or in case of organisms, is altering its place or “particle” density by swimming activity. Especially, suspension-feeders living fixed to the substrate, e.g. sponges and mussels, have developed special 199 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 199–219. ©2005 Springer. Printed in the Netherlands.
200 mechanisms to increase food ingestion by generating cilia driven water flows passing through a special filter apparatus retaining the food particles (e.g. Jørgensen 1990, Gili and Coma 1998, Riisgård and Randløv 1981, Okamura 1990, Shimeta and Jumars 1991). The scale determining suspension-feeding on an individual level is therefore dependent on 1) the distribution and patchiness of food particles 2) the size of the prey 3) the size of suspension-feeder 5) the active movement of the suspension-feeder and 6) the volume of medium which is processed by the suspension-feeder to meet its energy requirement (Jørgensen 1975, Jørgensen 1990). Suspension-feeders not only live as single individuals, but they often aggregate to form reefs or beds of large dimensions (e.g. Svane and Ompi 1993). Such an animal aggregation acts as a suspension-feeder at a higher level, which may be significant at the ecosystem level. Considering these aggregations of animals, the scale determined by suspension-feeders increases in relation to the water volume passing the animal aggregation and the exchange rate between animal community and the overflowing water. In addition to the size of the individual, that of the animal aggregation has also to be considered when assessing the significance of suspension-feeder communities at a higher level of organisation. At the lower end of the micro-scale (µm² to cm²) only suspensionfeeding processes of small individual organisms (e.g. ciliates, bivalve larvae or copepods), micro-communities (filtering capacity of organisms in biofilms on pebbles and small stones) and interactions between suspensionfeeders and micro- and meiofauna communities can be observed (Austen and Thrush 2001, Dittmann 1990, Seed 1996, Warwick et al. 1997). At this scale water volumes up to one or two dm³ per day may be processed. Laboratory methods in microcosms are suitable to record suspension-feeding processes at the micro-scale. Micro-scale observations allow only statements on the individual level and on the rating of the suspension-feeding of single organisms. Community measurements on this scale only consider micro- and meiofauna as well as small individual macrofauna and exclude the influence of larger organisms which may be significant and perhaps more important at the ecosystem level. Suspension-feeding by larger invertebrates often requires larger sampling areas and incubation vessels, from a microcosm for individual organisms to large volume mesocosms to assess the contribution of a natural density of suspension-feeders to the material budget of the adjacent community as well as plankton-suspension-feeder interactions. These microscale observations (cm² to m²) include also field experiments with suspension-feeders (e.g. exclusion experiments) which are used to describe the role of these organisms at the ecosystem level.
201 Increasing the scale from cm² to m² leads to considerable differences in the complexity and representativeness of groups of organism or environmental processes due to the size of the experimental design (Fig. 1) (Asmus et al. 1998, Lawrie and McQuaid 2001). Subsequently the measured signalseems tobe higher in those designs where a higher amount of processes is effective (close to nature designs) and a multifactorial analysis is required. At the lower end of the micro-scale individual processes are measured with a high resolution, whereas at the upper end of this scale the relevant ecosystem processes are included due to the size of test areas and may thus allow incorporating the results into ecosystem budgets and process assessments. Measurements on increasing scale from cores (cm2)
to flumes (m2)
bacteria benthic microalgae chemical processes phyto- and zooplankton nekton macrophytes macrofauna physical processes
Fig. 1. Relevance of different processes and ecological components in enclosures from small to large scale.
Enlarging the spatial scale of more than 1 m2 is necessary to investigate the relevance of suspension-feeders for spatial units up to one km². In this area the upper micro-scale changes to local scale (Schlüter and Ricklefs 1993). In this range most of the studies on communities should be carried out, because structural and functional aspects of a community are well represented. Although this scale is ideal to investigate sink/source functions and community exchange processes as well as prey/predator interactions at the ecosystem level, only few studies have been carried out considering this scale. At local scale variations within sites and systems can be investigated and interpreted by integrating multiresolution sampling designs and broad scale processes (Cummings et al. 1998, 2001, Moody and Woodcock 1995, Azovsky et al. 2000).
202 Mesoscale investigations focus on the functioning of total systems as lagoons, bays, defined basins or catchment areas of particular creeks, up to a size of about 1000 km² (Schlüter and Ricklefs 1993). McKindsey and Bourget (2000) showed that recruitment and structure of intertidal mussel beds varied with size of the system. The significance of suspension-feeder systems in these geographical units can be investigated experimentally either by black box approaches (e.g. import/export studies) or by modelling using information derived from the micro- and meso-scale experiments. Theoretically the mesoscale approach can also be extended to a regional scale, the biogeographical and to the global scale. These higher scales are only relevant assessing the significance of suspension-feeder processes for biodiversity studies, global material cycling, system comparisons and the importance for global change. They defy experimental control, but may be useful for modelling and conservation requirements (Connor et al. 1997, Schoch and Dethier 1996, Zacharias et al. 1999). Modelling the suspension-feeders´ influence at these upper scales requires information from meso- and macroscale investigations and a clear and careful method for scaling up. This study tries to assess the role of suspension-feeders at different spatial scales by following the influence of a single mussel to a mussel bed up to an embayment inhabited by suspension-feeder communities. Also the temporal dimension of scaling is explained by comparing suspensionfeeding processes in a short term, tidal, daily, seasonal, annual, inter-annual as well as long term context. Scaling is exemplified by the influence of the suspension-feeder guild in a tidal basin of the Wadden Sea (North Sea). The scaling problem is explained using biodiversity, productivity, filtration capacity and incorporation into the food web of this system as variables compared at different steps of ecological organisation.
RESULTS Biodiversity The total species number present in the worlds oceans can be assessed at 323,000 species (after Gruner 1993, see also Stork 1997). The total species number of suspension-feeders is unknown, but is estimated to be 75 000 (after Gruner 1993). This is nearly a quarter of all known aquatic species (Fig. 2). Most phyla present in the animal kingdom have developed suspension-feeders and in a lot of classes such as sponges and bivalves it is the dominant feeding type among species (Dame et al. 2001). Suspensionfeeders also occur over the total size range of animal species from microfauna to whales showing a maximum of species richness in a size range of 0.1 -1 g.
203 If we go from global scale to the temperate North East Atlantic shelf (1.1* 106 km²) we find 7500 (Costello et al. 2001) species and nearly 1000 species of suspension-feeders dominated by crustaceans, sponges and only in the third range by molluscs (Fig. 2, 3).
Global
NE Atlantic
species number 323,223
species number 7,535
Wadden Sea
Sylt-Rømø-Bay
species number 1,397
species number 1,155
Non Suspension Feeders
Suspension Feeders: Mollusca
Crustacea
Polychaeta
Porifera
Fig. 2. Numbers of suspension-feeders in relation to total species numbers.
In the Wadden Sea (about 6000 km²) crustaceans and molluscs dominate the suspension-feeders with 154 species from the total of 1397 species (Dankers et al. 1981). The total number of species in the Sylt-RømøBight ( 404 km²) is with 1155 close to that of the total Wadden Sea, however, the share of suspension-feeders of the total species number is less with 61 (unpublished species list of the Wadden Sea Station Sylt). 6 5
No. Sp.
4 3 2 1 0
Global NE Atlantic Shelf Shelf Wadden Sea SyltSylt Rømø-Ba ømø Bay area:361x106 1.1x106 6000 400 km m2 Suspension-feeders
Non Suspension- feeders
Fig. 3. Species number of suspension-feeders and non suspension-feeders from global to local scale.
204 Comparing the number of species on each of the scales the different groups are in a different state of investigation. Molluscs are popular species, which are very well taxonomically investigated, whereas the different groups of meiobenthos are underrepresented on a global and also on the North Atlantic scale. In the Wadden Sea but especially in the Sylt-Rømø-Bight, the last group is very well studied and dominates the total species number (e.g. Armonies and Reise 2000) so that macrobenthic species in general have lower shares in total diversity. Thus, variation in biodiversity of suspensionfeeders with spatial scale depends not only on the type of the ecosystem, but also varies with the degree of taxonomic knowledge. When going downscale from the bay area (404 km²) to the scale of the total mussel bed area (0.36 km2) of the total species number as well as number of suspension-feeder species continue to decrease from 1155 and 61 respectively to 161 and 26 in the mussel beds and only 66 and 10 in small mussel bed sections (1m²) (Fig. 4) (Asmus 1987, Dittmann 1990). In very small sediment cores, which cannot include a single mussel, only the sediment between the mussels is represented and hardly one species of the suspension-feeding guild is found. The total number of species at the lower end of the spatial scale is 20 to 25. Species density decreases from 8 species per 100 cm² to about 6 per 1 cm² (Dittmann 1990).
Biodiversityfrom macro- to microscale Sylt-Rømø-Bay 404x10 06 m²
mussel beds 5
36x10 0 m2
mussel bed section 100 cm²
mussel bed sediment 1 cm²
4 No.
3
Sp. 2
1
Suspensionfeeders Non Suspensionfeeders
0
Fig. 4. Species number of suspension-feeders and non suspension-feeders from local scale to micro-scale.
At every scale the species number of non suspension-feeders is much higher than that of suspension-feeders. Epibenthic suspension-feeder communities often possess hard shells, which attracts other particularly sessile organisms to settle on (Svane and Ompi 1993, Svane and Setyobudiandi 1996). The process of filtering material from the water column is followed by a rich detritus production, which again attracts other feeding guilds, such as detritus feeding polychaetes and oligochaetes (Commito and Boncavage 1989,
205 Commito and Dankers 2001). The wealth of invertebrates including the key organisms in such a system as well as the often rich habitat structure of mussel and oyster beds are basic requirements for predators such as crabs and fishes which also contribute to the biodiversity of those systems. Thus biodiversity of suspension-feeder systems, such as mussel beds, is characterised by a high number of species with other feeding modes, which dominate in general at each scale. Productivity While number of species in suspension-feeder systems is dominated by species with different feeding types, biomass in mussel beds is clearly dominated by suspension-feeders. If we compare the relative share of both groups in the intertidal area of the Sylt-Rømø tidal basin, we see that following a gradient from the high tide line to the low tide line where dense mussel beds occur, the dominance of suspension-feeder biomass increases (Asmus and Asmus 1990, Asmus 1994). More than 90 % of the animal biomass of mussel beds is due to suspension-feeders. In the deeper sand flat areas with inundation times of 5 to 8 hours per tide, suspension-feeders also represent more than 50% of the zoobenthic biomass. The dominant species is the endobenthic cockle Cerastoderma edule (Fig. 5). Because this type of a sandy tidal flat covers 67% of the total intertidal area, C. edule is the most important suspension-feeder in the system. The population of C. edule is sensitive to hard winters (Beukema 1985) and Sylt-Rømø-Bay Shares of communities in the intertidal area mud flats 3% mussel bed 0.3 % seagrass (SG) bed on sand 4 % SG bed on mud 8%
muddy sands 10% sandy beaches 5% sandy shoals3%
Arenicolaa-flat 67%
Shares of species in suspension feeder biomass
Shares of communities in su uspension feeder biomass SG bed on sand 2% SG bed on mud 4% musssel bed 13%
polychaetes 2%
Mya areanaaria 13%
Mytilus edulis 13%
muddy sa ands 9%
Arenicolaa-flat 71%
Cerastoderma edule 73%
Fig. 5. Arial shares (cycle above) and contribution of communities to total suspension-feeder biomass (left cycle) and shares of species to total suspension-feeder biomass (right cycle) in the Sylt-Rømø-Bay.
206 intensive predation by birds (Meire 1993), so that after a hard winter the biomass of C. edule is negligible, and for a while the total system shifts from a suspension-feeding to a system dominated by grazers and detritus feeders (Asmus 1994). At the upper intertidal zone, biomass of suspension-feeders is generally lower, because inundation time is too low to allow the filtering process. Here suspension-feeders such as Macoma balthica and Nereis diversicolor dominate, which are not obligatory suspension-feeders (Riisgård and Kamermans 2001). Secondary production of macrozoobenthos on the tidal flats of the Sylt-Rømø-Bay increases with tidal inundation. In areas close to the low tide line secondary production is dominated by suspension-feeders, whereas further up shore the share of suspension-feeders in secondary production shows a decrease (Asmus 1990). Suspension-feeders play a greater role in the productivity of a system compared to the biodiversity, because few suspension-feeding species may contribute significantly to biomass and secondary production, whereas their share in biodiversity is distinctly lower. However, suspension-feeders are often key species that enhance biodiversity and in that way increase the total productivity of a system. To explain the role of scaling is much more difficult for productivity studies compared to biodiversity studies because estimates of productivity in suspension-feeder systems focus mainly on the community aspect and on the geographical system aspect. Studies or even extrapolations either on the micro-scale or on larger scales like biogeographical regions and global scale are missing. The only way to get an impression on larger scales is to compare biomass and productivity studies on geographically different systems. By the comparison of two very different types of systems such as for example the Wadden Sea and the Antarctic shelf sea, where suspensionfeeders play a major role; some interesting relations between the productivity of these systems arise. In the Wadden Sea phytoplankton production and consumption by suspension-feeders are spatially closely connected and additional food is transported by the tides to the place where the consumers live. At the Antarctic shelf, suspension-feeders can only settle at the steep continental margins, because the shallow parts of the entire shelf region is scraped by the ice and therefore low in benthic biomass (Gutt et al. 1999). Primary productivity is restricted to the euphotic surface layer of Antarctic Seas and therefore the benthos can only use the production after this has sunk to the bottom. In the upper layer a rich zooplankton community may compete with the benthos for food sources. Total benthos in the Wedell Sea has an average biomass of 24 to 32 g afdw per m² (Brey and Gerdes 1999, Gerdes et al. 1992, Piepenburg et al. 2002), however the share of suspension-feeders is lower (1.8 -9 g afdw m-2).
207 If we focus on the fresh weight of the benthos more than a half of the biomass in that region would be due to suspension-feeders, because the dominating group is formed by sponges, which have a very high content of water in their tissue and therefore may show a lower accumulation of matter in biomass compared to bivalves. In contrast to the Wadden Sea the biomass of zooplankton is in the same order of magnitude (3g afdw m-2) like the lower range of benthic suspension-feeders. Benthic productivity of the Wedell Sea is low; probably the low temperature regime allows only a low material turnover, especially in the deep benthos (Brey and Gerdes 1999, Gerdes et al. 1992). Building benthic biomass in those areas takes a long time. In the shallow benthos average biomass and benthic secondary production are in the same order of magnitude; however pelagic consumers show a distinct higher secondary production than biomass and therefore are at least as active or even more active as the shallow benthos. In contrast to Antarctic and Arctic regions, the role of suspensionfeeders for productivity is less significant in tropical areas. Due to oligotrophic waters plankton development as well as particle content in the water is lower compared to higher latitudes, and the pelagic food sources are often not sufficient to supply the food requirements of the benthic animals. Thus, the strategy of symbiosis between suspension-feeders and monocellular algae helps to overcome this food shortage in many primarily suspension-feeding animals such as giant mussels Tridacna spp., tropical corals and sponges (Yonge 1936, Fankboner and Reid 1990, Klumpp et al. 1992, Hawkins and Klumpp 1995). In subtropical and warm temperate regions detritus feeding and predation seem to dominate productivity of heterotrophs compared to cold temperate and Polar Regions. In the Ria Formosa (South Portugal) more than 34 % of the animal production was based on detritus, 25% on phytobenthos and 17% on predation, so that phytoplankton contributes less than 24% to benthic secondary production (Sprung et al. 2001). In summary, suspension-feeders dominate the secondary productivity of benthos in most coastal systems with increasing importance from tropical to Polar Regions. Because suspension-feeders increase biodiversity they enhance system productivity as well.
Filtration capacity The scale determining suspension-feeding on an individual level is dependent on: 1) the distribution and patchiness of food particles, 2) the size of prey (particles), 3) the size of the suspension-feeder, 4) the active movement of the suspension-feeder and 5) the volume of medium which is processed by the suspension-feeder to meet its energy requirement.
208 Dame et al. (2001) and Riisgård (2001) report summaries on different filtration rates, filtration capacities and volumes of water filtered. Filtering by sponges is very effective with respect to particle size, but the volume of water processed by this group is quite low, compared to bivalves or hydrozoans (Riisgård et al. 1993). Dense colonies of hydrozoans, large anthozoans and corals may attain similar filtration rates per m², but they remove larger particles as food. Individual mussels generate a continuous water flow from the ambient water through their gills and back to the surrounding medium. The clearance rate per gram tissue weight increases with dry tissue weight (Møhlenberg and Riisgård 1978, 1979) following the formula: Clearance rate (l.h-1.g-11) = 7.45 x 0.66 where x is the dry tissue weight. A just settled recruit filters on average a water volume of 0.5 ml h-1, whereas a single adult mussel (1g dry tissue weight) processes about 4 l h-1. This inhalant flow meets the basic metabolic requirements supporting the mussel with food and oxygen. To improve the yield of these vital resources mussels are able to regulate the flow through their gills due to temperature and particle concentration, but also due to the food quality of the organic suspension. Phytoplankton is generally the main energy source for Mytilus edulis (Winter 1974, Williams 1981). To measure phytoplankton uptake by a mussel bed at a larger scale directly in the field a flume system was used. This flume includes 40 m² of a mussel bed in one 20 m long lane, and 40 m² of bare sediment in a parallel lane, from which all mussels had been removed (Asmus and Asmus 1991, Asmus et al. 1992). At the inflow, concentration of phytoplankton was measured by taking water samples every hour. When the water body has passed the flume, an additional sample was taken at the outflow of the flume. The inflow concentration of phytoplankton is higher than the outflow concentration. From this difference the uptake rate by the mussel bed was calculated. Uptake was the dominant process in the mussel bed. In the control lane with bare sediment inflow and outflow concentrations have irregular fluctuating differences. We therefore observe in the control lane uptake as well as release rates. From the uptake rates over the mussel bed an average in situ clearance rate of 6.4 m³ m-2 h-1 (SD = 4.1 ) ( range 2.1 - 11.7 m³ m-2 h-1 due to the phytoplankton concentration) could be estimated (Asmus et al. 1992). Lower clearance rates were measured when phytoplankton concentration were between 6 and 15 million cells l-1, whereas higher uptake rates were recorded when the concentration was between 12,000 and 25,000 cells l-1 (Asmus unpublished). The clearance rate estimated over the control lane was negative
209 or in one case small but positive because sedimentation of phytoplankton cells was observed during high tide caused by calm weather. Estimated clearance rates (m3 m-2 h-1) measured in the flume correspond to an average individual clearance rate of 1.3 l per individual mussel h-1 (range 0.4 to 2.3 l per individual mussel h-1) (Asmus et al. 1992). When clearance rates were estimated based on the individual sizes of mussels (Møhlenberg and Riisgård 1979) in the flume, a higher value of 2.44 l per individual mussel h-1 was found compared to those directly measured in the field. This may indicate that less phytoplankton was available in the field or that it was less attractive as food compared to algae offered in the laboratory. During the field experiments phytoplankton was dominated by Phaeocystis globosa which is reported by some authors to cause lower feeding and growth rates (Beukema and Cadée 1991). However, comparing different field situations and different phytoplankton compositions, uptake of phytoplankton is more dependent on phytoplankton concentration than on phytoplankton composition (Smaal et al. 1986, Dame and Dankers 1988, Prins and Smaal 1990). Individual clearance rates estimated from the flume experiments indicate a decreasing trend when the amount of phytoplankton carbon and the percentage phytoplankton carbon of total particulate organic carbon increases. The comparison of individual clearance rates at different spatial scales (chamber scale in the laboratory (few cm²) (Møhlenberg and Riisgaard 1979) and flume scale (40 m²) (Asmus et al. 1992), showed a higher rate measured at the laboratory scale compared to the field. However, this may be more an effect of both phytoplankton quantity and quality than a scale effect between laboratory and field, although this is difficult to separate (Asmus and Asmus 1993). Due to the tide the phytoplankton concentration and the physical and chemical conditions vary at large scale in the field in the water flowing over a mussel bed. Because all these natural fluctuations are ecosystem constituents, filtration rates measured at larger scales are reflecting the actual filtration, which is a valuable base for evaluating the filtration of suspension-feeders at higher scales. Direct measurements of filtration rates at larger scales than that of the flume measurements are missing in the Sylt-Rømø tidal basin. Even worldwide references on those measurements are rare and only done in systems where benthic suspension-feeders obviously play an outstanding role in material flow (Cloern 1982, Cahoon and Owen 1996, Cohen et al. 1984, Dolmer 2000). In these cases inflow-outflow differences of phytoplankton concentrations were related to the high filtration potential of suspensionfeeders. This also includes processes like sedimentation, grazing by zooplankton etc. overestimating the possible impact of benthic suspensionfeeders. Filtration processes on the scale of the total tidal basin have also to include the contribution of other suspension-feeders such as cockles, clams, slipper limpets and the recently established razor clams and pacific oysters. Cockles are faunal components of the sandy and muddy tidal flats where they
210 live in high but patchy abundances. These sandy tidal flats cover 67% of the intertidal area of the bight. Thus cockles dominate suspension-feeder biomass with 73 % (see also figure 5) and with this the filtration processes of the system. Filtration capacity of suspension-feeder systems thus varies with spatial scale from species dependent individual rates at the micro-scale to community dependent rates at the macro- local and regional scale. One key factor influencing the filtration capacity at different scales is the rate of water movement and thus the volume of water available to consumers. However, a further generalisation is not possible because of the poor knowledge of filtration processes on the microscale and on the large scales (e.g. total basins)
Food webs Scaling is of special importance for the role of suspension-feeders in the food web. At the micro-scale only few trophic interactions between suspension-feeders and their prey can be observed. Microbenthic food webs include several trophic levels at this scale. Increasing this scale from cm² to 1 m² leads to considerable differences in the representativeness of groups of organisms and environmental processes considered (Asmus et al. 1998). At the cm² scale trophic interactions can be analysed at an individual level with high resolution (Riisgård 2001). At the m² scale a food web can be constructed which includes all those compartments which are spatially close to each other, such as mussels and their attached community or the interaction between mussels and the microbial food web (Asmus 1987 Austin and Thrush 2001, Seed 1996). At the meso-scale, the macrobenthic food web represents further trophic levels. Starting from a point showing a food web with only few elements at the m² - level, more and more components can be included by enlarging this scale. A larger scale is necessary to consider large predators as birds and fishes. Even assessing the impact of men needs at least such a mesoscale approach. Focussing on the food webs of total systems as lagoons, bays, defined basins or catchment areas of particular creeks, the food web of suspensionfeeders can be investigated either by black box approaches or by modelling using information derived from meso- and macroscale experiments. This local scale approach is necessary for food web comparisons between systems and can be extended to a regional scale, biogeographical up to a global scale. These higher scales are only relevant assessing the significance of suspensionfeeding processes for biodiversity studies, global material cycling, system comparisons and the importance for global change. They defy experimental controlling but may be useful for modelling.
211 DISCUSSION The share of suspension-feeding species in total species decreases with spatial scale, suggesting a minor influence of this feeding type to total diversity of systems. Assessing total diversity of species is only possible by some approximations because the taxonomic knowledge still increases and the different taxonomic groups are specified to a different degree. Assigning trophic groups to species richness is only approximate, because the knowledge of the biology of most species is still low and for many regions the taxonomic knowledge of the species pool is poor. The relatively low share of suspension-feeders in total species richness does not indicate a minor role of this group for biodiversity. Suspension-feeders can increase biodiversity on a local and also on a regional scale, because they attract an array of other organisms of different trophic groups to use the special habitat for shelter, nursery, feeding ground and as firm settling structure (Dame et al. 2001, Dittmann 1990, Dankers 1993, Asmus 1987, Tsuchiya and Nishihara 1986, Peake and Quinn 1993). The spectrum of functions as well as the habitat creating potential is more significant in epibenthic suspension-feeder assemblages than in endobenthic communities. Therefore most epibenthic suspension-feeding species are typical key organisms and they have an important function as bioengineers creating special habitats (Dame et al. 2001). Mussel and oyster beds in the Wadden Sea are hard bottom oases in a vast desert of soft and sand bottoms, whereas dense assemblages of cockles may only attract direct predators, but do not show most of the functions characterising a key organism. It is hypothesised that two non-mutually excluding types of mechanisms are important to explain the link between diversity and ecosystem function (Giller and O´Donnovan 2002). The complementary mechanism shows that increasing diversity leads to positive species interactions by facilitation of resource partitioning. The selection mechanism occurs when due to increasing diversity, processes, such as interspecific competition or intrinsic growth differences may underlie the dominance of key species. In this case particular traits of those species would largely determine ecosystem function. More diverse mixtures of species have a higher chance of containing these particular key species. Suspension-feeder assemblages are key organisms as well as bioengineers increasing biodiversity by facilitation of resource partitioning (benthic-pelagic coupling makes material produced in the pelagic environment available to an array of food cycles linking a huge number of benthic organisms) (Paine 1969, Grimm 1995, Jones et al. 1994, Lawton and Jones 1995). At least on a local scale the second mechanism is also observed that suspension-feeders would largely determine ecosystem function and thus select the species assemblage associating with this key species. Because
212 epibenthic communities, settling on soft bottoms create hard bottom habitats, unique at muddy or sandy shores, they may increase species number for total embayments and coastal regions. However, the significance of suspension-feeder systems seems to decrease with decreasing latitude on a global scale and with increasing diversity of the species pool. In Arctic, Antarctic and cold temperate regions, suspension-feeders may show high productivities and biomasses compared to warm temperate, subtropical and tropical areas. The reason may be the seasonality of pelagic primary productivity of coldwater areas, which is enlarged by high nutrient availability at the beginning of the vegetation period (Heinrich 1962). Shallow water systems at these high latitudes are supported by a large seasonal phytoplankton import from adjacent open seas during spring. This additional food source seems to be used with a high efficiency in these systems; therefore dense suspension-feeder assemblages can develop especially in places where phytoplankton import forms a large additional input to local phytoplankton production, which is too low to sustain the high biomass. In warm temperate, subtropical and tropical regions coastal systems seem to depend more on system internal processes, such as detritus production by seagrass beds, salt marshes, mangrove forests, or imports from river runoffs. Pelagic food sources have a share of 51% on total food consumption in the cold temperate Sylt-Rømø-Bight compared with only 30 % of a morphologically similar intertidal system in the warm temperate region of the Portuguese Coast (Sprung et al. 2001). This difference is even more pronounced in tropical systems where the water of the oceans often shows oligotrophic conditions. To overcome the pelagic food shortage in tropical systems, especially suspension-feeders show the ability to take advantage from symbiotic algae to satisfy their energy requirements (Yonge 1936, Fankboner and Reid 1990, Klumpp et al. 1992, Hawkins and Klumpp 1995). Within biogeographical regions, suspension-feeder communities may show a varying significance due to the special environmental conditions and habitat configuration. In the Wadden Sea suspension-feeder communities dominate the pelagic-benthic coupling of phytoplankton uptake. Within the Wadden Sea there are regions where epibenthic mussel beds are totally absent or very small, as in those parts of the German Wadden Sea, which have strong currents and tidal ranges and are directly influenced by the big estuaries of the rivers Elbe and Weser. In these systems endobenthic suspension-feeders, such as C. edule or Mya arenaria may compensate the loss of epibenthic mussel beds. These endobenthic species have been reported to use resuspended microphytobenthos as main food source, as could be shown for an intertidal system at Marennes-Oleron on the French Atlantic coast (Sauriau et al. 2000). This means that endobenthic suspension-feeders may contribute to a much lower degree to benthic-pelagic coupling than epibenthic suspension-feeders.
213 However, feeding on resuspended diatoms is only possible in muddy intertidal areas where the diatom assemblage is easily resuspended, whereas on sandy bottoms the diatom assemblage is dominated by species living firmly attached to sand grains and are to a certain degree resistant to resuspension (Asmus and Asmus 1993). The lack of epibenthic suspension-feeder communities in some Wadden Sea systems is not caused by food shortage, but may be a result of harsh physical conditions of the environment, such as strong tidal currents and higher exposure of these areas to wave action during storms because of the lack of natural barrier islands. Epibenthic mussel beds in the Wadden Sea are highly dependent on physical shelter and therefore are out-competed by endobenthic communities (Nehls and Thiel 1993). Therefore climatic changes, increasing the frequency of extreme weather situations with heavy storms may lead to a retreat of epibenthic communities, such as mussel beds, but also seagrass beds in the Wadden Sea (Schanz et al. 2002). A typical characteristic of coastal systems is the high variability of habitats, their species composition as well as the morphology and the physical and chemical configuration of the environment. Therefore the different spatial scales which may be important for suspension-feeder communities are closely linked to strong temporal changes. In the Wadden Sea fundamental changes are reported especially for epibenthic communities in the past hundred years (Reise 1982, Reise et al. 1989, Riesen and Reise 1982, Wohlenberg 1935, 1937). On a regional scale, subtidal beds of the European oyster (Ostrea edulis) as well as sabellid worm reefs (Sabellaria spinulosa) were widely distributed along the tidal gullies (Reise et al. 1989, Hagmeier and Kändler 1927). The extinction of both communities is caused by a combination of a direct anthropogenic effect (overexploitation, destruction) and a changing environment partly through man made effects (eutrophication, coastal engineering, and climate change). The potential of suspension-feeding of these communities as well as their importance on the ecosystem level is unknown. However, it may be reconstructed from early maps (Wohlenberg 1935, 1937). The historical changes are teaching us that the Wadden Sea has a potential for a higher biodiversity and an even greater carrying capacity for different epibenthic suspension-feeding communities. Although the quantitative importance of the loss of these suspension-feeder systems is unknown, it may be probable that the ecosystem function in former times have been more efficient in benthic- pelagic coupling. The historical losses of suspension-feeder communities were partly compensated by an increased extension of subtidal mussel beds. As the mussel fishery expanded in the 80`s and 90`s, systems phase shifts. The effect of suspension-feeders invading the Wadden Sea in recent times has not yet been estimated. The endobenthic American razor clam (Ensis ( americanus) may reach similar biomasses in subtidal sandy areas as dense mussel or cockle beds. This species occupies sandy frequently perturbated areas, and seems to be a pioneer filter-feeder in those areas. The invaded epibenthic pacific oyster (Crassostrea gigas) inhabits intertidal mussel beds, where it overgrows mussels. Filtering rate of
214 this species is estimated to be higher as that of a blue mussel (Bernard and Noakes 1990), and thus dense settlements of these oysters may increase benthic-pelagic coupling. On the other hand the value as a prey for the higher trophic levels within the food web is lower, and will probably lead to serious changes in the upper part of the food web. On a local scale suspension-feeders may influence the food web and the energy and material flow of the particular system. A control of phytoplankton biomass could be shown for many enclosed systems worldwide (Cloern 1982, Cahoon and Owen 1996, Cohen et al. 1984, Dolmer 2000). While the filtering capacity of the suspension-feeder component depends on the aerial extent of the particular suspension-feeder communities, the impact of these groups on pelagic benthic coupling is higher in those suspension-feeders covering most of the area of the system. In the Sylt-RømøBight different communities contribute to different shares to the total suspension-feeder biomass of the whole intertidal area (Fig. 5). This figure reflects the situation of the bay between 1992-94. Typical sand flats ( (Arenicola flats) covering 67% of the intertidal area reveal 71% of the suspension-feeder biomass of the total bight, whereas mussel beds which cover less than 1 percent of the area, hold a share of 13%. Splitting the total suspension-feeder biomass into 6 dominant species, 73% of the biomass is covered by C. edule, and only 13% by M. edulis and the same percentage will be occupied by M. arenaria (Fig. 5). Suspension-feeding polychaetes are of minor importance. Recent figures would include also the new invaded species such as E. americanus and C. gigas, which increased in abundance in the last few years significantly (Reise 1998, Beukema and Dekker 1995). Other suspension-feeders occurring in this system are not obligate suspension-feeders (e.g. M. balthica, N. diversicolor), and are not considered in this budget. The share of species as well as communities contributing to the total suspension-feeder biomass on a local scale is varying in Wadden Sea systems especially after hard winters when adult stocks of the dominant species C. edule are depleted. While pelagic food sources (phytoplankton and detritus) dominate the food web of the Sylt-Rømø tidal basin, network analysis showed a dominance of detritus based food chain over grazing food chain by a ratio of 1.44:1.00 in both pelagial and benthal (Baird et al. 2004). Total consumption of all food resources by macrofauna in the Sylt-Rømø-Bay is dominated by benthic detritus-feeders followed by suspension-feeders and benthic grazers with the ratio 2.7: 2.6: 1. Considering the influence of suspension-feeders on the organic matter pool in the sediment, this trophic group may have also indirect effects on benthic detritus feeders and therefore have an important influence of the total food web. Summarising the results of this study suspension-feeders are significant components of coastal systems at the species, community and
215 ecosystem level, because they enhance biodiversity indirectly, productivity directly, have strong controlling functions in food webs and are part of strong links in complex systems.
REFERENCES Armonies W Reise K 2000 Faunal diversity across a sandy shore. Mar Ecol Prog Serr 196: 4957 Asmus H 1987 Secondary production of an intertidal mussel bed community related to its storage and turnover compartments. Mar Ecol Prog Serr 39: 251-266 Asmus H 1994 Benthic grazers and suspension-feeders: which one assumes the energetic dominance in Königshafen? Helgoländer Meeresunters 48: 217-231 Asmus H Asmus RM 1990 Trophic relationships in tidal flat areas: to what extent are tidal flats dependent on imported food? Neth J Sea Res 27: 93-99 Asmus H Asmus RM 1993 Phytoplankton - mussel bed interactions in intertidal ecosystems. In: Bivalve Filter-feeders in Estuarine and Coastal Ecosystem Processes, Dame RF (Ed) Springer, Heidelberg, pp 57-84 Asmus H Asmus RM Prins TC Dankers N Francés G Maaß B Reise K 1992 Benthicpelagic flux rates on mussel beds: tunnel and tidal flume methodology compared. Helgoländer Meeresunters 46: 341-361 Asmus RM Asmus H 1991 Mussel beds: limiting or promoting phytoplankton? J Exp Mar Biol Ecoll 148: 215-232 Asmus RM Jensen MH Jensen KM Kristensen E Asmus H Wille A 1998 The role of water movement and spatial scaling for measurement of dissolved inorganic nitrogen fluxes in intertidal sediments. Estuar Coast Shelf Scii 46: 221-232 Austen MC Thrush SF 2001 Experimental evidence suggesting slow or weak response of nematode community structure to algal suspension feeder. J Sea Res 46(1): 69-84 Azovsky AI Chertoprood MV Kucheruck NV Rybnikov PV Sapozhnikov FV 2000 Fractal properties of spatial distribution of intertidal benthic communities. Mar Bioll 136: 581590 Baird D Asmus H Asmus RM 2004 Energy flow of a boreal intertidal ecosystem, the SyltRømø-Bight. Mar Ecol Prog Serr in press Bernard FR Noakes DJ 1990 Pumping rates, water pressures, and oxygen use in eight species of marine bivalves molluscs from British Columbia. Can J Fish Aquat Scii 47: 13021306 Beukema JJ 1985 Zoobenthos survival during severe winters on high and low tidal flats in the Dutch Wadden Sea. In: Marine Biology of Polar Regions and Effects of Stress on Marine Organisms, JS Gray ME Christiansen (Eds). Wiley, pp 351-361 Beukema JJ Cadée GC 1991 Growth rates of the bivalve Macoma balthica in the Wadden Sea during a period of eutrophication: relationships with concentrations of pelagic diatoms and flagellates. Mar Ecol Prog Serr 68: 249-256 Beukema JJ Dekker R 1995 Dynamics and growth of a recent invader into European coastal waters: the American razor clam, Ensis directus. J Mar Biol Ass U K 75: 351-362 Brey T Gerdes D 1999 Benthic community productivity in the Magellan region and in the Wedell Sea. Sci. Marr 63 (Suppl 1): 145-148 Cahoon LB Owen DA 1996 Can suspension-feeding by bivalves regulate phytoplankton biomass in Lake Waccamaw, North Carolina ? Hydrobiologia 325: 193-200 Cloern JE 1982 Does the benthos control phytoplankton biomass in south San Francisco Bay? Mar Ecol Prog Serr 9: 191-202 Cohen RRH Dresler PV Phillips EJP Cory RL 1984 The effect of the Asiatic clam, Corbicula fluminea, on phytoplankton of the Potomac River, Maryland. Limnol Oceanogrr 29: 170-180
216 Commito JA Boncavage EM 1989 Suspension-feeders and coexisting infauna: an enhancement counter example. J Exp Mar Biol Ecol 125: 33-42 Commito JA Dankers NMJA 2001 Dynamics of spatial and temporal complexity in European and North American soft-bottom mussel beds. In: Ecological Comparisons of Sedimentary Shores, K Reise (Ed) Ecological Studies 151. Springer, Heidelberg, pp 3959 Connor DW Brazier DP Hill TO Northen KO 1997 Marine nature conservation review: marine biotope classification for Britain and Ireland. In: Littoral Biotopes. Version 97.06. JNCC Report., No. 229, Vol. 1, Joint Nature Conservation Committee, Peterborough. Costello MJ Emblow C White R 2001 European Register of Marine Species. A check list of the marine species in Europe and a bibliography of guides to their identification Patrimoines naturels 50: 463 p Cummings VJ Thrush SF Hewitt JE Turner SJ 1998 The influence of the pinnid bivalve Atrina zelandica (Gray) on benthic macroinvertebrate communities in soft-sediment habitats. J Exp Mar Biol Ecoll 228: 227-240 Cummings VJ Thrush SF Hewitt JE Funnell GA 2001 Variable effects of a large suspensionfeeding bivalve on infauna: experimenting in a complex system. Mar Ecol Prog Ser 209: 159-175 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Dame RF Bushek D Prins TC 2001 Benthic suspension-feeders as determinants of ecosystem structure and function in shallow coastal waters. In: Ecological Comparisons of Sedimentary Shores, K Reise (Ed), Ecological Studies 151, Springer, Heidelberg, pp 1137 Dame RF Dankers N 1988 Uptake and release of materials by a Wadden Sea mussel bed. J Exp Mar Biol Ecol 118: 207-216 Dankers N 1993 Integrated estuarine management – obtaining a sustainable yield of bivalve resources while maintaining environmental quality. In: Bivalve Filter-feeders in Estuarine and Coastal Ecosystem Processes, Dame RF (Ed) Springer, Heidelberg, pp 478-511 Dankers N Kühl H Wolff WJ 1981 Invertebrates of the Wadden Sea. Final report of the section ‘Marine Zoology’ of the Wadden Sea Working Group. Report 4 Balkema, Rotterdam, NL, 221p Dittmann S 1990 Mussel beds – amensalism or amelioration for intertidal fauna? Helgoländer Meeresunters 44: 335-352 Dolmer P 2000 Feeding activity of mussels Mytilus edulis related to near-bed currents and phytoplankton biomass. J Sea Res 44: 221-231 Fankboner PV Reid RGB 1990 Nutrition in giant clams (Tridacnidae). In: The Bivalvia – Proc. Mem. Symp. In Honour of Sir Ch. M. Yonge. Edinburgh 1986, Morton B (Ed). Hong Kong University Press, Hong Kong. pp 195-209 Gerdes D Klages M Arntz WE Herman RL Galeron J Hain S 1992 Quantitative investigations on macrobenthos communities of the southeastern Wedell Sea shelf based on multibox corer samples. Pol Biol 12(2): 291-391 Gili JM Coma R 1998 Benthic suspension-feeders: their paramount role in littoral marine food webs. TREE E 13: 316-321 Giller PS O´Donovan G 2002 Biodiversity and ecosystem function: do species matter? Biol Environ 102B: 129-139 Grimm NB 1995 Why link species and ecosystems? A perspective from ecosystem ecology. In: Linking Species and Ecosystems, CG Jones JH Lawton (Eds), Chapman and Hall, New York, pp 5-15
217 Gruner HE 1993 Lehrbuch der speziellen Zoologie Band I Wirbellose. G Fischer (Ed), Jena, Stuttgart, New York Teil 1-4: 2836 p Gutt J Buschmann A Dijkstra J Dimmler W Piepenburg D Teixidó N 1999 Study on benthic resilience of the macro- and megabenthos by imaging methods. In Arntz WE Gutt J (Eds) the expedition Antarktis XV/3 (EASIZ II) of RV Polarstern in 1998. Ber Polarforsch 301: 17-22 Hagmeier A Kändler R 1927 Neue Untersuchungen im nordfriesischen Wattenmeer und auf den fiskalischen Austernbänken. Wiss Meeresunters (Abt. Helgoland) 16: 1-93 Hawkins AJS Klumpp DW 1995 Nutrition of the giant clam Tridacna gigas (L.) by symbiotic alga toward total nitrogen requirements for tissue growth and metabolism. J Exp Mar Biol Ecol 190: 263-290 Heinrich AK 1962 The life histories of plankton animals and seasonal cycle of plankton communities in the oceans. J Cons Int Explor Merr 27: 15-24 Jones CG Lawton JH Shachak M 1994 Organisms as ecosystem engineers. Oikos 69: 373386 Jørgensen CB 1975 Comparative physiology of suspension-feeding. Annu Rev Physioll 37: 5779 Jørgensen CB 1990 Bivalve Filter Feeding: Hydrodynamics, Bioenergetics, Physiology and Ecology. Olsen and Olsen, Fredensborg, DK, 140p Klumpp DW Bayne BL Hawkins AJS 1992 Nutrition of the giant clam Tridacna gigas (L.). I. Contribution of filter feeding and photosynthesis to respiration and growth. J Exp Mar Biol Ecoll 155: 105-122 Lawrie SM McQuaid CD 2001 Scales of mussel bed complexity: structure, associated biota and recruitment. J Exp Mar Biol Ecoll 257: 135-161 Lawton JH Jones CG 1995 Linking species and ecosystems: organisms as ecosystem engineers. In: Linking Species and Ecosystems. CG Jones JH Lawton (Eds), Chapman and Hall, New York, pp 141-150 McKindsey CW Bourget E 2000 Explaining mesoscale variation in intertidal mussel community structure. Mar Ecol Prog Serr 205: 155-170 Meire P 1993 The impact of bird predation on marine and estuarine bivalve populations: a selective review of patterns and underlying causes. In: Bivalve Filter-feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed). Springer, Heidelberg, pp 197-243 Møhlenberg F Riisgård HU 1978 Efficiency in particle retention in 13 species of suspensionfeeding bivalves. Ophelia 17: 239-246 Møhlenberg F Riisgård HU 1979 Filtration rate using a new indirect technique, in thirteen species of suspension-feeding bivalves. Mar Bioll 54: 143-147 Moody A Woodcock CE 1995 The influence of scale and the spatial characteristics of landscapes on land-cover mapping using remote sensing. Landscape Ecoll 10: 363-379 Nehls G Thiel M 1993 Large-scale distribution patterns of the mussel Mytilus edulis in the Wadden Sea of Schleswig-Holstein: Do storms structure the ecosystem ? Neth J Sea Res 31: 181-187 Okamura B 1990 Behavioural plasticity in the suspension-feeding of benthic animals. In: Behavioural Mechanisms in Food Selection, Hughes RN (Ed). Springer-Verlag, Heidelberg, pp 637-660 Paine RT 1969 A note on trophic complexity and community stability. Am Natt 103: 91-93 Peake AJ Quinn GP 1993 Temporal variation in species-area curves for invertebrates in clumps of an intertidal mussel. Ecography 80: 269-277 Piepenburg D Schmid MW Gerdes D 2002 The Benthos of King George Island (South Shetland Islands, Antarctica): further evidence for a lack of latitudinal biomass cline in the southern Ocean. Pol Bioll 25: 146-158 Prins T C Smaal AC 1990 Benthic-pelagic coupling: the release of inorganic nutrients by an intertidal bed of Mytilus edulis. In: Trophic Relationships in the Marine Environment, Barnes M and Gibson RN (Eds). Proc 24th Europ Mar Biol Symp, Aberdeen Univ. Press, UK, pp 89-103
218 Reise K 1982 Long-term changes in the macrobenthic invertebrate fauna of the Wadden Sea: are polychaetes about to take over? Neth J Sea Res 16: 29-36 Reise K 1998 Pacific oysters invade mussel beds in the European Wadden Sea. Senckenberg maritt 28: 167-175 Reise K Herre E Sturm M 1989 Historical changes in the benthos of the Wadden Sea around the island of Sylt in the North Sea. Helgoländer Meeresunters 43: 417-433 Riesen W Reise K 1982 Macrobenthos of the subtidal Wadden Sea: revisited after 55 years. Helgoländer Meeresunters 35: 409-423 Riisgård HU 1998 Filter feeding and plankton dynamics in a Danish fjord: a review of the importance of flow, mixing and density-driven circulation. J Environ Man 53: 195-207 Riisgård HU 2001 On measurement of filtration rates in bivalves – the stony road to reliable data: review and interpretation. Mar Ecol Prog Serr 211: 275-291 Riisgård HU Randløv A 1981 Energy budget, growth and filtration rates in Mytilus edulis at different algae concentrations. Mar Bioll 61: 227-234 Riisgård HU Thomassen S Jakobsen H Weeks JM Larsen PS 1993 Suspension-feeding in marine sponges Halichondria panicea and Haliclona urceolus: effects of temperature on filtration rate and energy cost of pumping. Mar Ecol Prog Ser 96: 177-188 Riisgård HU Kamermans P 2001 Switching between deposit and suspension-feeding in coastal zoobenthos. In: Ecological Comparisons of Sedimentary Shores, Reise K (Ed). Ecological Studies 151, Springer-Verlag, Heidelberg, pp 73-101 Sauriau PG Kang CK 2000 Stable isotope evidence of benthic microalgae-based growth and secondary production in the suspension-feeder Cerastoderma edule (Mollusca, Bivalvia) in the Marennes-Oléron Bay. Hydrobiologia 440: 317-329 Schanz A Polte P Asmus H 2002 Cascading effects of hydrodynamics on an epiphyte-grazer system in intertidal seagrass beds of the Wadden Sea. Mar Biol 141: 287-297 Schlüter D Ricklefs RE 1993 Species diversity: an introduction to the problem. In: Species Diversity in Ecological Communities: Historical and Geographic Perspectives, Ricklefs, RE and Schlüter D (Eds.). University of Chicago Press, Chicago, pp 1-10 Schoch GC Dethier MN 1996 Scaling up: the statistical linkage between organismal abundance and geomorphology on rocky intertidal shorelines. J Exp Mar Biol Ecol 201: 37-72 Seed R 1996 Patterns of biodiversity in the micro-invertebrate fauna associated with mussel patches on rocky shores. J Mar Biol Assoc UK: 76: 203-210 Shimeta J Jumars PA 1991 Physical mechanisms and rates of particle capture by suspensionfeeders. Oceanogr Mar Bioll 29: 191- 257 Smaal AC Verhagen JHG Coosen J Haas HA 1986 Interaction between seston quantity and quality and benthic suspension-feeders in the Oosterschelde The Netherlands. Ophelia 26: 385-399 Sprung M Asmus H Asmus R 2001 Energy Flow in Benthic Assemblages of Tidal Basins: Ria Formosa (Portugal) and Sylt-Römö Bay (North Sea) Compared. In: Ecological Comparisons of Sedimentary Shores. Reise K (Ed), Springer, Heidelberg. pp 237-254 Stork NE 1997 Measuring Global Biodiversity and Its Decline. In: Biodiversity II, ML ReakaKudla DE Wilson EO Wilson (Eds). Joseph Henry press, Washington, DC, pp 41-68 Svane I Ompi M 1993 Patch dynamics in beds of the blue mussel Mytilus edulis L.: Effects of site, patch size, and position within a patch. Ophelia 37: 187-202 Svane I Setyobudiandi I 1996 Diversity of associated fauna in beds of the blue mussel Mytilus edulis L.: effects of location, patch size, and position within a patch. Ophelia 45: 39-53 Tsuchiya M Nishihara M 1986 Island of Mytilus as habitat for small intertidal animals: effects of Mytilus age structure on the species composition of the associated fauna and community organisation. Mar Ecol Prog Ser 31: 171-178
219 Warwick RM McEvoy AJ Thrush SF 1997 The influence of Atrina zelandica Gray on meiobenthic nematode diversity and community structure. J Exp Mar Biol Ecol 214: 235-247 Williams P 1981 Detritus utilisation by Mytilus edulis. Estuar coast Shelf Scii 12: 739-746 Winter JE 1974 Growth in Mytilus edulis using different types of food. Ber dt wiss Komm Meeresforsch 23: 360-375 Wohlenberg E 1935 Beobachtungen über das Seegras, Zostera marina L., und seine Erkrankung im nordfriesischen Wattenmeer. Nordelbingen 11: 1-19 Wohlenberg E 1937 Die Wattenmeer-Lebensgemeinschaften im Königshafen von Sylt. Helgoländer Meeresunters 1: 1-92 Yonge CM 1936 Mode of life, feeding, digestion and symbiosis with zooxanthellae in the Tridacnidae. Sci Rep Gr Barrier Reef Exped Br Mus 1: 283-321 Zacharias MA Morris MC Howes DE 1999 Large scale characterisation of intertidal communities using a predictive model. J Exp Mar Biol Ecol 239: 223-242
INVADERS IN SUSPENSION-FEEDER SYSTEMS: VARIATIONS ALONG THE REGIONAL ENVIRONMENTAL GRADIENT AND SIMILARITIES BETWEEN LARGE BASINS
Sergej Olenin, Darius Daunys Coastal Research and Planning Institute, Klaipeda University, Klaipeda, Lithuania
Abstract:
Biological invasions increasingly alter taxonomical and functional structure of benthic communities. Among the invasive benthic invertebrates, the suspension-feeders are the most widespread type. Species belonging to that trophic group constitute from half to two thirds of total invasive species in various European seas. The importance of the alien suspension-feeders is particularly obvious in evolutionary young, species poor brackish water bodies, such as the Baltic Sea. We analyzed changes in suspension-feeder systems formed by both alien and native species along the salinity and depth gradients from a river mouth (the Curonian Lagoon) down to the halocline area of the Baltic Proper. There was a clear shift in the biomass dominance from the alien species in the lagoon to the native suspension-feeders in the sea. Both the native and alien obligatory suspension-feeders occupied shallow coastal marine and lagoon habitats and did not form stable communities below the 30 m depth.
Key words: bioinvasions, trophic guilds, salinity gradient, functional homogenization
INTRODUCTION Biological invasions mix previously isolated organisms on global scale causing their genetic, taxonomic and functional homogenization (Leppäkoski et al. 2002a, Olden et al. 2004). This human-mediated addition to local fauna and flora is defined as xenodiversity (Gr. xenos – strange) to indicate diversity caused by non-indigenous, invasive species (Leppäkoski and Olenin 2000a). The influence of xenodiversity is traced at different hierarchical levels of biological organization, including functional/community level due to alterations of food webs and ecosystem functioning, emergence 221 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 221–237. ©2005 Springer. Printed in the Netherlands.
222 of novel functions, and changes in structure of trophic/functional guilds (e.g. Olenin and Leppäkoski 1999; Karatayev et al. 2002; Ojaveer et al. 2002; Baltic Sea Alien Species Database 2003, and references therein). The importance of biological invasions in alteration of trophic structure of benthic communities is especially obvious in evolutionarily young, species-poor water bodies, such as the Baltic Sea (Leppäkoski et al. 2002b). In this study, we consider the suspension-feeding systems formed by both alien and native species along environmental gradients from a river mouth (the Curonian Lagoon) down to the halocline area of the Baltic Proper. The studied habitats differ in topography, substrate, hydrological regime, biological communities as well as the level and type of human impact. Analysis of the structural changes along such gradients can give insights into how physical and biological features of habitats influence the suspensionfeeder systems and their susceptibility to invasion. Although many recent works address functional changes in benthic communities caused by invasion on a local scale (e.g. Zettler 1996; Karatayev et al. 2002) the broader; basin wide aspect of these changes is still lacking. In this study, we assess the trophic guild composition of invaders in different geographical regions, investigating how common suspension feeding is among the invasive benthic invertebrate species. Thus, we consider two aspects of structural change in the benthic communities, caused by the invasion of suspension-feeding invertebrates: 1) suspension-feeder systems formed by both invasive and native species at the scale of a regional habitat gradient (“invasive versus native”), and 2) suspension-feeding as one of the trophic types of invasive invertebrate groups at the basin wide scale.
MATERIALS AND METHODS Study Sites Along the Southeastern Baltic Environmental Gradient We considered suspension-feeder systems in seven habitats along an environmental gradient in the southeastern part of the Baltic Sea (Fig. 1). The Curonian Lagoon is a large shallow (mean depth 3.8 m) coastal water body connected to the Sea by a narrow (400 m) strait (Klaipeda port area), which is artificially deepened down to 14 m. In the open sea, the depths ranged from ca. 5 m in the coastal zone down to 80 m in the halocline area (Table 1). In the Lagoon, there is a gradient in mean annual salinity from the Klaipeda Strait (STR, Fig. 1) through the northern part (LAG) towards the central area (DEL). The later is under strong freshwater influence of the Nemunas River. Episodic inflows of the sea water cause irregular rapid (hours-days) salinity fluctuations in the Strait and to a less extent, in the
223 northern part of the Lagoon (Daunys 2001 and references therein). In the Sea, the salinity is relatively uniform and stable down to approximately 55 m, increasing in the halocline zone (ca. 55-80 m). Temperature range is most variable in the Curonian Lagoon and more stable in the halocline area. In summer, the coastal marine areas (COS, COH) are within a warm upper layer of water, while the intermediate depth zone (INT) lays beneath the summer thermocline (ca. 25-30 m) within the cold intermediate water layer (Olenin 1997a and references therein). In winter, the Lagoon (except the strait) is covered by ice. In the sea the ice occurs only as a narrow (less than several tens of m) stripe along the shoreline.
Fig. 1. Study sites in the southeastern part of the Baltic Sea: DEL (delta) – the central part of the Curonian Lagoon in front of the Nemunas River mouth; LAG (Lagoon) – the northern part of the Lagoon; STR (strait) – the outlet of the Curonian Lagoon (Klaipeda Strait); COS (coastal soft bottoms); COH (coastal hard bottoms); INT (intermediate depth zone); HAL (halocline area).
The main bottom sediments in the Lagoon g are sand and silt, on sites with shell deposits p (mainly ( y of invasive bivalve Dreissena ppolymorpha y p and native ggastropods of the ggenus Valvata). In the Klaipeda strait, the bottom sediments are greatly influenced by constant dredging for the waterway
224 maintenance. In the Sea, there is a well defined depth dependent gradient in distribution of bottom substrates, from sand (COS, Fig. 1) and mixture of stones and gravel (COH) in the coastal zone to silt in the intermediate zone (INT) and mud with clay in the halocline area (HAL). Wave exposure is the most important factor shaping stony and sandy bottom biotopes and bottom macrofauna communities in the coastal zone (COS, COH), where the wave height may reach up to 8 m. The waves and near-shore currents transport sand and gravel, that cause strong abrasive effects on benthic organisms in the uppermost part (<10 m depth) of the coastal slope (Olenin 1997b, and references therein). Eutrophication is recognized as the most serious environmental problem in the Curonian Lagoon and in the coastal areas of the southeastern Baltic (Olenina and Olenin 2002). Heavy blue-green algae blooms are usual phenomena in summer and early autumn, especially in the Lagoon. However, due to intensive mixing of water there is no oxygen deficiency in all studied sites except the halocline area, where the oxygen content drops sharply from 6-9.5 (saturation 70-100%) to 2 ml/l (saturation <20 %) causing permanent hypoxia in the lower part of the halocline (Olenin 1997a). Table 1. Environmental changes along the salinity and depth gradients from the Curonian Lagoon to the halocline area of the Baltic Sea*. The study sites abbreviations shown in Fig. 1.
DEL
Depth range, m 1-3
Salinity range, PSU <0.5
Temperature range, °C 0-24
LAG
1-3
0.0-3.0
0-24
STR
5-14
0.5-7.5
0-22
COS
5-30
6.0-8.0
COH
5-30
INT HAL
Study site
Bottom substrate
Wave exposure
Anthropogenic disturbance
Sand, silt, shell deposits Sand, silt, shell deposits Sand, moraine clay, artificial hard substrates
Moderate
Eutrophication
Weakmoderate Weak
Eutrophication
0-20
Sand
6.0-8.0
0-20
30-55
7.0-8.8
0-11
Stones, gravel Silt
55-80
8.0-10.5
3-6
Mud, clay
Strongmoderate Strongmoderate Weaknone None
Eutrophication, dredging, industrial and municipal wastes
Oxygen deficiency
* Based on Olenin 1997a; Daunys 2001 and references therein
Thus, there were two well defined environmental gradients in the study area: one related to the salinity change, from the fully lacustrine or riverine in the Lagoon’s delta area to the mesohaline conditions of the open
225 sea habitats, and another one associated with the depth dependent changes in hydrodynamic activity and bottom substrates, from the coastal areas down to the halocline zone. Sampling of Benthic Macrofauna The quantitative data used for the analysis of structural changes in suspension-feeder systems along the environmental gradient were extracted from the database on benthic macrofauna of the Curonian Lagoon and the southeastern part of the Baltic Sea (Olenin and Daunys, unpublished). In the Lagoon, the field material has been collected in May and October-November between 1980 and 2001 using a 0.025 m² Petersen-type grab. In the Baltic, collections were made in April and August between 1981 and 2002 using the 0.1 m² Van Veen type grab. On stony bottoms, the samples were obtained by scraping of macrofauna from the measured surface of boulders lifted by the grab or a SCUBA diver. All samples were washed through a 0.5 mm mesh sieve and preserved with 4 % formalin neutralized with NaHCO3. Further treatment of material was performed according to a standard procedure (HELCOM 1988). All groups of suspension-feeders were identified to species level. Biomass was determined as formalin wet weight (g m-2). A total of 549 samples were analyzed, quantitative data were averaged for each of the habitats. Trophic Classification of Benthic Invertebrates Trophic classification of benthic invertebrates found in the Baltic Sea was based on works of Turpaeva (1953), Fauchald and Jumars (1979), Järvekülg (1979), Kuznetzov (1980), Lee and Swartz (1980), TsikhonLukanina (1987) and Pearson (2001) with modifications discussed in our earlier papers (Olenin 1997a, c; Olenin and Leppäkoski 1999). Some species (e.g. polychaetes Nereis diversicolorr and Pygospio elegans, gastropods Valvata piscinalis and Viviparus viviparus, bivalve Macoma balthica, etc) are known to use more than one feeding mode, e.g. to get particles from the water column (by filtering or using mucus nets) and also collect detritus from the surface of the bottom sediments (Fauchald and Jumars 1979; Kuznetzov 1980; Tsikhon-Lukanina 1987). These species were defined as “facultative suspension-feeders” to distinguish them from those which exclusively suspension-feed, i.e. “obligatory suspension-feeders”, such as the bivalve Mytilus edulis trossulus, the barnacle Balanus improvisus or the bryozoan Electra crustulenta. In our analysis we considered only true benthic dwelling species, excluding the mobile nektobenthic crustaceans (such as amphipods Pontoporeia femorata, Bathyporeia pilosa, Chaetogammarus ischnus and Corophium volutatorr and the mysids Limnomysis benedeni and Paramysis lacustris).
226 To estimate the proportion of obligatory and facultative suspensionfeeders among invaders, we used recently published lists of invasive species in the Baltic Sea including Kattegat (Leppäkoski and Olenin 2000b; Baltic Sea Alien Species Database 2003), North Sea including Skagerrak (Reise et al. 2002), Black Sea including the Sea of Azov (Gomoiu et al. 2002) and the Caspian Sea (Aladin et al. 2002). Trophic classification of these species was based on the same principles as those used in the Baltic Sea. Statistical Analysis The similarity analysis of the suspension-feeder species composition in different habitats along the southeastern Baltic environmental gradients was performed using multidimensional scaling (MDS) procedure in Primer software (Plymouth Marine Laboratory, Clarke and Warwick 1994).
RESULTS Changes in Suspension-feeder Systems Along the Salinity and Depth Gradients in the Southeastern Baltic Sea
Species Composition The suspension-feeding benthic macrofauna in the Curonian Lagoon and the southeastern Baltic Sea was represented by 30 species: 17 of them were defined as obligatory and 13 as facultative (Table 2). All polychaetes, hydroids, crustaceans, bryozoans and four bivalves (Mytilus edulis trossulus, Cerastoderma lamarcki, Mya arenaria andd Macoma balthica) are euryhaline marine or brackish water species, while the rest of bivalves and all gastropods are of freshwater origin. There are six invasive species within the studied systems: three obligatory (Cordylophora caspia, Balanus improvisus, Dreissena polymorpha and M. arenaria) and two facultative (Marenzelleria viridis and Lithoglyphus naticoides) suspensionfeeders. The highest total species richness within the studied suspension-feeder systems was found in the northern part of the Lagoon (LAG, 20 species) and in the delta area (DEL, 19 species), mainly due to presence of the freshwater bivalves and gastropods (Fig. 2). In these habitats, the only brackish water suspension-feeders were the invasive species: the hydrozoan C. caspia and the polychaete M. viridis. The number of invasive species in these two habitats was comparatively low: 3 - 4 species. The highest ratio of invasive to native species was in the Klaipeda strait (STR): 5 to 6, respectively. In the marine habitats (COS, COH and INT) the number of invasive species was the same as inside the Lagoon (3 – 4 species), while in
227
Sabellariidae CRUSTACEA Cirripediae GASTROPODA Bithyniidae Lithoglyphidae Valvatidae
Viviparidae BIVALVIA Dreissenidae Mytilidae Cardiidae Myidae Tellinidae Unionidae
Sphaeriidae
F F F O O
Balanus improvisus
O
Bithynia tentaculata B. leachi Lithoglyphus naticoides Valvata klinensis V. piscinalis V. pulchella V. trochoidea Viviparus viviparus V. contectus
F F F F F F F F F
Dreissena polymorpha Mytilus edulis trossulus Cerastoderma lamarcki Mya arenaria Macoma balthica Unio tumidus U. pictorum Anodonta cygnea A. piscinalis Sphaerium corneum S. nitidum Pisidium amnicum Euglesa sp.
O O O O F O O O O O O O O
Electra crustulenta
O
X
DEL
Nereis diversicolor Pygospio elegans Marenzelleria viridis Fabricia sabella Manayunkia aestuarina
LAG
O
STR
Cordylophora caspia
COS
O F
COH
HYDROZOA Clavidae POLYCHAETA Nereidae Spionidae
Species
INT
TAXON/Family
HAL
Table 2. Composition of native and invasive (highlighted) obligatory (O) and facultative (F) benthic suspension-feeders in the southeastern Baltic habitats (HAL-DEL, see Fig. 1)
X
X
X
X
X
X X X
X X X X
X X X X X
X X X
X
X
X
X
X
X
X
X X
X X X X
X X X X
X X
X
X
X
X X X X X X X X X
X
X
X X X X X X X X
X X X X X X X X
X X
BRYOZOA Membraniporidae
X X X X X X X X X
X
228 Invasive
Native
No. of species
20 15 10 5 0 HAL
INT
COS
COH
STR
LAG
DEL
Fig. 2. Changes in number of invasive and native suspension-feeding species along the southeastern Baltic environmental gradient (HAL-DEL, see Fig. 1).
the halocline zone the suspension-feeding organisms were represented by two native species only: M. balthica and P. elegans, both being facultative suspension-feeders. Due to prevalence of the freshwater suspension-feeders LAG and DEL habitats clearly differed in species composition from the rest of the studied sites, where only marine organisms occurred (Fig. 3). The presence of marine suspension-feeders in the Klaipeda Strait determined the affinity of this habitat to the COS, COH and INT areas. Finally, the halocline zone showed clear dissimilarity due to absence of the most of suspension-feeders common in the upper marine benthic areas. Biomass Within the Lagoon, the biomass of suspension-feeders showed a clear decline trend from the delta area (1037±353 and 139±48 g/m2, invasive and native species, respectively) towards the strait (7.4±4.2 and 2.7±0.9 g/m2) (Fig. 4). The invasive species dominated, composing from 73 (in the strait) to 88% (delta) of total suspension-feeder biomass. The zebra mussel D. polymorpha was the biomass dominant species in the Lagoon (LAG and DEL areas), while the native unionids were subdominants. In the Klaipeda Strait, the biomass of the suspension-feeders was low; besides the invasive species
229 ((B. improvisus and M. arenaria) two native suspension-feeders (M. edulis trossulus and M. balthica) were dominant on sites. All four species are of marine origin, having no stable populations in the strait. In contrast, their biomass in the marine areas was higher by one-two orders of magnitude. Within the marine suspension-feeder systems, the highest biomass (925 and 97 g/m2, native and invasive species, respectively) was characteristic for the coastal hard bottoms, dominated mostly by the blue mussel M. edulis trossulus with the barnacle B. improvisus as the subdominant.
Fig. 3. The MDS plot based on suspension-feeding species composition in the study sites along the southeastern Baltic environmental gradient (HAL-DEL, see Fig. 1).
The Baltic tellin, M. balthica, a facultative suspension-feeder, was the most important biomass dominant species (up to 95% of total) throughout all soft bottom habitats, from the coast down to the halocline area. The invasive soft clam M. arenaria, an obligatory suspension-feeder, was the subdominant in the COS area; however it did not occur in the deeper (INT, HAL) habitats. Another obligatory suspension-feeder dwelling in the soft bottoms, the Lagoon cockle C. lamarcki also was found only in the coastal areas. SUSPENSION-FEEDERS AND OTHER TROPHIC TYPES OF INVASIVE BENTHIC MACROFAUNA
The total number of invasive benthic invertebrate species presently known from the Caspian, Black, Baltic and North Seas is 94. Their feeding types are as diverse as that of native macrofauna, including suspension and deposit feeders, herbivores, omnivores and predators. The invasive suspension-feeding macrofauna comprised 64 species: 41 obligatory and 23 facultative suspension-feeders (Table 2, Fig. 5).
230 Invasive
Native
1000
100
10
Biomass, g/m m2
10000
1 HAL
INT
COS
COH
STR
LAG
DEL
Fig. 4. Changes in biomass of invasive and native suspension-feeding species along the southeastern Baltic environmental gradient (HAL-DEL, see Fig. 1).
Table 2. Species richness in major taxonomic groups of invasive obligatory (O) and facultative (F) suspension-feeders in the Caspian, Black, Baltic and North Seas*. Caspian Taxon Anthozoa Hydrozoa Polychaeta Crustacea Gastropoda Bivalvia Ectoprocta Entoprocta Tunicata TOTAL
O
Black
F
O
3 1 2 2 1 1
1 2
Baltic F
3 2 2 2
O
North F
O
5 4 1 1
1 1 6 1
2 5 2
4 4 3 4 1 3 3
O
5 1 1 1
1 1
9
Total** F
4
12
6
16
1 9
23
8
4 12 5 4 1 8 4 2 1 41
F
11 6 2 4
23
*Based on: Leppäkoski and Olenin 2000b; Reise et al. 2002; Gomoiu et al. 2002; Aladin et al. 2002; Baltic Sea Alien Species Database 2003. ** Species common for two or more regions were counted once.
231 The most diverse groups among the invasive suspension-feeders were cnidarians (both anthozoans and hydroids, together comprising 16 species) and polychaetes (16 species), followed by bivalves (12 species) and crustaceans (10 species). In the later group, all obligatory suspension-feeders were represented by barnacles only ((Balanus amphitrite, B. improvisus, B. eburneus andd Elminius modestus). Seventeen invasive suspension-feeder species were common to at least two of the studied regions; four species (the barnacle Balanus eburneus, the mud snail Potamopyrgus antipodarum, bivalves Crassostrea gigas and M. arenaria) were found in three seas and one species ((B. improvisus) occupied all four regions. The total number of invasive suspension-feeders was the highest in the North Sea (31 species), 17 of them did not occur in the other
Fig. 5. The share of obligatory (OSF) and facultative (FSF) suspension-feeder species among other trophic types of invasive benthic invertebrates in various European Seas (n = total number of species). Other feeding types included predation, deposit feeding and grazing on macrophytes.
three regions. The number of such “unique” species in the Black and in the Baltic Sea was the same (11), while there were fewer (8) in the Caspian Sea. Although the share of suspension-feeders varied slightly from 61% in the Baltic to 72% in the North Sea, the general rule was that this feeding type
232 was the most common among the invasive bottom macrofauna species (Fig. 5).
DISCUSSION Native and Invasive Suspension-feeders: Similar and Dissimilar Distribution Patterns Along the Environmental Gradients Many works on trophic distribution along environmental gradients have shown that the diversity and biomass of the suspension-feeding macrofauna are the highest in coastal areas with active hydrodynamics, and both parameters gradually decline with increasing depth (e.g. Kuznetzov 1980; Dame 1996; Boaventura et al 1999 and references therein). In our study area, this model of suspension-feeder distribution along the depth gradient was true for both native and invasive species (Fig. 2 and 4, Table 2). For instance, the most indicative group, the obligatory suspension-feeders (either native or invasive) reached the highest biomass only in the Lagoon and coastal habitats, while in the intermediate depth zone they were insignificant and absent in the halocline area. In contrast to the depth/hydrodynamics related distribution there was no common pattern in native and invasive suspension-feeder systems in relation to the salinity gradient. The native species richness is essentially higher in the nearly lacustrine conditions of the Curonian Lagoon (DEL and LAG sites) than in the mesohaline environment of marine habitats. This result fits well into the classical Remane (1934) curve showing that within the gradient from the fresh to fully saline marine waters the minimum in species number corresponds to the salinity range of 5-8 PSU. In our case, this matches the salinity in COS, COH and INT habitats. Despite the higher salinity in the HAL habitat, the number of native suspension-feeder species was the lowest in this deeper zone due to depleted oxygen and weak hydrodynamic activity. Bonsdorff and Pearson (1999) also showed the overall gradual decrease in all functional groups of benthic macrofauna, including suspension-feeding species, from the fully marine Baltic approaches to its inner brackish water parts. Distribution of the invasive suspension-feeder species, nevertheless, did not follow the same pattern. Their highest number was found in Klaipeda Strait, i.e. where the salinity regime is most variable. However, here these species did not form stable populations being dependant on larvae transport either from the Sea or from the Lagoon (Daunys 2001). The number of invasive species in the remaining habitats, except the halocline zone, was nearly the same: 3 - 4 species (Fig. 6).
233
Fig. 6. Number of invasive suspension-feeder species plotted against the number of native suspension-feeder species in the southeastern Baltic benthic habitats (HAL-DEL, see Fig. 1).
This “breach” of the Remane’s rule by the invasive suspensionfeeders may be explained by their ecological plasticity, which is generally characteristic for invaders (e.g. Carlton 1996, Ruiz and Hewitt 2002). In the Baltic Sea, most of the alien benthic invertebrates, including the suspensionfeeders, originate from the brackish waters of the North American east coast, the Ponto-Caspian region and the South-East Asia (Leppäkoski et al. 2002b). Capability to cope with low salinity conditions is the common trait of these mostly estuarine species, and this probably explains why their distribution within the salinity range 0 – 8 PSU did not follow the native species pattern. Studies on invasive suspension-feeder systems across a broader salinity interval may determine if these generalizations are also true for other brackish water seas. Invasive Versus Native: Occupation of Empty Niches and Species Displacement We noticed a clear shift in the biomass dominance from the invasive species in the Curonian Lagoon (D. ( polymorpha) to the native suspensionfeeders in the Baltic Sea (M. edulis trossulus, M. balthica). D. polymorpha is known to inhabit fresh waters, where the niche of sessile byssate suspensionfeeders is usually unoccupied (Orlova 2002; Burlakova et al. 2005). In the oligohaline stable salinity coastal waters of the north-eastern Baltic, the zebra mussel may co-occur with the native M. edulis trossulus on sites, though it does not displace the later (Kotta and Møhlenberg 2002).
234 Another invasive suspension-feeding species within our study area, which was able to fill an empty niche, was the barnacle B. improvisus. This is the only sessile benthic animal which is able to withstand harsh conditions of constant wave action and abrasive effect of sand in the uppermost part of the coastal slope (Olenin and Daunys, unpublished). The barnacles form seasonal settlements on large boulders (at the depth < 1.5 m), which usually disappear in winter time. The third example is the hydroid Cordylophora caspia, which produces dense bush-like colonies on firm substrates both in the Lagoon and in the Sea. In spite of these three examples, our results do not support theories predicting that species-rich communities should be more resistant to invasions because of a more complete utilization of resources (see Ruiz and Hewitt 2002 for a comprehensive review). The number of invasive suspension-feeders was the same in the comparatively species-rich communities of the Curonian Lagoon and in the species-poorer Baltic marine communities. The significance of species invasion for functional changes might be much higher in species poor communities than in species rich ones. Variety of species of the same functional guild within a system (functional redundancy) may be a measure of the overall system’s ability to perform a given function. Extinction of a species in the Baltic Sea may cause the loss of entire process (e.g. biodeposition), while the same event, e.g., in the North Sea would mean only a minor shift in the species composition and overall role of the functional group. The invasion of new species in such species poor systems as the Baltic Sea in many cases means the increase in functional redundancy and denser “packing” of available niches. In the Baltic Sea, invasions have caused serious structural and functional changes, especially in its coastal Lagoons and inlets (Olenin and Leppäkoski 1999), however they have not yet triggered extinction of native species. INVASIVE SUSPENSION-FEEDERS AND FUNCTIONAL HOMOGENIZATION OF AQUATIC BIOTA Asmus and Asmus (2005) showed that the share of suspensionfeeding species in total species number is much lower than that of non suspension-feeders on different scales from global to local ones, suggesting a minor influence of suspension-feeding type to total diversity of systems. Our study, however, showed an opposite result in regard to the invasive benthic macrofauna. In all major European brackish water bodies (Baltic, Black and Caspian seas) as well as in the fully saline North Sea, suspension-feeding was the prevalent trophic type among benthic invertebrate invaders. It is unclear, how the species ability to filter-feed may promote their invasion success. Possibly, suspension-feeding as the most optimal foraging
235 strategy (Gili and Coma, 1998) adds to other common traits of invaders: ecological plasticity, profitable reproductive strategy, ability to use different substrates, etc (Ruiz and Hewitt 2002 and references therein). Increased pelagic food through eutrophication may also promote the relative success of suspension-feeder invaders, especially in the coastal areas, which are mostly exposed to new introductions. Further studies on these observations may help to better identify potential successful benthic invaders. Olden et al. (2004) consider ecological and evolutionary consequences of biotic homogenization caused by invasions, including the functional ones. Our findings indicate that the suspension-feeders are the prevalent group among invasive benthic invertebrates and assume the vector of that functional homogenization. If the rate and the scope of invasions remain at the recent high level then the role of suspension-feeding as a trophic type will grow in the future.
ACKNOWLEDGEMENTS This study was supported by the EU FW6 project EVK3-CT-200100065 CHARM “Characterization of the Baltic Sea Ecosystem: Dynamics and Function of Coastal Types” and EU FW6 IP 506675 ALARM “Assessing Large-scale environmental risks with tested methods”. We thank Erkki Leppäkoski (Åbo Akademi University, Turku, Finland) and an anonymous reviewer for their comments and suggestions to improve this manuscript.
REFERENCES Aladin NV Plotnikov IS Filippov AA 2002 Invaders in the Caspian Sea In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 351-359 Asmus H Asmus RM 2005 Significance of suspension-feeder systems on different spatial and temporal scales In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 199-220 Baltic Sea Alien Species Database 2003 S Olenin E Leppäkoski D Daunys (Eds) [online] URL: http://www.ku.lt/nemo/mainemo.htm (download 2003-10-02) Boaventura D Fonseca LC Teles-Ferreiraa C 1999 Trophic structure of macrobenthic communities on the Portuguese coast. A review of lagoonal, estuarine and rocky littoral habitats. Acta Oecologica 20: 407-415 Bonsdorff E Pearson TH 1999 Variation in the sublittoral macrozoobenthos of the Baltic Sea along environmental gradients: a functional-group approach. Austral J Ecology 24: 312326 Burlakova L Karatayev AY Padilla DK 2005 Functional changes in freshwater benthic communities after Dreissena polymorpha (Pallas) invasion In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 263-276 Carlton JT 1996 Pattern process and prediction in marine invasion ecology. Biol Cons 78: 97106
236 Clarke KR Warwick RM 1994 Change in Marine Communities. An Approach to Statistical Analysis and Interpretation. Plymouth Marine Laboratory, 144p Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Daunys D 2001 Patterns of the Bottom Macrofauna Variability and its Role in the Shallow Coastal Lagoon. Summary of Doctoral Dissertation. Klaipeda University, 43 p Fauchald K Jumars P 1979 The diet of worms: a study of polychaete feeding guilds. Oceanogr Mar Biol Ann Revv 17: 193-284 Gili J-M Coma R 1998 Benthic suspension feeders: their paramount role in littoral marine E 13: 316-321 food webs. TREE Gomoiu M-T Alexandrov T Shadrin N Zaitsev YP 2002 The Black Sea - a recipient, donor and transit area for alien species. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 341-350 HELCOM 1988 Guidelines for the Baltic monitoring program for the third stage. No. 27D. Part D. Biological determinants Järvekülg A 1979 Bottom Fauna of the Eastern Part of the Baltic Sea. Valgus Press, Tallinn, 382 p (In Russian) Karatayev A Burlakova L Padilla D 2002 Impacts of Zebra Mussels on Aquatic Communities and their role as ecosystem engineers. In: Invasive Aquatic Species of Europe – Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 433-446 Kotta J Møhlenberg F 2002 Grazing impact of Mytilus edulis and Dreissena polymorpha in the Gulf of Riga, Baltic Sea estimated from biodeposition rates of algal pigments. Ann Zoo Fenn 39: 151-160 Kuznetzov AP 1980 Ecology of Bottom Communities of Shelf Zones of the World Ocean. Trophic structure. Nauka Press, Moscow, 274 p (In Russian) Lee H Swartz RC 1980 Biological processes affecting the distribution of pollutants in marine sediments. Part II. Biodeposition and Bioturbation. In: Contaminants and Sediments, Vol. 2. RA Baker (Ed), Ann Arbor Science Publ., Ann Arbor, pp 555-606 Leppäkoski E Gollasch S Gruszka P Ojaveer H Olenin S Panov V 2002b The Baltic – A sea of invaders. Can J Fish Aq Sc 59: 1175-1188 Leppäkoski E Gollasch S Olenin S 2002a Alien Species in European Waters. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 1–6 Leppäkoski E Olenin S 2000a Xenodiversity of the European brackish water seas: the North American contribution. In: Marine Bioinvasions. Proc 1stt Natl Conf, J Pederson (Ed), Massachusetts Institute of Technology, Cambridge, pp 107-119 Leppäkoski E Olenin S 2000b Non-native species and rates of spread: lessons from the brackish Baltic Sea. Biological Invasions 2(2): 151-163 Ojaveer H Leppäkoski E Olenin S Ricciardi T 2002 Ecological impact of Ponto-Caspian invaders in the Baltic Sea, European inland waters and the Great Lakes: an interecosystem comparison. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 412-425 Olden JD Poff NL Douglas MR Douglass ME Fausch K.D 2004 Ecological and evolutionary consequences of biotic homogenization. TREE 19: 18-24 Olenin S 1997a Benthic zonation of the Eastern Gotland Basin. Neth J Aq Ecoll 30 (4): 265282 Olenin S 1997b Marine benthic biotopes and bottom communities of the southeastern Baltic shallow waters. In: Proc. 30 European Marine Biology Symposium LE Hawkins S
237 Hutchinson AC Jensen JA Williams M Sheader (Eds). University of Southampton Press, Southampton, UK, pp 243–249 Olenin S 1997c Comparative study of the southeastern Baltic coastal zone and the Curonian Lagoon bottom communities. In: Proc. 13 Baltic Marine Biologists Symposium, A Andrushaitis (Ed), University of Latvia Press, Riga, pp 151-159 Olenin S Leppäkoski E 1999 Non-native animals in the Baltic Sea: alteration of benthic habitats in coastal inlets and lagoons. Hydrobiologia 393: 233-243 Olenina I Olenin S 2002 Environmental problems of the southeastern coast and the Curonian Lagoon. In: Baltic Coastal Ecosystems. Structure, Function and Coastal Management. E Schernewski U Schiewer (Eds), Springer-Verlag, Berlin, pp 149-156 Orlova M 2002 Dreissena polymorpha: evolutionary origin and biological peculiarities as prerequisites of invasion success. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch, S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 127-134 Pearson TH 2001 Functional group ecology in soft-sediment marine benthos: the role of bioturbation. Oceanogr Mar Biol: An Annual Revv 39: 233-267 Reise K Gollasch S Wolff WJ 2002 Introduced marine species of the North Sea coasts. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 260-266 Remane A 1934 The Brackishwater fauna. Zoologische Anzeiger 16: 327-359 (In German) Ruiz GM Hewitt CL 2002 Toward understanding patterns of coastal marine invasions: A Prospectus. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 529-547 Tsikhon-Lukanina EA 1987 Trophology of Aquatic Mollusks. Nauka Press, Moscow 176 p (In Russian) Turpaeva E 1953 Feeding and feeding groups of marine benthic invertebrates. Doklady Akademii Nauk SSSR 65: 503-512 (In Russian) Zettler ML 1996 Successful establishment of the spionid polychaete, Marenzelleria viridis (Verrill, 1873), in the Darss-Zingst estuary (southern Baltic) and its influence on the indigenous macrozoobenthos. Arch Fish Mar Res 4: 273-284
CONTRASTING DISTRIBUTION AND IMPACTS OF TWO FRESHWATER EXOTIC SUSPENSION FEEDERS, DREISSENA POLYMORPHA AND CORBICULA FLUMINEA
Alexander Y. Karatayev1, Lyubov E. Burlakova1, and Dianna K. Padilla2 1
Department of Biology, Stephen F. Austin State University, Nacogdoches, TX, USA Department of Ecology and Evolution, Stony Brook University, Stony Brook, NY, USA
2
Abstract: Dreissena polymorpha and Corbicula fluminea are among the most aggressive freshwater invaders world wide, and often dominate water bodies they invade. They occur in similar habitats, however, their tolerance and preference for certain characteristics of freshwaters differ in important ways, and they can have different impacts on the environments they invade. We identify similarities and contrast differences between these species, and highlight important questions yet to be addressed, including: the ability to link short-term laboratory findings to large scale and long-term effects of invasion, the consequences of invasion by both species together rather than considering each in isolation, and identification of local versus system-wide effects when these non-native ecosystem engineers invade. Keywords: Zebra mussel, Asiatic clam, invasive species, physiological limits, habitats, food selectivity, filtration rate, predators, parasites, ecosystem impacts
INTRODUCTION The role of marine and estuarine bivalves as ecosystem engineers has long been recognised (reviewed in Dame 1993). However, most of this research has focused on native species, in environments where they are dominant and clearly play important roles. Although historically the role of estuarine and marine bivalves concentrated on pristine and relatively undisturbed habitats, recently, the focus has shifted to areas where they are over harvested or are lost due to disease or human disturbance, and the resultant dramatic changes in ecosystems as due to the loss of these important engineering species. The opposite situation occurs in fresh waters when suspension-feeding bivalves invade and cause dramatic changes in an environment. 239 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 239–262. ©2005 Springer. Printed in the Netherlands.
240 Invasive species are currently one of the greatest environmental threats around the world, and the total estimated annual cost of their impact in the US alone exceeds $125 billion (Pimentel et al. 2000). The zebra mussel ((Dreissena polymorpha) and the Asiatic clam (Corbicula fluminea) are among the most aggressive freshwater invaders worldwide (Morton 1979, Karatayev et al. 2002). These two species are not only extremely aggressive invaders, often dominating water bodies they invade, they are also very effective ecosystem engineers, altering both ecosystem structure and function (Phelps 1994, Karatayev et al. 1997, 2002, McMahon 1999). They change existing and provide new habitats for other organisms, affect trophic interactions and the availability of food for both pelagic species and other benthic species, and affect the rates of ecosystem processes, including mineralization of nutrients, oxygen availability and sedimentation rates (Mattice 1979, Morton 1997, Hakenkamp and Palmer 1999, Karatayev et al. 1997, 2002). As a consequence, direct feedbacks are created with other species that interact with or are impacted by these invaders, as well as indirect feedbacks through food chains, disturbance, succession, and other longer-term community and ecosystem processes. D. polymorpha is native to the fresh and brackish waters of the Caspian and Black Sea drainage basins (Mordukhai-Boltovskoi 1960, Starobogatov and Andreeva 1994). In the late 1700’s - early 1800's, zebra mussels spread through canals built for commerce to connect the Black and Caspian Seas with the Baltic (reviewed in Karatayev et al. 2003b). The range of D. polymorpha in Europe is still expanding, in 1993/4 zebra mussels were found in Ireland (Minchin 2000). Zebra mussels were first discovered in North America in Lake St. Clair in the mid-1980s (Hebert et al. 1989). Since their introduction, zebra mussels have spread throughout the Great Lakes, the Hudson, Ohio, Illinois, Tennessee, Mississippi and Arkansas rivers, as well as other lakes and rivers in 21 states and the provinces of Quebec and Ontario in Canada (McMahon and Bogan 2001). C. fluminea, native to Southeast Asia, Australia, and Africa, has successfully invaded North American freshwaters over the last 60 years (reviewed in McMahon 1999). First found in 1938 in the Columbia River, Washington, it subsequently spread throughout 36 continental states, Hawaii, and northern and central Mexico. C. fluminea was introduced into South America and Europe in the late 1970s (reviewed in McMahon 1999). In Europe, C. fluminea is in France, Portugal, Spain, Germany, Belgium, and the Netherlands. In 1998 C. fluminea was found in the U.K. for the first time (Howlett and Baker 1999). Both C. fluminea and D. polymorpha can create large populations in waterbodies they invade. However, C. fluminea are solitary, and burrow in sediments (Britton and Murphy 1977). In contrast, D. polymorpha attaches by
241 byssus to hard substrates and each other, often in high densities, and can create new 3-D habitat, providing not only food, but shelter for benthic invertebrates (reviewed in Karatayev et al. 1997, 2002). Moreover, although both species are suspension feeders, C. fluminea also feeds directly from the sediment (Reid et al. 1992). Therefore these two species are likely to have different ecosystem effects in waterbodies they invade. Our goal is to compare and contrast potential ecosystem impacts of these two powerful suspensionfeeders in Europe and North American waterbodies of various types and draw attention to important questions that are yet to be studied.
PHYSICAL ENVIRONMENT Dreissena and Corbicula occur in similar habitats; however, they differ in some important ways in their tolerance and preference for certain physical characteristics of freshwaters. Although most studies of physical tolerances and preferences are short-term laboratory experiments, the patterns described below are usually confirmed with field-based observations and experiments. Salinity Both D. polymorpha and C. fluminea can colonize brackish and fresh waters, however these two species differ in their upper salinity limit (Table 1). There are several subspecies of D. polymorpha, each with a different tolerance to salinity (reviewed in Karatayev et al. 1998). Although as a species D. polymorpha has a wide salinity tolerance, from fresh to 18‰, D. p. polymorpha, the subspecies that invaded Western Europe and North America, lives in salinities < 6.2 ‰. D. p. andrusovi populates areas of the Caspian Sea where salinities range from 2 - 12‰ (Shkorbatov et al. 1994). D. p. obtusicarinata and D. p. aralensis were the dominant benthic species in the Aral Sea and had an upper salinity limit of 18.4‰ (Lyakhnovich et al. 1994) and 17.6‰ (Khusainova 1958), respectively. By 1980, after an increase in salinity, both subspecies disappeared from the Aral Sea (Starobogatov and Andreeva 1994). In contrast to D. p. polymorpha, C. fluminea has a much higher salinity tolerance, up to 17‰ (Table 1). Temperature The lower temperature limit is slightly lower for D. polymorpha (0qC) than for C. fluminea (2oC) (Table 1), which will restrict the northern distribution of Corbicula. In regions with winter temperatures lower than 2oC, C. fluminea is usually restricted to areas heated by thermal power plants (Graney et al. 1980, French and Schloesser 1996). During warmer times of the year C. fluminea populations may expand out of heated water areas, however,
242 when winter water temperatures drop below 2oC most of the clams in unheated areas die (Graney et al. 1980, French and Schloesser 1996). 10 11qC is the minimal temperature for growth and development in both D. polymorpha and C. fluminea (Table 1). The upper temperature limit, however, is substantially higher for C. fluminea (37oC) than for D. polymorpha (33oC). Therefore, C. fluminea may spread much further south than D. polymorpha, while zebra mussels may colonize areas that are too cold for the Asiatic clam. Table 1. Environmental limits for Dreissena polymorpha and Corbicula fluminea Dreissena Corbicula References polymorpha fluminea Upper salinity limit (‰) 4 - 6.2 10 - 17 Reviewed in Karatayev et al. 1998, Evans et al. 1979 Lower temperature limit 0 2 Luferov 1965, Mattice 1979, (oC) Rodgers et al. 1979 Minimal temperature for 10 - 11 10 - 11 Reviewed in Karatayev et al. 1998, growth and development Joy 1985, Fritz and Lutz 1986, (oC) Boltovskoy et al. 1997 Upper temperature limit 31.5 - 33 36 - 37 Reviewed in Karatayev et al. 1998, (oC) Dreier and Tranquilli 1981, Britton and Morton 1982 Lower pH limit 7.3 - 7.4 5.6 Ramcharan et al. 1992, Burlakova 1998, Kat 1982 Density - rocky substrates 1580 - 5540 0 - 377 Reviewed in Karatayev et al. 1998, (m-2) Abbott 1979, Leff et al. 1990 Density - sand, silty sand 211 - 3930 54 - 1215 Reviewed in Karatayev et al. 1998, substrates (m-2) Abbott 1979, Belanger et al. 1985, Leff et al. 1990 Density - shelly substrates 1081 43 Reviewed in Karatayev et al. 1998, (m-2) Karatayev et al. 2003a Density - submerged 1246 - 3545 0 Reviewed in Karatayev et al. 1998, macrophytes (m-2) Karatayev et al. 2003a Density - silt (m-2) 0 - 64 3.6 Reviewed in Karatayev et al. 1998, Karatayev et al. 2003a Density in lakes (m-2) 6 - 3453 39 - 1278 Stanczykowska and Lewandowski 1993, Burlakova 1998, Beaver et al. 1991 Density in reservoirs (m-2) 838 - 3150 30 - 796 Lyakhov and Mikheev 1964, Lvova 1977, Burlakova 1998, Abbott 1979, Dreier and Tranquilli 1981, Karatayev et al. 2003a Density in rivers (m-2) 7 - 138 315 - 3206 Reviewed in Karatayev et al. 1998, Rodgers et al. 1979, Belanger et al. 1985, Boltovskoy et al. 1997 Density in streams (m-2) Usually absent 54 - 974 Lyakhnovich et al. 1994, Leff et al. 1990, Arias, 2004 Density in canals (m-2) 40000 2255 Reviewed in Karatayev et al. 61000 16688 1998, Eng 1979, Marsh 1985 Factors
243 pH D. polymorpha is limited to waters with neutral or alkaline pH < 7.3 (Ramcharan et al. 1992, Burlakova 1998). There are no published data on the pH limits for C. fluminea, however, some studies have found this clam in waters with relatively low pH. Populations of Corbicula are found in the Parana River delta, Argentina, where pH averages 6.9 and ranges from 6.5 7.2 (Boltovskoy et al. 1997), and the Ogeechee River, Georgia (USA), where pH ranges from 6.6 - 7.2 (Stites et al. 1995). However, in Mosquito Creek, Florida (USA) (pH 5.6), shell dissolution may be a major source of mortality for Corbicula over 3 years old (Kat 1982).
Oxygen Both D. polymorpha and C. fluminea are intolerant of even moderate hypoxia (reviewed in Karatayev et al. 1998, McMahon 1999), therefore, both species are usually restricted to littoral and sublittoral zones. They can also be found in well-oxygenated profundal areas (Fast 1971, Karatayev et al. 1998, McMahon 1999).
Substrates One of the main factors that affects the distribution and abundance of D. polymorpha (Zhadin 1946, Lyakhnovich et al. 1994, Karatayev et al. 1998) and C. fluminea (Leff et al. 1990, Karatayev et al. 2003a) is suitable substrate. In most waterbodies rock and sometimes sand can be the most suitable substrate for zebra mussel attachment (Table 1). However, in shallow parts of large lakes and reservoirs, even on suitable substrates, particularly sands, zebra mussels can be limited by water motion (reviewed in Karatayev et al. 1998). Shelly sediments and silty sand can also be suitable substrates for D. polymorpha. In addition, zebra mussels can be extremely abundant on submerged macrophytes. The poorest substrate for zebra mussels is silt. The best substrate for C. fluminea is sand, sometimes mixed with silt or clay (Table 1). Asiatic clams are in much lower densities on rocks and in silt. C. fluminea also usually avoids sediments under beds of submerged macrophytes (Karatayev et al. 2003a). Both species can be very abundant on sand and, silty sand and both avoid pure silt. However C. fluminea usually avoids rocks, the best substrate for zebra mussel attachment. In addition, in contrast to C. fluminea, D. polymorpha may aggregate in high densities on submerged macrophytes.
244 HABITAT Zebra mussels and Asiatic clams can be found in a wide range of types of waterbodies (reviewed in Karatayev et al. 1998, McMahon 1999), however, most work on D. polymorpha has been focused on factors affecting its presence and abundance in lakes. In contrast, most of the research on C. fluminea has been conducted in rivers. This difference may reflect the fact that in general zebra mussels usually form higher densities and play a much more important role in lakes and reservoirs (Lyakhnovich et al. 1994, Karatayev et al. 1998) and Asiatic clams, in contrast, are much more abundant in rivers and even small streams than in lakes (Britton and Morton 1982, McMahon 1983).
Lakes Trophic type affects the probability of finding zebra mussels in a lake. Zebra mussels are found most often in mesotrophic lakes, less often in oligotrophic and meso-oligotrophic lakes, least often in eutrophic lakes, and do not inhabit dystrophic lakes (Karatayev et al. 2003b). However, the highest densities of D. polymorpha are found in eutrophic lakes (Karatayev and Burlakova 1995b). C. fluminea also differ in abundance among lakes of different trophic types. Beaver et al. (1991) found that Asiatic clam abundance in Florida lakes generally increased with trophic state. Clam densities were 39 r 17 m-2 in oligotrophic lakes, 368 r 328 m-2 in mesotrophic lakes, and 1278 r 1047 m-2 in eutrophic lakes. In two hypertrophic lakes the density of clams averaged 198 m-2 (Beaver et al. 1991).
Reservoirs In reservoirs D. polymorpha colonizes all suitable substrates, often at high densities (reviewed in Karatayev et al. 1998). Especially high densities of D. polymorpha are found in reservoirs created by flooding forested areas, where zebra mussels colonize flooded stumps, trunks and branches of trees and brushwood. For example in Kamskoe Reservoir (Russia), the density of D. polymorpha in flooded forest areas was as high as 371,703 m-2 with a biomass density (total wet mass) of 11.4 kg m-2 (Gubanova 1968). C. fluminea may form high densities in certain areas of reservoirs, but their overall average density is usually lower than in lotic waters (Table 1).
245 Rivers In rivers, zebra mussels are usually limited by unidirectional water flow, disturbance due to water flow, suspended sediment, and limited suitable attachment substrates (Karatayev et al. 1998, Schneider et al. 2003). Another factor that reduces zebra mussel densities in rivers is disturbance due to periodic flooding (Lyakhnovich et al. 1994). Constant water flow can make it difficult for Dreissena local populations in rivers to increase in density, as zebra mussel’s planktonic larvae are swept downstream (Schneider et al. 2003). In contrast, C. fluminea larvae primarily crawl away rather than float in the plankton, thus they avoid being swept downstream in the river currents. Upon release from the maternal clam, they can slowly crawl along the bottom and, move upstream as easily as downstream (Britton and Morton 1982). In addition, C. fluminea may alternate filter and pedal feeding (Reid et al. 1992, Hakenkamp et al. 2001) and thus survive during periods of high concentrations of suspended matter that may inhibit their filtering activities. C. fluminea lives on the surface or buries as much as two centimeters below the surface (Britton and Murphy 1977). In contrast to D. polymorpha, they do not depend on hard substrates or stable sediment. Therefore, C. fluminea usually form higher densities in rivers than in lakes or reservoirs (Table 1). The average density of C. fluminea in Nacogdoches Reservoir is 15.6 r 5.3 m-2 (Karatayev et al. 2003a) and in small stream (< 5 m width) flowing from this reservoir is 462 r 133 m-2 (Arias 2004). In contrast, zebra mussels almost never form high densities in upper courses of large rivers and usually do not colonize small rivers and streams.
Canals Canals are distinct from lakes and reservoirs in that there is a constant, unidirectional water current which delivers nutrients and oxygen, and different from rivers because bottom sediments are much more stable and the concentration of suspended matter is much lower than in rivers, particularly during periodic floods. Both D. polymorpha and C. fluminea may have extremely high densities in canals (Britton and Morton 1982, McMahon 1983, Karatayev et al. 1998).
BIOLOGY Food Selectivity - size and quality Many of the effects of zebra mussels and Asiatic clams on freshwater ecosystems are linked to their filtering. They circulate water for respiration
246 and feeding, and remove particles from the water, which are either consumed, or bound as pseudofeces and expelled to the benthos. However, the size range of particles filtered by D. polymorpha is larger than for C. fluminea. The smallest particles that zebra mussel can filter are between 0.4 and 1.3 Pm (Sprung and Rose 1988, Silverman et al. 1995, Roditi et al. 1996) and the maximum particle size is between 750 µm (Ten Winkel and Davids 1982) and 1200 Pm (Horgan and Mills 1997). Although they filter all particles out of the water, they are very selective in which of these particles they consume (Baker et al. 2000). C. fluminea has a similar lower size limit (< 1 µm) for filtered particles (McMahon and Bogan 2001), however their upper size limit is much smaller, about 20 µm (Way et al. 1990). Boltovskoy et al. (1995) found that C. fluminea consume algae with a spherical diameter up to 50 µm, and with the largest dimension up to 170 µm. Both species can effectively remove detritus, bacteria and algae (reviewed in Mikheev 1994, McMahon 1999, Boltovskoy et al. 1995). In addition, D. polymorpha has been reported to filter small zooplankton (reviewed in Mikheev 1994, MacIsaac et al. 1995, Wong et al. 2003).
Filtration Rate Although many researchers have investigated the filtering of D. polymorpha (reviewed in Karatayev et al. 1997) and C. fluminea (Cohen et al. 1984, Lauritsen 1986, Leff et al. 1990, Way et al. 1990, Boltovskoy et al. 1995, Cahoon and Owen 1996, et al.), standardized methodology has not been used, and often experimental setups are not adequately described to permit direct comparisons of results. To compare estimates of filtration calculated by different authors, Karatayev et al. (1997) converted all available literature data to volume of water filtered (mL) per gram of zebra mussel wet total mass (WTM). These common units were chosen because they provide the best correlation with filtration rates across seasons, independent of the reproductive status of mussels (Karatayev 1983). They found a relatively narrow range g of measured filtration rates for D. polymorpha (from 35 to 110 mL g WTM-1 h-1, avg. = 64 mL g WTM-1 h-1), in spite of the fact that these studies were made by different researchers, for different waterbodies, and using different methods. Because very different methods have been used to study C. fluminea as well, it is similarly difficult to compare among g studies. Prokopovich (1969) measured a filtration rate of 20 mL g WTM-1 h-1 for C. fluminea. Similar rates (24.1 mL g WTM-1 h-1) were measured by Cohen et al. (1984). These data suggest that D. polymorpha filter at a much higher rate than C. fluminea. Silverman et al. (1995) found that on a mussel-dry-weight basis D. polymorpha cleared
247 bacteria 30 to 100 times faster than C. fluminea. However, on a gill surface area basis, the rate of bacteria clearance by C. fluminea was greater than that by D. polymorpha (Silverman ett al. 1997). Both species cleared bacteria many times faster than any of six unionid species examined.
Role as a Biofilter Because both zebra mussels and Asiatic clams occur in high densities over large areas, they can filter large volumes of water in short periods of time and deposit vast quantities of pseudofeces on the bottom. D. polymorpha populations have been estimated to filter the volume of water equivalent to that of an entire waterbody in 1.3 to 123 days, depending on the mussel density, mussel biomass and size of the waterbody (Table 2). However, some of these estimates may be suspect and are very dependent on the methods used for estimation and assumptions about mussel densities and size structure. Because C. fluminea populations tend to dominate smaller waterbodies, such as streams, they may filter the volume of water equivalent to that of the entire waterbody from 16 min to 4 days. Table 2. Estimated time for Dreissena polymorpha and Corbicula fluminea to filter the volume of water equivalent to that of the waterbody. Waterbody D. polymorpha Pyalovskoe Reservoir, Russia Uchinskoe Reservoir, Russia Chernobyl Nuclear Station Cooling Reservoir, Ukraine Two Dutch lakes Lake Lukomskoe, 1975, Belarus Lake Lukomskoe, 1990, Belarus Lake Naroch, Belarus Lake Myastro, Belarus Lake Batorino, Belarus Hudson River, USA Long Point Bay, Lake Erie, USA C. fluminea Potomac River, USA Upper Chowan River, USA Meyers Branch Stream, USA Clear Fork of the Trinity River, USA
Time (days) 20 45 5-6 15 - 30 17 45 123 17 54 1.2- 3.6 17 3-4 1 - 1.5 1 0.01
References Mikheev 1967 Lvova 1980 Protasov et al. 1983 Reeders et al. 1989 Karatayev and Burlakova 1995a Karatayev and Burlakova 1995a Karatayev and Burlakova 1995b Karatayev and Burlakova 1995b Karatayev and Burlakova 1995b Strayer et al. 1999 Petrie and Knapton 1999 Cohen et al. 1984 Lauritsen 1986 Leff et al. 1990 McMachon and Bogan 2001
D. polymorpha and C. fluminea are extremely efficient at filtering water, and filtered water is almost free of suspended matter. Non-ingested particles are deposited on the bottom as pseudofeces, and post-digested material is deposited as feces. Both provide rich carbon sources for organisms feeding on the benthos. For example, in the Potomac River (USA) C. fluminea may reduce phytoplankton abundance 40 - 60% (Cohen et al. 1984). In Lake
248 Esrom (Denmark), 9 - 18% of the net phytoplankton production is ingested and assimilated by D. polymorpha (Hamburger et al. 1990). Although there are no quantitative data about the deposition rates of seston by Corbicula at the scale of whole waterbodies, there have been many studies of Dreissena (reviewed in Karatayev et al. 1997). For example, in the Uchinskoe Reservoir (Russia) D. polymorpha deposits 1,071 g m-2 of seston annually (Lvova 1980). Prior to the invasion of zebra mussels, the annual deposition of sediment was only 470 g m-2. In the Pyalovskoe Reservoir (Russia), zebra mussels deposit more than 36,000 tonnes of seston per year (Mikheev 1967), and in the Volgograd Reservoir (Russia) zebra mussels mineralize about 700,000 tonnes of organic matter in one growing season (Spiridonov 1973). Deposition of large amounts of suspended matter by D. polymorpha significantly improves the food base for many benthic animals. In Mikolajskie Lake (Poland) the annual dietary requirement for all of the noncarnivorous animals is met by 16% of the seston deposited each year by bivalves, and D. polymorpha alone produce 160 of the 164.5 tonnes of dry seston deposited by all bivalves in this lake (Alimov 1981). In Lake Lukomskoe, all benthic suspension feeders filtered the volume equivalent of that of the lake in 15 years, and planktonic filterers filtered that same volume in 5 days prior to D. polymorpha invasion. After invasion, zooplankton abundance declined, and the time required to filter the equivalent of the volume of the lake increased to 17 days (Karatayev and Burlakova 1992, 1995a). In contrast, due to the presence of D. polymorpha, the filtering capacity of benthic invertebrates had increased 320 times by 1975, and the volume equivalent to the lake could be filtered in 17 days.
NATURAL ENEMIES
Predators 176 species of various predators are known to feed on zebra mussels, including fish (15 predators on planktonic larvae and 38 on attached mussels) and birds (36 species). In addition, copepods and cnidarians are known to consume veligers, and leeches, crabs, crayfish and rodents are reported to feed on adult zebra mussels (reviewed in Molloy et al. 1997). Few quantitative studies have been conducted on the impact of predators on D. polymorpha. In some cases, fish may consume > 80% of zebra mussel production (Lvova 1977, Yablonskaya 1985) and birds may eat 20 to 70% of annual D. polymorpha production (Mikulski et al. 1975, Stempniewicz 1974). However
249 there is no evidence of any long-term decline of zebra mussel populations due to the effects of predation (Molloy et al. 1997). Similar animals that feed on zebra mussels are reported to consume C. fluminea, including 14 species of fish, 13 species of ducks, racoons, crayfish, and flatworms (reviewed in Sickel 1986). There is some evidence suggesting that fish predation may be a major cause of reduction in C. fluminea density (Dreier and Tranquilli 1981, Robinson and Wellborn 1988). In Fairfield Reservoir, Texas, fish predation reduced C. fluminea abundance 29 fold (Robinson and Wellborn 1988).
Parasites 34 species of endosymbionts are known to be associated with zebra mussels, including ciliates, trematodes, mites, nematodes, leeches, chironomids, oligochaetes, and bacteria (reviewed in Molloy et al. 1997). At least six species of ciliates (Conchophthirus acuminatus, C. klimentinus, Hypocomagalma dreissenae, Sphenophrya dreissenae, S. naumiana, and Ophryoglena sp.) are known to be species specific. There is also some evidence that the trematodes Bucephalus polymorphus, Phyllodistomum folium, and P. dogieli, are quite specific to Dreissena. All of these parasites are found exclusively in Europe. Only nonspecific symbionts (e.g., nematodes, chironomids, oligochaetes, mites) are found in North American zebra mussels. Only one parasite, the trematode B. polymorphus, has been well documented as being seriously debilitating to zebra mussels (i.e., it destroys gonads) (Molloy et al. 1997). In contrast to the wide variety of endosymbionts found in zebra mussels, only two species are known to be associated with C. fluminea: the oligochaete Chaetogaster limnaei (Sickel and Lyles 1981) and a mite (authors unpublished data). The endosymbiotic fauna of C. fluminea in their native region may be more diverse. C. fluminea could be a second intermediate host of Echinistoma revolutum and may be a vector of echinostomiasis in humans (Anazawa 1929). There are no data on the effect of C. fluminea parasites on their abundance.
ECOSYSTEM IMPACTS Local Effects D. polymorpha attaches by byssus to hard substrates and each other, and can create new 3-D habitat, providing not only food, but shelter for bottom invertebrates. These effects on benthic communities are well documented (reviewed in Karatayev et al. 1997, 2002). Zebra mussels have positive effects on isopods (Wolnomiejski 1970, Karatayev and Lyakhnovich
250 1990, Kuhns and Berg 1999), larval chironomids (Wolnomiejski 1970, Botts et al. 1996, Stewart et al. 1998, Kuhns and Berg 1999), leeches (Wolnomiejski 1970), snails (Karatayev et al. 1983, Stewart et al. 1998, reviewed in Strayer 1999), amphipods (Karatayev et al. 1983, Karatayev and Lyakhnovich 1990, Botts et al. 1996, Stewart et al. 1998, Kuhns and Berg 1999, Riccardi 2003), oligochaetes (Afanasiev 1987, Botts et al. 1996), turbellarians (Botts et al. 1996), and hydrozoans (Botts et al. 1996, Stewart et al. 1998). Negative effects are documented for native unionids (reviewed in Schloesser et al. 1996, Karatayev et al. 1997, Burlakova et al. 2000, Riccardi 2003), chironomid larvae (Sokolova et al. 1980, Karatayev et al. 1983) and sphaeriid bivalves (Strayer et al. 1998, Lauer and McComish 2001, Mills et al. 2003). In contrast C. fluminea is solitary, burrows in sediments and does not change the surface of the sediments. Therefore its role in benthic communities may be much smaller. To date there is no evidence of effects of C. fluminea on benthic macroinvertebrates (Karatayev et al. 2003a) or on meiofauna (Hakenkamp et al. 2001). McMahon (1999) hypothesised that C. fluminea detrital feeding could negatively impact other burrowing detritovores. However, in experiments conducted in Lake Nacogdoches, where clams were placed at different densities in trays with sand and the benthic community was allowed to develop for 30 d, there was no difference in species composition or the density of benthic animals with or without live C. fluminea, independent of clam density (R. Mood, A. Karatayev and L. Burlakova, unpublished data). The shells of zebra mussels may accumulate in large quantities and alter the sediments and change benthic community (Karatayev et al. 2002). Similarly, C. fluminea dead shells may affect the benthos (Prokopovich 1969). In the experiments described above, amphipod densities were significantly higher in trays of sand with dead C. fluminea shells than pure sand without shells (R. Mood et al. unpublished data).
System-wide Effects Zebra mussels and Asiatic clams are functionally different than most benthic invertebrates in freshwater. They filter large volumes of water and transport material removed from the water column to the benthos, providing a direct link between processes in the plankton and those in the benthos (benthic-pelagic coupling). The shift of suspended matter from the water column to the bottom induces changes in all aspects of freshwater ecosystems they invade (reviewed in Morton 1997, Karatayev et al. 1997, 2002, McMahon 1999, Vanderploeg et al. 2002, Mayer et al. 2002, Mills et al. 2003) (Table 3).
251 To a large extent the overall impact of D. polymorpha and C. fluminea as suspension feeders on freshwater ecosystem may be similar, however, information is much more available for zebra mussels than Asiatic clams. The filtering activity of both species causes water transparency to increase and decreases seston concentration, BOD, and phytoplankton density (Table 3). With increased transparency, a greater portion of the lake bottom covered with macrophytes. Increased macrophyte beds may provide additional substrate for the zebra mussel attachment and thus increase D. polymorpha populations. In contrast, increased macrophyte beds may cover previously available substrate for C. fluminea and negatively affect their overall density in a waterbody. Table 3. The impact of Dreissena polymorpha and Corbicula fluminea on freshwater ecosystems. Parameter Water transparency Seston concentration
D. polymorpha Increases (reviewed in Karatayev et al. 1997, 2002, Vanderploeg et al. 2002) Decreases (reviewed in Karatayev et al. 1997, 2002, Vanderploeg et al. 2002)
BOD in the water Nutrients
Decreases (reviewed in Karatayev et al. 1997, 2002) Alters nutrient cycling (Johengen et al. 1995, Arnott and Vanni 1996, Makarewicz et al. 2000) Decreases density and chlorophyll content (reviewed in Karatayev et al. 1997, 2002, Vanderploeg et al. 2002)
Phytoplankton
Macrophyte coverage Periphyton Zooplankton Zoobenthos Fish
Increases (reviewed in Karatayev et al. 1997, 2002, Vanderploeg et al. 2002) Increases (Lowe and Pillsbury 1995). Decreases (reviewed in Karatayev et al. 1997, 2002) Increases (reviewed in Karatayev et al. 1997, 2002) Increases quantity of benthophages (reviewed in Karatayev et al. 1997, 2002)
C. fluminea Increases (Buttner 1986, Phelps 1994) Decreases (Buttner 1986, Leff et al. 1990, McMahon 1999) Decreases (Buttner 1986) Alters nutrient cycling (Beaver et al. 1991, Lauritsen and Mozley 1989) Decreases density and chlorophyll content (Cohen et al. 1984, Beaver et al. 1991) Increases (Phelps 1994, McMahon 1999) No data No data No data Increases (Phelps 1994)
No comparable data are available for the impacts of C. fluminea on periphyton, zooplankton, and benthic animal communities. In contrast, the impacts of D. polymorpha in these communities is well documented (Table 3). Zebra mussel filtering results in periphyton and benthic algal increases in both standing stock and primary productivity. Total zooplankton density and biomass decreases. Introduction of both C. fluminea and D. polymorpha may result in increased fish production (Table 3). Although much more data are available on the impact of zebra mussels on fish, generalizations are far from being
252 clear. Many authors have reported an enhancement of all benthic feeding fishes, even those that do not feed on zebra mussels, because zebra mussel invasion is often associated with an increase in biomass of native benthic invertebrates (e.g., Kharchenko and Protasov 1981, Lyakhnovich et al. 1988, Karatayev and Burlakova 1995a, Stewart and Haynes 1994). In contrast, planktivorous fishes could be negatively affected because of decreased phytoplankton abundance and associated decreases in zooplankton, competition with benthic species, and by increasing fish predation on larvae due to increased water transparency (Francis et al. 1996, Lozano et al. 2001). The decline in abundance and body condition in lake whitefish (Coregonis clupeaformis) in lakes Ontario and Michigan (USA) is believed to be related to a decline of Diporea hoyi, an important item in fish diets, following the appearance and proliferation of dreissenid mussels (Hoyle et al. 1999, Pothoven et al. 2001).
GENERAL FINDINGS AND FUTURE DIRECTIONS Although many generalisations can be made about the impacts of Asiatic clams and zebra mussels and their function in freshwater systems, specific predictable impacts are far from clear. The most important aspects of this problem and needed targets for future study are:
Methodological Problems Because these two species are invaders and can cause environmental and economic damage, many aspects about their biology and ecosystem impacts are simplified or exaggerated to draw attention to the problem of invasive species and their spread. Many of these generalisations and exaggerations are then repeated or assumed proven without scientific rigor. Scientists must be careful, especially when extrapolating from short-term laboratory experiments to large scale and long term effects of invaders. We need more studies that link these two approaches before we can draw accurate predictions or assess real impacts. The methodology used to determine impacts is also critical because different methods often yield different results. For example, filtering rates for both Dreissena and Corbicula when feeding on mixed plankton versus single species, as well as filtered versus unfiltered lake water (with seston concentration higher than the incipient threshold seston concentration) may differ, and measures in small volumes of still water are likely to be different than measures made in larger volumes and flowing water. Filtration rates are
253 also reported based on different units. They may be calculated on shell length, wet total mass (WTM, shell plus soft tissue), dry body mass (DBM, soft tissue only), or per ash-free dry mass (AFDM, soft tissue only), and it is not often clear how to convert among these different units. Former Soviet Union scientists generally calculate the filtration rate of D. polymorpha based on shell length or WTM (Lvova 1977, Karatayev and Burlakova 1995b), as do many other Europeans (Reeders and Bij de Vaate 1990, Wisniewski 1990), although some Europeans calculate filtering rate per DBM (Kryger and Riisgård 1988). The majority of North American scientists also calculate the filtration rate of zebra mussels per DBM (Aldridge et al. 1995) or per AFDM (Fanslow et al. 1995, Lei et al. 1996). Common units are essential for cross-study comparisons. We suggest that for zebra mussels the most appropriate units to use are mL of water filtered per g WTM per hour. WTM is very easy to measure, even in the field, and for individuals is much more highly correlated with filtration rate than other measures such AFDM or DBM, which vary greatly with season and reproductive condition (Karatayev 1983). We also recommend that field estimates of filtering rates for Dreissena and Corbicula be calculated as a function of WTM, not density. Different sized mussels will filter at different rates, and similar densities of mussels with different size frequency distributions will have dramatically different filtering rates and therefore their ecosystem impact may vary widely (Young et al. 1996). In any case, appropriate conversions among measures need to be established.
C. fluminea and D. polymorpha Co-effects To date, there are no data on the co-effects of D. polymorpha and C. fluminea invasion on aquatic communities. Both of these invaders continue to spread throughout North America and Europe, and increasingly they are both found in the same freshwater bodies. The effects of both Dreissena and C. fluminea may be additive, or we may see synergistic effects, where their impacts are much more than would be expected by the impacts of either species alone.
C. fluminea vs. D. polymorpha Distribution Although both of these invaders are frequently lumped together because they are bivalves and invade fresh waters, the ecology of D. polymorpha and C. fluminea are different, and therefore their ecosystem impacts are likely to be different. D. polymorpha are more abundant in lakes and large rivers and do not occur in high densities in streams. In contrast, in addition to lakes and large rivers, C. fluminea may be extremely abundant in small streams. Therefore, we may expect to find both similarities and
254 differences in the ecosystem response to the presence of these two invaders. Although they overlap, each has areas where the other is less abundant and they rarely compete for space. The presence of one will not necessarily eliminate the other, and the relative interactions between the two will depend on characteristics of the system, e.g., streams vs. lakes vs. canals.
C. fluminea vs. D. polymorpha Local Effects The impacts of Asiatic clams and zebra mussels, or any other biological agent, are likely to be most intense close to individuals. The biology and natural history of the zebra mussel and Asiatic clam are different. D. polymorpha can live only on the surface of the sediments where they attach to hard substrates creating structure, and providing food and shelter for benthic species. In contrast C. fluminea lives solitary, burrows in sediments and does not alter the surface of sediments. However the accumulation of dead shells of both species may have a similar effect by altering substrate and thereby affecting the benthic community. Moreover, although both species are suspension feeders, C. fluminea also collects food particles from the sediment. Therefore C. fluminea may compete for food with benthic infauna.
C. fluminea vs. D. polymorpha System-wide Effects Depending on water mixing rates, lake morphology, and turnover rates, the effects of suspension feeders on aquatic ecosystems will vary greatly (Ackerman et al. 2001) and may be very local in deep water lakes (ReedAndersen et al. 2000). Although D. polymorpha and C. fluminea impacts on the environment may be similar, feed backs may be different for different invaders. Invasion of both species may increase macrophyte coverage of waterbodies they invade, however increased macrophyte community will provide additional substrate for zebra mussels attachment and therefore may cause further increase of D. polymorpha population size and their impact on ecosystem. In contrast, increased macrophyte coverage may decrease habitat available for C. fluminea and therefore may cause a decrease of Asiatic clam population size and their impact on ecosystem.
Freshwater vs. Marine Bivalves Marine and estuarine bivalves have long been recognized as ecosystems engineers (reviewed in Dame 1993). Invasive zebra mussels and
255 Asiatic clams function similarly in fresh waters. Contrasting the impact of freshwater invasive bivalves on cosystems they recently colonized vs. estuarine, and marine bivalves in ecosystems were they were naturaly dominant but are recently lost, may help us to undestand the role of these important suspension feeders as ecosystem engineers in various waterbodies. Both Dreissena and Corbicula are important freshwater invaders and, as a consequence, have been the focus of much research. However, we still have to learn a great deal about both their biology as well as their impacts on ecosystems that they invade. It is clear from the direct comparison of these two species there remain many “missing pieces” of the picture. We hope that our review will help to focus future efforts such that we will be able to construct the “whole picture” for these two aggressive invaders and their effects on the freshwater ecosystems. These two invaders will continue to provide important information about the capabilities of suspension feeders as well as the functioning of freshwater ecosystems. In addition, these species may be important models that will help us predict the spread and impacts of future invaders.
ACKNOWLEDGEMENTS We would like to acknowledge the support provided by Stephen F. Austin State University (Faculty Research Grant # 14123 to AYK, LEB and DKP, 2003 - 2004).
REFERENCES Abbott TM 1979 Asiatic clam (Corbicula fluminea) vertical distributions in Dale Hollow Reservoir, Tennessee. In: Proceedings, First International Corbicula Symposium, JC Britton (Ed). Fort Worth, Texas, USA, Oct. 13-15, 1977. pp 111-118 Ackerman JD Loewen MR Hamblin PF 2001 Benthic-pelagic coupling over a zebra mussel reef in Western Lake Erie. Limnol Oceanogrr 46: 892-904 Afanasiev SA 1987 The differences in oligochaete distribution in periphyton on substrate with different structure. In: Water Oligochaeta. Proceedings of the 6th All Union Symposium, Riga, Lithuania. pp 38-41 (in Russian) Aldridge DW Payne BS Miller AC 1995 Oxygen consumption, nitrogenous excretion, and filtration rates of Dreissena polymorpha at acclimation temperatures between 20 and 32oC. Can J Fish Aquat Scii 52: 1761-1767 Alimov AF 1981 Functional Ecology of Freshwater Bivalve Moluscs. Nauka Press, Leningrad. 248 p (in Russian) Anazawa K 1929 On a human case of Echinostoma revolutum and its infection route. J Formosan Med Assoc 288:221 Arnott DL Vanni MJ 1996 Nitrogen and phosphorus recycling by the zebra mussel ((Dreissena polymorpha) in the western basin of Lake Erie. Can J Fish Aquat Sci 53: 646-659
256 Arias J 2004 Population dynamics and growth rate of Corbicula fluminea in two lotic systems of East Texas. Master of Science Thesis, Stephen F. Austin State University, Nacogdoches, Texas, 73 p Baker SM Levinton JS Evan WJ 2000 Particle transport in the zebra mussel, Dreissena polymorpha (Pallas). Biol Bull 199: 116-125 Beaver JR Crisman TL Brock RJ 1991 Grazing effects of an exotic bivalve (Corbicula fluminea) on hypereutrophic lake water. Lakes Reserve: Res Manage 7: 45-51 Belanger SE Farris JL Cherry DS Cairns J 1985 Sediment preference of the fresh-water Asiatic clam, Corbicula fluminea. Nautilus 99: 66-73 Boltovskoy D Izaguirre I Correa N 1995 Feeding selectivity of Corbicula fluminea (Bivalvia) on natural phytoplankton. Hydrobiologia 312: 171-182 Boltovskoy D Correa N Cataldo D Stripeikis J Tudino M 1997 Environmental stress on Corbicula fluminea (Bivalvia) in the Parana River Delta (Argentina): complex pollutionrelated disruption of population structures. Arch Hydrobiol 138: 483-507 Botts PS Patterson BA Schloesser DW 1996 Zebra mussel effects on benthic invertebrates: physical or biotic? J N Am Benthol Soc 15: 179-184 Britton JC Morton B 1982 A dissection guide, field and laboratory manual for the introduced bivalve Corbicula fluminea. Malacol Rev 1-82 Britton JC Murphy CE 1977 New records and ecological notes for Corbicula manilensis in Texas. Nautilus 91: 20-23 Burlakova LE 1998 Ecology of Dristan polymorpha (Pallas) and its role in the structure and function of aquatic ecosystems. Candidate Dissertation, Zoology Institute of the Academy of Science, Republic Belarus, Minsk, Belarus 168 p (in Russian) Burlakova LE Karatayev AY Padilla DK 2000 The impact of Dreissena polymorpha (Pallas) invasion on unionid bivalves. Int Rev Hydrobiol 85: 529-541 Buttner JK 1986 Biology of Corbicula in catfish rearing ponds. In: Proceedings of the Second International Corbicula Symposium, JC Britton (Ed). American Malacological Union, Hattiesburg. Am Malacol Bull 2 (Special ed.) pp 211-218 Cahoon LB Owen DA 1996 Can suspension feeding by bivalves regulate phytoplankton biomass in lake Waccamaw, North Carolina? Hydrobiologia 325: 193-200 Cohen RRH Dresler PV Phillips EJP Cory RL 1984 The effect of the Asiatic clam, Corbicula fluminea, on phytoplankton of the Potomac River, Maryland. Limnol Oceanogrr 29: 170-180 Dame RF 1993 Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes. Springer-Verlag, Heidelberg, 579 p Dreier H Tranquilli JA 1981 Reproduction, growth, distribution, and abundance of Corbicula in an Illinois cooling lake. Ill Nat Hist Surv Bull 32: 378-393 Eng LL 1979 Population dynamics of the Asiatic clam, Corbicula fluminea (Müller), in the concrete-lined Delta-Mendota canal of Central California. In: Proceedings, First International Corbicula Symposium, JC Britton (Ed). Fort Worth, Texas, USA, Oct. 1315, 1977. pp 39-78 Evans LPJr. Murphy CE Britton JC Newland LW 1979 Salinity relationships in Corbicula fluminea (Müller 1774). In: Proceedings, First International Corbicula Symposium, JC Britton (Ed). Fort Worth, Texas, USA, Oct. 13-15, 1977. pp 193-214 Hakenkamp CC Palmer MA 1999 Introduced bivalves in freshwater ecosystems: the impact of Corbicula on organic matter dynamics in a sandy stream. Oecologia 119: 445-451 Hakenkamp CC Ribblett SG Palmer MA Swan CM Reid JW Goodison MR 2001 The impact of an introduced bivalve (Corbicula fluminea) on the benthos of a sandy stream. Freshwat Biol 46:491-501 Hamburger KP Dall C Jonasson PM 1990 The role of Dreissena polymorpha Pallas (Mollusca) in the energy budget of Lake Esrom, Denmark. Verh Int Verein Limnol 24: 621-625
257 Hebert PDN Muncaster BW Mackie GL 1989 Ecological and genetic studies on Dreissena polymorpha (Pallas): a new mollusc in the Great Lakes. Can J Fish Aquat Sci 46: 1587-1591 Horgan MJ Mills EL 1997 Clearance rates and filtering activity of zebra mussel (Dreissena ( polymorpha): implications for freshwater lakes. Can J Fish Aquat Sci 54: 249-255 Howlett D Baker R 1999 Corbicula fluminea (Müller): New to UK. J Conchol 36:83 Hoyle JA Schaner T Casselman JM Dermott R 1999 Changes in lake whitefish (Coregonus clupeaformis) stocks in eastern Lake Ontario following Dreissena mussel invasion. Great Lakes Res Rev 4: 1430-1441 Graney RL Cherry DS Rodgers JH Cairns J 1980 The influence of thermal discharges and substrate composition on the population-structure and distribution of the Asiatic clam, Corbicula fluminea, in the New River, Virginia. Nautilus 94: 130-135 Gubanova IF 1968 The influence of waste water on epifauna of flooded wood in Kamskoe Reservoir. In: First Conference on Studying Waterbodies of the Volga Basin, NA Dzubav (Ed). Abstracts. Tolyatti, Russia. pp 231-232 (in Russian) Fanslow DL Nalepa TF Lange GA 1995 Filtration rates of the zebra mussel (Dreissena ( polymorpha) on natural seston from Saginaw Bay, Lake Huron. J Great Lakes Res 21: 489-500 Fast AW 1971 The invasion and distribution of the Asiatic clam (Corbicula manilensis) in a Southern California Reservoir. Bull Southern California Acad Sci 70: 91-98 Francis JT Robillard SR Marsden JE 1996 Yellow perch management in Lake Michigan: a multi-jurisdictional challenge. Fisheries (Bethesda) 21: 18-20 French JRP Schloesser DW 1996 Distribution and winter survival health of Asian clams, Corbicula fluminea, in the St Clair River, Michigan. J Freshwat Ecol 11: 183-192 Fritz LW Lutz RA 1986 Environmental perturbations reflected in internal shell growth patterns of Corbicula fluminea (Mollusca, Bivalvia). Veligerr 28: 401-417 Johengen TH Nalepa TF Fahnenstiel GL Goudy G 1995 Nutrient changes in Saginaw Bay, Lake Huron, after the establishment of the Zebra Mussel ((Dreissena polymorpha). J Great Lakes Res 21: 449-464 Joy JE 1985 A 40-week study on growth of the Asian clam, Corbicula fluminea (Müller), in the Kanawha River, West-Virginia. Nautilus 99: 110-116 Karatayev AY 1983 Ecology of Dreissena polymorpha Pallas and its effects on macrozoobenthos of the thermal power plant's cooling reservoir. Candidate Dissertation, Zoology Institute of Academy of Science Belarusian SSR, Minsk, Belarus 178 p (in Russian) Karatayev AY Burlakova LE 1992 Changes in trophic structure of macrozoobenthos of an eutrophic lake, after invasion of Dreissena polymorpha. Biol Vnutr Vod/Biol Inland Waterr 93: 67-71 (in Russian) Karatayev AY Burlakova LE 1995a The role of Dreissena in lake ecosystems. Russian J Ecol 26: 207-211 Karatayev AY Burlakova LE 1995b Present and further patterns in Dreissena population development in the Narochanskaya lakes system. Vestsi Akademii Navuk Belarusi. Seriya Biyologichnikh Navuk 3: 95-98 (in Belorussian) Karatayev AY Burlakova LE Padilla DK 1997 The effects of Dreissena polymorpha (Pallas) invasion on aquatic communities in Eastern Europe. J Shellfish Res 16: 187-203. Karatayev AY Burlakova LE Padilla DK 1998 Physical factors that limit the distribution and abundance of Dreissena polymorpha (Pall.). J Shellfish Res 17: 1219-1235 Karatayev AY Burlakova LE Padilla DK 2002 Impacts of zebra mussels on aquatic communities and their role as ecosystem engineers. In: Invasive Aquatic Species of Europe: Distributions, Impacts and Management. Monoraphiae Biologicae Series, E Leppäkoski S Gollasch S Olenin (Eds.), Kluwer Scientific Publishers, Dordrecht, pp 433-446 Karatayev AY Burlakova LE Kesterson T Padilla DK 2003a Dominance of the Asiatic clam, Corbicula fluminea (Müller) in the benthic community of a reservoir. J Shellfish Res 22(2): 487-493
258 Karatayev AY Burlakova LE Padilla DK Johnson LE 2003b Patterns of spread of the zebra mussel ((Dreissena polymorpha (Pallas)): the continuing invasion of Belarussian lakes. Biol Invasions 5(3): 213-221 Karatayev AY Lyakhnovich VP 1990 Effect of Dreissena polymorpha Pallas on benthic crustaceans (Gammarus lacustris Sars, Pallasea quadrispinosa Sars and Asellus aquaticus L.) in Lukomskoe Lake. In: Species within their Range: Biology, Ecology, and Productivity of Aquatic Invertebrates, NN Khmeleva et al. (Eds). Navuka i Tekhnika Press, Minsk. pp 123-125 (in Russian) Karatayev AY Tishchikov GM Karatayeva IV 1983 The specific community of benthic animals associated with Dreissena polymorpha Pallas. Biol Vnutr Vod/Biol Inland Waterr 61:18-21 (in Russian) Kat P 1982 Shell dissolution as a significant cause of mortality for Corbicula fluminea (Bivalvia: Corbiculidae) inhabiting acidic waters. Malacological Review 15:129-134 Kharchenko TG Protasov AA 1981 On consortia in water ecosystems. Gidrobiologicheskii Zhurnall 17:15-20 (in Russian) Khusainova NZ 1958 The Biological Distinctive Features of Some Abundant Bottom Invertebrates Used by Fishes in the Aral Sea. Kazakhstan State University Press, AlmaAta 116 p (in Russian) Kryger J Riisgård HU 1988 Filtration rate capacities in 6 species of European freshwater bivalves. Oecologia 77: 34-38 Kuhns L Berg M 1999 Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel ((Dreissena polymorpha) invasion in southern Lake Michigan. J Great Lakes Res 25: 910-917 Lauer TE McComish TS 2001 Impact of zebra mussels ((Dreissena polymorpha) on fingernail clams (Sphaeriidae) in extreme Southern Lake Michigan. J Great Lakes Res 27(2): 230-238 Lauritsen DD 1986 Filter-feeding in Corbicula fluminea and its effect on seston removal. J N Am Benthol Soc 5: 165-172 Lauritsen DD Mozley SC 1989 Nutrient excretion by the Asiatic clam Corbicula fluminea. J N Am Benthol Soc 8: 134-139 Leff LG Burch JL McArthur JV 1990 Spatial distribution, seston removal, and potential competitive interactions of the bivalves Corbicula fluminea and Elliptio complanata, in a coastal plain stream. Freshwat Biol 24: 409-416 Lei J Payne BS Wang SY 1996 Filtration dynamics of the zebra mussels, Dreissena polymorpha. Can J Fish Aquat Scii 53: 29-37 Lowe RL Pillsbury RW 1995 Shifts in benthic algal community structure and function following the appearance of zebra mussel ((Dreissena polymorpha) in Saginaw Bay, Lake Huron. J Great Lakes Res 21: 558-566 Lozano SJ Scharold JV Nalepa TF 2001 Recent declines in benthic macroinvertebrate densities in Lake Ontario. Can J Fish Aquat Sci 58(3): 518-529 Luferov VP 1965 Organisms living anabiotically frozen in ice of the littoral zone of Rybinskoe Reservoir. Tr Inst Biol Vnutr Vod Akad Nauk SSSR 8(11): 151-154 (in Russian) Lvova AA 1977 The ecology of Dreissena polymorpha (Pall.) in Uchinskoe Reservoir. Candidate Dissertation, Moscow State University, Moscow, USSR 114 p (in Russian) Lvova AA 1980 Ecology of zebra mussel (Dreissena ( polymorpha polymorpha (Pall.). Trans Hydrobiol Soc RAN 23: 101-119 (in Russian) Lyakhnovich VP Karatayev AY Mitrakhovich PA Guryanova LV Vezhnovets GG 1988 Productivity and prospects for utilizing the ecosystem of Lake Lukoml, thermoelectric station cooling reservoir. Soviet J Ecoll 18: 255-259 Lyakhnovich VP Karatayev AY Andreev NI Andreeva SI Afanasiev SA Dyga AK Zakutskiy VP Zolotareva VI Lvova AA Nekrasova MY Osadchikh VF Pligin YV
259 Protasov AA Tischikov GM 1994 Living conditions. In: Freshwater Zebra Mussel Dreissena polymorpha (Pall.) (Bivalvia, Dreissenidae). Systematics, Ecology, Practical Meaning, JI Starobogatov (Ed). Nauka Press, Moscow. pp 109-119 (in Russian) Lyakhov SM Mikheev VP 1964 Distribution and density of Dreissena in the Kuibyshevskoe Reservoir seven years after its construction. Tr Inst Biol Vnutr Vod Akad Nauk SSSR 7(10): 3-18 (in Russian) Marsh PC 1985 Secondary production of introduced Asiatic clam, Corbicula fluminea, in a Central Arizona Canal. Hydrobiologia 124: 103-110 Makarewicz JC Bertram P Lewis TW 2000 Chemistry of the offshore surface waters of Lake Erie: pre- and post-Dreissena introduction (1983-1993). J Great Lakes Res 26 (1): 8293 MacIsaac HJ Lonnee CJ Leach JH 1995 Suppression of microzooplankton by zebra mussels: Importance of mussel size. Freshwat Biol 34: 379-387 Mattice JS 1979 Interactions of Corbicula sp. with power plants. In: Proceedings, First International Corbicula Symposium, JC Britton (Ed), Fort Worth, Texas, USA, Oct. 1315, 1977, pp 119-138 Mayer CM Keats RA Rudstam LG Mills EL 2002 Scale-dependent effects of zebra mussels on benthic invertebrates in a large eutrophic lake. J N Am Benthol Soc 21(4): 616-633 McMahon RF 1983 Ecology of an invasive pest bivalve, Corbicula. In: The Mollusca. Vol. 6. Ecology, WD Russel-Hunter (Ed), Academic Press, Inc., pp 505-561 McMahon RF 1999 Invasive characteristics of the freshwater bivalve Corbicula fluminea. In: Nonindigenous Freshwater Organisms. Vectors, Biology, and Impacts, R Claudi JH Leach (Eds.), Lewis Publishers, Ann Arbor, Michigan, pp 315-343 McMahon RF Bogan AE 2001 Mollusca: Bivalvia. In: Ecology and Classification of North American Freshwater Invertebrates, 2ndd Edition, JH Thorp AP Covich (Eds.), Academic Press, Inc., pp 331-430 Mikheev VP 1967 The nutrition of zebra mussels ((Dreissena polymorpha Pallas). Summary of the Candidate Dissertation, State Research Institute for Lakes and Rivers Fishery Industry, Leningrad 22 p (in Russian) Mikheev VP 1994 Ration of Dreissena in natural conditions. In: Freshwater Zebra Mussel Dreissena polymorpha (Pall.) (Bivalvia, Dreissenidae). Systematics, Ecology, Practical Meaning, JI Starobogatov (Ed). Nauka Press, Moscow, pp 129-132 (in Russian) Mikulski J Adamczak B Bittel L Bohr R Bronisz D Donderski W Gizinski A Luscinska M Rejewski M Strzelczyk E Wolnomiejski N Zawislak W Zytkowicz R 1975 Basic regularities of productive processes in the Ilawa Lakes and in the Goplo Lake from the point of view of utility values of the water. Pol Arch Hydrobiol 22: 101-122 Mills EL Casselman JM Dermott R Fitzsimons JD Gal G Holeck KT Hoyle JA Johannsson OE Lantry BF Makarewicz JC Millard ES Munawar IF Munawar M O’Gorman R Owens RW Rudstam LG Schaner T Stewart TJ 2003 Lake Ontario: food web dynamics in a changing ecosystem (1970-2000). Can J Fish Aquat Sci 60: 471-490 Minchin D 2000 Dispersal of zebra mussel in Ireland. Verh Internat Verein Limnol 27: 15761579 Molloy DP Karatayev AY Burlakova LE Kurandina DP Laruelle F 1997 Natural enemies of zebra mussels: predators, parasites and ecological competitors. Rev Fish Scii 5: 27-97 Mordukhai-Boltovskoi FD 1960 Caspian Fauna in the Azov and Black Sea Basins. Academia Nauk Press, Moscow (in Russian) Morton B 1979 Corbicula in Asia. In: Proceedings, First International Corbicula Symposium, JC Britton (Ed) Fort Worth, Texas, USA, Oct. 13-15, 1977, pp 1-54 Morton B 1997 The aquatic nuisance species problem: a global perspective and review. In: Zebra Mussels and Aquatic Nuisance Species, FM D'Itri (Ed). Ann Arbor Press, Inc., Chelsea, Michigan Petrie S Knapton R 1999 Rapid increase and subsequent decline of zebra and quagga mussels in Long Point Bay, Lake Erie: possible influence of waterfowl predation. J Great Lakes Res 25: 772-782
260 Phelps HL 1994 The Asiatic clam (Corbicula fluminea) invasion and system-level ecological change in the Potomac River Estuary near Washington, DC. Estuaries 17: 614-621 Pimentel D Lach L Zuniga R Morrison D 2000 Environmental and economic costs of nonindigenous species in the United States. Bioscience 50: 53-65 Pothoven SA Nalepa TF Schneeberger PJ Brandt SB 2001 Changes in diet and body condition of lake whitefish in southern Lake Michigan associated with changes in benthos. N Am J Fish Managg 21: 876-883 Prokopovich NP 1969 Deposition of clastic sediments by clams. J Sediment Petrol 39: 891901 Protasov AA Afanasiev SA Ivanova OO 1983 The distribution and role of Dreissena polymorpha in the periphytone of Chernobyl water cooling reservoir. In: Molluscs: Systematics, Ecology and Patterns of Occurrence. Abstracts of the 7th Meeting on the Molluscs Study. Nauka Press, Leningrad. pp 220-222 (in Russian) Ramcharan CW Padilla DK Dodson SI 1992 Models to predict potential occurrence and density of the zebra mussel, Dreissena polymorpha. Can J Fish Aquat Scii 49: 26112620 Reed-Andersen T Carpenter SR Padilla DK Lathrop RC 2000 Predicted impact of zebra mussel (Dreissena polymorpha) invasion on water clarity in Lake Mendota. Can J Fish Aquat Sci 57: 1617-1626 Reeders HH bij de Vaate A 1990 Zebra mussels (Dreissena ( polymorpha): a new perspective for water quality management. Hydrobiologia 200/201: 437-450 Reeders HH bij de Vaate A Slim FJ 1989 The filtration rate of Dreissena polymorpha (Bivalvia) in three Dutch lakes with reference to biological water quality management. Freshwat Biol 22: 133-141 Reid RGB McMahon RF Foighil DO Finnigan R 1992 Anterior inhalant currents and pedalfeeding in bivalves. Veligerr 35: 93-104 Riccardi A 2003 Predicting the impacts of an introduced species from an invasion history: an empirical approach applied to zebra mussel invasion. Freshwat Bioll 48: 972-981 Robinson JV Wellborn GA 1988 Ecological resistance to the invasion of a freshwater clam, Corbicula fluminea: Fish predation effects. Oecologia (Berl.) 77: 445-452 Roditi HA Caraco NF Cole JJ Strayer DL 1996 Filtration of Hudson River water by the zebra mussel ((Dreissena polymorpha). Estuaries 19: 824-832 Rodgers JHJr. Cherry DS Dickson KL Cairns JJr 1979 Invasion, population dynamics and elemental accumulation of Corbicula fluminea in the new river at Glen Lyn, Virginia. In: Proceedings, First International Corbicula Symposium, JC Britton (Ed). Fort Worth, Texas, USA, Oct. 13-15, 1977. pp 99-110 Schloesser DW Nalepa TF Mackie GL 1996 Zebra mussel infestation of unionid bivalves (Unionidae) in North America. Am Zool 36: 300-310 Schneider DW Stoeckel JA Rehmann CR Blodgett KD Sparks RE Padilla DK 2003 A developmental bottleneck in dispersing larvae: implications for spatial population dynamics. Ecol Lett 6 (4): 352-360 Shkorbatov GL Karpevich AF Antonio PI 1994 Ecological physiology. In: Freshwater Zebra Mussel Dreissena polymorpha (Pall.) (Bivalvia, Dreissenidae). Systematics, Ecology, Practical Meaning, JI Starobogatov (Ed). Nauka Press, Moscow, pp 67-108. (in Russian) Sickel JB 1986 Corbicula population mortalities: factors influencing population control. Am Malacol Bulll 2 (Special Ed.):89-94 Sickel JB Lyles MB 1981 Chaetogaster limnaei (Oligochaeta: Naididae) inhabiting the mantle cavity of the Asiatic clam, Corbicula fluminea, in Barkley Lake, Kentucky. Veligerr 23: 361-362
261 Silverman H Achberger EC Lynn JW Dietz TH 1995 Filtration and utilization of laboratorycultured bacteria by Dreissena polymorpha, Corbicula fluminea, and Carunculina texasensis. Biol Bull 189: 308-319 Silverman H Nichols SJ Cherry JS Achberger E Lynn JW Dietz TH 1997 Clearance of laboratory-cultured bacteria by freshwater bivalves: differences between lentic and lotic unionids. Can J Zooll 75: 1857-1866 Sokolova NY Izvekova EI Lvova AA Sakharova MI 1980 Structure, distribution and seasonal dynamics of benthic densities and biomass. Tr Vses Gidrobiol O-va 23: 7-23 (in Russian) Spiridonov YI 1973 The mineralization by zebra mussels in Volgograd Reservoir. Tr. Kompleksn Eksped Sarat Univ Izuch Volgogr Sarat Vodokhran 3: 131-134 (in Russian) Sprung M Rose U 1988 Influence of food size and food quality on the feeding of the mussel Dreissena polymorpha. Oecologia 77: 526-532 Stanczykowska A Lewandowski K 1993 Thirty years of studies of Dreissena polymorpha ecology in Mazurian Lakes of Northeastern Poland. In: Zebra Mussels: Biology, Impacts, and Control, TF Nalepa DW Schloesser (Eds). Lewis Publishers, Ann Arbor, Michigan, pp 3-33 Starobogatov JI Andreeva SI 1994 Distribution and history. In: Freshwater Zebra Mussel Dreissena polymorpha (Pall.) (Bivalvia, Dreissenidae). Systematics, Ecology, Practical Meaning, JI Starobogatov (Ed.). Nauka Press, Moscow, pp 47-55 (in Russian) Stempniewicz L 1974 The effect of feeding of the coot (Fulica ( atra L.) on the character of the shoals of Dreissena polymorpha Pall. in the Lake Goplo. Acta Univ Nicolai Copernici, Ser Mat Przyrr 34: 83-103 Stewart TW Haynes JM 1994 Benthic macroinvertebrate communities of Southwestern Lake Ontario following invasion of Dreissena. J Great Lakes Res 20: 479-493 Stewart TW Miner JG Lowe RL 1998 Quantifying mechanism for zebra mussel effects on benthic macroinvertebrates: organic matter production and shell-generated habitat. J N Am Benthol Soc 17: 81-94 Stites DL Benke AC Gillespie DM 1995 Population dynamics, growth, and production of the Asiatic clam, Corbicula fluminea, in a blackwater river. Can J Fish Aquat Scii 52: 425437 Strayer DL 1999 Effects of alien species on freshwater mollusks in North America. J N Am Benthol Soc 18: 74-98 Strayer DL Caraco NF Cole JJ Findlay S Pace ML 1999 Transformation of freshwater ecosystems by bivalves. A case study of zebra mussels in the Hudson River. BioScience 49: 19-27 Strayer DL Smith LC Hunter DC 1998 Effects of the zebra mussel (Dreissena polymorpha) invasion on the macrobenthos of the freshwater tidal Hudson River. Can J Zool 76: 419-425 Ten Winkel EN Davids C 1982 Food selection by Dreissena polymorpha Pallas (Mollusca: Bivalvia). Freshwat Bioll 12: 553-558 Vanderploeg HA Nalepa TF Jude DJ Mills EL Holeck KT Liebig JR Grigorovich IA Ojaveer H 2002 Dispersal and emerging ecological impacts of Ponto-Caspian species in the Laurentian Great Lakes. Can J Fish Aquat Scii 59(7): 1209-1228 Way CM Hornbach DJ Millerway CA Payne BS Miller AC 1990 Dynamics of filter feeding in Corbicula fluminea (Bivalvia, Corbiculidae). Can J Zool 68: 115-120 Wisniewski R 1990 Shoals of Dreissena polymorpha as a bio-processor of seston. Hydrobiologia 200/201: 451-458 Wolnomiejski N 1970 The effects of Dreissena polymorpha Pall. aggregation on the differentiation of the benthonic macrofauna. Zecz nauk UMK K 25: 31-39 Wong WH Levinton JS Twining BS Fisher N 2003 Assimilation of micro- and mesozooplankton by zebra mussels: a demonstration of the food web link between zooplankton and benthic suspension feeders. Limnol Oceanogr 48: 308-312 Yablonskaya EA 1985 The Caspian Sea: Fauna and Biological Productivity. Nauka Press, Moscow (in Russian)
262 Young BL Padilla DK Schneider DW Hewett SW 1996 The importance of size-frequency relationships for predicting ecological impact of zebra mussel populations. Hydrobiologia 332 (3): 151-158 Zhadin VI 1946 The traveling shellfish Dreissena. Priroda 5: 29-37 (in Russian)
FUNCTIONAL CHANGES IN BENTHIC FRESHWATER COMMUNITIES AFTER DREISSENA POLYMORPHA (PALLAS) INVASION AND CONSEQUENCES FOR FILTRATION
Lyubov E. Burlakova1, Alexander Y. Karatayev1, and Dianna K. Padilla2 1
Department of Biology, Stephen F. Austin State University, Nacogdoches, TX, USA Department of Ecology and Evolution, Stony Brook University, Stony Brook, NY, USA
2
Abstract: Dreissena is extremely abundant in waters it invades, and dramatically changes benthic invertebrate communities in terms of total biomass, species composition, and the relative abundance of functional groups. We analyzed the relative abundance of feeding functional groups of the benthic community before and after zebra mussel invasion in three Belarussian lakes, four lakes after invasion only, and one lake in the same region that has not been invaded. After invasion, benthic structure was dominated by one trophic group – filterers. This group accounted for greater than 96% of the total biomass of benthic invertebrates. We found that the relative abundance of feeding functional groups in the rest of the benthic community, without including Dreissena biomass, was also different in lakes examined before and after zebra mussel invasion. Before invasion and in the un invaded lake, planktonic invertebrates filtered a volume equivalent to the volume of the lake within few days, and were more than 200 times more effective than benthic filterers, which would take about 4 years to filter an equivalent volume. After Dreissena invaded the lakes, the total average biomass of all benthic invertebrates (including zebra mussels) increased more than 20 times. The filtration efficiency of the benthic community increased greater than 70 times, and the time required to filter the volume of the lake was not significantly different than that for zooplankton. These dramatic changes will alter the relative roles of the plankton and benthos in a variety of ecosystem functions, especially the movement of carbon from the plankton to the benthos. Key words: Zebra mussels, benthic community, trophic structure, feeding functional groups, filtration efficiency.
INTRODUCTION The zebra mussel, Dreissena polymorpha Pallas (1771), continues to spread throughout the freshwaters of Eurasia and North America, and new lakes and rivers are constantly being invaded (Kinzelbach 1992, McMahon and Bogan 2001, Minchin et al. 2002, Karatayev et al. 2003). Species of 263 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 263–275. ©2005 Springer. Printed in the Netherlands.
264 Dreissena are the only bivalves in freshwater to attach to hard substrates and possess a dispersing planktonic larval stage. D. polymorpha is extremely abundant in waters it invades, is frequently competitively dominant over native freshwater fauna, and has large impacts on all parts of the ecosystem, especially benthic animals (reviewed in Karatayev et al. 1997, 2002). Characterizing the feeding functional group composition in lakes before and after invasion by zebra mussels can provide insights into how benthic communities respond to invasion. The feeding functional group approach enables a quantitative assessment of the degree of dependence of the invertebrate biota on particular food resources, and the linkages between food sources and morphological and behavioral adaptations (Merritt and Cummings 1996). Although the effect of D. polymorpha invasion on species composition and abundance within the benthic community has been documented for certain lakes, to date there are few studies of resultant changes in the trophic structure of communities (Sokolova et al. 1980a, Karatayev and Burlakova 1992). When zebra mussels invade, they create a large population of effective suspension-feeders that can cause radical changes in the benthic community (Lvova-Kachanova and Izvekova 1978, Sokolova et al. 1980a, Karatayev and Burlakova 1992, Karatayev et al. 1997). Native suspensionfeeders can be out-competed by D. polymorpha, and decrease in abundance, while animals feeding on the sediments can increase in abundance (Sokolova et al. 1980a, Karatayev and Burlakova 1992, Karatayev et al. 1994). In this study, we assessed the impacts of invasion by zebra mussels on the structure of the benthic communities by examining feeding functional groups within the benthic community of 8 Belarussian lakes. By comparing the structure of communities without zebra mussels with those that have been invaded, we can assess the impacts of invasion on trophic processes and some aspects of ecosystem function.
MATERIALS AND METHODS Study Sites and Sampling To study the trophic structure of benthic invertebrates in lakes with and without Dreissena, we used data collected from glacial lakes (Karatayev et al. 2003), as part of a larger survey of Belarussian lakes. The Republic of Belarus is situated between Poland and Russia, and was part of the former Soviet Union. A variety of chemical, geological, physical, and biological data were collected in mid-summer for each of these lakes by the Lakes Research Laboratory of the Belarussian State University. We collected additional data for some lakes in the summer of 1998 and 1999. For three lakes we have data
265 before and after invasion, for four lakes we have data after invasion only, and we have data for one lake, in the same region, that has not been invaded (Table 1). To determine the species composition, density and biomass of benthic invertebrates 7 – 16 samples were collected from each lake, depending on the lake size. Sample sites were selected to maximize coverage of the lake bottom and include all major habitat types. For all benthic samples we used a Petersen grab for hard substrates and an Eckman grab for soft substrates (sample area 0.025 m2). Samples were washed through a 500 Pm mesh. Retained macroinvertebrates were preserved with 10% neutral buffered formalin. All macroinvertebrates were identified to the lowest possible taxon, counted and weighed to the nearest 0.0001 g after being blotted dry on absorbent paper (wet mass). For three lakes, Myadel, Boginskoe, and Svir, we had data both before and after invasion, which allowed us to do paired comparisons (Table 1). We also had data for four additional lakes, Bolshie Shvakshty, Volchin, Dolzha, and Bolduk that have been invaded by zebra mussels. All of these lakes were invaded by Dreissena between 1980 and the mid-1990s, however, unfortunately, we do not know the exact years of invasion. Table 1. Limnological parameters of the studied Belarussian lakes. Lake
Year Studied Surface Volume Maximum Secchi Trophic Before After area (106 m3) depth (m) depth (m) status invasion invasion (km2) 1973 1999 1.2 9.12 16.2 2.4 eutrophic Boginskoe Svir 1980 1998 22.3 104.3 8.7 1.8 eutrophic 1980 1998 16.4 102.0 24.6 4.8 mesotrophic Myadel Dolzha 1998 1.0 5.4 13.7 2.6 eutrophic Bolshiye Shvakshty 1998 9.6 22.3 5.3 3.1 eutrophic Volchin 1998 0.5 7.9 32.9 3.8 mesotrophic Bolduk 1999 0.8 11.9 39.7 4.5 oligomesotrophic Ikazn 1973 not 2.4 7.9 8.4 1.4 eutrophic invaded
Functional Groups We used the classification scheme by Merritt and Cummins (1996) for functional feeding groups. For invertebrates that were identified to species, we used data from Izvekova (1975), Sokolova et al. (1980b) and Monakov (1998, 2003) to assign feeding functional group. For invertebrates identified to genus, we used Merritt and Cummins (1996) and Thorp and Covich (2001). However, some species and genera fit into more than one group. For example, some collectors are known to filter-feed and gather (e.g., Microtendipes cloris, Bithynia tentaculata, B. leachi, Tanytarsus sp.)
266 (Izvekova 1975, Merritt and Cummins 1996, Monakov 1998). As these species were abundant in the lakes sampled, we considered them as “filtering + gathering collectors”, a sub-group within the group “collectors”. To determine if there were changes in the relative abundance of feeding functional groups associated with zebra mussel invasion, we compared the relative proportions of biomass of each functional group before and after zebra mussel invasion. Functional Consequences Shifts in the feeding functional groups in a lake will have functional consequences for both the processing of benthic carbon, and for links between the benthic and the planktonic communities (benthic-pelagic coupling). To estimate the filtration capability of the zooplankton community in each lake we used Kryuchkova's (1989) estimate that zooplankton can filter 120 mL mg wet mass-1d-1 in a cladoceran dominated eutrophic lake. To estimate the filtration capacity of the benthos, we used the literature values for the filtration rates, based on wet total mass, of individual species (Izvekova 1975, Alimov 1981, reviewed in Monakov 1998) weighted by the average biomass (wet total mass (body plus shell), g m-2) of that species in the lake. For species whose filtration rates were not known, we used the rates for the closest related taxon whose rate was known. The total filtration capacity of the benthic community was estimated by multiplying the filtration rates by the average biomass of each taxon determined to be in the filtering collectors functional group. As it is very difficult to determine the relative proportion of time a species filters or gathers (references in Monakov 1998), the impact of members of the filtering + gathering collectors was divided in half. For zebra mussels, we used a filtration rate of 58 mL g-1 h-1 (literature estimates range from 35 - 110 mL g-1 h-1, Karatayev et al. 1997). To estimate the total filtration capacity of zebra mussels, we multiplied the average biomass of zebra mussels (g m-2) by this filtration rate. In this way we were able to compare the filtration rate of the entire community of zooplankton with the amount of filtration for the entire benthic community. Statistical Analyses To compare the relative abundance of trophic groups in lakes with and without zebra mussels, we used either a t-test or Mann-Whitney U test on percentage data for each lake (Zar 1996). To compare the structure of functional trophic groups of benthic community before and after zebra mussel invasion we used Fisher-Freeman-Halton test (a generalization of the Fishers Exact test to r by c contingency table) with a Monte Carlo estimate of the Pvalue to test for homogeneity in contingency tables. Effects were considered
267 statistically significant at P < 0.05. Analyses were performed with StatXact-4 (version 4.0.1, Cytel Software Corp.) and Statistica software (STATISTICA version 6, StatSoft, Inc. 2001). When multiple tests were conducted on the same data, we used a sequential Bonferroni correction (Rice 1989) to adjust the critical alpha considered for statistical significance. Where appropriate, we present the critical alpha (Į) with the results of each statistical test.
RESULTS Feeding Functional Group Composition Before Invasion Collectors (filterers, filterers + gatherers and gatherers) dominated the benthic macroinvertebrate community in lakes uninhabited by zebra mussels, and comprised approximately 70% of the total biomass of the community. Filterers + gatherers were the largest group (50% of total biomass). Predators constituted about 17% of the total biomass, comparable to shredders, scrapers and scavengers combined (Table 2). Table 2. Relative proportions of feeding functional groups of benthic macroinvertebrates (% of total biomass) in Belarussian lakes with and without zebra mussels (average ± SE). Feeding functional group Collectors: Filterers Collectors: Filterers + Gatherers Collectors: Gatherers Shredders Scarpers Scavengers Predators
Lakes without zebra mussels (n = 4) 4.2 ± 0.9 50.4 ± 5.1 14.5 ± 5.2 2.5 ± 2.3 9.6 ± 3.5 2.3 ± 1.1 16.5 ± 1.8
Zebra mussel invaded lakes excluding zebra including zebra mussels (n = 7) mussels (n = 7) 7.6 ± 2.0 96.7 ± 0.8 34.4 ± 8.5 21.9 ± 3.2 0.3 ± 0.3 6.0 ± 1.5 7.4 ± 3.2 22.5 ± 5.2
1.5 ± 0.7 0.7 ± 0.2 < 0.01 0.2 ± 0.1 0.3 ± 0.2 0.7 ± 0.1
After Invasion, excluding Dreissena: Lakes with before and after data For the three lakes where we had before and after invasion data, the changes in functional group structure (excluding Dreissena biomass) were significant (Lake Boginskoe: P = 0.0008, Į = 0.017; Lake Myadel: P = 0.004, adjusted critical Į = 0.025; Lake Svir: P = 0.044, Į = 0.05; Fisher-FreemanHalton test) (Table 3).
268 Table 3. Relative proportions of feeding functional groups of benthic macroinvertebrates (% of total wet biomass) excluding Dreissena biomass in three Belarussian lakes studied before (B) and after (A) zebra mussel invasion. Feeding functional group Collectors: Filterers Collectors: Filterers + Gatherers Collectors: Gatherers Total Collectors Shredders Scarpers Scavengers Predators
Lake Myadel B A 5.7 2.4
Lake Boginskoe B A 5.1 10.6
Lake Svir B A 1.6 12.4
36.0 29.8
36.4 12.7
55.9 7.9
51.3 21.4
50.9 11.1
51.0 13.6
71.5 0.0 2.3 5.4 20.8
51.5 0.0 4.5 3.7 40.3
68.9 0.8 14.8 0.8 14.7
83.3 0.0 4.0 5.4 7.3
63.6 0.0 16.4 1.9 18.1
77 0.0 7.7 2.5 12.8
After Invasion, excluding Dreissena: Pooled data from all lakes The pooled data for all invaded (7 invaded lakes - Myadel, Bolshie Shvakshty, Boginskoe, Svir, Bolduk, Dolzha, and Volchin) and uninvaded (4 uninvaded lakes - Myadel, Boginskoe, Svir and Ikazn) lakes were tested for differences in the relative biomass of different feeding functional groups. For these pooled data, the changes in the functional trophic groups after zebra mussel invasion were not significant (P = 0.068, adjusted critical Į = 0.013, Fisher-Freeman-Halton test). Collectors remained more than 60% of total community biomass (excluding D. polymorpha), and the relative abundance of filterers + gatherers did not change significantly (P = 0.12, adjusted critical Į = 0.017, t-test). The portion of collectors-gatherers and filterers also did not significantly change (P > 0.30, adjusted critical Į = 0.025) (Table 2). The change in the relative abundance of all types of collectors before and after invasion was also not significant (P = 0.065, Fisher-Freeman-Halton test adjusted critical Į = 0.010). The portion of scavengers increased, but this change was not significant (P = 0.039, Mann-Whitney U test, adjusted critical Į = 0.008) and the proportion of predators also did not change (P = 0.36, ttest, critical Į = 0.05). After Invasion, including Dreissena If D. polymorpha is considered with the rest of the benthic community, the trophic structure of the benthic community was characterized by an extremely high dominance of one trophic group – collectors filterers, which accounted for > 96 % of the total biomass of benthic invertebrates in lakes populated by zebra mussels (Table 2). The relative proportions of feeding
269 functional groups (in terms of biomass) in the benthic community before and after zebra mussel invasion were significantly different (P << 0.001, FisherFreeman-Halton test). Filtration Capacity of Planktonic versus Benthic Community After Dreissena invasion, the total average biomass of the benthic community including zebra mussels increased 22 times, from 10.3 ± 3.8 (mean ± SE) to 225.5 ± 18.9 g m-2 (P = 0.002, adjusted critical Į = 0.025, paired d t-test). Before the zebra mussel invasion, planktonic invertebrates filtered the equivalent of the volume of the lake in 4 - 9 days (6.5 ± 1.1) (Table 4). The benthic community, however, would take 0.2 - 10 years (3.8 ± 1.9) to filter the same volume, an average 213 times slower than for zooplankton. After invasion, the filtration ability of benthic community increased > 70 times. The time required for the benthic community to filter the volume Table 4. Mean biomass of zooplankton and benthic communities, and the time required to filter a volume of water equal to the lake volume before and after zebra mussel invasion. Parameter
Lake Bolshie Shvakshty
Before Dreissena invasion Zooplankton*: biomass (g m-3)
Zoobenthos:
days to filter biomass (g m-2) days to filter
0.9 7 9 20. 0 63
Lake Svir
Lake Volchin
1 .55 5 1 0.1 5 9
After Dreissena invasion Zooplankton*: biomass (g m-3) days to filter Zoobenthos excluding Dreissena: biomass (g m-2) days to filter
2
0.5 2 16
.78
22. 7 7
6.2
3 1 3 9
Dreissena:
biomass (g m-2) days to filter
16 3 10
2 38 1 4
Zoobenthos including Dreissena: biomass (g m-2) days to filter
18 5 4
2 54 1 0
*data from Karatayev and Makritskaya (1999)
1 .03 8 1 .2 3 688 1 .37 6 9 .1 3 03 1 92 5 5 2 02 4 7
Lake Dolzha
2 .08 4 9 .8 1 732 1 .25 7 4 .4 5 06 2 57 1 5 2 61 1 4
270 equivalent to that of the lake decreased to 19 ± 10 days, and was not significantly different from that for the planktonic community (8 ± 3 days, P = 0.40, paired t-test). The total biomass of benthic invertebrates excluding Dreissena did not change significantly after zebra mussel invasion (P = 0.41, paired t-test). There was also no significant difference in the biomass of zooplankton before and after zebra mussel invasion (1.4 ± 0.3 g m-3 vs. 1.5 ± 0.5 g m-3, paired t-test, P = 0.88), nor in the time to filter the equivalent of the volume of the lake (6.5 ± 1.1 vs. 7.9 ± 2.8, P = 0.65, t-test). The high filtration rate for the zoobenthos in Lake Bolshie Shvakshty was attributed to an unusually large biomass of chironomids, which can have a very high filtration rate (> 1,700 mL g-1 hr-1, Izvekova 1975).
DISCUSSION Feeding Functional Group Composition We found a dramatic shift in the benthic trophic structure after D. polymorpha invasion. The structure of feeding functional groups in the community including Dreissena was overwhelmingly dominated by collectors-filterers. D. polymorpha was the dominant benthic species in terms of biomass. These results are consistent with findings from other lakes and reservoirs in the former Soviet Union including Uchinskoe Reservoir, Russia (Lvova-Kachanova and Izvekova 1978, Sokolova et al. 1980a, Sokolova et al. 1980b) and Lake Lukomskoe, Belarus (Karatayev and Burlakova 1992, Karatayev et al. 1997). The invasion of D. polymorpha in Uchinskoe Reservoir resulted in the replacement of the dominant species, the chironomid filter-feeder Glyptotendipes paripes, and drastic changes in the relative abundance of different species and trophic groups (Lvova-Kachanova and Izvekova 1978, Sokolova et al. 1980a, 1980c). Following the invasion of D. polymorpha in Lake Lukomskoe the benthic community was characterized by an exceedingly high dominance of filterers, which accounted for 95% of the total benthic animal biomass (Karatayev and Burlakova 1992). As a result, the trophic structure of the littoral zone was impoverished, and the remaining trophic groups contributed relatively little to the total biomass. Similar patterns were found for six other waterbodies across the Former Soviet Union, where Dreissena comprised > 93% of the total biomass of benthic community (Karatayev et al. 1994, 1997). Other studies of benthic communities in zebra mussel beds have shown dramatic differences in the density and biomass of associated taxa compare to substrates without mussels (reviewed in Karatayev et al. 1997,
271 2002). The creation of new habitat is perhaps the most important effect that zebra mussels have on the benthic community. Several experimental studies have shown that the structure created by zebra mussels provides refuge for a variety of species, and this impact is seen even in the absence of the biological activity of filtering mussels (Slepnev et al. 1994, Ricciardi 1997, Stewart et al. 1998, 1999). Another possible mechanism through which zebra mussels may impact benthic community functional structure is through trophic impacts. Some studies have reported increases in the relative abundance of collectors due to enrichment of the benthos with organic substances from feces and pseudofeces created by zebra mussels or decreases in filterers due to competition for food with zebra mussels (Izvekova Lvova-Katchanova 1972, Stewart et al. 1998, Berezina 1999). All of these changes may be obvious within zebra mussel beds and areas with high mussel densities. However, in all lakes with zebra mussels much of the bottom is not covered by mussels, especially in the profundal zone (Stánczykowska and Lewandowski 1993, Burlakova, Karatayev, personal observations). Therefore, the overall effect on the entire benthic community might be much less pronounced than in areas with high zebra mussel densities. When the average structure of benthic trophic groups was pooled across lakes with and without zebra mussels, differences in the relative abundance of the feeding functional groups did not significantly change after zebra mussel invasion. However, we had relatively few lakes in our sample, limiting the power of our statistical tests. With greater sample sizes, the quantitative changes in scavengers could be significant. These results suggest that the major change in the community was the addition of zebra mussels rather than the displacement of other functional groups. However, the data for individual lakes indicated significant changes in several functional groups before and after zebra mussel invasion, but the changes were not in concert, and were often in opposite directions for the same functional group. Thus, these changes were masked when the data were pooled. These results do, however, highlight the importance and power of before and after data. To fully understand the impacts of zebra mussels on communities we need much more community data on lakes before and after invasion, and, unfortunately, these data are rare. Suspension-feeders are an integral component of aquatic ecosystems. They feed upon a very dilute food resource and convert previously dispersed, minute materials to larger animal biomass, i.e., their own bodies (Wallace and Merritt 1980). The overpowering dominance of collectors-filterers after zebra mussel invasion will drive changes in ecosystem function as they greatly enhance the rates of deposition of both organic and inorganic material on the bottom and thus build a direct connection between the planktonic portion of the water body and the benthos (benthic-pelagic coupling) (reviewed in Karatayev et al. 2002). In addition, Dreissena as well as many associated benthic animals are prey for benthivorous fishes (reviewed in Molloy et al.
272 1997, Karatayev et al. 1994). They may also provide an important path for moving energy from the benthic community to higher trophic levels.
Zebra Mussel as a Biofilter The role of bivalves in and impacts on aquatic ecosystems has long been recognised for marine and estuarine ecosystems (reviewed in Dame 1993, 1996). Bivalves can affect nutrient cycling by consuming particulate and dissolved organic matter and excreting inorganic nutrients. They affect community structure (both in the water column and on the benthos) and can influence community stability, diversity and interspecies links (Dame 1996). Zebra mussels create high densities over large areas in lakes and efficiently filter large volumes of water. They deposit substantial amounts of pseudofeces and feces on the bottom. Thus, they play the same ecosystem engineering role as marine bivalves (Karatayev et al. 2002). We found that before invasion by zebra mussels, the planktonic community filtered the equivalent to the volume of each lake within a few days, and were on average > 200 times more effective than the benthic community, which took four years to filter the same volume. We found no significant changes in the total biomass or filtration capacity of the zooplankton community in lakes after zebra mussel invasion. However, the total average biomass of the benthic community, including zebra mussels, increased 22 times, and filtration ability of the benthic community increased > 70 times. Consequently, the time required to filter the volume of each lake for the benthos was no different than that for the zooplankton community. The impacts of increased benthic filtration on the ecosystem will depend on the size of the Dreissena population, lake morphometry and rate of water exchange, and will be more pronounced in littoral zone where zebra mussels are most dense and less in profundal areas of deep lakes where zebra mussels are rare or absent. Our results are consistent with the findings of other studies. During the summer, D. polymorpha has been estimated to filter the volume of water equivalent to that of an entire waterbody from 5 to 90 days (Mikheev 1967, Stanczykowska 1977, Lvova et al. 1980, Protasov et al. 1983, Reeders et al. 1989, Karatayev and Burlakova 1995a, Petrie and Knapton 1999). After D. polymorpha invaded Lake Lukomskoe (Belarus), the filtration capacity of the benthic community increased 320 times, and the time to filter the equivalent of the volume of the lake decreased from 15 years to 17 days. At the same time zooplankton abundance declined, and the time required for the zooplankton community to filter the equivalent of the volume of the lake increased from 5 to 17 days (Karatayev and Burlakova 1992, 1995b).
273 In most freshwater ecosystems, benthic production is driven by the slow rain of suspended organic material to the bottom and to a lesser extent by the filtration activity of bottom suspension-feeders where most species feed on detritus or other benthic organisms. Consequently, the typical benthic freshwater system is considered to be detritus dominated, rather than relying on large amounts of primary productivity or direct links to planktonic processes. Usually, the benthos are not capable of controlling processes or dynamics in the planktonic system. Zebra mussels, filtering vast amount of water in a short period of time, provide a direct link between processes in the plankton and those in the benthos and by their deposition of pseudofeces and feces, provide a direct conduit for primary productivity in the water column to the benthos. Thus, they are able to control pelagic processes by removal of particulate matter, increasing water transparency and hence the volume of the photic zone, impacting phytoplankton standing stock, and, therefore, they can influence planktonic trophic interactions (reviewed in Karatayev et al. 2002). As a result, the role of the benthic community in lakes populated by zebra mussels increases tremendously and the benthos become capable of controlling processes and dynamics in the planktonic system and, therefore, the whole freshwater ecosystem. All of these effects are the direct result of changes in the trophic structure of the benthic community after zebra mussel invasion and the overwhelming dominance of one trophic group – filterers.
ACKNOWLEDGMENTS We would like to acknowledge our colleagues and former students of Belarussian State University Igor Rudakovsky, Galina Vezhnovets, Nataliya Lisovskaya, Yulyana Shilenko, Elena Makritskaya, Andrei Usov, Sergej Mastitsky and Vladimir Volosyuk for help in data collection and species identification. We appreciate the help of Dr. Tatyuna Zhukova and the staff of Narochanskaya Biological Station of Belarussian State University during field research. This study was supported by grants from Belarussian State University (Research Grant # 444/50 to AYK and LEB) and Stephen F. Austin State University (Faculty Research Grant # 14123 to AYK, LEB and DKP).
REFERENCES Alimov AF 1981 Patterns of the Growth of Freshwater Bivalve Molluscs. Nauka Press, Leningrad, 248 p (in Russian) Berezina NA 1999 Peculiarities of development of macrozoobenthos communities under influence of Dreissena polymorpha Pallas in experimental mesocosms. Zhurnal Obshchei Biologiii 60: 189-198 Dame RF 1993 Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes. Springer-Verlag, Heidelberg, 579 p
274 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Izvekova EI 1975 The nutrition and feeding links of the most abundant species of Chironomid larvae in the Uchinskoe reservoir. Summary of the Candidate Dissertation, Moscow State University, Moscow, USSR (in Russian) Izvekova EI Lvova-Kachanova AA 1972 Sedimentation of suspended matter by Dreissena polymorpha Pallas and its subsequent utilization by chironomid larvae. Pol Arch Hydrobioll 19: 203-210 Karatayev AY Burlakova LE 1992 Changes in trophic structure of macrozoobenthos of a eutrophic lake, after invasion of Dreissena polymorpha. Biologiya Vnutrennikh Vod. Inform. Byull 93: 67-71 Karatayev AY Burlakova LE 1995a Present and further patterns in Dreissena polymorpha (Pallas) population development in the Narochanskaya lakes system. Vestsi Akademii Navuk Belarusi. Seriya biyalagichnikh navuk 3: 95-99 (in Belarussian) Karatayev AY Burlakova LE 1995b The role of Dreissena in lake ecosystems. Russian J Ecol 26: 207-211 Karatayev AY Burlakova LE Padilla DK 1997 The effects of Dreissena polymorpha (Pallas) invasion on aquatic communities in Eastern Europe. J Shellfish Res 16: 187-203 Karatayev AY Burlakova LE Padilla DK 2002 Impacts of Zebra Mussels on aquatic communities and their role as ecosystem engineers. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds). Kluwer, Dordrecht, pp 433-446 Karatayev AY Burlakova LE Padilla DK Johnson LE 2003 Patterns of spread of the zebra mussel ((Dreissena polymorpha (Pallas)): the continuing invasion of Belarussian lakes. Biological Invasions 5(3): 213-221 Karatayev AY Lyakhnovich VP Afanasiev SA Burlakova LE Zakutsky VP Lyakhov SM Miroshnichenko MP Moroz TG Nekrasova MY Skalskaya IA Kharchenko TG Protasov AA 1994 The place of species in ecosystem. In: Freshwater Zebra Mussel Dreissena polymorpha (Pall.) (Bivalvia, Dreissenidae). Systematics, Ecology, Practical Meaning, JI Starobogatov (Ed). Nauka Press, Moscow, pp 180-195 (in Russian) Karatayev AY Makritskaya EN 1999 Zooplankton of lakes in Naroch Region. In: The Results and Future of Aquatic Ecology Research. Proceedings of the International Conference on Aquatic Ecosystems. Minsk, BSU Press, Belarus, pp 108-114 (in Russian) Kinzelbach R 1992 The main features of the phylogeny and dispersal of the zebra mussel Dreissena polymorpha. In: The Zebra Mussel Dreissena polymorpha: Ecology, Biological Monitoring and First Applications in the Water Quality Management, D Neumann and HA Jenner (Eds.). Gustav Fisher, Stuttgart, pp 5-17 Kryuchkova NM 1989 Trophic Relationships Between Zoo- and Phytoplankton. Nauka Press, Moscow (in Russian), 124 p Lvova AA Izvekova EI Sokolova NY 1980 The role of benthic animals in the conversion of organic matter and in the processes of waterbody self-cleaning. Trudy Vsesoiuznogo Gidrobiologicheskogo Obshchestva 23: 171-177 (in Russian) Lvova-Kachanova AA Izvekova EI 1978 Zebra mussel and chironomids from Uchinskoe reservoir. In: Plant and Animal Life in Moscow and its Environments, TN Dunaeva et al. (Eds). Moscow University Press, Moscow, pp 119-121 (in Russian) McMahon RF Bogan AE 2001 Mollusca: Bivalvia. In: Ecology and Classification of North American Freshwater Invertebrates, 2ndd Edition, JH Thorp and AP Covich (Eds). Academic Press, New York, pp 331-430 Merritt RW Cummins KW 1996 An Introduction to the Aquatic Insects of North America (3rd. ed.). Kendall/Hunt Publishing Co., Dubuque, IA, 862 p
275 Mikheev VP 1967 The nutrition of zebra mussels ((Dreissena polymorpha Pallas). Summary of the Candidate Dissertation. State Research Institute for Lakes and Rivers Fishery Industry, Leningrad, USSR (in Russian) Minchin D Lucy F Sullivan M 2002 Zebra mussel: impacts and spread. In: Invasive Aquatic Species of Europe - Distribution, Impacts and Management, E Leppäkoski S Gollasch and S Olenin (Eds). Kluwer, Dordrecht, pp 135-148 Monakov AV 1998 The Feeding of Freshwater Invertebrates. Russian Academy of Sciences. A. N. Severtsov Institute of Ecological and Evolutionary Problems, Moscow, 318 p (in Russian) Monakov AV 2003 Feeding of Freshwater Invertebrates Belgium, Kenobi Prod, 400 p Molloy DP Karatayev AY Burlakova LE Kurandina DP Laruelle F 1997 Natural enemies of zebra mussels: predators, parasites and ecological competitors. Rev Fish Sci 5(1): 27-97 Petrie S Knapton R 1999 Rapid increase and subsequent decline of zebra and quagga mussels in Long Point Bay, Lake Erie: Possible influence of waterfowl predation. J Great Lakes Res 25: 772-782 Protasov AA Afanasiev SA Ivanova OO 1983 The distribution and role of Dreissena polymorpha in the periphytone of Chernobyl water cooling reservoir. In: Molluscs: Systematics, Ecology and Patterns of Occurrence. Abstracts of the 7th Meeting on the Investigation of Molluscs, Nauka Press, Leningrad, pp 220-222 (in Russian) Reeders HH bij de Vaate A Slim FJ 1989 The filtration rate of Dreissena polymorpha (Bivalvia) in three Dutch lakes with reference to biological water quality management. Freshwater Biologyy 22: 133-141 Ricciardi A Whoriskey FG Rasmussen JB 1997 The role of the zebra mussel (Dreissena ( polymorpha) in structuring macroinvertebrate communities of hard substrata. Can J Fish Aqua Sci 54: 2596-2608 Rice WR 1989 Analysing tables of statistical tests. Evolution 43: 223-225 Slepnev AE Protasov AA Videnina YL 1994 Development of a Dreissena polymorpha population under experimental conditions Hydrobiol. J 30: 26-33 Sokolova NY Izvekova EI Lvova AA Sakharova MI 1980a Structure, distribution and seasonal dynamics of benthic densities and biomass. Trudy Vsesoiuznogo Gidrobiologicheskogo Obshchestva 23: 7-23 (in Russian) Sokolova NY Izvekova EI Lvova AA Sakharova MI 1980b The ecology of mass species of bottom invertebrates. Trudy Vsesoiuznogo Gidrobiologicheskogo Obshchestva 23: 39131 (in Russian) Sokolova NY Izvekova EI Lvova AA Sakharova MI 1980c Features of benthos forming in small waterbodies using Uchinskoe Reservoir as an example. Trudy Vsesoiuznogo Gidrobiologicheskogo Obshchestva 23: 161-170 (in Russian) Stanczykowska A 1977 Ecology of Dreissena polymorpha (Pall.) (Bivalvia) in lakes. Pol Arch Hydrobioll 24: 461-530 Stánczykowska A Lewandowski K 1993 Thirty years of studies of Dreissena polymorpha ecology in Mazurian Lakes of northeastern Poland. TF Nalepa DW Schloesser (Eds). Zebra Mussels: Biology, Impacts, and Control, Lewis Publishers, Boca Raton Stewart TW Gafford JC Miner JG Lowe RL 1999 Dreissena-shell habitat and antipredator behavior: combined effects on survivorship of snails co-occurring with molluscivorous fish. J N Am Benthol Soc 18: 274-283 Stewart TW Miner JG Lowe RL 1998 Quantifying mechanism for zebra mussel effects on benthic macroinvertebrates: organic matter production and shell-generated habitat. J N Am Benthol Soc 17: 81-94 Thorp JH Covich AP 2001 Ecology and Classification of North American Freshwater Invertebrates (2ndd Ed.) Academic Press, New York, 950 p Wallace JB Merritt RW 1980 Filter-feeding ecology of aquatic insects. Ann Rev Entomoll 25: 103-132 Zar HJ 1996 Biostatistical Analysis. (3rdd Ed.) Prentice Hall, New York, 662 p
DOES THE INTRODUCTION OF THE PACIFIC OYSTER CRASSOSTREA GIGAS S LEAD TO SPECIES SHIFTS IN THE WADDEN SEA?
Aad Smaal1, Marnix van Stralen2, Johan Craeymeersch1 1
Netherlands Institute for Fishery research, Centre for Shellfish research, Yerseke, The Netherlands 2 MarinX Consultancy, Elkerzee, The Netherlands Abstract: Over centuries dramatic changes have occurred in the species composition of the Wadden Sea, a shallow coastal sea bordering the North Sea. Natural dynamics as well as direct and indirect anthropogenic influences have resulted in the introduction and the disappearance of important benthic populations. Historic records and extensive surveys show large variability in benthic suspension-feeder stocks. Infaunal species like the cockle (Cerastoderma edule) are extremely variable over time and space, hence show a typical resilient response. Mussel (Mytilus edulis) beds seem to be more stable over time. Once lost, mussel beds need more time to re-establish bed structures. It is hypothesized that infaunal populations have a high resilience, while epifauna species are characterized by resistance to changes as they form structures like reefs or beds. On the basis of this hypothesis the consequences of new introductions can be evaluated. It can be expected that the recent introduction of the resistant reef-building epifaunal Pacific oyster Crassostrea gigas, will lead to shifts in benthic suspensionfeeder populations and eventually will develop a new stable state for the Wadden Sea that potentially offers less food for birds. This situation may deviate considerably from the actual nature conservation objectives that focus on the role of the Wadden Sea as one of Europe’s most important wetlands for migratory bird populations. Keywords: mussels, cockles, oysters, resistance, resilience, fishery.
INTRODUCTION The Wadden Sea (8000 km2) is a shallow estuarine ecosystem that stretches from Denmark to the Netherlands (Fig. 1). About 50 % of the Wadden Sea consists of tidal flats, and these are important habitats for benthic populations and their predators, predominantly wader birds. About 400 macrobenthic species occur in the Wadden Sea (Petersen et al., 1996). Cockles (Cerastoderma edule) and mussels (Mytilus edulis) dominate macrobenthos abundance. These suspension-feeder populations show a high natural variability over time and space, reflecting impacts of storms, 277 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 277–289. ©2005 Springer. Printed in the Netherlands.
278
Fig. 1. The Wadden sea area including transport routes of Crassostrea gigas (after Reise, 2004)
severe winters and variable recruitment success. As they are exploited species, there are also fishing or harvesting influences. These species are typically rstrategists showing high fecundity, broadcast spawning and a high potential for colonizing areas. Mussel stocks are less variable than cockle stocks, which might be related to the ability of the mussels to form bed structures. However, when these bed structures disappear due to storms or harvesting the restoration of bed structures may take quite a while (Dare et al 2004). Historic records show both extirpations (Wolff 2000) and introductions (Reise et al. 2002) of species in the Wadden Sea. With regard to benthic suspension-feeders the decline of the flat oyster (Ostrea edulis) in the period 1920 – 1940 is well-described (Dijkema 1997; Hagmeier 1941). Introductions of Mya arenaria (ca 1250), Mytilopsis leucophaeta (1835, brackish water species), Mercenaria mercenaria (1864), Crepidula fornicata (1887), Petricola pholadiformis (1890) and Ensis americanus (1978) have recently been reviewed by Reise et al (2002). According to Wolff (2000), the main causes of extermination are overexploitation and habitat destruction. To date, there is little evidence that introductions have driven native species to extinction (Reise et al. 2002). A relatively new species for the area is C. gigas
279 that was first observed on hard substrates along the dikes in 1983 (Bruins 1983). It has expanded rapidly since and also colonized tidal flats areas. In contrast to the introduced infaunal species it might be expected that the oyster, that builds reef structures on the tidal flats, will cause shifts in suspensionfeeder populations and potentially will bring the system to a new stable state. This potentially new state raises questions for nature management as present objectives focus on dominant bird populations that depend on the existing suspension-feeders such as cockles and mussels and not on those depending on oysters. On the basis of new data on the abundance and distribution of the introduced Pacific oyster in relation to the dynamics of existing benthic suspension feeder populations, we want to test the hypothesis that the rapid proliferation and the dominance of the Pacific oyster will result in shifts of suspension-feeders and a change in the state of the Wadden Sea ecosystem.
MATERIALS AND METHODS Annual Assessment of Shellfish Stocks From 1990 onward, each spring a survey is carried out by the Netherlands Institute for fishery Research (RIVO) to estimate the total standing stock of dominant bivalve species of the tidal flats of the Dutch Wadden Sea. The primary aim of the survey is to establish whether the stock size exceeds a given threshold that is conserved as food for wader birds. If the stock is below the threshold no cockle or mussel seed fishery is allowed. Following a stratified approach samples are taken at 1500 stations in the Wadden Sea (Fig. 2). The sampling grid is based on data from previous surveys. Also, results of explorative cockle surveys by fishermen just prior to our survey are used as well as aerial surveys prior to sampling in order to identify mussel and oyster bed structures. Sampling is done with a cockle-fishing vessel with a draught of 50 cm, allowing sampling of intertidal areas at high water. Samples are taken with an adapted mechanical cockle dredge that allows sampling of 0.5 m2/sample. Stations higher in the intertidal zone are sampled with a handheld sampling device (0.1 m2), operating on a small boat, or at low water with a corer (0.1 m2). Sampling depth is 7 cm, and mesh size is 5 mm. Quantitative sampling is realized for 15 species (Table 1). Samples are sorted, analyzed and wet weight is registered on board and data are directly stored in a database. In addition to sampling, mussel and oyster beds are quantified by satellite positioning (GPS) of the location and size of the individual beds, and by an estimate of composition and density. These data are stored in a database and analyzed with an Arcview geographic information system (GIS).
Ameland TERSCHELLING
Vlieland
Texel
Fig. 2. RIVO survey area of tidal flats in the Dutch Wadden Sea, with stratified sampling stations.
280
SCHIERMONNIKOOG
281 Table 1. Dominant benthic suspension-feeder species in the Dutch Wadden Sea with densities and standing stock in 2002, on the basis of an extensive spring survey on the tidal flats. (*) introduced species, see text Species Cerastoderma edule Crassostrea gigas* Donax vittatus
N/m2
106 kg
18.988 0.003
168.70 0.07
0.047
0.08
Ensis americanus*
2.061
0.76
Macoma balthica
51.910
50.78
Mya arenaria*
11.940
9.70
Mytilus edulis
34.140
50.30
Petricola pholadiformis*
0.007
0.11
Tellina tenuis
0.091
0.02
RESULTS AND DISCUSSION Dominant Populations The top 10 benthic suspension-feeder species for 2002 as they are found in our annual survey of the tidal flats of the Dutch Wadden Sea (Table 1). Densities vary over several orders of magnitude with Macoma as the most abundant species. In terms of biomass dominant species are the cockle (C. edule) and the mussel (M. edulis). Total stocks show large fluctuations over time (Fig. 3) and there is considerable spatial heterogeneity as shown in Fig. 4A and B for cockle and mussel beds. Large-scale fishery of cockles exists in the Dutch Wadden Sea – it was abandoned in the German and Danish Wadden Sea, while mussels are fished and cultured along the whole Wadden Sea (Smaal 2002, Kamermans and Smaal 2003). Among the top 10 list there are 4 introduced species: M. arenaria, P. pholadiformis, E. americanus and C. gigas. M. arenaria is the oldest record of anthropogenic impacts, related to transatlantic expeditions of the Vikings (Petersen et al. 1992, Wehrmann et al, 2000). It is a dominant species in the Wadden Sea and generally not considered as exotic. In contrast to the US East Coast where it originates from, no commercial exploitation occurs of Mya in NW Europe. Ensis has increased its biomass rapidly since first observations in 1978 near Helgoland, where it was introduced in the ballast water of ships. The species expanded through larval drift at about 100 km/year both in western and northern direction (Armonies and Reise 1999, Armonies 2001). It is now also dominant along the Dutch North Sea coast (RIVO data). Biomass is underestimated due to
282 insufficient sampling gear. At present 3 ships exploit the species in this area. The species is an important prey for Eider ducks (Reise 2004). Petricola was introduced more than 100 years ago by shipping and is less dynamic, has a low density and is not commercially exploited. The abundance of the Pacific oyster has increased dramatically on the tidal flats of the Dutch Wadden Sea and spatial distribution of Pacific oysters in the Dutch Wadden Sea is given in Fig 4C. Beds were found near Texel, in the central area, near Ameland and near Schiermonnikoog. The data in Table 1 shows the results of the 2002 survey. From our recent survey of 2004, total stock was calculated as 11.5 million kg wet weight forming at least 400 ha of bed area of > 200 g/m2. New beds were found over the whole Dutch Wadden Sea and also in the Ems estuary.
Proliferation of Crassostrea gigas The proliferation of the Pacific oyster in Europe started after the severe winter of 1962/63 when the species was introduced to restore oyster cultures (flat oyster, Ostrea edulis) that almost became extinct due to the frost. Apart from unsuccessful introductions in the Wadden Sea in 1913 in Lower Saxony (Wehrmann et al. 2000), it was not until the Pacific oyster was introduced in 1964 in the Oosterschelde estuary (SW Netherlands) that they developed rapidly (Drinkwaard 1999). Since 1974 culture trials were done in Lower Saxony, but none of these introductions resulted in an economically feasible population (Wehrmann et al. 2000). In the eastern Wadden Sea a culture started near Sylt in 1986 (Reise 1998). It is obvious that the occurrence of wild Pacific oysters in the Wadden Sea is due to aquaculture introduction, but it is remarkable that proliferation occurred over quite large distances. First records of wild Pacific oysters date from 1983 near Texel, the western part of the Dutch Wadden Sea, and from the size it was estimated that these oysters had been there since 1977. The oysters may have been introduced accidentally by mussel boats that frequently sail between the Oosterschelde and the Wadden Sea, although the only transport of shellfish is from the Wadden Sea to the Oosterschelde and not vice versa. Also larval drift may have bridged the gap, as a distance of 150 km can be reached by larvae within 20 – 25 days on the basis of a residual current of 0.1 m/s, while the life span of pelagic larvae is up to 30 days (Wehrmann et al. 2000, but see also Cadée 2000). Further records date from 1998 in the Ems-Dollard area, mainly on hard substrates on the dikes (Tydeman 1999) and along the Lower Saxony coast, particularly on mussel beds (Wehrmann et al. 2000). Around the island of Sylt first observations of wild Pacific oysters are from 1991 near the culture area, and in 1995 wild beds were observed south of Sylt and later on also in northern areas (Fig. 1). Apparently the local culture
2004
2002
2000
1998
1996
1994
A. Cockles
1992
500 450 400 350 300 250 200 150 100 50 0
1990
mln kg fresh weight
283
mln kg fresh weight
70 60
B. Mussels
50 40 30 20 10
1996
1997
1998
1999
2000
2001
2002
2003
1996
1997
1998
1999
2000
2001
2002
2003
1995
1994
1993
1992
1991
90 80 70 60 50 40 30 20 10 0
1995
1994
1993
1992
1991
C. Macoma
1990
mln kg fresh weight
1990
0
Fig. 3. Variations in stock size of major benthic suspension-feeder populations in the Dutch Wadden Sea (A) cockles, (B) mussels and (C) Macoma
284
A. Cockles
B. Mussels
Fig. 4. Spatial distribution of cockles (2002), mussels (2002) and Pacific oysters (2004). In Figs. 4A and B, the intertidal area is grey. In Fig. 4C, the sampling stations are shown in grey.
285 population was the source for proliferation near Sylt, while the other Wadden Sea areas have been colonized by long distance larval dispersal with the Oosterschelde population as a resource.
Dynamics of Crassostrea gigas The dynamics of the oyster population is clearly related to recruitment success. Observed increases in abundance always followed years with good recruitment, and these correspond with extended periods of high summer temperature (Wehrmann et al. 2000). This was documented for the Oosterschelde in 1976, 1982, 1986, 1987, 1989, 1992 (Drinkwaard 1999) and a series of recent years. From aerial photographs, a reconstruction was made of the expansion of oyster beds in the Oosterschelde from 1980 onwards. It shows a development from 30 ha (1980) to 300 ha (1990) and is now 700 ha (2000) (Kater and Baars 2003). Meanwhile sublittoral stocks have been estimated on the basis of side scan sonar survey as approximately 700 ha (Kater et al. 2002). Climate events like severe winters or warm summers, or biotic processes like predation or diseases do not seem to hamper the increase of the population. Warm summers are considered as the main vector for extension of the population (Reise 2004). In 2002 a good spatfal occurred and this is reflected in the dramatic increase of biomass as recorded in the Wadden Sea, from 0.07 million kg in 2002 to 11.5 million kg in 2004. In 2003 summer was extremely warm and high larval concentrations were measured in a monitoring program in the Dutch Delta area (Kamermans, pers. com.), hence a further biomass increase is to be expected. For the Dutch Wadden Sea, early observations are quite limited and most records are from 1995 onwards (Dankers et al. 2004). It is only since 2002 that oysters have become apparent in our annual spring survey in a quantitative way.
Further Developments It is expected that the Pacific oyster will expand further on the tidal flats in the Wadden Sea. From our survey we are able to quantify the standing stock and the oyster bed area in the Dutch Wadden Sea, and a rapid increase in only a few years is evident. Observations from the Oosterschelde estuary show a continuous expansion on the tidal flats with a doubling of covered area in 10 years time. Proliferation started on hard substrates along the dikes in the seventies but by 1980 about 30 ha of bed area could be quantified from aerial
286 photographs of tidal flats. The colonialization of tidal flats starts with settlement of larvae on shell fragments and alike, and then the oyster shells form substrate for new spat. Hence a positive feedback exists when oysters have settled and start colonizing an area. This mechanism obviously results in the formation of reefs and dense oyster beds. We also observed the occurrence of adult oysters in areas that had no oysters the year before. Apparently, wave driven transport of individual oysters or clumps of oysters occurs and new areas can be colonized in the vicinity of existing beds through this mechanism. As shown, recruitment is the driving force of pacific oyster expansion and this is particularly successful in warm summers. As climate change may result in an increase of summer temperatures in the Wadden Sea area, it is quite likely that the expansion rate will increase. This fits in the observation that recruitment of benthic species changes under the influenced of temperature rise (Beukema 1992). Without control measures it is expected that the Pacific oyster will become a dominant species in the Wadden Sea and will change the landscape from tidal flats with dominant infauna and mussel beds to areas that are at least partly covered with oyster reefs. The oyster has a large filtration and biodeposition capacity, hence will play a dominant functional role as well in the Wadden Sea. As shown by Dame (1996, 2005), oyster reefs are complex systems, characterized by multiple feedbacks and a high level of selforganization. These feedbacks stabilize the reef system and create resistance to changes. Eventually a new stable state of the ecosystem may be established.
Potential Impacts and Measures The distribution of the Pacific oysters on the tidal flats may change the landscape. As mentioned, oysters form reefs and dense beds, and outcompete local infauna. Particularly in the Wadden Sea, oysters also settle on existing mussel beds. This was observed in the Dutch Wadden Sea (Dankers et al. 2004) and in Lower Saxony (Wehrmann et al. 2000) and Sleswich Holstein (Reise 1998). It is possible that mussel beds will gradually be taken over by the oysters. The Pacific oyster has a large filtration capacity and filters on average 5 l/g/h (Bougrier et al. 1995) but values up to 25 l/g/h have been recorded (K. Troost pers. comm, Ren et al 2000). As a consequence, there is a potential competition for food with other suspension-feeders and this may have impact on dominant mussel and cockle populations. Current research is focused on the potential of oysters to filter bivalve larvae. It is hypothesized that the oyster reefs may filter large quantities of larvae from the water column, hence reduce the recruitment success of species like macoma, cockle and mussel.
287 Also the competition for food may have impact on the early life stages of suspension-feeders as they also depend on phytoplankton. If the Pacific oysters have a strong competition potential when compared to other benthic suspension-feeders, a decrease in the population size of the other species can be expected. This may have consequences for the food chain, as the Wadden Sea has large bird populations feeding on bivalves like cockles, macoma and mussels. These birds, like - despite its name - the oystercatcher, are not able to feed on the Pacific oyster. The question is therefore what may limit the Pacific oyster. As it is an exotic species it can be expected that it may take quite a while before a natural regulation process will evolve. As shown, predation is not a serious threat so far for the Pacific oyster in this area, as major bird populations cannot handle the species. In France the predatory gastropod Ocinebrellus inornatus has impact on cultured oyster stocks, but this species does not occur in the Wadden Sea. Harvesting Pacific oysters is at present not allowed due to nature conservation directives. The quality of the oysters for consumers is excellent. Therefore there is a chance to harvest good quality product and meanwhile manage the size and spatial distribution of the stock. However, once the oysters have developed reefs the product quality for the consumers decreases dramatically due to clumping, increase of shell size and decrease of meat content. Combining management and harvesting requires therefore timely decisions.
CONCLUSIONS The Pacific oyster is rapidly becoming a dominant suspension-feeder on the tidal flats of in the Wadden Sea. The oyster reefs form a complex system with multiple feedbacks, a high level of self-organization, and a high resistance. As they colonize both mussel-beds and infaunal habitats, and they have a large filtration and biodeposition capacity, the species has a strong competition potential. Shifts in benthic suspension-feeder population may be expected and as a consequence food availability for dominant bird populations may decrease. In contrast to earlier invasions of resilient infaunal suspensionfeeders like E. americanus, the introduction of C. gigas may result in a new stable state in the Wadden Sea ecosystem with large areas of oyster reefs and lower stocks of other benthic suspension-feeders and related bird populations.
Acknowledgements The authors are grateful to Richard Dame for comments and improvements of the manuscript, and to Josien Steenbergen, Divera Baars and
288 Joke Kesteloo for help with data collection and analysis. Data collection was done under contract of the WOT Programme of the Dutch Ministry of LNV. NATO ARW is acknowledged for financial support of the first author to participate in the Nida workshop.
REFERENCES Armonies W 2001 What an introduced species can tell us about the spatial extension of benthic populations. Mar Ecol Prog Ser 209: 289-294 Armonies W Reise K 1999 On the population development of the introduced razor clam Ensis americanus near the island of Sylt (North Sea). Helgoländer Meeresunters 52: 291-300 Beukema JJ 1992 Expected changes in the Wadden Sea benthos in a warmer world: lessons from periods with mild winters. Neth J Sea Res 30: 73-79 Bougrier S Geairon P Deslous-Paoli JM Bacher C Jonquieres G 1995 Allometric relationships and effects of temperature on clearance and oxygen consumption rates Crassostrea gigas (Thunberg). Aquaculture 54: 143-154 Bruins RBW 1983 Crassostrea gigas op Texel. Correp Blad Ned Malacol Verr 215: 14361438 (in Dutch) Cadée G 2000 Japanse oester (Crassostrea gigas) populaties tussen Oudeschild en Mok, Texel. Het Zeepaardd 60: 260-269 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press Boca Raton, FL, 254 p Dame RF 2005 Oyster reefs as complex ecological systems. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 331-344 Dankers NMJA Dijkman EM Jong ML de Kort G de Meijboom A 2004 De verspreiding en uitbreiding van de Japanse oester in de Nederlandse Waddenzee. Alterra rapport 909, 51 p(in Dutch) Dare PJ Bell MC Walker P Bannister RCA 2004 Historical and current status of cockle and mussel stocks in the Wash, CEFAS, Lowestoft, 85 p Dijkema R 1997 Molluscan fisheries and culture in The Netherlands. NOAA Technical Report NMFS 129: 115-136 Drinkwaard AC 1999 Introductions and development of oysters in the North Sea: a review. Helgoländer Meeresunters 52: 301-308 Hagmeier A 1941 Die intensive Nutzung des nordfriesischen Wattenmeeres durch Austernund Muschelkultur. Z Fischerei 39: 105-165 Kater B Baars D 2003 Reconstructie van oppervlakten van litorale japanse oesterbanken in de Oosterschelde in het verleden en een schatting van het huidige oppervlak. RIVO rapp C017/03 Kater B Baars D Perdon J van Riet M 2002 Het inventariseren van sublitorale oesterbestanden in de Oosterschelde mbv side scan sonar. RIVO rapp C058/02, 26 p Kamermans P Smaal AC 2003 Mussel culture and cockle fishery in the Netherlands: finding a balance between economy and ecology. J Shellfish Res 21: 509-517 Petersen GH 1996 Red list of macrofaunal benthic invertebrates in the Wadden Sea. Helgoländer Meeresunters 50(suppl): 69–76 Petersen KS Rasmusen KL Heinemeier J Rud N 1992 Clams before Columbus? Nature 359: 679 Reise K 1982 Long-term changes in the macrobenthic invertebrate fauna of the Wadden Sea: are the polychaetes about to take over? Neth J Sea Res 16: 29-36
289 Reise K 1998 Pacific oysters invade mussel beds in the European Wadden Sea. Senckenbergiana maritima 28: 167-175 Reise K 2004 Contribution to Wadden Sea quality status report, in press Reise K Gollasch S Wolff WJ 2002 Introduced marine species of the North Sea coasts. In: In: Invasive Aquatic Species of Europe: Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 260-266 Ren JS Ross AH Schiel DR 2000 Functional descriptions of feeding and energetics of the Pacific oysterr Crassostrea gigas in New Zealand. Mar Ecol Prog Serr 208 : 119-130 Smaal AC 2002 European mussel cultivation along the Atlantic coast: production status, problems and perspectives. Hydrobiologia 484: 89-98 Tydeman P 1999 Japanse oesters in de Eemshaven. Het Zeepaard 59(2): 58-64 Wehrmann A Herlyn M Bungenstock F Hertweck G Millat G 2000 The distribution gap is closed – first records of naturally settled Pacific oysters Crassostrea gigas in the East Frisian Wadden Sea, North Sea. Senckenbergiana maritima 30: 153-160 Wolff WJ 2000 Causes of extirpations in the Wadden Sea, an estuarine area in The Netherlands. Cons Bioll 14(3): 876-885 Wolff WJ Reise K 2002 Oyster imports as a vector for the introduction of alien species into northern and western European coastal waters. In: In: Invasive Aquatic Species of Europe: Distribution, Impacts and Management, E Leppäkoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 193-205
ONE ESTUARY, ONE INVASION, TWO RESPONSES: PHYTOPLANKTON AND BENTHIC COMMUNITY DYNAMICS DETERMINE THE EFFECT OF AN ESTUARINE INVASIVE SUSPENSION-FEEDER
Janet K. Thompson United States Geological Survey, Menlo Park, CA, US Abstract: Invasive suspension-feeding bivalves have reduced phytoplankton biomass in many aquatic systems, which has resulted in loss of trophic complexity in some systems. Using an example of one invasive bivalve in San Francisco Bay, Potamocorbula amurensis, the causes of differing system level responses are explored. San Francisco Bay, similar to of other shallow, turbid, non-nutrient limited, but low productivity systems, is likely to be most stressed by the loss of primary producers. While the northern bay has lost primary production following the invasion of P. amurensis, the southern bay (SB) has not and these differences are shown to be due to the different mechanisms responsible for the seasonal turbidity in the systems. Because the period of lowest turbidity in SB is coincident with the period of lowest bivalve grazing, the southern bay has not seen a reduction in its high magnitude but short spring bloom. A method for predicting if a shallow, turbid and nutrient replete estuary might lose phytoplankton production with a sudden increase in suspension-feeders is explored. Keywords: suspension-feeder, bivalve, exotic, grazing, San Francisco Bay, phytoplankton
INTRODUCTION Estuaries are recipients of a high rate of human-mediated invasions by non-indigenous species. Ruiz et al. (2000) report a minimum of 374 successful invasions of algae and invertebrates, more than 100 successful invasions of fish, and over 200 successful invasions of vascular plants in North American coastal marine communities. Anthropogenic transport of species is recent in North America, so it is expected that these numbers are not large, relative to the historical invasions of Europe (Leppakoski et al. 2002). Because human mediated invasions are believed to be accelerating in many geographic areas on all continents (Ruiz and Hewitt 2002), it is important to understand how 291 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 291–316. ©2005 Springer. Printed in the Netherlands.
292 they affect aquatic ecosystems. One group of invasive organisms, those that suspension-feed, are of particular interest because they have changed estuaries at the system scale by reducing phytoplankton biomass (Alpine and Cloern 1992, Cohen et al. 1984) resulting in declines at higher trophic levels (Kimmerer 2002, Ojaveer et al. 2002, Shiganova 1998), by changing substrate character and thereby changing the suitability of the bottom for commercially important species (Chauvaud et al. 1998), and by altering nutrient regeneration rates (Chauvaud et al. 2000, Ojaveer et al. 2002) thereby potentially affecting phytoplankton community composition and growth rates. However, there have also been many estuaries with a rich history of invasive suspensionfeeder species that have not responded with these types of system level effects. Because of these disparate responses, invasions of suspension-feeders in aquatic systems can be viewed as system-scale experiments that allow us to explore what characteristics allow systems to be vulnerable to suspensionfeeder control of the food web. We can predict, based on an elegant and simple relationship relating average annual system clearance time by bivalves (CT) to water mass residence time (RT) developed by Smaal and Prins (1993), if the phytoplankton have the potentiall of being controlled by bivalve suspensionfeeders in most systems. The question that we are interested in here is slightly different: Is it likely that an increase in benthic suspension-feeding will control the phytoplankton biomass of a system? To address this question, we should include a combination of factors that are required for phytoplankton growth and accumulation: sufficient nutrients, sufficient light, and grazing loss rates that are low, relative to phytoplankton growth rates. We will limit our need to account for all of these factors by focusing on that class of estuaries whose phytoplankton and trophic webs are most vulnerable to overgrazing - estuaries that have low primary production despite non-limiting nutrients, due to a combination of light limitation and grazing pressure. As shown by Monbet (1992) and Cloern (2001) these types of systems are broadly distributed and common. We begin by exploring the vulnerabilities of estuarine food webs to invasive suspension-feeders by examining the specifics of an invasion of one bivalve suspension-feeder in the San Francisco Bay. We will see that the physical characteristics of the different parts of this system resulted in different phytoplankton responses to the invader. Using this example as a foundation, I will then develop a relationship to test when phytoplankton in turbid, low production estuaries are most vulnerable to significant reduction in biomass with invasive benthic suspension-feeders. I will conclude with a discussion on what we have learned about estuarine ecosystems that may or may not have changing, alternate ecosystem states.
293 The Estuary as an Environment for Benthic Suspension-feeders and for Phytoplankton Estuarine ecosystems are vulnerable to change by invasive suspension-feeders due to the characteristics of a successful invader, the inherent vulnerability of primary producers to suspension-feeders in shallow systems, and the low primary production of many estuaries (Cloern 2001, Heip et al. 1995). All benthic species must survive a range of environmental stresses, such as rapidly changing salinity and turbidity, in order to be successful in the estuarine environment. However, successful invasive species may be particularly tolerant and resilient to the wide temporal and spatial ranges of physical stressors having survived the selection process during transit to, and inoculation into a new system. Once established, those invasive benthic suspension-feeders that can expand into the full range of physical environments and can tolerate the largest range of physical stresses induced by tidal action can benefit from these same physical forces. Food supply for the benthos is increased in tidal regions due to: (1) increased vertical mixing rates, (2) advective transport of food sources and (3) in some cases increased water column residence time resulting in higher primary producer growth and accumulation.
Figure 1 Map of San Francisco Bay with place-names used in text.
The physical environment can limit primary production in estuaries relative to coastal systems and most freshwater systems (Cloern 2001).
294 Although phytoplankton in some estuaries are nutrient limited, many estuaries receive high concentrations of nutrients with little effect on primary production (Heip et al. 1995) reflecting the importance of other physical factors in estuarine phytoplankton growth. Light limitation is so common in estuaries that many estuaries have photic depths that are limiting to phytoplankton growth in the deepest portions of the system and during at least some portions of the year in the shallow water due to a combination of tidal mixing, wind resuspension of sediments, and freshwater inflow. As we have begun to understand more about the origin and variability of phytoplankton blooms (Lucas et al. 1999a, b and Heip et al. 1995) and estuarine turbidity (Cloern 1999, Monbet 1992), we see that the shallow water regions and the exchange between shallow and deep water in each system may determine the temporal and spatial variability of primary producers in estuarine systems. Thus any natural or anthropogenic interruption of this exchange may further limit primary production.
San Francisco Bay: Physical Setting and Exotic Species History San Francisco Bay (SFB) is a temperate, middle-latitude estuary located in the center of a large urban center (>8 million people), and is the natural terminus for 70% of the state’s annual freshwater runoff. However, 80% of the freshwater needs occur in the southern portion of the state and thus about 50% of the freshwater that historically flowed into SFB is now being retained for urban and rural consumption and hydroelectric power (Peterson et al. 1985). The bay is very shallow (median depth is 2.8 m at MSL), mesotidal, and has unequal, semi-diurnal tides. The northern bay can be divided into two sections (Figure 1), North Bay (NB: Honker, Suisun, and Grizzly Bays), through which the majority of freshwater enters, and San Pablo Bay (SPB), the next down-bay system. The most southerly bay, South Bay (SB), is the most saline system, except for the bay at the ocean interface (Central Bay), due to the diversion of most of the freshwater from the southern watershed. The two major systems, NB and SB, are therefore hydrologically and hydrodynamically very different. Freshwater flow peaks in NB in early winter and again in spring with the release of snowmelt water from reservoirs and varies in magnitude between years and seasons. High freshwater flow results in short residence times (§1 day) and high suspended loads in winter and spring. High turbidities also occur in summer due to the resuspension of the sediment deposited during freshwater runoff, by semi-diurnal tides and summer diurnal winds. Large interannual variability in precipitation (Figure 2) is responsible for ill-defined “average” conditions in NB. Salinity ranges
295 from 0-25 psu and the salinity gradient is substantial throughout NB, which is classified as a river-dominated partially mixed estuary. In contrast to the NB, the majority of the freshwater flowing into SB is from sewage effluent, and during very wet years, the southerly advection of the snowmelt runoff. Residence times are longer than in NB (in excess of 14 days during summer and fall, Gross 1997, Walters et al. 1985) and the turbidity, although less than in NB, is still sufficient to restrict phytoplankton growth in the narrow channels (Cloern et al. 1985). The SB is classified as a lagoonal estuary (salinity range of 15-30 psu). Persistent stratification is rare in either system for periods longer than about two weeks except in limited areas of NB where channels have sharp bathymetric gradients (Monismith et al. 2000). Neither system is very productive (20 g C m-2 yr-1 in NB and 150-200 g C m-2 yr-1 in SB, Alpine and Cloern 1992 and Cloern 1987) or nutrient limited. Primary production is predominantly light limited with significant top down control by benthic grazers in both systems (Cloern 2001). The major organic carbon source throughout the system is phytoplankton (Jassby et al. 1993). San Francisco Bay may be unique due to the dominance and number of invasive species in the system: there are a minimum of 234 known invasive species, and an additional 125+ cryptogenic species in this system (Cohen and Carlton 1998) with an average of one new species arriving every 24 weeks since 1970 (Cohen and Carlton 1995). The discovery of gold in the foothills and mountains east of San Francisco Bay in 1849 resulted in an influx of 48,000 people into the area by ship in just the first two years of the gold rush. The introduction of organisms living on the ships and in their dry ballast is likely responsible for many of the early invasions, although the largest pulse of invasions prior to 1940 occurred with the near continuous import of oyster spat from the western Atlantic (1869-1910) and the eastern Pacific (19321939) for culturing (Bonnott 1935). The rapid increase in invasive species that began in the 1940’s, following the conversion of dry ballast to wet ballast in most ships, is convincingly argued to be due to ballast water inoculations by Carlton (1985) and Cohen and Carlton (1998). The majority of the identified invasive species are invertebrates (147, Cohen and Carlton 1995) and the majority of those are suspension-feeders. Some of the invasive bivalve suspension-feeders became sufficiently successful that they were commercially harvested until a combination of poor meat quality and contamination made them unpalatable. Given this history of species invasions, it was not surprising to find a new bivalve, Potamocorbula amurensis, a suspension-feeding, euryhaline corbulid, in the NB in 1986. It was however, quite surprising to watch the speed with which it spread and its apparent tolerance of a wide range of environmental conditions (Carlton et al. 1990). Today, we find P. amurensis in the full sediment, depth, and salinity range of the estuary. Although we do not believe that it can reproduce in fresh water, it can osmoregulate as a larva
296 (10-30 psu at 2 hours after fertilization; Nicolini and Penry 2000) and juveniles and adults are commonly found at the freshwater boundary where the animals are routinely exposed to fresh water during some tidal conditions. Although P. amurensis was first found in NB, it quickly spread to the SB in 1988 (Thompson unpublished data), where it became a dominant bivalve by 1990. However, it has not been as interannually persistent in SB as in NB.
Data Sources and Methods Water column chlorophyll a values used in this paper are courtesy of the US Geological Survey and the California Department of water resources. The data are available on the following databases: http://sfbay.wr.usgs.gov/access/wqdata/ http://sarabande.water.ca.gov:8000/~bdtdb/ Biomass estimates of P. amurensis in NB that were used for estimated grazing rate were based on benthic samples loaned to the author by the California Department of Water Resources from the longest sampled benthic station in the estuary. Although only one station has been sampled consistently over the 30 year time period, other spatially intensive surveys of the benthos have shown P. amurensis biomass at this station to be reasonably representative of the biomass at other shallow water locations in Grizzly Bay (Hymanson 1991). Biomass and grazing rate estimates for SB suspensionfeeders represent an average of 41 to 76 stations and were reported in Thompson (1999). Grazing rates are based on measured filtration rates for P. amurensis (Cole et al. 1992) and Venerupis japonica (O’Riordan et al. 1995) and literature values for Mya arenaria and Musculista senhousia and all have been seasonally corrected for temperature, maximum concentration boundary layer development as suggested by O’Riordan et al. (1995), and a feeding frequency of 67%; this percentage represents the intermittent feeding behavior that was observed by the author in the field and in a laboratory flume for these species.
SAN FRANCISCO BAY – A TWO ESTUARY LABORATORY FOR STUDIES OF SUSPENSION-FEEDER PROCESSES P. amurensis has been blamed for the demise of the phytoplankton bloom (Figure 2) in the northern-most bays and has been shown to be capable of consuming the phytoplankton at a sufficient rate to be effectively controlling phytoplankton biomass in these systems (Alpine and Cloern 1992,
297 Thompson et al. in review). Although P. amurensis has also successfully invaded SPB and SB, we have not seen similar, significant declines in phytoplankton biomass (Figure 3) in these systems. The basic mechanisms for phytoplankton growth are similar for all embayments in this nutrient replete but light-limited estuary. The shallow portions throughout the bay are the most productive with chlorophyll a concentrations and primary production
80
Chlorophyll a (Pg/L) g
70
Grizzly Bay
60 50 40 30 20 10 0 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00
20000 16000
3
Freshwater Flow (m m/s)
18000 14000 12000 10000 8000 6000 4000 2000 0 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00
Figure 2. a) Phytoplankton biomass in Grizzly Bay prior to the invasion of P. amurensis in late 1986 and after, and b) hydrograph of same period. Data from state of California database at http://sarabande.water.ca.gov:8000/~bdtdb/
298 being two to three times higher in the shallow water than in the channels (Cloern et al. 1985). Field and modeling studies have shown that turbidity limits phytoplankton growth in the channels, and system-wide blooms cannot be maintained by phytoplankton production in the channel (Cloern 1996, Lucas et al. 2001a, in review). In addition, top down control is less likely to occur in the channel due to the reduced vertical mixing rates. In spite of these similarities in the embayments, the responses of these systems to P. amurensis grazing have been quite different. Mechanisms of Phytoplankton Control in North Bay The three freshest bays (HB, SUB, and GB) have historically had a small magnitude phytoplankton bloom that lasted through summer and fall of most years. The primary production for this system was controlled by a combination of residence time and turbidity. Cloern (1996) suggests that during periods of high freshwater flow, the residence time declines sufficiently in NB to eliminate the annual bloom (see Figure 2, years 1983 and 1986, Figure 4a). With normal freshwater flow, residence time increases
Figure 3. Integrated annual chlorophyll a concentrations in each embayment prior to (first box plot), and after (second box plot) the invasion of P. amurensis into the estuary. *** Integrated chlorophyll a values are significantly different (p0.001) for the two time periods in an embayment.
299
Chlorophyll a ( Pg/L/bloom) g
a.
7000 6000 5000 4000
p pre-1987
3000 2000 post-1987
1000 0 0
2E+11
4E+11
6E+11
8E+11
Integrated Freshwater Inflow- m3/bloom
Chlorophyll a ( Pg/L/yr)
b.
4000 3500 3000 2500 2000 1500 1000 500 0 0
5E+11
1E+12
1.5E+12
2E+12
2.5E+12
3E+12
3
Integrated Freshwater Inflow- m /yr Figure 4. Relationship between integrated freshwater inflow and phytoplankton biomass in (a) NB (integrated over the bloom period) and (b) SB (integrated annually).
and turbidity declines in late spring, thereby allowing a slow growing bloom to develop. Cloern et al. (1983) also hypothesized that there was likely to be an optimum freshwater flow that would allow phytoplankton growing in the shallows to be tidally exchanged with the deeper water and still be retained in the embayment, thereby allowing the phytoplankton biomass to accumulate. It can be seen that there is now no significant relationship between freshwater flow and productivity since P. amurensis invaded the system (Figure 4a). P. amurensis grazing rates (Figure 5), peak in fall and are at a minimum in the
300 winter and early spring in a pattern that is opposite to that of chlorophyll a concentration. The minimum bivalve biomass in winter and early spring throughout the shallow reaches of NB, SPB, and SB is believed to be due to predation by birds that immigrate into SFB each fall, as part of the Pacific Flyway migration (Poulton et al. 2002). Although benthic grazing appears to be controlling phytoplankton biomass in NB, it is unlikely that the other mechanisms of residence time and light availability are irrelevant processes.
Mechanisms of Phytoplankton Control in South Bay SB phytoplankton blooms are short (usually 2-4 weeks) and have the highest peak chlorophyll a concentrations in the bay (60-80 µg/L). The mechanisms for bloom development in SB have been well studied (Cloern 1996, Thompson et al. in review) and modeled (Cloern 1982, Lucas et al. 1999a, 1999b, in review, May et al. 2003). As in other portions of the bay, light is limiting and blooms are mostly generated in the shallow water during periods of decreased tidal and wind mixing. Persistent stratification, which occurs only during periods with high freshwater flow, tends to increase the magnitude of the bloom (Figure 4b). Similar to what we see in NB, phytoplankton biomass increases during periods when bivalve grazing rates are at their lowest levels, each winter/spring (Figure 5b). Bivalve recruitment occurs in spring in SB and bivalves may grow sufficiently during the bloom, to limit the magnitude and duration of the bloom (see 1995, Figure 5).
A Summary of Differences – Why Do the Phytoplankton Bloom in South Bay? The factors that control turbidity in NB are, in order of importance, freshwater inflow (Schoellhamer 2002), tidal resuspension at fortnightly time scales, and tidal and wind wave resuspension at the diurnal time scale (Ruhl et al. 2001). The high turbidity during the high river flow period (January through April) limits the period for phytoplankton growth to summer and fall, which is the period when P. amurensis populations begin building and peaking in biomass. The SB system has more tidal energy (tidal range of 2m), and as defined by Monbet (1995) is fairly typical of tidally energetic estuaries in that tidal resuspension (at fortnightly and semi-diurnal scales) is the main contributor to turbidity. However, early season freshwater inflow can also bring highly turbid water into SB as can diurnal wind resuspension in summer periods (Schoellhamer 1996). This leaves a short period when the
301 phytoplankton are not light limited in SB and that, by chance, corresponds to the period when shallow-water bivalve suspension-feeders are at their annual minimum biomass.
12 2 9 6 1 3 0
0 1991
ƇChlorophyll a (P g/L)
b.
1992
1993
1994
1995
5
70 60
4
50 40
3
30
2
20 1
10 0 1991
Ƒ Benthic Grazing (m/d)
3
15
Ƒ Benthic Grazing (m/d)
ƇChlorophyll a (P g/L)
a.
0 1992
1993
1994
1995
Figure 5. Phytoplankton biomass (chlorophyll a concentration) and grazing rates of benthic bivalves in (a) NB and (b) SB.
CAN WE PREDICT WHEN A NEW SUSPENSION-FEEDER WILL EXERT TOP-DOWN CONTROL ON THE PHYTOPLANKTON IN A TURBID SYSTEM?
302 The large trophic changes that have resulted from Dreissena sp. and Corbicula sp. invasions in North American and European ecosystems, in addition to the invasion described here, highlight the necessity of understanding which systems are most at risk of large trophic changes from invasive suspension-feeders. My goal is to use what we have learned in these two systems to (1) determine if the CT/RT model developed by Smaal and Prins (1993) is sufficiently sensitive to answer this question and if not, (2) to develop a method to predict the response of phytoplankton in turbid, nutrient rich estuaries to an increase in suspension-feeding. The best method to answer this question is, of course, to develop a multi-dimensional hydrodynamic model that is connected to a biological process model. Given that few systems have such models or the data needed to validate the models, there is a need to develop simple tools that address some of our basic questions.
Figure 6. The clearance time (CT days), residence time (RT days), primary production turnover time (PPT days) relationship developed by Dame and Prins (1998) with new San Francisco Bay data added to the plot.
The ideal requirements for such a tool include that it be simple and intuitive, that it use clear, easily derived parameters that accurately portray the important ecological processes, and that the findings be consistent with our observations. The Smaal and Prins (1993) CT/RT model was described earlier (hereafter referred to as the CR model) and was developed to address questions about when commercially important bivalves had reached the
303 carrying capacity of a system, ie when bivalve consumption was in balance with a combination of system produced and imported primary producers. The major conclusion from that study was that in shallow, closed systems in which bivalve clearance time (CT, the time for suspension-feeders to filter the volume of water in the system) is less than the residence time for the volume of water in the system (RT), the suspension-feeders had the potential of regulating the biomass of primary producers. Dame and Prins (1998) added a measure of the turnover of primary producers (ie. primary production time, PPT, an ecosystem scale measure of the annual rate at which phytoplankton biomass is turned over which they found to be similar to cell doubling time) to the CR model, thereby extending the model to incorporate the rate of growth of the primary producers (hereafter referred to as the CRP model). The 2-D space on this plot is the CR plot and as shown on Figure 6, the drop lines on the parameter points can be used to evaluate these systems using their earlier model. By adding PPT, the relationship now has the advantage of incorporating factors that control phytoplankton growth rate such as nutrient and light limitation and the effect of “new” or imported phytoplankton. One difficulty with a 3-D plot is to know which relationship takes precedence. For example, although the Western Wadden Sea is in the CT/RT space for bivalves exceeding the carrying capacity of a system, the PPT is less than CT and RT and thus may represent a phytoplankton system capable of compensating for some of the consumption losses. To help our interpretations, we can look at this plot as a series of 2-D plots. For example if transport and mixing are not important then the bivalves should be at their maximum biomass if CT=PPT and below their system biomass if CT>PPT. If CT
304 Cloern (1982) which has been suggested to be too high (ie CT was too low). The average CT shown here, based on data from a much more spatially and temporally intensive study than that used in the Cloern (1982) estimate, does not differ substantially from the earlier estimates; the annual average used here is 1.6 days, and a growth season estimate of 1.1 (not plotted) is comparable to the Cloern estimate of 0.7 days (Table 1). Using the new data, CT is still much smaller than RT, except for the subset of data representing the bivalves living in the shallow water during the spring bloom. CT is equal to, or half of PPT, which supports the conclusion that SB bivalves need imported food to maintain their biomass Since the sources of carbon into SB are limited (60% of the labile carbon is estimated to come from local phytoplankton and 28% from local benthic microalgae) and the flux of material from outside the system is only 11% of the total 120 Gg C yr-1 (Jassby et al. 1993), we might conclude that based on the CRP estimation, this is a system out of balance, and that the bivalves are beyond the carrying capacity of the system. However, our observations show that the benthic-pelagic system appears to be closely in balance except during spring bloom periods, when, as shown in Figure 5b, the bivalve biomass in the shallow water is extremely low. The most compelling available data that can support this balance are (1) the population biomass of suspension-feeding bivalves, as reflected in the grazing rates shown in Figure 5, continually increase from the time of recruitment until the period of increased predation in late fall, and (2) the growth production of the bivalve suspension-feeders (10-15 gC m-2 yr-1, Thompson 1999 and unpublished data for 1991-1993) appears to be in balance, assuming a 10% trophic assimilation efficiency (Dame 1996), with the net primary production of in situ phytoplankton (160 gC m-2 yr-1). Thus the CR and CRP models do not appear to answer our question accurately for SFB although we found that using seasonal and location specific values did improve its prediction capability. The major practical limitation of both of these models is the required use of residence time (RT). RT is the time it takes a water parcel to leave a domain for the first time and is best estimated with a numerical model. RT varies with location, bathymetric heterogeneity, the initial phase of the tide from which the model was started, and does not take into account the oscillation of the water parcel in and out of a domain due to tidal currents. For these reasons, Monsen et al. (2002) recommend that the total amount of time that a water parcel stays in a domain (the exposure time) may be a more appropriate time scale for geochemical and biologic processes such as phytoplankton growth dynamics in bathymetrically complex, tidal systems. Exposure time is, due to its aforementioned characteristics, longer than RT, but like RT can only be accurately estimated with a numerical model. Both of these time scales are thus not easily or immediately available for many systems. The lack of a 3-D model for NB and for much of SB has resulted in a range of estimated RT values (Table 1) which
305 may account for some of the difficulties in using this model in SFB. RT may also not be the appropriate parameter to determine when benthic grazers control the biomass of phytoplankton in shallow, turbid systems such as SFB. The new parameter to be explored here can be easily estimated and can be used with less spatial ambiguity than RT for a system and thus satisfies two of our criteria. Z/TR (mean water depth at mid tide/tidal range) is an analog for the vertical mixing rate and tidal exchange and can be estimated for a system as a whole or for specific locations without the use of a numerical model. Like previous relationships we assume that the water column is not stratified but by using a relationship that describes tidal mixing, we do not assume that the systems are equally well mixed vertically which is realistic and important in turbid systems. Systems with low Z/TR values should have high vertical mixing rates, high horizontal transport rates, and potentially high horizontal exchange rates with outside systems dependent on the morphology of the system and its inlet; for example systems with broad intertidal flats, adjacent to deep regions that cover large areas may exchange rapidly with the deep region, but exchange with the ocean will depend on the tidal prism and the dimensions of the inlet. Increased vertical mixing should increase the food flux to the benthic feeders, increase the turbidity of the water column depending on the sediment composition, and increase the rate at which phytoplankton cells are advected through the euphotic zone. Thus, when considered with constant CT and PPT, a low Z/TR should represent a system where benthic grazers are likely to dominate, whereas a high Z/TR value is likely to indicate a more pelagic system. A plot of CT, PPT and Z/TR in 3-D space shows that the position of the pelagic systems is consistent with our expectations (Figure 7). The systems with the highest Z/TR and thus lowest vertical mixing are the three systems known to have a dominant pelagic foodweb (Delaware, Narragansett, and Chesapeake Bays). In addition the three points representing the spatially average conditions in NB (NFP) prior to the invasion of P. amurenis, are shown accurately as a pelagic system. The NB shallow data (SNFP) prior to the invasion of P. amurensis shows Z/TR to be near unity, which when combined with the high values of CT and PPT, is a good example of what an extremely turbid, shallow water, pelagic system looks like in this parameter space.
306
Table 1. A comparison of parameters for the systems discussed. Letters in parentheses are abbreviations used in figures. Bivalve Primary CT/ Data Spatial Water Depth System Period
Context
Mass Residence Time RT (d) 1 65.8 7.1 26 12 12
/Tidal Range
Clearance Time CT (d)
Production Time PPT (d)
PPT
1.5 1.4 1.7 6.9 2.1
0.7 490 2.7 25 12.4 >3100
0.8 16.9 10 1.7 .6 20
0.9 29 .3 14.9 19.7 155
12
2.1
11
20
0.6
14
2.4
1.6
2.5
0.6
12
1.2
1800
10
180
North Inlet (NI) Carlingford Lough (CF) Marennes-Oléron (M) Narragansett Bay (NB) Ria de Arosa (RA) North SFB Pre-P. amurensis (NFP) North SFB Post-P. amurensis (NF) South SFB (SF)
annual annual annual annual annual annual
North SFB Pre-P. amurensis (SNFP) North SFB Post-P. amurensis (SNF) South SFB (SSF) Sylt (S) North SFB Pre-P. amurensis (NFPG) North SFB Post-P. amurensis (NFG) South SFB (SFG)
annual
spatial average spatial average spatial average shallow
annual
shallow
12
1.2
3.8
10
0.4
annual growth bloom
shallow
14 0.5 16
1.3 0.7 2.1
1.7 4 >3100
1.7 1.0 20
1 4 155
16
2.1
1.1
20
0.1
7
2.4
2.2
5.0
0.4
2.6 1.2 6.7 10 1.2
3.7 5.8 1278 325 1800
3.1 1.8 7 2.5 10
1.2 3.2 183 130 180
Oosterschelde (O) Western Wadden Sea (W) Delaware Bay (DB) Chesapeake Bay (CB) North SFB Pre-P. amurensis (SNFPG) North SFB Post-P. amurensis (SNFG) ( South SFB (SSFG)
annual annual
bloom bloom
spatial average spatial average spatial average
growth growth growth growth bloom
shallow
40 10 97 22 16
bloom
shallow
16
1.2
7.8
10
0.8
bloom
shallow
7
1.3
22
2.5
8.8
307
Figure 7. Clearance time (CT days), primary production turnover time (PPT days), and tidal mixing ratio (depth/tidal range), for all systems shown in Dame and Prins (1998) and SFB. NB points are shown as solid circles
A plot of a subset of this same data (Figure 8) allows us to bisect the 2-D CT-PPT space to show the relative values of bivalve suspension-feeding rate and primary production rate for a variety of systems. The differences in this plot compared to the previous CRP plot reflect the different uses of the plots. Most of the points in this plot are near the line, where we expect grazing rate to balance primary production rate, or to the left of the line, where the bivalve clearance rate is predicted to be insufficient to control a phytoplankton bloom. Similar to the CRP plot, points to the right of the line represent systems where the bivalves are controlling phytoplankton accumulation (blooms) in the system. The third parameter Z/TR, a mixing parameter, is indicative of large vertical and horizontal mixing when the ratio is §1. Thus any point to the right of the line with a small mixing parameter is likely to be accurately portrayed as a system where benthic consumers control phytoplankton biomass. These types of systems are represented here by the NB system (SNFG and SNF) which is presently controlled by P. amurensis. The other system represented in this space, the Marennes-Oléron system is unique within those systems represented here as the benthic bivalves control the phytoplankton through grazing but are dependent on the resuspended microphytobenthos (Dame and Prins 1998) to maintain their biomass. Thus the mixing parameter in this system (Z/TR=1.7) represents mixing at a rate sufficient to deliver pelagic food to the bivalves and to resuspend the microalgae. Phytoplankton in systems within this CT/PPT sector, but with a mixing parameter near 2 or greater appear to be sufficiently separated from the
308 bottom that the phytoplankton can still accumulate; this is true of the SB system as shown in Figure 5b and represented in Figure 8 as SFG and SF.
Figure 8. Clearance time (CT days), primary production turnover time (PPT days), and tidal mixing ratio (depth/tidal range), for a subset of systems (CT<28days) shown in Dame and Prins (1998) and SFB.
Systems in which CT and PPT are in balance may have phytoplankton blooms or not, depending on the mixing parameter. For example, the bivalves in the North Inlet (NI) and in the shallow water in SB (SSF) are known to reduce if not control the phytoplankton in their systems and their small mixing parameter as shown in Table 1 and Figure 8 tips their balance toward benthic controlled systems. The Sylt, is the only system with a mixing parameter <1, which is indicative of a tidally flushed system. It’s presence in the sector where phytoplankton blooms are possible is consistent with the observation of several researchers that bivalves in the Sylt do not totally consume the phytoplankton brought in by tide (summarized by Dame 1996). The systems of most interest to evaluate with the new model are those that do not meet the initial model criteria (turbid, nutrient rich) or those in which observations do not match predictions. The Oosterschelde is a system that does not meet our initial criteria since it is oligotrophic, responsive to nutrient regeneration by the bivalves, and is not turbid. Herman and Scholten (1990) report that phytoplankton in this system is controlled by bivalve suspension-feeders. The prediction for the Oosterschelde (O) based on Figure 8 does not meet these observations. The point is near the balance point and
309 the mixing parameter (>2) should “tip” the prediction toward a system with limited phytoplankton blooms which is apparently not the case today. Based on the limited data presented here, the new model appears to predict phytoplankton bloom success for systems that meet the criteria of being nutrient rich and light limited. We found that (1) phytoplankton blooms (ie seasonal net accumulation of phytoplankton) in systems that meet these criteria, CTPPT, and a mixing parameter between 1 and 2 are likely to be controlled by the benthic suspension-feeders; (2) systems with PPT less than CT and a mixing parameter between 1 and 2 may experience seasonal or location specific phytoplankton blooms (eg Western Wadden Sea); and (3) systems with very large mixing parameters are pelagic based systems independent of CT and PPT in the systems examined here. It remains to be seen if this model is an improvement of the CRP model for the questions posed by Dame and Prins (1998) or for the questions posed here. The new model appears to represent the shallow turbid systems tested here more accurately than the CRP model and has the advantage of not being dependent on RT. The major difficulty with the model is the lack of an intuitive feel for the Z/TR parameter. In the data tested here the threshold for the designation of a system having increased potential for phytoplankton control due to horizontal and vertical mixing was one with a mixing ratio of less than 2 and greater than 1.
SAN FRANCISCO BAY: AN ESTUARY WITH MULTIPLE AND CHANGING ALTERNATE STATES? Dame (1996) questioned if a phase shift between a system dominated by pelagic suspension-feeders to one dominated by benthic suspension-feeders could occur in systems not artificially altered, such as those created by culturing oysters in the Marennes-Oléron (Heral et al. 1988 in Dame 1996). He suggested that such a shift might add stability to ecosystems “over longer time periods and with a greater variety of environmental cycles” due to the shorter food chain, greater stored biomass, longer life spans, and more energy efficient food delivery mechanisms of benthic bivalve suspension-feeders relative to pelagic suspension-feeders. The possibility of an ecosystem phase shift in NB and SB is discussed below, followed by comments, based on this study, on the advantages and disadvantages of a system moving from a pelagic dominated system to a benthic dominated system. Deep sediment cores from SFB are depauperate of suspension-feeding bivalves in the euryhaline portions of this system (NB). Macoma balthica, a non-obligate suspension-feeder, is the only infaunal bivalve present in these cores in the non-marine portions of the system. In addition to M. balthica, the
310 native oyster, Ostreola conchalphila, was present throughout SB and in the deep portions of San Pablo Bay where the more saline water moved up the bay (Bonnot 1935). It is likely that the NB has been a pelagic system for many hundreds if not thousands of years. The classification of SB is less clear. Although O. conchalphila is likely to have dominated the suspension-feeding in this system it is unknown how important pelagic suspension-feeders were in this system.
North Bay – A Shift to the Benthos with Some Ramifications The NB system has shifted from a system dominated (by biomass and grazing pressure) by pelagic suspension-feeders to one that is now dominated by benthic suspension-feeders. This shift displays at least two of the defining characteristics of a shift in ecosystem state as described by Scheffer et al. (2001): (1) The contrast between states was due to a shift in dominance among organisms with different life forms, ie a change from pelagic to benthic suspension-feeders that have different population and individual organism characteristics. (2) The state shift was triggered by a stochastic event, ie the random introduction of a new species. (3) Feedbacks that stabilize the different states involve biological, chemical and physical mechanisms. Although less well defined than the earlier characteristics, three feedbacks favor the benthic system. First, the elimination of the annual phytoplankton bloom has limited the success of some pelagic suspension-feeders thereby reinforcing the dominance of the benthic suspension-feeders. Second, the highly variable salinity in this system limits the success of other, nonsuspension-feeding benthic organisms with the potential to compete with P. amurensis due to its exceptional osmoregulatory capabilities. Third, the efficient filtration and assimilation of bacteria and phytoplankton by P. amurensis, in addition to its larger body size and longer life span relative to the pelagic suspension-feeders, allows it to survive periods when phytoplankton biomass is low. Due to the high nutrient concentrations and slow growth rate of primary producers in this system (Cloern et al. 1985) nutrients are not limiting and the nutrient regeneration feedback between benthic suspension-feeders and primary producers, seen in many other systems, is not relevant in this system. The system response to this shift has not been as dramatic as we might expect, given the dependence of this system on phytoplankton as a carbon source (Jassby et al. 1993). Trophic responses to the decline in phytoplankton in NB have included population declines in two copepod species (Eurytemora ( affinas, Acartia spp.), one rotifer (Synchaeta bicornis) a trophically important native mysid shrimp ((Neomysis mercedis, Orsi and Mecum 1996), and possibly one species of fish (Kimmerer 2002). The lack of a larger trophic
311 response may indicate, as suggested by Kimmerer (2002), that NB is an example of a complex aquatic food web that was and is weakly coupled and thus the food web is stabilized due to the weak trophic interactions. It is also possible that the NB ecosystem today can be described as in an unstable equilibrium (as defined by Scheffer 2001) and that this shift in state may be reversed or solidified by future perturbations to the system. If NB is an alternate state system, capable of shifting between benthic and pelagic dominance following extreme stochastic events, then its ability to shift back to a pelagic system is dependent on its resilience, ie its ability to absorb or resist perturbations in its present state. Most estuaries are resilient (Choi and Patten 2001; note they redefine classical resilience as resistance) and thus prone to change in state given extreme perturbations. The extreme perturbations likely to occur in this system, extreme climate events, introduction of non-indigenous species, and catastrophic infrastructure damage to water retention and transport facilities due to seismic activity, all have the potential of changing the state of the system. Determining the potential of returning to the previous pelagic dominance in this system is not possible at present due to our lack of understanding of several critical factors. First, the strength of the feedbacks in the pelagic system and the types of changes occurring in the pelagic system need to be understood. For example, the reason for the catastrophic decline in N. mercedis but not in other pelagic suspension-feeders is unknown. Second, we need to know if this shift in state has been aided by subtle anthropogenic changes in the system that, it combination with this new ecosystem state, make a reversal unlikely. Scheffer et al. (2001) comment that all models of systems with alternate stable states show that gradual changes to the landscape, such as increased nutrient loading in aquatic systems, have little effect on the state of the systems but do change the stability domain. The stability domain, or the resilience of the current ecosystem state, responds to these gradual changes thereby altering the possible response to a disturbance. The anthropogenic changes to this system that were likely to have affected the resilience of SFB historically and continue to affect it today include the following: (1) changes to the hydrograph with the continual removal and entrainment of freshwater thereby reducing the average intra-annual salinity variability; (2) altering and removing freshwater and saltwater marshes thereby removing the “buffer” zone between watershed processes and the estuary; (3) input of nutrients at sufficient levels to eliminate nutrient limitation on phytoplankton growth; (4) input of high sediment loads associated with hydraulic mining that increased the turbidity of the system and left a legacy of fine sediments in the rivers and bay; and (5) input of trace elements and pesticides at sufficient levels to affect viability of some organisms. Of these stressors, we have recently seen that the change in freshwater flow has left us with a system less capable of responding to large climatic events (droughts and floods). As described earlier, the
312 phytoplankton bloom in NB was dependent on residence time and thus on freshwater flowrate with optimum flow being something between the extremes. We have seen the phytoplankton bloom eliminated during high freshwater flow years and during the second year of a 2 year drought, when salinities became sufficiently high for the freshwater intolerant bivalve Mya arenaria, to colonize NB and shift the system to a benthic suspension-feeder system for one year (Nichols 1985). I suggest that the system has been, since our studies began in the 1960’s and prior to the arrival of P. amurensis, moving between three states: one where low hydrologic residence time doesn’t allow a bloom to develop, a second state where extended drought conditions allow freshwater intolerant benthic suspension-feeders to dominate, and a third where hydrologic residence time and salinity intrusion are in “balance” such that phytoplankton blooms can develop and freshwater intolerant benthic grazers can not colonize. The invasion of P. amurensis altered this system because, unlike M. arenaria, it did not leave with the return of the freshwater at the end of the drought.
South Bay – An Intensification of Benthic Dominance?
The SB system has been dominated by benthic suspension-feeders since we began studying the system in the 1960’s. Prior to European settlement in the area in 1849, the native oyster (Ostreola conchalphila) was the dominant bivalve suspension-feeder in SB as noted by their presence in shell middens that date back 3-4000 years (Bonnot 1935). Large scale culture of the Eastern Oyster (Crassostrea virginica) began in SFB in 1879. Because C. virginica is unable to reproduce in this system, 1-2 year old oysters were delivered to SFB by railcar, placed in “grow-out” beds that covered about 25% of the area in the southern extreme of SB (south of the bridge shown in South Bay in Figure 1) for 12-18 months and fattened in much smaller beds on the western shore north of the bridge. An estimate of the grazing rate of the oysters in the grow out beds (based on data in Skinner (1962) and Barrett (1963)) shows that although the filtration pressure was no doubt higher during this period (1875-1905) than in previous years, the system was unlikely to be overstocked (Figure 8). The native oyster declined in abundance during this period, possibly due to overfishing, competition for food with the more filtration efficient Crassostrea species (ostia are larger in O. conchalphila than in C. virginica and d C. gigas), or it’s inability to adapt to high turbidity generated by hydraulic mining in the watershed. Since 1875, there has been a series of suspension-feeding bivalves introduced with the oysters, which make up the benthic bivalve community in SB today. So one can argue that SB may have always been a benthic dominated system that has only added
313 benthic species as new non-indigenous species arrived. However, it is doubtful that O. conchalphila, known to be intolerant of a variety of environmental conditions common to SB including low salinity, high turbidity and high temperature was as uniformly distributed in time and space as are the combination of species present today. So benthic suspension-feeders are likely to be more dominant in the system today than in the past but a shift in ecosystem state is not apparent. Dame (1996) questioned what the disadvantages and advantages of a shift from a pelagic to a benthic system might be in the natural world. An unexpected ecosystem effect of the P. amurensis invasion in NB and SPB is a change in the trophic transfer of selenium, a reproductive and developmental toxin to vertebrate animals. Selenium occurs naturally in the watershed in this system but is also concentrated and released in agricultural drain water. The combination of the dramatic increase in benthic biomass (Nichols et al. 1990) of a desirable and easily consumed organism and P. amurensis’’ ability to accumulate selenium (Schlekat et al. 2000, 2002) has resulted in a new, contaminated food source for benthivores. Some of P. amurensis’’ major predators, such as the Green Sturgeon and Surf Scoter have shown increased Se concentrations coincident with their consumption of P. amurensis (Linville et al. 2002). Although pelagic suspension-feeders are also exposed to selenium and other contaminants, unless the contaminants are magnified by the food web, the short life span, and small accumulation potential of pelagic suspension-feeders means that they may have less impact on higher order animals.
ACKNOWLEDGEMENTS This research was funded by the Priority Ecosystem Science and Toxics Program of the U.S. Geological Survey. Additional student funding was received through an EPA STAR grant. None of this work could have been done without the ongoing assistance of Francis Parchaso.
REFERENCES Alpine AE Cloern JE 1992 Trophic interactions and direct physical effects control phytoplankton biomass and production in an estuary. Limnol Oceanogrr 37: 946-955 Barrett EM 1963 The California Oyster Industry. Calif. Dept. of Fish and Game Fish Bulletin 123 Bonnott P 1935 The California oyster industry. Calif. Fish and Game 21(1): 65-80 Carlton JT 1985 Transoceanic and interoceanic dispersal of coastal marine organisms: the biology of ballast water. Oceanogr Mar Biol Ann Revv 232: 313-371
314 Carlton JM Thompson JK Schemel LE Nichols FH 1990 Remarkable invasion of San Francisco Bay (California, USA) by Asian clam Potamocorbula amurensis. Introduction and dispersal. Mar Ecol Prog Serr 66: 81-84. Chauvaud L 1998 La coquille Saint Jacques en rade de Brest: un modele biologique d’etude des reponses de la faune benthique aux fluctuations de l’environnement. These de doctorat, universite de Bretagne Occidentale, Brest Chauvaud L Jean F Ragueneau O Thouzeau G 2000 Long-term variation of the Bay of Brest ecosystem: Benthic-pelagic coupling revisited. Mar Ecol Prog Serr 200: 35-48. Choi JS Patten BC 2001 Sustainable development: lessons from the paradox of enrichment. Ecosystem Health 7(3): 163-177 Cloern JE 1982 Does the benthos control phytoplankton biomass in South San Francisco Bay? Mar Ecol Prog Serr 9: 191-202. Cloern JE 1987 Turbidity as a control on phytoplankton biomass and productivity in estuaries. Cont Shelf Res 7: 1367-1381 Cloern JE 1996 Phytoplankton bloom dynamics in coastal ecosystems: A review with some general lessons from sustained investigation of San Francisco Bay, California. Rev Geophys 34: 127-168 Cloern JE 1999 The relative importance of light and nutrient limitation of phytoplankton growth: a simple index of coastal ecosystem sensitivity to nutrient enrichment. Aquat Ecoll 33: 3-16 Cloern JE 2001 Our evolving conceptual model of coastal the eutrophication problem. Mar Ecol Prog Serr 210: 223-253 Cloern JE Alpine AE Cole BE Wong RLJ Arthur JF Ball MD 1983 River discharge controls phytoplankton dynamics in the northern San Francisco Bay estuary. Est Coast Shelf Sci 16: 415-429 Cloern JE Cole BE Wong RLJ Alpine AE 1985 Temporal dynamics of estuarine phytoplankton: a case study of San Francisco Bay. Hydrobioll 129: 153-176 Cohen AN Carlton JT 1995 Biological study: Non-indigenous aquatic species in a United States estuary: a case study of the biological invasions of the San Francisco Bay and Delta. US Fish Wild and Nat. Sea Grant Prog Reportt PB96-166525, Springfield, Virginia, USA, 273pp. Cohen AN Carlton JT 1998 Accelerating invasion rate in a highly invaded estuary. Science 279: 255--558 Cohen RRH Dressler PB Phillips EJ Cory RL 1984 The effect of the Asiatic clam, Corbicula fluminea on phytoplankton of the Potomac River. Limnol Oceanogrr 29: 170-180 Cole BE Thompson JK Cloern JE 1992 Measurement of filtration rates by infaunal bivalves in a recirculating flume. Mar Bioll 113: 219-225. Dame RF 1996 Ecology of Marine Bivalves: An Ecosystems Approach. CRC Press, Boca Raton, FL, 254 pp Dame RF Prins TC 1998 Bivalve carrying capacity in coastal ecosystems. Aquat Ecoll 31: 409-421 Gross E 1997 Numerical modeling of hydrodynamics and scalar transport in an estuary. PhD thesis, Stanford University Heip DHR Goosen NK Herman PMJ Kromkamp J Middelburg JJ Soetaert K 1995 Production and consumption of biological particles in temperate tidal estuaries. Ocean Mar Biol Ann Revv 33: 1-149. Herman PMJ Scholten H 1990 Can suspension-feeders stabilize estuarine ecosystems. In: Trophic Relationships in the Marine Environment, M Barnes RN Gibson (Eds), Aberdeen Univ. Press, pp 104-116 Hymanson Z 1991 Results of a Spatially Intensive Survey for Potamocorbula amurensis in the Upper San Francisco Bay Estuary. Interagency Technical Report Number 30, Calif Dept of Water Resources, Sacramento CA
315 Jassby AD Cloern JE Powell TM 1993 Organic carbon sources and sinks in San Francisco Bay: variability induced by river flow. Mar Ecol Prog Serr 95: 39-54 Kimmerer WJ 2002 Effects of freshwater flow on abundance of estuarine organisms: physical effects or trophic linkages. Mar Ecol Prog Serr 243: 39-55 Leppakoski E Gollasch S Olenin S 2002 Invasive Aquatic Species of Europe: Distribution, Impacts and Management. Kluwer Academic Pub. Dordrecht, The Netherlands 583 pp Linville RG Luoma SN Cutter L Cutter GA 2002 Increased selenium threat as a result of invasion of the exotic bivalve Potamocorbula amurensis into the San Francisco BayDelta. Aqua Toxic 57: 51-64 Lucas LV Koseff JR Cloern JE Monismith S Thompson J 1999a Processes governing phytoplankton blooms in estuaries. I. The local production-loss balance. Mar Ecol Prog Serr 187: 1-15. Lucas LV Koseff JR Cloern JE Monismith S Thompson J 1999b Processes governing phytoplankton blooms in estuaries. . II. The role of horizontal transport. Mar Ecol Prog Serr 187: 17-30 May CL Koseff JR Lucas LV Cloern JE Schoellhamer DH 2003 Effects of spatial and temporal variability of turbidity on phytoplankton blooms. Mar Ecol Prog Serr 254: 111-128. Monbet Y 1992 Control of phytoplankton biomass in estuaries: a comparative analysis of microtidal and macrotidal estuaries. Estuaries 15: 563-571 Monismith SG Kimmerer W Burau JR Stacey MT 2002 Structure and flow-induced variability of the subtidal salinity field in northern San Francisco Bay. J Phys Ocean 32: 3003-3019 Monsen NE Cloern JE Lucas LV 2002 A comment on the use of flushing time, residence time, and age as transport time scales. Limnol Oceanogrr 47: 1545-1553 Nichols FH 1985 Increased benthic grazing: An alternative explanation for low phytoplankton biomass in northern San Francisco Bay during the 1976-1977 drought. Est Coast Shelf Sci 21: 379-388 Nichols FH Thompson JK Schemel LE 1990 Remarkable invasion of San Francisco Bay (California, USA) by the Asian clam Potamocorbula amurensis. 2. Displacement of a former community. Mar Ecol Prog Serr 66: 95-101 Nicolini MH Penry DL 2000 Spawning, fertilization, and larval development of Potamocorbula amurensis (Mollusca: Bivalvia) from San Francisco Bay, California. Pac Sci 54: 377-388 Ojaveer H Leppakoski E Olenin S Ricciardi A 2002 Ecological impact of Ponto-Caspian invaders in the Baltic Sea, European inland waters and the Great Lakes: An interecosystem comparison. In: Invasive Aquatic Species of Europe: Distribution, Impacts and Management. E Leppakoski S Gollasch S Olenin (Eds), Kluwer Academic Pub. Dordrecht, pp 412-425 O’Riordan CA Monismith SG Koseff JR 1995 The effect of bivalve excurrent jet dynamics on mass transfer in a benthic boundary layer Limnol Oceanogrr 40:330-344 Orsi JJ Mecum WL 1996 Food limitation as the probable cause of a long-term decline in the abundance of Neomysis mercedis the opossum shrimp in the Sacramento-San Joaquin estuary. In: San Francisco Bay: The Ecosystem. JT Hollibaugh (Ed), AAAS, San Francisco, California USA pp 375-401 Peterson DH Smith RE Hager SW Harmon DD Herndon RE Schemel LE 1985 Interannual variability in dissolved organic nutrients in North San Francisco Bay estuary. Hydrobiol 129: 37-58 Poulton VK Lovvorn JR Takekawa JY 2002 Clam density and scaup feeding behavior in San Pablo Bay, California. Condorr 104: 518-527
316 Ruhl CA Schoellhamer DH Stumpf RP Lindsay CL 2001 Combined use of remote sensing and continuous monitoring to analyze the variability of suspended-sediment concentrations in San Francisco Bay, California. Est Coast Shelf Sci 53: 801-812 Ruiz GM Fofonoff PW Carlton JT Wonham MJ Hines AM 2000 Invasions of coastal marine communities in North America: apparent patterns, processes, and biases. Ann Rev Ecol System 31: 481-531 Ruiz GM Hewitt CL 2002 Toward understanding patterns of coastal marine invasions: a prospectus. In: Invasive Aquatic Species of Europe: Distribution, Impacts and Management. E Leppakoski S Gollasch S Olenin (Eds) Kluwer Academic Pub Dordrecht, The Netherlands, pp 529-547 Scheffer M 2001 Alternative attractors of shallow lakes. The Scientific World 1: 254-263 Scheffer M Carpenter S Foley JA Folk C Walker B 2001 Catastrophic shifts in ecosystems. Nature 413: 591-596 Schiganova TA 1998 Invasion of the Black Sea by the ctenophore Mnemiopsis leidyi and recent changes in pelagic community structure. Fish Ocean 7: 305-310 Schlekat CE Dowdle PR Lee BG Luoma SN Oremland RS 2000 Bioavailability of particleassociated SE to the bivalve Potamocorbula amurensis. Environ Sci Technoll 34: 45044510 Schlekat DW Lee BG Luoma SN 2002 Assimilation of selenium from phytoplankton by three benthic invertebrates: effect of phytoplankton species. Mar Ecol Prog Serr 237: 79-85 Schoellhamer DH 1996 Factors affecting suspended-solids concentrations in South San Francisco Bay, California. J Geophys Res 101: 12087-12095 Schoellhamer DH 2002 Variability of suspended-sediment concentration at tidal to annual time scales in San Francisco Bay, USA. Cont Shelf Res 22: 1857-1866 Skinner JE 1962 Fish and wildlife resources of the San Francisco Bay area. Calif. Dept. of Fish and Game, Water Proj Branch Reportt 1, 226pp. Smaal AC Prins TC 1993 The uptake of organic matter and the release of inorganic nutrients by bivalve suspension feeder beds. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed), Springer-Verlag, Heidelberg, pp 273-298 Thompson JT 1999 The effect of infaunal bivalve grazing on phytoplankton bloom development in South San Francisco Bay. PhD Thesis, Stanford University Walters RA Cheng RT Conomos TJ 1985 Time scales of circulation and mixing processes of San Francisco Bay waters. Hydrobiol 129: 13-36
DEVELOPMENT OF HUMAN IMPACT ON SUSPENSION-FEEDING BIVALVES IN COASTAL SOFT-BOTTOM ECOSYSTEMS
Wim J. Wolff Marine Biology, University of Groningen, PO Box 14, 9750 AA Haren, The Netherlands Abstract: Suspension-feeding bivalves often may occur in large concentrations ('beds') on tidal flats. This makes them attractive for human consumers and the archaeological record shows collection of bivalves by coastal populations already tens of thousands of years ago. In modern time human interference with coastal stocks of bivalves intensified. This paper describes the successive steps in this development: 1. Local exploitation and local consumption, leading to reduction of average age, average size and small shifts in species composition. 2. Local exploitation coupled to remote markets. This leads to the same changes as at 1, but also may lead to extirpation of local populations. 3. Relaying of imported bivalves to restock overexploited beds. This may result in destruction of genetic adaptations, and it has been demonstrated that it results in the importation of parasites and diseases. 4. Cultivation of bivalves with either spat from natural sources or from hatcheries. Bivalve culture usually results in increased harvests compared to open fisheries; it may lead to overstocking with effects on the remainder of the ecosystem. 5. Introduction of new, exotic species, either to be cultivated or as an unplanned introduction of 'weed' species. Keywords: suspension-feeders, oysters, mussels, bivalves, shellfish cultivation, human impact
INTRODUCTION Among the many different forms of suspension-feeders bivalve molluscs often dominate tidal benthic environments. Especially on tidal flats benthic bivalves may be abundant. This abundance attracts many different predators, including fish, birds, and mammals. Also humans discovered benthic bivalves as a rich food source very early in history. Human-made shell middens in South Africa are believed to be at least 100,000 years old. This
317 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 317–330. ©2005 Springer. Printed in the Netherlands.
318 contribution discusses the role humans have played and continue to play in shaping populations of benthic bivalves and their ecosystems. It starts with humans simply collecting locally available bivalves for food. It finishes with culture systems based on imported non-indigenous species. It is shown that all these forms of human impact influence the bivalve community as well as the ecosystem they are part of.
CHARACTERISTICS OF AN UNEXPLOITED POPULATION: THE BLOODY COCKLE ANADARA SENILIS S AT THE BANC D’ARGUIN, MAURITANIA The Banc d’Arguin is an area of shallow water and tidal flats at the Atlantic coast of Mauritania, West Africa (about 20o N). Because of the absence of rivers and the rare occurrence of rain, salinities at the Banc d’Arguin are 38 psu and higher. Due to its tidal range of about 2 m at spring tides the Banc d’Arguin comprises vast areas of intertidal flats. The greater part, approximately 412 km2, of these usually very muddy flats is covered with seagrass (Zostera noltii) meadows (Wolff and Smit 1990). The macrobenthic fauna of these flats is dominated by a large suspension-feeding bivalve Anadara senilis (Wolff et al. 1987). In 1985-86 Anadara was estimated to have an average biomass of 5.5 g ash-free dry mass (AFDM) m-2 which is about 40% of the total benthic biomass at the tidal flats (Wolff et al. 1993). The Banc d’Arguin area borders on the Sahara and, hence, has an extremely low human population density. The few people living there are nomads and fishermen. These fishers target only fish species that are caught from traditional sailing vessels. So, the tidal-flat benthos is virtually undisturbed and unexploited. The Banc d’Arguin is nearly without pollution, also because the nearest river discharges downstream the coastal current 300 km to the South. In this virtually pristine environment in 1985-86, Anadara had a remarkable population structure (Wolff et al. 1987). Individuals 13 to 19 years old dominated the population, with smaller numbers 5 to 12 years and 20 to 30 years of age. Some specimens may even have been older than 30 years. In 1985-86 newly settled spat was extremely rare among the adult population and only found in small numbers on a sandy beach. It was hypothesized that for its reproduction Anadara, which is elsewhere considered a brackish-water species, is dependent on temporary low salinities during infrequent periods of precipitation. However, rainfall is not measured at the Banc d’Arguin, so it was assumed that precipitation at the Banc d’Arguin is correlated with the amount of rain in the Sahel area further to the South. Because precipitation in the Sahel governs the discharge of the Senegal River, this discharge was used
319 as a proxy. Doing so, a significant linear correlation (r = 0.62; p < 0.01) was established between the year-class strength of A. senilis at the Banc d’Arguin and the discharge of the Senegal River at Baker (Wolff et al. 1987). The nearabsence of year-classes younger than 13 years apparently was a consequence of the extreme Sahel drought in the 1970’s and 1980’s.
EXPLOITED POPULATIONS OF ANADARA SENILIS In 2001 and 2002 we visited the Banc d’Arguin again. Remarkable enough, we still found small numbers of Anadara over 30 years old, which apparently were representatives of the population peak found in 1985-86. The great majority of the population in 2001 and 2002, however, consisted of the year-classes 0 to 5 with a few individuals up to 15 years old. The 0-year-class was particularly abundant with 80-90% of the total numbers of individuals (unpublished observations Hilde Boer and Wim J. Wolff; de Vries 2001). We assume that the return of normal rain conditions in the Sahel also has benefited Anadara at the Banc d’Arguin. This unexploited, pristine population offers the possibility of determining natural mortality rates. Only a few predators are able to extract A. senilis from its very thick and strong shell. The oystercatcher ((Haematopus ostralegus) is able to cut the retractor muscles to open the shell (Swennen 1990). Large gastropods belonging to the genus Cymbium also open the shell but the mechanism is unclear (Wolff and Montserrat submitted). Finally the pufferfish Ephippium guttiferum is able to crush the thick shell (Vonk 2001). Although our infrequent observations and the unbalanced population structure of Anadara make estimation of the annual mortality rate difficult, we have nevertheless arrived at estimates of respectively 6.5%, 8% and about 10% per year (de Vries 2001; Swennen 1990; Wolff et al. 1987). In former millennia, much larger numbers of people than today (Vernet 1993) inhabited the Sahara and also the coastal area of the Banc d’Arguin. Because vegetation is very sparse in the Banc d’Arguin area archaeological remains and shell middens are easy to find. In 2001, we sampled a number of shell middens. In a typical sample, Anadara predominates; other recognizable species are large gastropods of the genera Semifusus and Cymbium. This is in agreement with the present abundance of these species at the tidal flats. For Anadara, we determined the age structure of the samples. 0- and 1-year-class individuals are virtually lacking, suggesting that we sampled indeed human-collected shell middens. Second year-class specimens were usually most frequent and the older year-classes decreased in abundance in relation with their age (Fig. 1). These age distributions enabled us to calculate the mortality rates of these exploited
320
Frequency
populations; we arrived at annual rates ranging from 27 to 46% (de Vries 2001). Differences between the present unexploited population of Anadara and the exploited populations in the same area from the past may be summarized as follows. The youngest year-classes apparently escaped exploitation, probably they were too small to be profitably collected. Exploitation had a clear effect on the age structure: in the unexploited population individuals could become more than 30 years old, whereas in the exploited population animals reached their age limit at about 10 years. This result is reflected in the average size of the population; the exploited individuals on average are much smaller than the unexploited ones. We cannot determine whether the density or the biomass of the two populations differed. However, it is likely that the exploited population was able to support the exploitation because adult and probably reproducing animals were common. I conclude that the main effects of exploitation of Anadara at the Banc d’Arguin in the past were a reduction of the age of the population and a decrease of the average size.
35 30 25 20 15 10 5 0 1
2
3
4
5
6
7
8
9
10
11
12 Age
Figure 1. Frequency of different year-classes of Anadara senilis in a sample from a prehistoric shell midden at the Banc d'Arguin, Mauritania. From de Vries, 2001.
321 COLLECTION OF BIVALVE SUSPENSION-FEEDERS FROM TIDAL FLATS ELSEWHERE Bivalve suspension-feeders have been or are still being collected along the shores of almost all parts of the world. De Boer (2000) studied this at Inhaca Island, Mozambique. Women and children searched for crabs and shellfish at the intertidal flats of this island. Women were more efficient than children as their catch was heavier, and the mean weight per animal was larger. Their prey choice was in accordance with optimal foraging theory. Species with profitability’s lower than the mean intake rate generally were excluded from the diet. However, the women did not simply use prey weight as their criterion for profitability, but instead preferred prey according to their relative value ranks (e.g., weight, taste). Women maximised the cumulative relative value ranks, instead of total weight (de Boer et al. 2002). Of the at least 117 species present the women collected in total 64 different species per tidal period, their mean catch amounting to 30.5 g AFDM (ash-free dry mass) m-2 per person. Their average harvest was 0.12 g AFDM m-2 yr-1; this is 2% of the average biomass of 6.0 g AFDM m-2 present. Crabs constituted nearly 75% of the prey caught. Bivalves were only 13% of total AFDM (de Boer and Prins 2002). Next de Boer et al. (2000) compared the recent human predation with former predation pressure by investigating recent and abandoned shell middens at Inhaca Island. It was assumed that recent middens reflected a higher predation pressure than the abandoned middens because of population growth. It appeared that older, abandoned middens had higher species diversity than recent middens. The absence of the mussel Perna perna from recent middens was believed to be related to overexploitation. The mean shell size was significantly smaller in recent middens for 5 species and the conspicuous surface-dwelling snail Terebralia palustris showed the largest size reduction. It appears that a seemingly low predation pressure, nevertheless, has a noticeable effect on the composition of benthic communities. Man has exploited intertidal resources in Mozambique for at least 12,000 years. Elsewhere in Africa exploitation has occurred for at least 100,000 years (see de Boer 2000 for references). We assume that de Boer's findings apply also to those other cases of exploitation of tidal flats in Africa.
FIRST STAGE OF HUMAN IMPACT We consider the exploitation of intertidal flats as exemplified by our African cases the first stage of human impact on bivalve suspension-feeders.
322 The impact consists of a slight reduction of community species diversity, a reduction of average shell size, and a reduction of the average age of the bivalve population. No doubt, this is not an African phenomenon; it will apply to similar cases all over the world, including the famous Danish kjøkkenmøddingerr (“kitchen middens”). In many cases it will have occurred for 10,000s of years. Yonge (1960) concludes that in Europe “oysters were consumed on many coasts during the centuries between Roman and modern times but with scanty local populations and lack of transport to the inland towns fishing was not intense and the natural oyster beds could easily afford all that were collected.” I want to stress that an important characteristic of this first stage of impact is that local collection of shellfish was connected to local consumption.
SECOND STAGE OF HUMAN IMPACT Later in history locally collected shellfish were transported to population centres elsewhere. Also technically more advanced fisheries developed. Yonge (1960) records that shells of European flat oysters (Ostrea edulis) originating from the Thames estuary have been found in Roman settlements throughout England and even in Rome. He assumes that these beds continued to be exploited after the Romans left Britain. From the Middle Ages until the 20th century the beds were exploited by a well-regulated fishery supplying annually hundreds of millions of oysters to the London markets. However, towards the 20th century these beds and other ones elsewhere in Britain became overexploited. Many beds disappeared and other ones showed strongly reduced yields. Yonge (1960) considers the tremendous town demand for oysters which was easily met by rail and sea transport as a major cause for the decline. Dijt (1961) and van der Vlis (1975) describe the demise of the hardly regulated oyster fishery in the Dutch Wadden Sea near Texel. European flat oysters were fished from at least 1700. During the 18th century millions of oysters were caught annually and transported by sailing vessels to be sold in the cities of Amsterdam and Hamburg. Overfishing became noticeable in the 19th century and around 1930 the flat oyster was extirpated from the Dutch Wadden Sea. Also elsewhere flat oysters were extirpated. Moebius (1870) still could make his famous description of the oyster bed as a biocenosis, but 60 years later the fishery came to an end in the German Wadden Sea. Again the local catch was sold at markets far removed from the beds, also in this case aided by the presence of railroads (Riesen and Reise 1982; Reise and Schubert 1987; Reise et al. 1989). But overexploitation of oyster beds is not a European phenomenon. Rothschild et al. (1994) describe how the overexploitation of Chesapeake Bay
323 beds of the Eastern oyster Crassostrea virginica in the 19th and 20th century led to a very strong reduction of the stock, also because the mechanical harvest of these reef-forming oysters destroyed the habitat needed for the larvae to recruit. Again we observe that local catch is sold at markets far away thanks to the railroad system. This is the general pattern of the second stage of human impact. The combination of local fisheries and remote markets apparently has no negative feedback stabilising the fishery. Scarcity leads to higher market prices and this provides an incentive for overexploitation and finally extirpation.
THIRD STAGE OF HUMAN IMPACT
One method to counteract the effects of overexploitation is the import of seed oysters from elsewhere to restock local beds. This practice ecologically resembles the import of brood oysters for relaying and fattening. Thus the Romans brought oysters from the Adriatic Sea to the shore of the Tyrrhenian Sea near Naples for growing and fattening (Yonge 1960). In northern Europe as early as 1714 seed oysters were imported from Denmark to restock the beds in the western Dutch Wadden Sea (Hoek 1911; van der Vlis 1975). About 1750 seed oysters from the British Isles were relaid in the Dutch Oosterschelde estuary (Baster 1759). Oyster beds in the East-Frisian part of the German Wadden Sea were reinforced with young oysters from different origins between 1773 and 1806 (Wehrmann et al. 2000). Yonge (1960), Neudecker (1985) and de Jonge (1990) describe how more recently European flat oysters were frequently exchanged between culture areas in Denmark, Germany, The Netherlands, Belgium, Scotland, England, Ireland and France during the 19th and 20th century. For example, Korringa (1951a) records that Ostrea edulis were imported from the West coast of France into the Dutch Oosterschelde estuary in 11 years between 1890 and 1935. Korringa (1957) drew attention to the impact this practice of importing seed oysters might have on genetic adaptation of local stocks. European flat oysters release well developed larvae with a short pelagic phase probably resulting in rather limited dispersal. This may have resulted in the development of stocks adapted to local conditions. Although never actually demonstrated it seems quite possible that large-scale oyster imports have destroyed such adaptations. Stauber (1950) as well as Loosanoff and Nomejko (1951) made similar suggestions for the American eastern oyster Crassostrea virginica. The existence of local physiological genotypes indeed has been demonstrated for this species (Gaffney 1996) although on a larger scale than separate estuaries. On the other hand, Barber et al. (1991) demonstrated that
324 oysters from different origin after several generations of culture in Delaware Bay continued to display different reproductive timing. So, the earlier suggestions about loss of genetic adaptation of local stocks cannot be confirmed. Also diseases and parasites can be imported with imported seed oysters or other shellfish. Korringa (1951b) describes how the parasitic copepod Mytilicola intestinalis was imported with seed mussels from the German Wadden Sea into the Dutch Oosterschelde estuary in 1948. The pathogenic Bonamia ostreae was imported with European flat oysters from California into Spanish waters in 1977 (Cigarria and Elston 1997) and from Brittany, France, into the Netherlands in 1980 (van Banning 1988). The diseases caused by Perkinsus marinus and Haplosporidium nelsoni in American eastern oysters have been spread along the Atlantic coast by oyster transfers (Carlton and Mann 1996). The third stage of human impact acts through the import of bivalves for relaying. It may destroy genetic adaptations to local conditions and it has been demonstrated to result in the importation and subsequent spread of diseases and parasites. 40 harvest in millions
35
Introduction of oyster culture
30 25 20 15 10 5 0 1870
1871
1872
1873
1874
1875 year
Figure 2. Increase of oyster (O. edulis) harvest in millions after the introduction of culture plots closed for fisheries in the Oosterschelde estuary. After Dutch fisheries statistics.
CULTIVATION: THE FOURTH STAGE OF HUMAN IMPACT The fourth stage of human impact on populations of bivalve suspension-feeders consists of cultivation. There exists a gradual transition between relaying of bivalves for growth and fattening and bivalve culture
325 based on the collection of spat and subsequent cultivation. As discussed before relaying was practised already by the Romans. It is possible that the Romans also collected spat by means of submersed faggots (Yonge 1960). The same author describes how in 17th century English cultures the spat were collected from natural beds to be relaid on special fattening beds. In 1858 in France, however, Professor J. Coste 'invented' a newer form of culture of the European flat oyster based on the collection of spat on artificial collectors. The idea was rapidly picked up, so the new way of oyster culture spread to most European countries. Later, it was followed by “cultures” of other bivalve species: other species of oysters, mussels, Manilla clams etc. Although many of these cultures are still dependent on natural spatfall, the new practice proved to be highly successful. In the Dutch Oosterschelde estuary the introduction of culture practices in 1870 led to a 30-fold increase of the oyster harvest within five years (Fig. 2). Cultivation of bivalves includes the risk of overstocking. Already in 1902 Hoek concluded that slow growth and low meat percentages of European flat oysters in the Dutch Oosterschelde estuary, were caused by overstocking of the beds. Overstocking of cultivated bivalves may also affect non-cultivated species in the same estuary (Smaal and Prins 1993). In mesocosm experiments Wolff et al. (in prep.) also found indications for this effect. The ecosystem effects of overstocking, if any, will be reversible, however, because it will always be possible to reduce the stocks of the cultivated species.
INTRODUCTION OF EXOTIC SPECIES: THE FIFTH STAGE OF HUMAN IMPACT Introduction of exotic bivalves for cultivation is a long-standing practice. As early as the 1860’s live C. virginica were shipped from the American Atlantic coast to Arcachon Bay and Normandy, France. Shortly thereafter the same species was exported to England by the millions (Carlton and Mann 1996, Miller 2000). Some of these oysters were directly consumed, but another part was relaid on beds in European waters. With these oysters, at least five American invertebrate species were introduced in Britain, although C. virginica itself failed to establish itself (Wolff and Reise 2002). After the Second World War these imports came to an end. In about the same period the same C. virginica was introduced to the Pacific coast of North America and even to Hawaii. Here C. virginica was able to establish itself; it also brought several other species with it (Carlton and Mann 1996). Another species imported on a large scale is the Pacific oyster Crassostrea gigas. Around 1900 it was imported from Japan into British Columbia, Canada. In the 1960s it was introduced into France and the
326 Netherlands both from British Columbia and from Japan. Again many other exotic species were introduced with these oysters (Wolff and Reise, 2002). These and other imports have resulted in a strongly differing set of mollusc suspension-feeders in many estuaries. For example, historically, three native species of bivalve suspension-feeders occurred in estuaries around the North Sea: O. edulis, M. edulis, and C. edule. Today, however, the same estuaries are inhabited by the native M. edulis and C. edule, as well as the introduced d C. gigas, Mya arenaria, Ensis directus, Mercenaria mercenaria (only in the southern estuaries), and Crepidula fornicata (Reise et al 1999; Wolff, submitted). Remarkably, some introductions have failed. Both the Portuguese oyster Crassostrea 'angulata' (in reality an early introduced stock of the Pacific oyster) and the American Atlantic oyster C. virginica never established in the North Sea estuaries, although millions of individuals have been imported (Reise et al. 1999; Wolff submitted). The case of the Portuguese oyster suggests that physiological differences exist, especially with regard to temperature, between C. ‘angulata’’ reported to originate from Taiwan (Boudry et al. 1998, O’Foighil et al. 1998), and the ‘real’ C. gigas from Japan. How stable is this new suspension-feeder community? One might expect competition between the different species, notably for space and food. Also it seems possible that the different species influence each other by ingestion of each other's larvae. Competition for space will no doubt occur as Pacific oysters occupy a part of the tidal flats with dense beds. Here, Mytilus and Cerastoderma no longer occur. In the Wadden Sea, it has been observed that mussel beds are being taken over by the oysters (Reise 1998). However, this form of competition probably is less important, since the oyster beds occupy only a small part of the estuary. Competition for food and/or ingestion of larvae might be more important. Observations on stable isotope composition demonstrate that the species concerned apparently have a very similar food choice (Riera et al. 2002). Pacific oysters seem to be strong suspension-feeders; this might give them an advantage over other species (Table 1). In any case, monitoring data in the Oosterschelde estuary and the saline Lake Grevelingen suggest that increasing populations of C. gigas correlate with decreasing populations of native suspension-feeders (Smaal, pers. comm., Troost pers. comm.). If this is a causal relationship, it suggests that the introduction of an exotic species may shift the state of the receiving ecosystem considerably. This is considered to be the fifth stage of human impact on bivalve suspension-feeder populations.
327 Table 1. Densities and filtration rates per individual and per m2 for bivalve suspension-feeders in Dutch coastal waters. After Bougrier et al. 1995, Dupuy et al. 2000, Foster-Smith 1975, Gerdes 1983, Møhlenberg and Riisgård 1979, Smaal and Twisk 1997, Walne 1972, Winter 1973 and unpub. data Netherlands Institute for Fisheries Research (courtesy Karin Troost). _
Species
Density
Filtration rate
(n.m-2)
(l.h-1.ind-1)
(l.h-1.m-2)
Cerastoderma edule
2,000
1.0 - 3.4
2,000 -
Mytilus edulis
6,000
0.4 - 9.1
2,000 - 55,000
10,000
3.0 - 25.0
30,000 - 250,000
Crassostrea gigas
7,000
THE COMPARATIVE ROLE OF HUMAN IMPACT We have discussed five stages of human impact on populations of suspension-feeding bivalves: 1. Local exploitation and local consumption, leading to a reduction of average age and average size as well as to small shifts in species composition. These changes are reversible. 2. Local exploitation coupled to remote markets. This leads to the same changes as at 1, but also may lead to extirpation of local populations. The latter phenomenon is in practice nearly irreversible. 3. Relaying of imported bivalves to restock overexploited beds. This may result in destruction of genetic adaptations, and it has been demonstrated that it results in the importation of parasites and diseases. These changes are irreversible. 4. Cultivation of bivalves with either spat from natural sources or from hatcheries. Shellfish culture usually results in greatly increased harvests compared to open fisheries, and it even may lead to overstocking with effects on the remainder of the ecosystem. These effects are reversible, however. 5. Introduction of new, exotic species, either to be cultivated or as an unplanned introduction of 'weed' species. If such new species have an impact on other species and on the receiving ecosystem in general, these effects are normally irreversible. Now what could be the next stage? Two possibilities come to mind. One possibility is the genetic modification of cultivated species, and another one is
328 a more or less industrial cultivation of bivalve suspension-feeders in culture systems isolated from the natural ecosystem.
REFERENCES Barber BJ Ford SE Wargo RN 1991 Genetic variation in the timing of gonadal maturation and spawning of the eastern oyster, Crassostrea virginica (Gmelin). Biol Bulll 181: 216221 Baster J 1759 Natuurkundige Uitspanningen, behelzende eenige waarnemingen over sommige zee-planten en zee-insecten, benevens derzelver zaadhuisjes en eijernesten, eerste stukje. J. Bosch, Haarlem Boudry P Heurtebise S Collet B Cornette F Gerard A 1998 Differentiation between populations of the Portuguese oyster, Crassostrea angulata (Lamarck) and the Pacific oyster, Crassostrea gigas (Thunberg), revealed by mtDNA RFLP analysis. J Exp Mar Biol Ecol 226: 279-291 Bougrier S. Geairon P Deslous-Paoli JM Bacher C Jonquières G 1995 Allometric relationships and effects of temperature on clearance and oxygen consumption rates of Crassostrea gigas (Thunberg). Aquaculture 134: 143-154 Carlton JT Mann R 1996 Transfers and world-wide introductions. In: The Eastern Oyster Crassostrea virginica. VS Kennedy RIE Newell AF Eble (Eds), Maryland Sea Grant Book, College Park, pp 691-706 Cigarría J Elston R 1997 Independent introduction of Bonamia ostreae, a parasite off Ostrea edulis, to Spain. Diseases Aquat Organisms 29: 157-158 de Boer WF 2000 Between the Tides. The Impact of Human Exploitation on an Intertidal Ecosystem, Mozambique. PhD thesis, University of Groningen, 268 p de Boer WF Blijdenstein AF Longamane F 2002 Prey choice and habitat use of people exploiting intertidal resources. Environm Conserv 29: 238-252 de Boer WF Pereira T Guissamulo A 2000 Comparing recent and abandoned shell middens to detect the impact of human exploitation on the intertidal ecosystem. Aquat Ecoll 34: 287-297 de Boer WF Prins HHT 2002 Human exploitation and benthic community structure on a tropical intertidal flat. J Sea Res 48: 225-240 de Jonge JA 1990 Het Zeeuwse goud. Geschiedenis van de Mossel- en Oestercultuur. De Bataafsche Leeuw, Amsterdam. 144 p de Vries, AB 2001 The population structure, growth and biomass of the bloody cockle Anadara senilis on the Banc d’Arguin, Mauritania (In Dutch). MSc thesis, University of Groningen, 17 pp + appendices. Dijt MD 1961 Texelse oesters, aantekeningen over opkomst en verval. Visserijnieuws 14: 99102 Dupuy C Vaquer A Lam-Höai T Rougier C Mazouni N Lautier J Collos Y Le Gall S 2000 Feeding rate of the oyster Crassostrea gigas in a natural planktonic community of the Mediterranean Thau Lagoon. Mar Ecol Prog Serr 205: 171-184 Foster-Smith RL 1975 The effect of concentration of suspension on the filtration rates and pseudofaecal production for Mytilus edulis L., Cerastoderma edule (L.) and Venerupis pullastra (Montagu). J Exp Mar Biol Ecoll 17: 1-22 Gaffney PM 1996 Biochemical and population genetics. In: The Eastern Oyster Crassostrea virginica. VS Kennedy RIE Newell AF Eble (Eds), Maryland Sea Grant Book, College Park, pp 423-441
329 Gerdes D 1983 The Pacific oyster Crassostrea gigas. Part I. Feeding behaviour of larvae and adults. Aquaculture 31: 195-219 Hoek PPC 1902 Rapport over de oorzaken van den achteruitgang in hoedanigheid van de Zeeuwsche oester. Ministerie van Waterstaat, Handel en Nijverheid, 's-Gravenhage. Hoek PPC 1911 Rapport over de schelpdiervisscherij en schelpdierteelt in de Noordelijke Zuiderzee. Verslag van den Staat der Nederlandsche Zeevisscherijen over 1910. Den Haag, 162 p Korringa P 1951a The shell of Ostrea edulis as a habitat. Arch Néerl Zooll 10: 32-152 Korringa P 1951b Over Mytilicola intestinalis (Copepoda Parasitica) en enkele andere ongewenste vreemdelingen in onze wateren. Vakblad Bioll 31: 63-74 Korringa P 1957 Water temperature and breeding throughout the geographical range of Ostrea edulis. Ann Bioll 33: 1-15 Loosanoff VL Nomejko CA 1951 Existence of physiologically different races of oysters, Crassostrea virginica. Biol Bull 101: 151-156 Miller AW 2000 Assessing the Importance of Biological Attributes for Invasion Success: Eastern oyster (Crassostrea virginica) Introductions and Associated Molluscan Invasions of Pacific and Atlantic Coastal Systems. PhD thesis Univ California Los Angeles, 192 p Möbius K 1977 Die Auster und die Austernwirthschaft. Wiegandt, Hempel and Parey, Berlin. 126 p Møhlenberg F Riisgård HU 1979 Filtration rate, using a new indirect technique, in thirteen species of suspension-feeding bivalves. Mar Bioll 54: 143-147 Neudecker T 1985 Untersuchungen zur Reifung, Geschlechtsumwandlung und künstlichen Vermehrung der Pazifischen Auster Crassostrea gigas in deutschen Gewässern. Veröffentl Inst Küsten- Binnenfisch Hamburgg 88: 1-212 O’Foighil D Gaffney PM Wilbur AE Hilbisch TJ 1998 Mitochondrial cytochrome oxidase I gene sequences support an Asian origin for the Portuguese oyster Crassostrea angulata. Mar Bioll 131: 497-503 Reise K 1998 Pacific oysters invade mussel beds in the European Wadden Sea. Senckenbergiana maritt 28: 167-175 Reise K Gollasch S Wolff WJ 1999 Introduced marine species of the North Sea coasts. Helgoländer Meeresunters 52: 219-234 Reise K Herre E Sturm M 1989 Historical changes in the benthos of the Wadden Sea around the island of Sylt in the North Sea. Helgoländer Meeresunters 43: 417-433 Reise K Schubert A 1987 Macrobenthic turnover in the subtidal Wadden Sea: the Norderaue revisited after 60 years. Helgoländer Meeresunters 41: 69-82 Riera P Stal LJ Nieuwenhuize J 2002 į13C versus į15N of co-occurring molluscs within a community dominated by Crassostrea gigas and Crepidula fornicataa (Oosterschelde, The Netherlands). Mar Ecol Prog Ser 240: 291-295 Riesen W Reise K 1982 Macrobenthos of the subtidal Wadden Sea: revisited after 55 years. Helgoländer Meeresunters 35: 409-423 Rothschild BJ Ault JS Goulletquer P Heral M 1994 Decline of the Chesapeake Bay oyster population: a century of habitat destruction and overfishing. Mar Ecol Prog Serr 111: 29-39 Smaal AC Prins TC 1993 The uptake of organic matter and the release of inorganic nutrients by bivalve suspension-feeder beds. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed) Springer, Heidelberg, pp 271-298 Stauber LA 1950 The problem of physiological species with special reference to oysters and oyster drills. Ecology 31: 109-118 Swennen C 1990 Oystercatchers feeding on giant bloody cockles on the Banc d’Arguin, Mauritania. Ardea 78: 53-62
330 van Banning P 1988 Management strategies to control diseases in the Dutch culture of edible oysters. Am Fish Soc Spec Publl 18: 243-245 van der Vlis JA 1975 ‘tLandt van Texsel. Den Burg Vernet R 1993 Préhistoire de la Mauritanie. Centre Culturel Français, Nouakchott. 427 p Vonk JA 2001 Predation by Fish on the Intertidal Flats of the Banc d'Arguin, Mauritania. MSc thesis University of Groningen. Walne PR 1972 The influence of current speed, body size and water temperature on the filtration rate of five species of bivalves. J Mar Biol Ass UK K 52: 345-374 Wehrmann A Herlyn M Bungenstock F Hertweck G Millat H 2000 The distribution gap is closed. First record of naturally settled Pacific oysters Crassostrea gigas in the East Frisian Wadden Sea, North Sea. Senckenbergiana maritt 30: 153-160 Winter JE 1973 The filtration rate of Mytilus edulis and its dependence on algal concentration, measured by a continuous automatic recording apparatus. Mar Bioll 22: 137-328 Wolff WJ Duiven AG Duiven P Esselink P Gueye A Meijboom A Moerland G Zegers J 1993 Biomass of macrobenthic tidal flat fauna of the Banc d’Arguin, Mauritania. Hydrobiologia 258: 151-163 Wolff WJ Gueye A Meijboom A Piersma T Sall MA 1987 Distribution, biomass, recruitment and productivity of Anadara senilis (L.) (Mollusca: Bivalvia) on the Banc d’Arguin, Mauritania. Neth J Sea Res 21: 243-253 Wolff WJ Reise K 2002 Oyster imports as a vector for the introduction of alien species into northern and western European coastal waters. In: Invasive Aquatic Species of Europe: Distributions, Impacts and Management, E Leppaköski S Gollasch S Olenin (Eds), Kluwer, Dordrecht, pp 193-205 Wolff WJ Smit CJ 1990 The Banc d’Arguin, Mauritania, as an environment for coastal birds. Ardea 78, 17-38 Yonge CM 1960 Oysters. Collins, London, 209 p
OYSTER REEFS AS COMPLEX ECOLOGICAL SYSTEMS
Richard Dame Marine Science, Coastal Carolina University, Conway, SC, USA Abstract:
Aggregations of suspension-feeding organisms like oyster reefs, mussel beds and worm reefs are prominent systems in coastal environments. The fundamental properties of these systems are reviewed and indicate that they are complex systems that are highly optimized and evolutionarily selected for high productivity. Such systems are unstable when faced with a never experienced situation. In the case of oyster reefs, catastrophic collapse usually involves anthropogenic factors such as pollution, sediment loading, and over-harvesting. Strategies for examining these complex systems and developing compatible management approaches must include continued experimentation, learning, and adaptation by managers.
Keywords: oyster reefs, complexity, productivity, highly optimized tolerance, management
INTRODUCTION Ancient philosophers to modern cosmologists observed that our universe tends toward increasing complexity over time. In general, across scales ranging from nuclear particles to galaxies complexity increases with time. Why? Basically, complexity arises because there are large energy gradients in our universe that open non-equilibrium systems take advantage of within the laws of thermodynamics to build evermore complex structures (Chaisson 2001). All living systems are open to their environment and in a state of thermodynamic non-equilibrium. Furthermore, natural selection results in ecosystems that are organized for high complexity and production (Leigh and Vermeij 2002). The main objective of this paper is to identify the
331 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 331–343. ©2005 Springer. Printed in the Netherlands.
332 properties of benthic suspension-feeder systems that may be the result of complexity. Using intertidal eastern oyster reefs as examples, the general properties of complex systems will be defined and compared to data from intertidal oyster reefs. Finally, the imperative of using a complex systems approach to frame the strategies that are used to manage these resources will be explored.
ECOLOGICAL COMPLEXITY Dense aggregations of benthic suspension-feeders are common to many shallow water environments. Typically called reefs or beds, these systems often play such dominant structural and functional roles in ecosystems that they are frequently categorized as keystones components or ecosystem engineers (Dame 1996). Reefs and beds of suspension-feeders are thermodynamically open systems that are connected to their external
Fig. 1. The intertidal oyster reef or network. Strength of a link as quantified by energy or material flux is implied by thickness of arrows. Oys = oysters; P = oyster parasites; Bv = other bivalves; Barn = barnacles; D = detritus; Wrm = various worms (data from Dame 1979).
environment by material and energy fluxes that include the processes of feeding, excretion, reproduction, predation, resuspension and nutrient cycling (Dame 1996). Other organisms within these aggregations are connected to the
333 Table 1. Basic characteristics or properties of complex systems with intertidal oyster reef examples. Characteristic
Oyster Reef
Open systems
Energy/Matter/Information Flow
Dame (1996)
Multiple interacting components
Functional groups
Dame and Patten (1981)
Hierarchical
Organisms/Populations/Systems
Dissipative structures
Organisms/Populations/Systems
Non-equilibrium
Living system
Dame (1972a, 1976), Dame and Patten (1981), Dame et al. (1989) Dame (1972a, 1976), Dame and Patten (1981) This paper
Positive feedback
Reef structure Part of N-cycle
Negative feedback
External environment
This paper Dame (1996), Newell et al (2002) Dame (1972a, 1996)
Self-organization & non-linear
Shell & biomass maximized
This paper
Emergence
Reef Structure Flow network structure Turbulent flow Externally forced Plankton vs. Benthos
This paper Dame and Patten (1981) Dame (unpublished) Dame et al. (2002) Newell (1988)
Scaling relationships
Individuals/Clumps/Reefs
Surprise
Remove oysters/Little change
Dame 1972a, b, Brown et al (2002), Commito and Rusignuolo (2000), Kostylev and Erlandsson (2001), Lawrie & McQuaid (2001) Dame et al. (2002)
Highly optimized tolerance (HOT)
System collapse Anthropogenic stresses
Multiple states
Reference(s)
Carlson and Doyle (1999, 2002), Dame et al. (2002), This paper
suspension-feeders as well as to each other and include primary producers, grazers, sediment processors, decomposers and predators (Dame and Patten 1981). Thus, the bed or reef is a network (Fig. 1) or subsystem of inter- and intra-connected organisms or components with different functional attributes within a larger system or ecosystem. The ecosystem is itself composed of a
334 network of inter- and intra-connected subsystems. It is from the interrelationships between the components within and between organizational scales (organism, aggregation and ecosystem) that behavior or properties emerge. In general terms, the properties described above are the definition of a complex ecological system. Some of the more common properties of ecological complexity are listed in Table 1 with intertidal oyster reef examples. The following narrative describes and explains these properties.
Non-equilibrium and Dissipative Structures Living systems continuously utilize gradients of useable energy to do work. As a consequence of the 2ndd Law of Thermodynamics, entropy or the amount of disorder in the universe increases while locally the living system lives and builds orderly structure. Structures that produce entropy are said to be dissipative. Because thermodynamic equilibrium is defined as the state when no entropy is being created and when there is no net flow of energy, living systems are dissipative structures in a state of non-equilibrium. From a biological or living system perspective, dead systems exist in thermodynamic equilibrium. Therefore, as oysters and oyster reefs are living systems consuming usable energy, building orderly structures and producing entropy, they are non-equilibrium, dissipative structures. Hierarchy The inherent differences in the consumption and dissipation of energy by functional groups and species leads to the development of hierarchies (food chains and webs). The dominant species or functional group that exerts the greatest control over usable resources imposes a cascade of top-down control measures on the rest of the system. At each higher level in the hierarchy there is a decreased dissipation rate (Johnson 1995) and a progression from rapid turnover at the bottom to low turnover at the top. In the intertidal oyster reef (Fig. 2), oysters are the dominant species in the dominant functional group (suspension-feeders) and they are the longest-lived organisms within their system (Dame and Patten 1981). Thus, changes in oyster abundance potentially can cause changes elsewhere in the system.
Feedback, Self-organization and Emergence Examples of feedback loops are easily seen within intertidal oyster reefs. Predator-prey relationships between organisms on the surface of the reef
335 and in the sediments supporting the reef are well-documented (Dame and Patten 1981). In addition, feedback resulting from nutrient cycling within oyster reefs and between oyster reefs and other components of the estuary is also evident (Dame 1996). These types of feedback depend more on the relationships between components than the components themselves. Given the number and variety of feedback pathways in ecological systems, these relationships can extend beyond simple feedback to higher order nonlinear processes. This internal structure of feedback relationships has the property of self-organization that allows this internal structure to change and better interact with its environment (Manson 2001).
Fig. 2. A process/functional model of an intertidal oyster reef (see Dame and Patten 1981 for a similar description in joules).
A second type of feedback involves feedback between internal structure and the external environment. This kind of feedback is exemplified by the development of the physical structure of the intertidal reef (Fig. 3). It is well known that oyster larvae are gregarious and prefer to settle on adult oysters or clean oyster shell. The oyster larvae even orientate their shells parallel to tidal currents in order to remove suspended particles more
336 efficiently from the currents (Lawrence 1971). Each generation of larvae tends to settle on top of the preceding generation so that the reef grows vertically
Fig. 3. The emergence of a reef’s structure from (A.) larval settlement and orientation, (B.) clumping and (C.) oyster reef (see text for explanation).
337 (Bahr and Lanier 1981). Eventually, the top of the reef encounters the external negative feedback of high temperatures and extended exposure during the summertime and this causes high oyster mortality on the top of the reef (Dame 1996, Bushek pers. com). Thus, the shape of an intertidal oyster reef emerges from the interaction of positive and negative feedbacks. There is no grand master plan, only a few simple rules that govern when, where and how the larvae settle and how the oysters interact with their physical environment. The multitude of feedback mechanisms acting in concert within and between systems is also non-linear in character, meaning the behavior of the system to a forcing parameter may not always give a similar response. In complex systems jargon, this behavior is often referred to as “surprise.” In our recent creek scale experiment comparing nutrient fluxes in creeks with oysters to creeks with oysters removed, a seemingly linear experiment, we were “surprised” to find no differences among the different categories of creeks. Using simple linear spreadsheet models, we predicted less ammonium in creeks with oysters removed (Dame et al. 2002), but can only speculate that the network of feedback mechanisms in the creeks somehow compensated for or reconfigured itself to the removal of the oysters.
Multiple or Alternate States Freshwater ecosystems, particularly lakes, appear to exist in alternate states (Scheffer et al. 1993, Dent et al. 2002). Internal feedback mechanisms tend to keep the system in a particular state, while small incremental changes in external environmental feedbacks or anthropogenic forcings can result in a sudden change in the systems state by overwhelming internal stabilizing mechanisms (Scheffer et al. 2001, Scheffer and Carpenter 2003). If the new state is stabilized by its own set of internal feedback mechanisms, it may be difficult if not impossible to reverse the state change simply by reversing the external forcing (Dent et al. 2002). Although I know of no evidence supporting alternate states in intertidal oyster reefs dominated by the eastern oyster Crassostrea virginica, there is evidence of such changes in subtidal C. virginica reefs in Chesapeake Bay (Newell 1988) and in reefs once dominated by intertidal and subtidal European oysters, Ostrea edulis, in Marennes Oléron and the Wadden Sea (Reise et al. 1989, Héral 1993).
338
Fig. 4. A hypothetical representation of two potentially dominant states in an estuary and the external forcing stresses
In the cases where a system state change occurred, the external forcings of over-harvesting, pollution, increased suspended sediments and disease were usually implicated (Fig. 4) and reversal of the forcing to achieve the original state were usually ineffective. This irreversibility is common in living systems because the extensive feedback networks reconfigure to new conditions essentially generating a new system that is enhanced by its own feedback network. Also, traditional restoration tries to re-establish the environmental conditions prior to the observed change of state, but this approach has had limited success when anthropogenic stresses are involved (Suding et al. 2004).
Scaling, Self-similarity, Fractals Scaling relationships that are self-similar or fractal over a broad range of spatial or temporal scales is a type of emergent ecological phenomena (Brown et al. 2002). These relationships are generally characterized by simple power functions. The majority of the early work on biological scaling was directed at variation among individuals or allometry (Brown and West 2000). Studies on the metabolism or oxygen consumption of intertidal oysters in
339 relation to temperature and body size fall into this category (Dame 1972a, b). At scales greater than the individual (population, assemblage and ecosystem); ecologists studying suspension-feeders have focused on spatial and temporal variations in system structure. At these scales, fractal geometry has been used to describe the structure and complexity of animal associations. Although several groups have investigated the fractal-like structure of mussel beds (Commito and Rusignuolo 2000, Kostylev and Erlandsson 2001, Lawrie and McQuaid 2001), I know of no similar published work on intertidal oyster reef, but it is highly likely that such relationships also exist for oyster reefs.
Conceptualization of Complex Systems – SOC versus HOT Early development in the area of complex systems grew out of the physical sciences and was conceptualized by simple models typically composed of homogeneous subcomponents. Complexity was associated with the emergence of dissipative structures and spatial patterns in nonequilibrium systems (Kauffman 1993, Bak 1996). These models were loosely grouped under the terms of self-organized criticality (SOC) and complex adaptive systems (CAS). These approaches generally begin with the idea that many complex systems reside at a boundary between order and disorder, and focus on internal properties that are generic, homogeneous and self-similar. SOC systems generally have a low yield or productivity and use external forcing to initiate mechanisms that give rise to complexity internally (Carlson and Doyle 1999). Proponents of SOC/CAS also tend to downplay detailed differences in subcomponent behavior and interconnectivity. Highly optimized tolerance (HOT) or selection for high productivity in ecological terms was recently introduced as another way to conceptualize and study complexity (Carlson and Doyle 1999 & 2002). This approach emphasizes the highly structured, non-generic, self-dissimilar internal relationships and the robust yet fragile external behavior of the system. These systems are highly productive and evolve in a manner that rewards successful strategies depending on the external stimulus. A comparison of the properties of SOC and HOT showed that the two approaches predict not just different but exactly opposite features of complex systems (Carlson and Doyle 2002). Basically, HOT suggests that random interconnections of subcomponents do not tell us much about the complexity of actual living systems, that the details of the subcomponents matter enormously, and that generic phase transitions play a peripheral role in system complexity. When Carlson and Doyle (2002) assessed the properties of SOC and HOT with data from living systems, the results were the same as when compared with HOT (Table 2).
340 If we compare published data on intertidal oyster reefs to the main properties of the HOT approach there is a strong congruence with HOT (Table 2). Intertidal oyster reefs are highly structured and productive systems that are composed of non-generic subunits that have self-dissimilar internal relationships. With millions of years of evolutionary history, these systems have evolved in a manner that rewards successful strategies in response to their external environment. They are robust systems, yet they are also fragile when faced with environmental circumstance they have never experienced, for example anthropogenic stresses and diseases. Trophic cascades are Table 2. A comparison of selected properties of HOT (highly optimized tolerance), SOC (selforganized criticality) and oyster reef systems (Adapted from Carlson and Doyle 2002). Property
SOC
Internal configuration
Generic, homogeneous , self-similar
Structured, heterogeneous, self-dissimilar
External behavior (robustness)
Generic
Robust yet fragile
Energy flow (production)
Low
Feedback structure Trophic cascades
HOT
Oyster reef
Reference
Trophic web, functional groups, selfdissimilar Dominant system in estuaries preindustrialization
Dame and Patten (1981)
High
High
Dame (1976)
Simple
Detailed
Detailed
Dame and Patten (1981)
Common
Rare (barriers)
Rare (barriers?)
(not observed)
Newell (1988)
probably rare and these systems have most likely evolved barriers against them. In ecological systems, complexity is almost always internal and involves a great deal of evolved redundancy and structure to allow these systems to behave in a reasonably predictable fashion in spite of uncertainties in their environment (Carlson and Doyle 1999). To understand how these systems work and how to ecologically manage them will require new approaches because the details probably do make a difference.
MANAGEMENT OF COMPLEX ECOLOGICAL SYSTEMS Single species management of organisms particularly marine fisheries have been, in general, dismal failures. The ideal of sustainable harvest of any
341 fishery has recently been challenged by Pauly et al. (2002) who argue that fishing induces serial depletions in stocks that have been long masked by the improved technology, geographic expansion, and exploitation of species lower in the food web. Based on the preceding review of ecological complexity as related to suspension-feeding bivalves and intertidal oyster reefs in particular, the evidence implies that it is the disruption of feedback networks within systems that change the system. Because most ecological systems have not had time to evolve mechanisms to adapt themselves to anthropogenic stresses, these stresses can lead to serious damage if not extinction. From a management perspective the situation may be more insidious, Pandolfi et al. (2003) found in a global study of coral reef geology that the reef ecosystems began to deteriorate as soon as the first Homo sapiens arrived at the site. Accepting that most ecological systems are in fact complex systems and that some if not all of these systems have the properties of HOT, what kind of management strategies should we use? It is imperative that a whole or ecosystems approach be the core framework for managing. First, all available information on the system should be synthesized into conceptual and/or simulation models. This information should be used as a basis for designing a routine monitoring program with both spatial and temporal parameters. Then important levels of organization or scale should be identified. Properties that are invariant across scales as well as fast and slow variables need to be recognized. Second, as a number of managerial approaches have had little success, i.e., top down, governing board, etc., Carpenter et al. (1999) suggest an approach that involves the manager in the system or sees the manager and the management strategy as an integral part of the system needs to be carefully considered. By being part of the system network, the manager continually revises the management plan and in a sense adapts to/with the system as it changes through time (Carpenter et al. 1999). If the system is in a HOT state the expectation of hypersensitivity to unforeseen perturbations requires serious scientific investigations of potential perturbations before they occur in order to gauge the potential resilience of the system (Carlson and Doyle 1999).
CONCLUSIONS Why should we care if intertidal oyster reefs or other suspensionfeeding systems are complex or not complex? If these systems are complex and I believe the evidence supports this assertion, then scientifically we must start to look beyond linear explanations to non-linear descriptions of systems behavior. Unexpected and often sudden changes in living systems, such as changes to alternate states, loss of species with strong links and invasions of
342 species with even stronger links, must be accepted as viable possibilities. Restoration of degraded systems is difficult and requires a much greater depth of knowledge and understanding than is currently evident. The prospects for research on systems dominated by suspension-feeders (strong links) are bright.
REFERENCES Bahr LM Lanier WP 1981 The ecology of intertidal oyster reefs of the South Atlantic coast: A community profile. FWS/OBS-81/15, US Fish and Wildlife Service, 105 p Bak P 1996 How Nature Works: The Science of Self-Organized Criticality. Copernicus, New York, 212 p Brown JH Gupta VK Li B-L Milne BT Restrepo C West GB 2002 The fractal nature of nature: power laws, ecological complexity and biodiversity. Phil Trans R Soc Lond B 357: 619-626 Brown JH West GB 2000 Scaling in Biology Oxford University Press, New York, 352 p Carlson JM Doyle J 1999 Highly optimized tolerance: A mechanism for power laws in designed systems. Phys Rev E 60: 1412-1427 Carlson JM Doyle J 2002 Complexity and robustness. PNAS S 99: 2538-2545 Carpenter S Brock W Hanson P 1999 Ecological and social dynamics in simple models of ecosystem management. Cons Ecoll 3(2):4 [online] URL: http://www.consecol.org/vol3liss2/art4 Chaisson EJ 2001 Cosmic Evolution: The Rise of Complexity in Nature. Harvard University Press, Cambridge, MA, 275 p Commito JA Rusignuolo BR 2000 Structural complexity in mussel beds: the fractal geometry of surface topography. J Exp Mar Biol Ecoll 255: 133-152 Dame RF 1972a The ecological energies of growth, respiration and assimilation in the intertidal American oyster, Crassostrea virginica. Mar Bioll 17: 243-250 Dame RF 1972b Comparison of various allometric relationships in intertidal and subtidal American oysters. Fish Bulll US 70: 1121-1126 Dame RF 1976 Energy flow in an intertidal oyster population. Est Coast Mar Sci 4: 243-253 Dame RF 1979 The abundance, diversity and biomass of macrobenthos on North Inlet, South Carolina, intertidal oyster reefs. Proc Nat Shellfish Assoc 69: 6-10 Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Dame RF Bushek D Allen DM Lewitus AJ Edwards D Koepfler ET Gregory L 2002 Ecosystem response to bivalve density reduction. Aqua Ecoll 36: 51-65 Dame RF Patten BC 1981 The analysis of energy flows in an intertidal oyster reef. Mar Ecol Prog Ser 5:115-124 Dame RF Spurrier JD Wolaver TG 1989 Carbon, nitrogen and phosphorus processing by an oyster reef. Mar Ecol Prog Serr 54: 249-256 Dent CL Cumming GS Carpenter SR 2002 Multiple states in river and lake ecosystems. Phil Trans R Soc Lond B 357: 635-645 Héral M 1993 Why carrying capacity models are useful tools for management of bivalve culture. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed). Springer-Verlag, Heidelberg, pp 455-477 Johnson L 1995 The far-from-equilibrium ecological hinterlands. In: Complex Ecology: The Part-Whole Relation in Ecosystems, Patten BC Jorgensen SE (Eds), Prentice Hall, NJ, pp 51-103 Kauffman SA 1993 The Origins of Order: Self-Organization and Selection in Evolution. Oxford University Press, New York, 734 p
343 Kostylev V Erlandsson J 2001 A fractal approach for detecting spatial hierarchy and structure in mussel beds. Mar Bioll 139: 497-506 Lawrence DR 1971 Shell orientation in recent and fossil oyster communities from the Carolinas. J Paleo 45: 347-349 Lawrie SM McQuaid CD 2001 Scales of mussel bed complexity: structure, associated biota and recruitment. J Exp Mar Biol Ecoll 257: 135-161 Leigh EG Vermeij GJ 2002 Does natural selection organize ecosystems for the maintenance of high productivity and diversity? Phil Trans R Soc Lond B 357: 709-718 Manson SM 2001 Simplifying complexity: a review of complexity theory. Geoforum 32: 405414 Newell RIE 1988 Ecological changes in Chesapeake Bay: are they the result of overharvesting the American oyster, Crassostrea virginica? In: Understanding the estuary: advances in Chesapeake Bay research , MP Lynch EC Krome (Eds), Chesapeake Bay Research Consortium, Solomon’s, Maryland, pp 536-546 Newell RIE Cornwell JC Owens MS 2002 Influence of simulated bivalve deposition and microphytobenthos on sediment nitrogen dynamics: A laboratory study. Limnol Oceanogrr 47: 1367-1379 Pandolfi JM Bradbury RH Sala E Hughes TP Bjorndal KA Coke RG McArdle D McClenachan L Newman MJH Peredes G Warner RR Jackson JBC 2003 Global trajectories of the long-term decline of coral reef ecosystems. Science 301: 955-958 Pauly D Christensen W Genetta S Pitcher TJ Sumaila UR Walters CJ 2002 Towards sustainability in world fisheries. Nature 418: 689-695 Reise K Herre E Sturm M 1989 Historical changes in the benthos of the Wadden Sea around the island of Sylt in the North Sea. Helgoländer Meeresunters 43: 417-433 Scheffer M Carpenter S 2003 Catastrophic regime shifts in ecosystems: linking theory to observation. TREE E 18: 648-656 Scheffer M Carpenter S Foley JA Walker B 2001 Catastrophic shifts in ecosystems. Nature 413: 591-596 Scheffer M Hospar SH Meijer M-L Moss B Jeppesen E 1993 Alternative equilibria in shallow lakes. TREE E 8: 275-279 Suding KN Gross KL Houseman GR 2004 Alternate states and positive feedbacks in E 19: 46-53 restoration ecology. TREE
THE ROLES OF SUSPENSION-FEEDERS IN ECOSYSTEMS: SYNTHESIS AND CONCLUSIONS
Participants of the NATO Advanced Research Workshop NIDA, LITHUANIA, OCTOBER 2003
INTRODUCTION Suspension-feeding organisms are ubiquitous in marine, estuarine and freshwater environments, but there are few comparisons of their structural and functional roles in these very different systems. In this report, we present a synthesis of the Working Groups that were an important component of the NATO Advanced Research Workshop held in Nida, Lithuania in October of 2003. Some of the questions addressed by the Working Groups were: (1) What is, or how do we define suspension-feeding? (2) What are the structural and functional roles of pelagic suspension-feeders? (3) What roles do suspension-feeders play in specific systems and why are suspension-feeders important in a given system (both natural and human dominated)? (4) Why are suspension-feeders well suited to be invaders in aquatic ecosystems?
Suspension-feeders In aquatic systems, currents generated by winds and tides provide the energy necessary to initiate and maintain organic particles, including phytoplankton, microzooplankton, bacteria, and detrital particles, in suspension. “Suspension-feeder” is a generic term for organisms that obtain their food by indiscriminately capturing these suspended particles from the water column.. Mechanisms of particle capture are highly diverse among the major groups of suspension-feeders (e.g., crustaceans, polychaetes, molluscs, ascidians, fish etc), each having evolved very different mechanisms of particle capture (Shimeta and Jumars 1991). For example, some species of ascidians simply use mucus covered structures to entrap all suspended particles and then 345 R.F. Dame and S. Olenin (eds.), The Comparative Roles of Suspension-Feeders in Ecosystems, 345–353. ©2005 Springer. Printed in the Netherlands.
346 ingest the particles and mucus together. Others, most notably among the molluscs, use elaborately ciliated structures to capture particles (Ward et al. 1998, Ward and Shumway 2004). Most species are “active” suspensionfeeders that use energy to induce a water current across their feeding structures (Newell 1979). Other species are “passive” suspension-feeders that simply orientate their feeding structures in the prevailing ambient water flow. Some nektonic and benthic suspension-feeders (e.g., barnacles) possess the ability to switch between active and passive capture of food particles (Kajak et al. 1975, Newell 1979, Unger et al. 1984). Although there is generally indiscriminate capture of particles from the water column, most suspension feeders have developed elaborate mechanism for sorting captured particles, thereby ensuring that the most nutritious particles are ingested (Ward and Shumway 2004)
Coastal and Marine Systems Marine systems are defined as those environments with stable salinities over 30 and located in or adjacent to the open ocean (Remane and Schlieper 1971, Symposium on the Classification of Brackish-Waters: The Venice System 1959). Marine pelagic suspension-feeders include planktonic and nektonic animals. Meroplanktonic and holoplanktonic forms are important components of some systems where they graze the phytoplankton and remineralize particulate organic materials. Marine nektonic suspensionfeeders or motile macroinvertebrates, fish and mammals (Bushek and Allen 2005) play an additional role of actively transporting energy and materials between different systems and subsystems (Lundberg and Moberg 2003). Benthic suspension-feeders play important roles in benthic-pelagic coupling, but the roles of pelagic suspension-feeders in pelagic-benthic coupling are not well studied. Marine benthic suspension-feeders function, e.g., oysters and barnacles, as sinks for suspended particles and sources of dissolved nutrients (Asmus et al. 1992, Chauvaud et al. 2000). Some epibenthic species, such as oyster and mussels, also provide structure for larval settlement, nursery habitats and macroalgae attachment. Within their systems, these organisms have a large capacity for storing energy and materials as biomass (Dame 1996).
Estuarine Systems
Estuaries are defined as semi-enclosed tidal bodies of water that connect water runoff from the land to the sea across a salinity gradient.
347 These usually highly productive and complex systems are often heavily impacted by human activities. The role of benthic suspension-feeders in estuaries has been extensively investigated (Dame 1993). In estuaries, benthic suspension-feeders often dominate their systems through a number of strong-links or feedback pathways in a variety of important processes including benthic-pelagic coupling (Alpine and Cloern 1992, Dame et al. 2001), nutrient regeneration (Dame et al. 1989, Smaal and Prins 1993, Chauvaud et al. 2000, Ojaveer et al. 2002, Newell et al. 2005), primary production (Asmus and Asmus 1991 and 1993, Prins and Escaravage 1995), sedimentation (Kautsky and Evans 1987, Chauvaud et al. 1998, Dame et al. 2001, Kotta et al. 2005), and biocomplexity (Dame 2005). In addition to their involvement in natural processes, benthic suspension-feeders in estuarine systems are often cultivated for food use by higher trophic levels including humans (Shpigel 2005), used in eutrophication control and mitigation (Shpigel 2005), and employed as bio-monitors or sentinels of environmental contamination (Borchardt et al. 1988, Phillips and Rainbow 1994, Kramer 1994, Wolowicz et al. 2005).
Freshwater Systems Freshwater systems are defined as standing and enclosed (ponds and lakes) or flowing (streams and rivers) aquatic systems with salt concentrations less than 0.5 (Remane and Schlieper 1971). Both benthic and planktonic suspension-feeders can play important roles in these systems, by changing both abiotic and biotic ecological parameters (Ostroumov 2002 and 2005), but their effects are usually limited to shallow areas. However, there are several lake-wide complex ecosystem effects recorded as a result of pelagic suspension-feeders and their impacts may substantially differ in shallow and open (deep) parts of a standing water-body (Ojaveer 2005). In deep stratified lakes pelagic suspension-feeders can graze phytoplankton in the euphotic zone (Ojaveer 2005). In rivers, the presence of dense populations (Tudorancea and Gruia 1968) of unionid clams suggests that suspension-feeders may be important. The exception to the freshwater situation described above is that of species such as the Ponto-Caspian zebra mussel, Dreissena polymorpha, and Asiatic clam, Corbicula fluminea. These species are invaders in many freshwater ecosystems worldwide and they can quickly dominate a system and establish a strong benthic-pelagic coupling that can dramatically change the biological and physical characteristics of freshwater systems (Phelps 1994, Karatayev et al. 2002, 2005).
348 Differences and Similarities In general, marine systems are dominated by phytoplankton primary productivity, but their relative productivity compared to other aquatic systems is low except in coastal upwelling zones (Vinogradov and Shushkina 1987, Nixon 1988). Suspension-feeders rarely dominate these systems (Table 1). Table 1. A comparison of the relative importance of different modes of suspension-feeding in aquatic environments. + = low , ++ = moderate, and +++ = high affect. Mode
Marine
Estuarine
Freshwater
Benthic
++
+++
+
Planktonic
+
+
+
Nektonic
++
+
+
Invaders
+
++
+++
Estuarine systems have highly productive planktonic and vascular plant components and are often dominated by benthic suspension-feeders (Dame 1996, Herman et al. 1999). Extensive work by Wolff (1973) found that marine and freshwater species rarely co-occur and there was no overlap
Table 2. Comparison of the relative importance of suspension-feeder processes in aquatic systems. . + = low , ++ = moderate, and +++ = high affect. Process Benthic-Pelagic Coupling Production/ Eutrophication Sedimentation
Marine
Estuarine
Freshwater
+
+++
+
++
+
++
+
Contaminants/ Biomagnification Aquaculture Mobile Links
++ + ++
+
349 among marine and freshwater suspension-feeders along a salinity gradient in several northern European estuaries. Later work (Wolff 1999) determined that before the invasion of these estuaries by exotics no suspension-feeding molluscs occurred in their brackish zones and there existed a clear spatial gap between the marine and freshwater species. Benthic-pelagic coupling is commonly a strong foodweb link in estuarine systems dominated by benthic suspension-feeders (Dame 1993, 1996). Due to intense human influence, estuarine systems seem to be susceptible to invasions by exotic suspension-feeders that force these systems into alternate states, which are difficult to change or switch back to the original state (Dame 2004). Freshwater systems are characterized by higher pelagic productivity than marine systems and tend to be dominated by planktonic crustaceans (Ojaveer 2004). These systems also seem to be very sensitive to successful invasion by benthic suspension-feeders (Olenin and Daunys 2004). .
MAJOR QUESTIONS AND INFORMATION GAPS Cross-system Comparisons The marine, estuarine and freshwater environments not only have very different physical parameters and species, but the relative importance of suspension-feeders appears to vary in relation to process. In many estuaries, resident benthic suspension-feeders have virtually disappeared and their functions replaced by invading benthic suspension-feeders (San Francisco Bay and the Wadden Sea) or by pelagic suspension-feeders (Chesapeake Bay – Newell 1988). In slow moving or closed freshwater systems, invading benthic suspension-feeders often fill an unoccupied niche and develop the properties of system engineers that quickly dominate their environment (Dame 1996). Exactly why and how these suspension-feeders are successful is a major question of interest. Cross-system comparisons should illuminate the roles of suspension-feeders and provide information on why benthic suspensionfeeders are successful invaders.
Relative Importance of Pelagic and Benthic Suspension-feeders Pelagic suspension-feeders in marine and estuarine environments have seldom been studied as a functional group and their roles as well as impacts on ecosystems are mainly speculative. In addition, these animals can be either meroplanktonic or holoplanktonic as well as nektonic. These forms are either
350 passively or actively mobile and capable of moving both vertically and horizontally in the aquatic environment. Their mobility allows them to not only to transfer materials between different environments, but also to translocate their functional capabilities into other habitats. It also gives these suspension-feeders the potential to form strong links between different systems and subsystems. Benthic suspension-feeders are thought to be longer lived and more stabilizing than pelagic suspension-feeders on ecosystems (Herman and Scholten 1990). In the role of mobile processors, pelagic suspension-feeders may also be essential to ecosystem continuity or history. Furthermore, the role of benthic and pelagic suspension-feeders in biodeposition may differ between environments and result in differences in nutrient cycling between different systems (Newell et al. 2005). Thus, it is appropriate to study benthic and pelagic suspension-feeders in a comparative and comprehensive framework.
Before-After Experiments and Invasive Suspension-feeders Because many invasive suspension-feeders are exotic and dominate the environments they take over, investigators always seem to be playing catch-up in determining their influence on system scale processes. In addition to constructing and maintaining long term data sets, researchers need to conduct short term experiments using Before-After/Control-Incident (BACI) designs (Underwood 1994) or the like in order to develop an under-standing of how invading suspension-feeders influence their new system.
Alternative States in Suspension-feeder Dominated Systems Suspension-feeder systems, particularly those that develop emergent structure (reefs) are non-linear, non-equilibrium systems that may develop alternate states in response to species loss or species invasion (Dame 2004). Because of their non-linearity, the development of these systems is difficult to predict and new strategies need to be devised to manage and reconstruct these systems appropriately. REFERENCES Alpine AE Cloern JE 1992 Trophic interactions and direct physical effects control phytoplankton biomass and production in an estuary. Limnol Oceanogrr 37: 946-955
351 Asmus H Asmus R 1993 Phytoplankton-mussel bed interactions in intertidal ecosystems. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed), Springer-Verlag, Heidelberg, pp 57-84 Asmus H Asmus R Prins T Dankers N Frances G Maas B Reise K 1992 Benthic-pelagic flux rates on mussel beds: tunnel and tidal flume methodology compared. Helgoländer Meeresunters 46: 341-361 Asmus R Asmus H 1991 Mussel beds: limiting or promoting phytoplankton. J Exp Mar Biol Ecoll 148:215-232 Borchardt T Burchert S Hablizel H Karbe L Seitner R 1988 Trace metal concentrations in mussels: comparison between estuarine coastal and offshore regions in the southeastern North Sea from 1983 to 1986. Mar Ecol Prog Serr 42: 17-31 Bushek D Allen DM 2005 Motile suspension feeders in estuarine and marine ecosystems. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 53-72 Chauvaud L 1998 La coquille Sanit Jacques en rade de Brest: un modele biologique d’etude des reponses de la faune benthique aux fluctuations de l’environnement. These de doctorat, universite de Bretagne Occidentale, Brest Chauvaud L Jean F Ragueneau O Thouzeau G 2000 Long-term variation of the Bay of Brest ecosystem: Benthic-pelagic coupling revisited. Mar Ecol Prog Serr 200: 35-48 Dame RF 1993 Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes. Springer-Verlag, Heidelberg, 579 p Dame RF 1996 Ecology of Marine Bivalves: An Ecosystem Approach. CRC Press, Boca Raton, FL, 254 p Dame RF 2005 Oyster reefs as complex ecological systems. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, 331-344 pp Dame RF Bushek D Prins TC 2001 Benthic suspensioin-feeders as determinants of ecosystem structure and function in shallow coastal waters. In: Ecological Comparisons of Sedimentary Shores, K. Reise (Ed), Springer, Berlin, pp. 11-37 Dame RF Spurrier J Wolaver TG 1989 Carbon, nitrogen and phosorus processing by an oyster reef. Mar Ecol Prog Serr 54: 249-256 Herman PMJ Middelburg JJ van de Koppel J Heip CHR 1999 Ecology of estuarine macrobenthos. In: Advances in Ecological Research: Estuaries, DB Nedwell and DG Raffaelli (Eds), Academic Press, London, pp 195-231 Herman PMJ Scholten H 1990 Can suspension feeders stabilise estuarine ecosystems? In: Trophic Relationships in the Marine Environment. M Barnes and RN Gibson (Eds), Aberdeen University Press, Aberdeen, p 104-116 Kajak Z Rybak JI Spodniewska I Godlewska-Lipowa WA 1975 Influence of the planktonivorous fish Hypophthalmichthys molitrixx (Val.) on the plankton and benthos of eutrophic lake. Pol Arch Hydrobioll 22: 301-310 Karatayev AY Burlakova LE Padilla DK 2002 Impacts of zebra mussels on aquatic communities and their role as ecosystem engineers. In: Invasive Aquatic Species of Europe: Distributions, Impacts and Management. E Leppäkoski, S Gollasch S Olenin (Eds), Kluwer Scientific Publishers, Dordrecht, pp 433-446 Karatayev AY Burlakova LE Padilla DK 2005 Contrasting distribution and impacts of two freshwater exotic suspension-feeders, Dreissena polymorpha and Corbicula fluminea. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 239-262 Kautsky N Evans S 1987 Role of biodeposition by Mytilus edulis in the circulation of matters in a Baltic coastal ecosystem. Mar Ecol Prog Ser 38: 201-212 Kotta J Orav-Kotta H Field I 2005 Measurements on the variability in biodeposition and estimates of grazing pressure of suspension-feeding bivalves in the northern Baltic Sea.
352 In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 11-30 Kramer KJM 1994 Biomonitoring of Coastal Waters and Estuaries. CRC Press, Boca Raton, FL, 327 p Lundberg J Moberg F 2003 Mobile link organisms and ecosystem functioning: Implications for ecosystem resilience and management. Ecosystems 6: 87-98 Newell RC 1979 Biology of Intertidal Animals. Marine Ecological Surveys, Faversham, UK. Newell RIE 1988 Ecological Changes in Chesapeake Bay, Are they the result of overharvesting the Eastern oyster (Crassostrea virginica)? In: Understanding the Estuary, MP Lynch EC Krome (Eds). Chesapeake Research Consortium Publication 129 (CBP/TRS 24/88), Gloucester Point, VA, USA. pp 536-546. Available from www.vims.edu/GreyLit/crc129.pdf Newell RIE Fisher TR Holyoke RR Cornwell JC 2005 Influence of Eastern oysters on N and P regeneration in Chesapeake Bay, USA. In: The Comparative Roles of SuspensionFeeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 93-120 Nixon, S 1988 Physical energy inputs and the comparative ecology of lake and marine ecosystems. Limnol Oceanogrr 33: 1005-1025 Ojaveer, H 2005 Impact of suspension feeding nekton in freshwater ecosystems: patterns and mechanisms. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 73-92 Ojaveer H Leppakoski E Olenin S Ricciardi A 2002 Ecological impact of Ponto-Caspian invaders in the Baltic Sea, European inland waters and the Great Lakes: An interecosystem comparison. In: Invasive Aquatic Species of Europe: Distributions, Impacts and Management. E Leppakoski S Gollasch S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 412-425 Olenin S Daunys D 2005 Invaders in suspension-feeder systems: Variations along the regional environmental gradient and similarities between large basins. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht. pp 221-238 Ostroumov SA 2002 Inhibitory analysis of top-down control: New keys to studying eutrophication, algal blooms and water self-purification. Hydrobiologia 469: 117-129 Ostroumov SA 2005 Suspension-feeders as factors influencing water quality in aquatic ecosystems. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht. pp 147-164 Phelps HL 1994 The Asiatic clam (Corbicula fluminea) invasion and system level ecological change in the Potomac River Estuary near Washington D.C. Estuaries 17: 614-621 Phillips DJH Rainbow PS 1994 Biomonitoring of Trace Aquatic Contaminants (2nd edition), Chapman and Hall, London, 371p Prins TC Escaravage V 1995 Nutrient cycling and phytoplankton dynamics in relation to mussel grazing in a mesocosm experiment. Ophelia 41: 289-315 Remane A Schlieper C 1971 The Biology of Brackish Water. Wiley Interscience, New York, 372 p Shpigel M 2005 Bivalves as biofilters and valuable by-products in land-based aquaculture systems. In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 183-198 Shimeta J Jumars PA 1991 Physical mechanisms and rates of particle capture by suspensionfeeders. Oceanogr Mar Biol Ann Revv 29: 191-257 Smaal; AC Prins TC 1993 The uptake of organic matter and the release of inorganic nutrients by bivalve suspension feeder beds. In: Bivalve Filter Feeders in Estuarine and Coastal Ecosystem Processes, RF Dame (Ed), Springer-Verlag, Heidelberg, pp 271-298
353 Tudorancea C Gruia L 1968 Observations on the Unio crassus population from the Nera river. Trav Muse d:Histoire Naturelle “Grigore Antipa” 8:381-394 Underwood AJ 1994 On beyond BACI: sampling designs that might reliably detect environmental disturbances. Ecol Appl 4: 3-15 Unger PA Lewis WM McClean DH 1984 Nonvisual feeding in a visual planktivore Xenomelaniris venezuelae. Oecologia 64: 280-283 Vinogradov ME Shushkina EA 1987 Functioning of Plankton Communities of the Epipelagial Ocean. Nauka Press, Moscow, 240 p Ward JE Sanford LP Newell RIE MacDonald BA 1998 A new explanation of particle capture in suspension-feeding bivalve molluscs. Limnol Oceanogr 43(5): 741-752. Ward JE Shumway SE 2004 Separating the grain from the chaff: particle selection in suspension- and deposit-feeding bivalves. J Exp Mar Bio Ecol 300: 83-130 Wolff WJ 1973 The estuary as a habitat. An analysis of data on the soft-bottom macrofauna of the estuarine area of the rivers Rhine, Meuse and Scheldt. Zool Verhandel (Leiden) 126: 1-242 Wolff WJ 1999 Exotic invaders of the meso-oligohaline zone of estuaries in the Netherlands: Why are there so many? Helgoländ Meeresunters 52: 393-400 Woáowicz M Smolarz K Sokoáowski A 2005 Neoplasia in estuarine bivalves: effect of feeding behaviour and pollution in the Gulf of GdaĔ a sk (Baltic Sea, Poland). In: The Comparative Roles of Suspension-Feeders in Ecosystems, RF Dame S Olenin (Eds), Kluwer Academic Publishers, Dordrecht, pp 165-182
INDEX Black Sea, 57, 63, 68, 70, 71, 137, 138, 226, 236, 240, 259, 316 Brachiopoda, 151 Brackish, 346, 352 Brevoortia, 54, 56, 60, 68, 69 Bryozoa, 151
A Acartia, 32, 39, 62, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 310 Adriatic Sea, 323 Africa, 56, 59, 115, 240, 317, 318, 321 Alternate states, 343 Anadara, 185, 318, 319, 320, 328, 330 Antarctic, 63, 87, 206, 207, 212 Antarctic shelf sea, 206 Anthropogenic, 162, 224, 291, 333 Aquaculture, 119, 127, 134, 178, 179, 196, 197, 288, 328, 329, 348 Arctic, 180, 207, 212 Arenicola, 214 Arthropoda, 53, 56 Atlantic, 49, 56, 61, 65, 66, 69, 70, 71, 94, 118, 168, 178, 203, 204, 212, 289, 295, 318, 324, 325, 326, 329, 342 Aurelia, 54, 56, 68, 69, 71
C Cascading effects, 92, 218 Caspian Sea, 226, 231, 235, 240, 241, 261 Cerastoderma edule, 3, 6, 7, 152, 168, 181, 182, 185, 205, 218, 277, 281, 327, 328 Chesapeake Bay, 63, 69, 70, 93, 94, 97, 99, 101, 103, 104, 105, 106, 107, 108, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 153, 161, 168, 179, 305, 306, 322, 329, 337, 343, 349, 352 Chironomid, d 274 Chlamys, 9, 125, 134 Chlorophyll, 17, 38, 40, 106, 109 Choptank River, 95, 97, 99, 105, 106, 107, 108, 109, 110, 112, 113, 117, 118 Ciliates, 38, 39, 47, 48 Clams, 197, 288 Clearance, 12, 15, 17, 21, 22, 33, 36, 107, 139, 150, 161, 208, 257, 261, 306, 307, 308 Clupea, 56, 59, 68, 69 Cnidaria, 53, 56, 68, 69, 161 Cockles, 209, 277 Complex, 163, 339, 342 Complexity, 339, 342 Copepod, 38, 39, 46, 143 Corals, 151 Corbicula, 114, 116, 120, 215, 239, 240, 241, 242, 243, 247, 248, 251, 252, 253, 255, 256, 257, 258, 259, 260, 261, 302, 314, 347, 351, 352 Cordylophora, 226, 227, 234 Crabs, 321 Crassostrea gigas, 3, 6, 7, 32, 49, 50, 116, 122, 134, 167, 185, 189, 190, 192, 193, 196, 197, 213, 231, 277, 278, 281, 282, 285, 288, 289, 325, 327, 328, 329, 330 Crassostrea virginica, 94, 97, 99, 117, 118, 119, 120, 152, 160, 161, 167, 179, 185, 312, 323, 328, 329, 337, 342, 343, 352 Crepidula, 278, 326, 329
B Balanus, 151, 225, 226, 227, 231 Baleen whales, 55, 57, 58 Baltic, 11, 12, 13, 24, 25, 27, 28, 29, 50, 63, 64, 68, 69, 71, 118, 154, 161, 162, 165, 166, 168, 171, 172, 173, 175, 176, 178, 179, 180, 181, 182, 221, 222, 223, 224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 240, 315, 351, 352, 353 Beds, 282 Belarus, 247, 256, 257, 264, 270, 272, 274 Benthic, 29, 50, 55, 65, 66, 118, 119, 120, 160, 207, 215, 216, 217, 218, 225, 229, 236, 255, 258, 261, 269, 293, 312, 314, 346, 348, 349, 350, 351 Benthic-pelagic coupling, 119, 217, 255, 314, 349, 351 Biodeposition, 12, 17, 18, 19, 20, 106, 236 Biodeposits, 98, 99, 191 Biodiversity, 162, 202, 216, 218 Bioinvasions, 236 Bivalve, 1, 7, 9, 27, 31, 49, 51, 93, 94, 103, 116, 128, 130, 134, 161, 184, 195, 197, 215, 216, 217, 255, 256, 273, 300, 306, 314, 316, 317, 321, 329, 342, 351, 352
355
356 Ctenophores, 57 Curonian Lagoon, 221, 222, 223, 224, 225, 226, 232, 233, 234, 237
D Daphnia, 75, 78, 79, 83, 85, 86, 87, 90, 153 Denitrification, 100, 116, 119 Density, 242, 327 Dermo, 94, 107, 115 Diaptomus, 83 Diatoms, 38, 178 Dinoflagellates, 38 Disturbance, 158 Donax, 281 Dreissena, 9, 11, 12, 17, 28, 150, 156, 161, 163, 164, 223, 226, 227, 235, 236, 237, 239, 240, 241, 242, 245, 247, 248, 249, 251, 252, 253, 255, 256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 267, 268, 269, 270, 271, 272, 273, 274, 275, 302, 347, 351
E Echinodermata, 151 Ecosystem, 27, 28, 49, 51, 116, 118, 134, 155, 158, 159, 160, 163, 181, 215, 216, 217, 235, 236, 256, 273, 274, 288, 313, 314, 315, 316, 328, 329, 342, 351, 352 Emergence, 333, 334 Energy budget, 29, 218 Ensis, 213, 215, 278, 281, 288, 326 Epibenthic, 204, 213 Estonia, 11, 73 Estuary, 109, 118, 161, 260, 293, 314, 352 Etiology, 172 Eutrophication, 224, 348
F Faeces, 86 Feedback, 334, 340 Feeding, 9, 27, 28, 49, 57, 90, 118, 134, 146, 161, 164, 199, 216, 217, 237, 256, 267, 268, 270, 275, 328, 329 Filterers, 267, 268 Filtration rate, 29, 161, 217, 252, 257, 258, 327, 329 Fish, 28, 29, 49, 64, 68, 69, 70, 71, 75, 76, 77, 79, 80, 89, 90, 91, 117, 134, 146,
178, 179, 180, 181, 187, 188, 197, 215, 236, 237, 251, 255, 257, 258, 259, 260, 261, 275, 313, 314, 316, 330, 342 Food web, 210 France, 122, 134, 153, 167, 188, 240, 287, 323, 324, 325 Freshwater, 160, 161, 254, 255, 259, 260, 261, 273, 274, 275, 294, 337, 347, 348, 349 Functional change, 235, 263
G Gatherers, 267, 268 Generalized Watershed Loading Functions, 107 Germany, 153, 199, 240, 323 Gracilaria, 119, 189, 196 Gradient, 222 Grazing, 15, 28, 31, 69, 236, 256, 296, 303 Gulf of Finland, 13, 16, 29 Gulf of Gdansk, 168, 170, 171, 172, 173, 175, 176, 179, 181 Gulf of Riga, 13, 16, 27, 28, 236 Gut clearance, 139
H HAB, 64 Herring, 68 Highly optimized, 333, 339, 342 HOT, 333, 339, 340, 341 Hudson River, 117, 119, 247, 260, 261
I Ingestion, 50 Invaders, 235, 348, 352
J Jellyfish, 62, 70
K Klaipeda Strait, 222, 223, 228, 232
L Lake Baikal, 152, 163 Lake Batorino, 247
357 Lake Boginskoe, 267, 268 Lake Bolshie, 270 Lake Dolzha, 269 Lake Grevelingen, 326 Lake Lukomskoe, 247, 248, 270, 272 Lake Mendota, 83, 92, 260 Lake Michigan, 75, 83, 91, 257, 258, 260 Lake Myadel, 267, 268 Lake Nacogdoches, 250 Lake Naroch, 247 Lake Svir, 267, 268, 269 Lake Tuakitoto, 152 Lake Vaeng, 84 Lake Volchin, 269 Lake Warniak, 84 Lakes, 236, 240, 244, 256, 257, 258, 259, 261, 264, 267, 275, 315, 352 Laminaria, 125 Lanternfishes, 57 Lithoglyphus, 226, 227 Lithuania, 1, 221, 255, 345 Locomotion, 141
M Macoma, 166, 168, 171, 179, 180, 181, 182, 206, 215, 225, 226, 227, 281, 283, 309 Management, 31, 70, 115, 118, 197, 235, 236, 237, 257, 274, 275, 289, 315, 316, 330, 351, 352 Marennes-Oleron, 212 Marenzelleria, 226, 227, 237 Marine, 11, 27, 28, 29, 31, 49, 50, 53, 61, 73, 97, 115, 116, 117, 118, 119, 121, 146, 154, 159, 160, 179, 188, 215, 216, 217, 226, 236, 237, 254, 274, 288, 314, 317, 331, 342, 346, 348, 351, 352 Mauritania, 318, 320, 328, 329, 330 Menhaden, 60 Mercenaria, 102, 116, 117, 119, 152, 168, 178, 179, 182, 185, 278, 326 Mesocosm, 31, 34, 37 Mesoscale, 202 Metabolism, 161 Metals, 173, 181 Microzooplankton, 38, 80 Mobile link, 352 Model, 9, 17, 90, 118, 124, 125, 126, 134 Modelling, 1, 134, 202 Motile suspension-feeders, 53, 62 MSX, 94, 107, 115
Multiple states, 333, 342 Musculista, 100, 296 Mussel, 9, 47, 115, 117, 160, 170, 211, 215, 216, 257, 259, 260, 261, 272, 274, 277, 278, 288, 351 Mya, 119, 152, 166, 168, 178, 179, 180, 181, 182, 185, 212, 226, 227, 278, 281, 296, 312, 326 Mytilopsis, 278 Mytilus edulis, 3, 6, 7, 9, 11, 12, 17, 27, 28, 29, 31, 32, 49, 50, 118, 119, 122, 152, 154, 156, 161, 162, 163, 164, 166, 167, 168, 178, 179, 180, 185, 208, 216, 217, 218, 219, 225, 226, 227, 236, 277, 281, 327, 328, 330, 351
N N: P ratio, 75, 85, 99 Neoplasia, 165, 167, 171, 179, 180, 182, 353 Nereis, 119, 206, 225, 227 Nitrogen, 93, 99, 106, 117, 118, 119, 120, 148, 255 Non-equilibrium, 333, 334 North Inlet, 108, 152, 306, 308, 342 North Sea, 62, 63, 69, 70, 175, 177, 181, 199, 202, 218, 226, 229, 230, 231, 234, 237, 277, 281, 288, 289, 326, 329, 330, 343, 351 Nutrients, 81, 148, 251
O Oligochaeta, 80, 84, 255, 260 Oosterschelde, 33, 34, 48, 49, 119, 122, 152, 163, 218, 282, 285, 288, 306, 308, 323, 324, 325, 326, 329 Open systems, 333 Ostrea, 151, 167, 178, 180, 185, 213, 278, 282, 322, 323, 328, 329, 337 Ostreola, 167, 310, 312 Oyster, 98, 116, 117, 118, 119, 179, 288, 289, 312, 313, 323, 328, 330, 333, 340, 351
P Particle, 9, 28, 54, 164, 256 Particles, 58, 98, 164 PCB, 174
358 Pelagic, 55, 68, 70, 114, 118, 212, 348, 349 Perna, 185, 321 Petricola, 278, 281, 282 Phaeocystis, 31, 37, 38, 41, 46, 47, 48, 49, 50, 209 Phosphorus, 93, 99 Phytoplankton, 27, 37, 76, 81, 84, 91, 110, 117, 160, 197, 208, 215, 251, 274, 293, 297, 298, 300, 301, 307, 314, 351 Poland, 84, 165, 168, 171, 177, 179, 181, 248, 261, 264, 275, 353 Polyculture, 187 Porifera, 151 Portugal, 121, 207, 218, 240 Potamocorbula, 32, 152, 291, 295, 314, 315, 316 Potamocorbula amurensis, 32, 152, 291, 295, 314, 315, 316 Potomac River, 116, 215, 247, 256, 260, 314, 352 Predators, 248, 267, 268 Productivity, 205, 258, 261 Pseudofeces, 98 Pygospio, 225, 227
R Rangia, 114 Reefs, 332, 333 Reproduction, 143, 256 Rhodomonas, 33, 35, 36, 77 Roach, 119 Rotifers, 151 Russia, 137, 147, 152, 161, 163, 164, 244, 247, 248, 257, 264, 270
S Sabellaria, 213 San Francisco Bay, 32, 50, 116, 152, 160, 161, 215, 291, 292, 293, 294, 295, 302, 314, 315, 316, 349 Scales, 124, 217, 343 Sea of Azov, 226 Sediment toxicity, 174 Sediments, 173, 174, 236 Selection, 58, 59, 217, 342 Selective feeding, 9 Self-organization, 333, 334 Seston, 4, 9, 99, 107, 134, 251 Sharks, 71 Shifts, 258, 266, 287
Spatial scales, 121 Sungo Bay, 124, 125, 126, 128, 130, 133, 134 Surfactants, 156, 162 Survival, 143 Sylt, 199, 203, 204, 205, 206, 209, 212, 214, 215, 218, 219, 282, 285, 288, 306, 308, 329, 343 Sylt-Rømø-Bight, 203, 204, 212, 214, 215
T Tapes, 9, 152, 189, 190, 197 Tellina, 281 Temora, 31, 38, 44, 50 Texel, 282, 288, 322 The Netherlands, 31, 119, 152, 163, 218, 277, 288, 289, 315, 316, 317, 323, 329 Top-down, 50, 71, 90, 92 Transparency, 74, 80, 81 Tridacna, 207, 217 Trinity River, 247 Trophic cascade, 340 Trophic interaction, 313, 350 Trophic structure, 235, 236 Tumor, 179, 180 Turbidity, 314
U Unio, 156, 162, 175, 180, 227, 353 USA, 53, 64, 67, 70, 83, 93, 97, 115, 116, 117, 118, 119, 152, 153, 178, 239, 243, 247, 252, 255, 256, 259, 260, 263, 314, 315, 316, 331, 352
V Valvata, 223, 225, 227 Venerupis, 296, 328 Viviparus, 225, 227
W Wadden Sea, 48, 116, 152, 160, 199, 202, 203, 204, 206, 207, 211, 212, 213, 214, 215, 216, 217, 218, 277, 278, 279, 4, 281, 282, 283, 285, 286, 287, 288, 289, 303, 306, 309, 322, 323, 324, 326, 329, 330, 337, 343, 349
359 Water column, 296 Water purification, 149 Water quality, 147 Wedell Sea, 206, 207, 215, 216
X Xenobiotics, 148 Xenodiversity, 236
Z Zebra mussel, 239, 240, 244, 249, 250, 251, 256, 260, 263, 267, 272, 273, 274, 275 Zooplankton, 49, 71, 78, 82, 152, 160, 251, 269, 274