Lecture Notes in Earth Sciences Edited by Somdev Bhattacharji, Gerald M. Friedman, Horst J. Neugebauer and Adolf Seilach...
15 downloads
817 Views
21MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Lecture Notes in Earth Sciences Edited by Somdev Bhattacharji, Gerald M. Friedman, Horst J. Neugebauer and Adolf Seilacher
16 H. Wanner U. Siegenthaler (Eds.)
Long and Short Term Variability of Climate
Springer-Verlag Berlin Heidelberg NewYork London Paris Tokyo
Editors PD Dr. Heinz W a n n e r Universit&t Bern, G e o g r a p h i s c h e s Institut Hallerstrasse 12, C H - 3 0 1 2 Bern, Switzerland PD Dr. Ulrich Siegenthaler Universit&t Bern, Physikalisches Institut Sidlerstrasse 5, C H - 3 0 1 2 Bern, Switzerland
ISBN 3 - 5 4 0 - 1 8 8 4 3 - 6 Springer-Verlag Berlin Heidelberg N e w York ISBN 0 - 3 8 7 - 1 8 8 4 3 - 6 Springer-Verlag N e w York Berlin Heidelberg
Library of Congress Cataloging-in-Publication Data. Long and short term variability of climate / H. Wanner, U. Siegenthater, eds. p. cm.-(Lecture notes in earth sciences; 16) Papers presented at a symposium held in Bern, Oct. 10-11, 1986, organized by the Swiss Commission for Climate and Atmospheric Research. Includes index. ISBN 0-38?-18843-6 (U.S.) 1. Climatic changesCongresses. I. Wanner, Heinz. I1.Siegenthaler, U. (Ulrich), 1941-. II1.Schweizerische Naturforschende Gesellschaft. Schweizerische Kommission fur Klima- und Atmosph~.renforschung. IV. Title: Variability of climate. V. Series. QC981.8.C5L65 1988 551.6-dc lg 88-6542 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only permitted under the provisions of the German Copyright Law of September 9, 1965, in its version of June 24, 1985, and a copyright fee must always be paid. Violations fall under the prosecution act of the German Copyright Law. © Springer-Verlag Berlin Heidelberg 1988 Printed in Germany Printing and binding: Druckhaus Beltz, Hemsbach/Bergstr. 2132/3140-543210
Lake
of
walking
Constance, on t h e
winter
frozen
lake.
1830:
The
people
of
Rorschaeh
enjoy
PREFACE
This
volume
held
at
includes
Bern
organized
on
by
organized held
by
this in
Birkh~user
to
a
series.
The
their
The
symposium
them
together
form,
was
of
Atmospheric
planned
the
R. were
possible
one
had been
C.
this to
seine
FrShlich; lectures
appeared
submit like
met
natural
Lecture
our
by H.
Commission
for
C.
of
the of
thanks
Academy by
FrShlich,
A.-C.
involved
preparation
Swiss
symposium
ed.
mainly
Rickli,
administration
The
ready
and
Klima,
a n d we w o u l d
Research),
SchGpbach
express
were
Climate
(Das
it
in
was
second
symposium and
which
Notes
their
papers
to t h a n k
them
collaboration.
president
the
The
reception,
authors
for
Pfister.
1985).
for
first
1983
good
in c a m e r a - r e a d y
the
a symposium
1986,
the
StSrungen,
Verlag,
publish
in
at
11,
was
commission;
und
very
and
Commission It
Bern
Ver~nderungen
presented
10
Swiss
Research.
also
with
October
the
Atmospheric
papers
in
to all
of
providing
G.
the
book.
(then
Climate Furrer
Vogel-Clottu
conference. this
Oeschger
and and
and
organization U. We
Neu
assisted
should
like
December
1987
E. and in to
of them.
Sciences the
made
the
necessary
symposium financial
support.
Bern,
C.
H.
Wanner
U.
Siegenthaler
CONTENTS
Introduction
OBSERVATIONAL
Variability C. F R ~ H L I C H
STUDIES
of the
Solar
"Constant"
Hemispheric and Global Temperature P.D. J O N E S a n d P.M. K E L L Y
Climatic Information in W i d t h a n d D e n s i t y F.H. S C H W E I N G R U B E R
Data 18
of the Past H u n d r e d Y e a r s of C o n i f e r G r o w t h R i n g s 35
V a r i a t i o n s in the S p r i n g - S u m m e r C l i m a t e of C e n tral E u r o p e f r o m the H i g h M i d d l e A g e s to 1850 C. P F I S T E R
57
N o r w e g i a n S e a D e e p W a t e r V a r i a t i o n s over the L a s t C l i m a t i c Cycle: P a l e o - o o e a n o g r a p h i c a l Implications J.C. D U P L E S S Y , L. L A B E Y R I E a n d P.L. B L A N C
83
MODELLING
STUDIES
Numerical Models H. G R A S S L
of C l i m a t e
S e n s i t i v i t y of P r e s e n t - D a y Astronomical Forcing C. T R I C O T a n d A. B E R G E R
117
Climate
to
C a u s e s a n d E f f e c t s of CO 2 V a r i a t i o n s the G l a c i a l - I n t e r g l a c i a l C y c l e s U. S I E G E N T H A L E R
Subject
Adresses
Index
132
During 153
172
175
INTRODUCTION
The
awareness
only
the
growing
that
mankind
local but
also
interest
in
which
experimental
techniques,
climatic
patterns has made
answered, related causes
new
The of
also
of
in
the
problems
natural
climate.
have
thereby
the
of
research and
novel
time,
questions
recognized.
on
climate have been
One
is a b o u t the
not
information
same
some
change
against
are
have
all
question
the n a t u r e
background
from
up
deal
reviews
more
of
the
the
knowledge
character
studies
to
with
of
the
obseryational
variations
going
been
volume
papers
The
year-to-year
At
improved
wealth
While
climatic
this
others
contribution.
past.
modify
to a s t r o n g l y
Strengthened
of a
and
led
and
which
be v i e w e d .
to
Some
topic,
influence
use
yielded
variations,
contributions
current
made
much progress.
changes must
to
climate has
research.
have
to a n t h r o p o g e n i o
man-made
able
climate
activities,
modelling
is
the g l o b a l
variabilitY
of
cover
an
the
range
glacial-interglacial
instrumental
data
to
to
results
a
original from
contrast, from
proxy
records.
The
question
has
long
available
whether
been when
atmosphere.
data
of
are
obtained
unequivocal the
year.
by
with it
be
the
with or
since
been
of
it
shows
to
of
short
radiometers
t r e n d of a
outside the
time
became
"constant"
as
detect the
decrease
and The
satellite
well
as
the
solar of
the
operating
By u s i n g
radiometers
possible
with
only
with
solar
absolute
a long-term
1980
deals
spacecraft.
high-precision
varies data
undertaken
FrShlich of
output
reliable
could
C.
rockets
has
evidence
period
energy
but
c a r r i e d out
balloons,
measurements
For
paper
variability
measurements on b o a r d
sun's
measurements
The
longer-term
the
considered,
spot first
constant.
-0.019
%
per
P.D.
Jones
and
temperatures temperature activity, El
probable well
as
the
of
of t h e
ENSO
0.1
at s o m e
time
eruption
or
for
A
effect
within
between
30
critical
and
exert
regional paid
a
the
pares -
the
The
is
plained gruber
by
reduction cates in
that
years
the
Pfister
data
records
the
end
of
the
a
next
of
1270 of the
warm
the
the
the v o l c a n i c responsible in
the
and
of
one
Switzerland 1945
are
than being
The a u t h o r
for
more
than
temperature maps
and
he com-
of t e m p e r a t u r e width
factor
anomabe
He
growth
also
governed
by of
ex-
Schwein-
of c l e a r
1954.
(July
Europe.
cannot alone.
a phase and
for
density
irregularities for
is
trees.
patterns
strongly
field
site
formation
attention
maximum
climatic
strong
large
ring
tree-ring
in
The
den-
of c o l o u r e d
anomaly
between
of that
shows
like
anDmalies
and
patterns
indi-
deficits the
last
dendroelimatological
re-
years.
investigates
between
harvest til
open
as
it o c c u r s
problems
persisting
a set
width
observed
the
and are
He
great
signals
temperature
that
and
individual
changes
using of
ring
growth
well
within
period
By
influence be
and
climatic
precipitation.
may
search
C.
the can
summer
growth
between
demonstrates
the
duration
variability
on growth
reason,
density
but
of
The
after
Schweingruber.
sites
pattern
obvious,
possibilities
F.
that
of
maximum
coincidence
lies
For
abrupt
spatial
on
effects,
factors
interannual
influence
anomalies.
September),
two
the most
eruptions
6 months
immediately The
the
the by
incorporate
precipitation
similar of
the
temperature
temperature.
order
years
of
stronger
that
years
have
and
are
volcanic
the solar
records.
into
selection
emphasizes
the
event.
%
given
conditions.
to
three
cold 50
insight is
both
the t w o
or
mean
in
activity
phenomenon
global
for
variations
explosive
hemispheric
is of
temperature
droclimatology factors
0.2 °C, on
warm
hemispheric
in
hemispheric
causes
volcanic
(ENS0)
Large
phenomenon
to
maximum
scale.
that
dioxide,
variations
time
annual
possible
find
Oscillation
of
year
mean
discuss
They
carbon
Southern
causes
100
calculated
and
variations.
/
I to
Kelly 1861
increasing
Ni~o
order
P.M.
since
and this
the warm
"Little period
weather
1425
of
patterns
and
compares
period
with
Ice the
Age". High
of
the
vegetation
tree-ring
corresponding
Although Middle
and
data
continuous Ages
are
grape un-
proxy
not
yet
available, than
once
became the
he
every
very
High Middle and
was
when
of
long-term
average
the
in the
few
The
past
Blanc
is
which
much
the
180/160
changes
ratio
lysis
had
there, today
during
the
of
a
is
Europe!
two de-
began
compared of
the
at the
with the
has
impact
of
1420
the
strong
question
Europe
human
One
certainly
view
Central
for
of the
about
Europe
In
they
m a r k e d b y the
was
centuries,
by
1400
climate
only
month
more
measured
form
ari-
witnessed
or
just
mean
re-
135,000 years
They
now
Norwegian
not
have a
as
is
for
sea to
with
water
careful
isotope
ana-
the
the
the of
of
disen-
fact
freezing
significantly
function
be u s e f u l
conti-
temperature
using
near
low
ratios
able by
with
the
between
Sea,
been
table a
by
been
and temperature
present
that will
have
temperature
water
to
isotope
were modified
The
d u r i n g the
water
ocean
oxygen
about
studies.
lower of
fractionation
the
therefore
was
a n d P.L.
cycles,
sediment
f r o m the
authors
from
ocean
Labeyrie
amounts
the
of ice v o l u m e
time.
last
in
however,
deep-water
ratio
L.
water
large
transferred
isotope
cores
ocean
in o c e a n
because
French
can
tope
Duplessy,
glacial-interglacial
recorded
the
The
and
glacial
is
variability.
Central
by
J.C.
the
which,
effects
that
of
been
sediment
point
by
180/160
sediments,
the
and
induced
learned
is
affecting
of
took
millenium
excursions
present,
This
carbonate.
After
the w a r m
enormous
last
earlier
were
been
at
ice.
deep-sea
tangle
years
history
ratio
nental
an
1339.
occurred
fluctuations.
has
than
for
the p a p e r
isotope
ice age
and
of
by
the
Western
weather
natural
subject
oxygen
for
from
Ice Age"
advanced
reported here
and
watershed"
shift
in W e s t e r n
was
anomalies
occurrence
"climatic The
within
and
1269
single
"Little
harvest
ses w h e t h e r
present
A
to the
August
anomalies
a
characterized
the w i n e
end
not
century.
years
positive
between
and
century.
Age
warmest
that
decade
fourteenth
cades
show
rare,
seventeenth
early
the
can
lower
oxygen time
for
isothe
stratigraphic
studies.
The
part
three
problems mate. points
of
papers.
He
and
the H.
book Grassl
some
distincts
to the
dealing
with
discusses
important
results
meteorological
three principal
climate
basic of and
sources
modelling
includes
applications,
numerical climate
of e r r o r s ,
models
the
main
of
cli-
applications which
and
are n u m e -
rical
and
parameterization
incomplete
equations.
climate
model
climate
system,
not
available,
ceeds
the
to
because
agree global
that
much
more
The
difficult
to are
the
is
now
C.
sorbed mate.
and
U.
A.
the
still
have
While
will
lead
4.5
K,
modify
re-
modelto it
processes
particles
ex-
efforts
and
a
the
a is
(e.g.
the
cli-
for
air
the
that ice
explain
why
glaciations to
climate
alone.
The
chan~es
in
ocean
events.
glacial
ice
the
ages,
the
ac-
earth's
variation
been
radiation
relevant
absorbed years
of
the
considered
the
more
the
of
glacial-interglacial
instead is
for
and
so ab-
cli-
radiation discuss
as o n e
is of
of
which, to
cores.
C02
CO 2
may
for the
the
C02
however, appears
consider
interactive
the
about
model
in
both
climate
and
contri-
result
that a
re-
this would
hemispheres,
Milankovitch
have
shown
percent
largely
importance;
necessary
complex
30
Climate model
been
variations may
measure-
have
significantly The
synchronous from
was
have
special
of
They
Holocene.
level
temperatures.
were
it
implication
ice
d u r i n g the
understand
Thus,
the
of
Hemisphere
cause
cycles
simultaneously
the
has
200,000
polar
lower
age
Southern
hard
in
than
the
CO 2 v a r i a t i o n s
the
the
which
concentration
the
is
of
geometry
Generally,
discusses
of
that
of
past
occluded
cold
the
the
computed
the
ice age
indicate the
theory
contribution.
sult
the
1.5
chemical
consider
have
in
climatic
that of
system will
content
by
is s t i l l
a model
climate
atmosphere
surface,
cooling
fact
how
cause
Berger
Berger
atmospheric
to
of
Sie~enthaler on
during
studies
and
latitudes
made the
buted
model
atmosphere.
CO 2
in
accepted.
top
earth's
in t h e i r
Finally,
lower
the
aerosol
variations
widely
the
different
that
aware
Strong
between
Milankovitch,
the
the
Tricot
Tricot
ments
and
of s u c h
the
the
estimate
fundamental
at
at
results
or
which
insolation far.
of
increase
reactions)
astronomical,
cycles,
question how
to
well
caused
compartments
resources.
of
errors
are
relevant
complexity
doubling
as
ocean-atmosphere
composition
a
well
its c o m p o n e n t s .
cording orbit
the
temperature
photochemical and
coupled
the
changing
lers
all
computing
to a n s w e r
the
mean
mate
testad
as
modellers
include
available
been made act
Although
should a
errors
must
been for the
a
t h e o r y of have
been
i n i t i a t e d by unterstanding carbon
cycle
climate-CO 2 system.
Obviously,
the
survey.
think,
We
articles strate
dealing
very
faced with. some
papers
with
well We h o p e
important
in
this
however,
the
a
book
that
do
not
represent
constitute
number
of t o p i c a l
issues
climate
research
that
this volume
contributions
of
term variability
Bern,
1987
U.
may provide
current
a
complete
a collection
problems
long and short
December
they
and
an
research
is
thus
illu-
presently
insight in
Europe
of c l i m a t e .
Siegenthaler
of
H. W a n n e r
into on
VARIABILITY OF THE SOLAR "CONSTANT"
C.Frohlich Physikalisch-Meteorologisches O b s e r v a t o r i u m World Radiation Center CH-7260 D a v o s Doff, Switzerland
1.
Introduction
Since the first clear evidence of c h a n g e s in the solar "constant" S0 from the records of the Active Cavity R a d i o m e t e r for Irradiance Monitoring (ACRIM, Willson, 1979) on the Solar M a x i m u m Mission (SHM) a n d of the Hickey-Frieden r a d i o m e t e r (Hickey et al, 1980) on NIMBUS 7 proving t h a t the s u n is indeed a "variable" star, the interest on solar i r r a d i a n c e v a r i a b i l i t y on all t i m e s c a l e s h a s v e r y m u c h i n c r e a s e d (Willson, 1984: Frohlich, 1987). A t m o s p h e r i c physicists and climatologists are concerned, b e c a u s e of p o s s i b l e e f f e c t s on t h e e a r t h ' s e n e r g y balance. Solar physicists, on the other hand, became interested, b e c a u s e global c h a n g e s of the solar o u t p u t h a v e been d o u b t e d for a long time and their reality obviously leads to some revision of the u n d e r s t a n d i n g of the b e h a v i o u r of the s u n .
2. Solar Irradiance
Measurements
The solar "constant" is the solar irradiance at I a s t r o n o m i c a l unit (I A.U.= m e a n s u n - e a r t h distance) i n t e g r a t e d over the whole spectrum. I n s t r u m e n t s for the a c c u r a t e m e a s u r e m e n t of this q u a n t i t y are the so-called a b s o l u t e r a d i o m e t e r s (e.g. Kendall et al., 1970: Geist, 1972: Willson, 1979; B r u s a et al., 1986) which are also used as reference i n s t r u m e n t s for the calibration of operational r a d i o m e t e r s in meteorological networks. They are all b a s e d on the m e a s u r e m e n t of a heat flux t h r o u g h an electrically calibrated heat flux t r a n s d u c e r . The radiation is a b s o r b e d in a cavity which e n s u r e s a high a b s o r p t i v i t y (typically >99.95~.) over the spectral r a n g e of interest for solar r a d i o m e t r y (200 n m - 10 ~m). The heat flux t r a n s d u c e r consists of a t h e r m a l i m p e d a n c e a n d of t h e r m o m e t e r s (e.g. thermopile, resistors) to sense the t e m p e r a t u r e difference a c r o s s it. Heat developed in the cavity is c o n d u c t e d to the heat sink of the i n s t r u m e n t a n d the resulting t e m p e r a t u r e dif-
ference across the thermal impedance is sensed. The sensitivity of the heat flux t r a n s d u c e r is calibrated by shading the cavity and measuring the t e m p e r a t u r e difference while dissipating a known a m o u n t of electrical power in a heater element which is mounted inside the cavity. In the so-called active mode of operation an electronic circuit maintains the t e m p e r a t u r e signal constant by accordingly controlling the power fed to the cavity heater - independent of the mode, that is whether the cavity is shaded or irradiated. The substituted radiative power is then equal to the difference in electrical power as m e a s u r e d d u r i n g the s h a d e d a n d i r r a d i a t e d periods respectively. In the ideal case of a perfect substitution of radiative by electrical power, the irradiance S would simply be: S = (P,-P~)/A w h e r e P. a n d P, is the electrical p o w e r d i s s i p a t e d with the cavity shaded and irradiated respectively, and A is the area of the detector. However, there are many deviations from this ideal behaviour and the 1/A term will have to be replaced by a more elaborate expression accounting for these effects. The process of experimentally determining the size of t h e s e effects is called e x p e r i m e n t a l c h a r a c t e r i z a t i o n (Brusa et al., 1986). The uncertainty of the characterization determines the absolute accuracy of the radiometer which is of the order of -+0.2~ for present state-of-the-art solar radiometry.
,
03
E" C
O
o~
O
E 0
q~ o .%.-'--"
o
-
C O
0
CO
cw~l z
d! ~
z:
67 68 69 70~71
72 73 74 75 76 77 78 79 80 81 82 83 84
F i g u r e 1: Measured values of total solar irradiance 1967 to 1983 (for the labels see text). The full curve labeled with crosses represent the result of the satellite m e a s u r e m e n t s (one data point every month) for 1978 to 1980 from NIMBUS-7 (Hickey et al, 1982) and for 1980 to 1985 from SHH/ACRIH (Willson et al, 1986). For the discussion of the trends see section 4.
The solar r a d i a t i o n is depleted in the earth's a t m o s p h e r e by absorption and scattering, which depends strongly on the wavelength. Thus accurate determinations of So can only be made from high altitude balloons (above 35 km), rockets or spacecrafts. Determinations of $o from mountain tops were performed by the Smithsonian Institution under the leadership of Abbot (e.g. Abbot, 1942) continuing the pioneering work of Langley. Although sophisticated methods were applied to correct for the atmospheric extinction the results only marginally revealed the small solar " c o n s t a n t " v a r i a t i o n s (e.g. F o u k a l et al., 1977, Hoyt, 1979). Direct m e a s u r e m e n t s f r o m balloons, r o c k e t s a n d s p a c e c r a f t s s t a r t e d in the late sixties and have been continued to present with a g a p b e t w e e n 1971 a n d 76. The r e s u l t s are s h o w n in Fig.l: t h e Soviet balloon flights KN~, KN2 and KN3 (Kondratyev & Nikolsky, 1970, 1979), the X-15 rocket airplane flight DRW ( D r u m m o n d et al, 1968), radiometry on the Mariner Vl and VII spacecraft PLA ( P l a m o n d o n , 1969), the balloon flights of the Denver University group MUI to MU4 (Murcray et al, 1969: Kosters & Murcray, 1981), the balloon flight WIL of Willson (1973), the NASA calibration rocket flights ABI, AB2, and AB3 (e.g. Willson, 1981), the PHOD/WRC balloon flights WRI, WR2, and WR3 (Brusa, 1983) and the spacecraft m e a s u r e m e n t s on NIMBUS 7 (Hickey et al, 1982) and on SMM (Willson, 1984). These data are supplemented by the results from two rocket flights with PMO and ACR instruments: WR4 & 5 and AB4 & 5, and u p d a t e d data from SMM (Willson et al, 1986). This s u m m a r y d e m o n s t r a t e s the i m p r o v e m e n t s achieved in absolute radiometry especially since 1980. The s c a t t e r b e t w e e n the individual r e s u l t s in the late sixties is mostly instrumental, whereas the variability after 1980 is mostly of solar origin.
N
z
'`
s
Eo •--
s
a~
~z
i'` la
ts
2~
22
z',
UT-hours, Day 269 1984
c~
=-
r'l "'I 'T'I 2
~
6
8
1B
UT-hours,
12 Day
~4 189
i[ ~G
18
=-
2@
22
24
1980
Figure 2: SHM/ACRIM individual solar irradiance m e a s u r e m e n t s (every 131 s) d u r i n g one d a y in 1980 (lower panel) a n d 1984 (upper panel). The periodically missing data are due to the modulation by the spacecraft orbit around the earth with a period of 94 to 96 minutes.
3. Variability
of the Solar
"Constant"
The solar irradiance variability on time scales from a few minutes to several months is illustrated by the time series shown in Fig.2 to 4 for two different periods d u r i n g the solar activity cycle: 1980 (lower panels) around the maximum and ~984 (upper panels) close to the minimum of the solar cycle 21. Fig.2 shows the variability during one
c~
260
c~
262
26~
266
268
Eo
27~
272
274
276
278
280
Day 1984
A
~
N
N
i
180
a82
184
le6
188
i90
192
19~
19s
198
200
Day 1980
Figure
3:
days
SMM/ACRIM
in
J.980
solar
(lower
irradiance
panel)
and
data
1984
(orbital
(upper
means)
during
20
during
180
panel).
¢~
148
160
180
200
E
220
240
260
280
300
320
188
208
Z28
248
Day 1984
G8
8~
188
128
148
~68
Day 1980
Figure 4: days
SMM/ACRIM solar irradiance data (orbital means) in 1980 (lower panel) and 1984 (upper panel).
10
day. These short-term variances have a mean peak-to-peak (p-p) amplitude of about 200 ppm (parts per million) and are very similar during b o t h periods. For the 20 d a y s period s h o w n in Fig.3, however, the b e h a v i o u r in 1980 is quite d i f f e r e n t from the one in 1984 with p-p variations of 0.06~. and 0.03~. respectively. Also the main periods of the variability are quite different. This is even more pronounced for the period of 180 d a y s shown in Fig.4: the 1984 variance remains at the same level whereas the 1980 variance reaches several tenths of a percent. The short-term variations of Fig.2 are mainly due to solar press u r e oscillations (e.g. W o o d a r d , 1984, Frdhlich et al., 1984) a n d p a r t l y due to g r a n u l a t i o n . S o m e of the variability of Fig.3 m a y be caused by internal gravity oscillations with periods from several hours to d a y s (e.g. Fr~Jhlich et al., 1984: Frdhlich, 1986). The following discussion will mainly concentrate on the variations shown as time series in Fig.4 and on the trends indicated in Fig.1. The variability of the solar irradiance on time scales of days has been discussed by several a u t h o r s (e.g. Willson et al, 1981: Hickey et al, 1982) and several models have been established for the explanation of the variance mainly by sunspot blocking and facular e n h a n c e m e n t (e.g. H u d s o n et al, 1982: Schatten et al, 1982: Hoyt & Eddy, 1983~ Foukal & Lean, 1986: Pap, 1986). Host of these models are tested against the records of ACRIH/SHH and of H-F/NIMBUS-7. One issue in this context is the question whether the energy blocked by the s u n s p o t s is immediately balanced by the emissions in faculae (e.g. C h a p m a n , 1984) or whether the blocked energy has to be stored below the active regions and emerges only slowly over periods of months or years (e.g. F oukal et al., 1983). Even if the energy were exactly balanced, the irradiance at 1 A.U. would still vary because of the different spatial distribution on the solar surface and the different angular emission pattern of the two features. Recent results indicate that the facular contribution to So is at least comparable to that of spots, when integrated over m o n t h s (Foukal & Lean, 1986). This issue is very important for our u n d e r s t a n d ing of the behaviour of active regions and for adequately modelling the solar i r r a d i a n c e m o d u l a t i o n which in t u r n is needed to u n d e r s t a n d climate changes forced by solar variability. Fig. 5 shows the power spectra of ACRIM/SMM data in the frequency range up to 10 ~Hz (11.6 ~Hz corresponds to a period of I day) for 1980 and 1984. These m e a s u r e m e n t periods are before failure and after repair of the accurate pointing system of the SHH spacecraft and cover 9 and 8 m o n t h s respectively. The difference in the spectra is mainly due to the difference of the activity level of the sun during these periods. The two major peaks at low frequencies with periods of 51.4 and 23.5 d a y s are reduced by more than a factor of ten to a broad peak centered around a 17-days period in 1984 (half power points at periods of 46.3 and 10.6 days respectively). The period of 51.4 days is also found in the o c c u r r e n c e of high e n e r g y flares (Rieger et al, 1984), in the Zi/rich sunspot n u m b e r and in solar diameter data (Delache et al, 1985). Although the power spectrum o£ the projected sunspot area in 1980 shows a significant peak at 27 days, the peak in irradiance is shifted to
11
23.5 days. Cross-spectral analysis of the two spectra also reveals a very weak coherence between irradiance and sunspot area at 27 days (FrShlich, 1984, Foukal & Lean, 1986). Furthermore, the phase between the signals from sunspots and irradiance at 27 days indicates that it is more likely an enhancement which could be due to faculae than a depletion by spots. Obviously, the differences in spatial distribution of spot and faculae on the solar surface and their different evolution in time make that the individual contributions to the total irradiance signal can no longer be distinguished. The depletion of the irradiance due to sunspot blocking seems also to depend on the age o£ the spot and not only on its projected area: young and active spots have a stronger influence than old and passive spots and indeed a frequency analysis of the evolution of young and active spots in 1980 shows the same period of days as the ACRIM/SHH irradiance (Pap, 1986). Other significant peaks in the spectra are found at 7.0, 4.8, 3.4 and 1.3 days. The 4.8 and 1.3 days periods are found in the spectra of both years. In the 1984 spectrum also m a n y significant peaks between 5 and 9 ~Hz similar to the 1.3 days peak are found, the origin of which is still unknown.
23.5
4
~
8
7.0d
SMM/ACRIM 1980
i t~
Day 49-325
N ~ 3.4d
SMM/ACRIM 1984
Day123-366
7.0d
i
[2°
4.8d
,.io i
.
2
4 6 FREQUENCT ~HICROHERTZJ
8
.
.
.
~B FREQUENCT f~CROHERTZ)
Figure 5: Comparison of power spectra of ACRIH irradiance data during 1980 (277 days, left panel) and 1984 (244 days, right panel). The label R3 refers to Delache et al., 1985, and is a period found in the occurence of flares (Rieger et al., 1984). Note the lack o£ a 27-days peak present in power spectrum of the 1980 sunspot data.
Table 1 summarizes the distribution of the variance in the power spectra of 1980 and 1984. Host of the variance is concentrated in the range below about 2 ~Hz (more than 97% in 1980 and 92% in 1984) and it is also here where the biggest change in variance by nearly a factor of 7 (2.6 in amplitude) from 1980 to 1984 occurs. In the range from 2 to 5.8 ~Hz the amount is less than I% of the total variance and also the c h a n g e is m u c h smaller (factor of 2.8 in v a r i a n c e a n d 1,7 in amplitude). Above 5.8 wHz the variance is for both years very small relative
12
to the total and of the s a m e m a g n i t u d e for both years. In s u m m a r y the solar activity influences the variance of the total irradiance significantly, especially at low frequencies.
Table I: Variance of Solar Irradiance to 80 ~Hz in 1980 and 1984. Range Frequency MHz
Period days
for the frequency
Variance ppm 2 1980
1984
-
110
177000
27200
5.6
-
110
2.0
-
5.6
10
1.2
-
2.0
-
40
0.29
-
1.2
-
80
0.14
-
0.3
172000 1480 276 791 535
24900 518 236 770 635
0.1
-
80
0.14
0.1
-
2.1
2.1
-
5.8
5.8
-
10 40
range
from
0.1
Standard Deviation ppm 1980 1984 421
164
416
158
38.5
22.8
16.6
15.4
28.1
27.7
23.1
25.2
4. Long-term Trends
Trends purportedly found in the early m e a s u r e m e n t s of S0 by the Smithsonian Institution were generally doubted on the basis of the l a r g e a t m o s p h e r i c c o r r e c t i o n s involved. D e t e r m i n a t i o n s , m a d e occasionally from aircraft, balloons, X-15 rocket aircraft, and mariner satellites in the late 1960's seemed also too uncertain in both calibration and intrinsic error to allow comment on real variations in So during that period. The modern satellite data together with spot measur e m e n t s from sounding rockets and balloons, however, allow for the first time to assess confidently possible trends in So. Critical reviews of m e a s u r e m e n t s of So made after 1967 have been given e l s e w h e r e (FrShlich, 1977: Fr~hlich & E d d y , 1984" Fr~hlich, 1987). Host of the earlier v a l u e s h a v e been a d j u s t e d from original published values to conform to a common standard, the World Radiometric R e f e r e n c e (WRR). This w a s done in the m a n n e r d e s c r i b e d earlier by Fr6hlich (1977). In addition, to i n s u r e u n i f o r m i t y the a t m o s p h e r i c correction for all balloon m e a s u r e m e n t s was recomputed using the scheme adopted in the reduction of the PHOD/WRC results (Brusa, 1983). The results of the 1980 experiment of the University of Denver (HU4) can be directly compared with the results of the rocket experiments AB2 and AB3 and the balloon fights WR1 and WR2 using the NIMBUS 7 record for interpolation. Thus an absolute value can be attributed to MU4 independent of atmospheric transmission correction. As MU1 in 1969 and HU4 in 1980 were carried out at the s a m e altitude and with the same instrument, the calibration for MU4 can be transferred to HUl making use of the difference of 0.38 per cent between the two determinations reported
13
by K o s t e r s and H u r c r a y (1981). The result is labeled HUT in Fig.1. The close a g r e e m e n t between HUI, HU2 and HUT d e m o n s t r a t e s the stability of the D e n v e r i n s t r u m e n t a t i o n and s u p p o r t s the u p w a r d trend. The linear regression analysis to the spot m e a s u r e m e n t s before 1981 s h o w n in Fig. 1 s u g g e s t s an increase of the solar c o n s t a n t until 1980 at a rate of 0.029 per cent per year. This t r e n d is significantly different from zero at the 99.9 per cent confidence level. It is of the s a m e sign as the c h a n g e of 0.38 per cent between 1969 a n d 1980 noted by K o s t e r s and H u r c r a y , 1981, a l t h o u g h the slope is only a b o u t threeq u a r t e r s as great, Higher-order analysis gives an i m p r o v e d fit to the composite data, shown as the curved line in Fig. I and indicating a m a x i m u m a r o u n d 1979. One m u s t bear in mind, however, t h a t most of the d a t a t a k e n in the early p a r t of the set were the results of inherently l e s s - r e l i a b l e balloon m e a s u r e m e n t s w h i c h could be i n f l u e n c e d by a c o m m o n s y s t e m a t i c o v e r e s t i m a t i o n of the s t r a t o s p h e r i c t r a n s m i t t a n c e . In this case, one would have to a s s u m e either an a n o m a l o u s (high) concentration of s t r a t o s p h e r i c ozone a b o u t 1.5 times the climatological value - or an increased opacity due to an e n h a n c e d a b u n d a n c e of highaltitude aerosol. The latter might e n s u e from a major volcanic eruption, a l t h o u g h there was none r e p o r t e d in this period. All this s e e m s
1370
980 .....
; .....
198t ; .....
; ......
t982 ; .....
; .....
t983 ; .....
; .....
t984 ;,,:,,;
.....
1985 ;
t368
0 C
"0 m
1366 0 IROCKET/ACR SMM/ACRIM I ROCKET/PMOD 8ALLOON/PHOD
0 • [] •
t364
500 Days
iO00 since
Jan.
1.
1980
Figure 6: Time series of S H H / A C R I H daily m e a n results for the period from 1980 to 1985. The linear least s q u a r e fit s h o w n has a slope of -0,019~ per year. I n d e p e n d e n t total irradiance o b s e r v a t i o n s by s o u n d i n g rocket and balloon e x p e r i m e n t s show good a g r e e m e n t with ACRIH results (from Willson et al., 1986).
14
unlikely (see also Eosters & Murcray, 1981). Furthermore, the results from Mariner and the X-15 should be exempt from atmospheric effects and they support the lower values of the early balloon measurements. Thus it is concluded that the low values of So from the late 1960's are most probably real. For the period since 1980 the ACRIM data have been used to determine the trend as shown in Fig.6. A full discussion of this result is given by Willson et al., 1986. The linear fit for this period is calculated from the daily means of the ACRIM data and yields a trend of 0.019 per cent per year. This trend is confirmed by the NIMBUS 7 d a t a and the spot m e a s u r e m e n t s during this period and is the first clear evidence of a long-term trend of the solar constant. The extant m e a s u r e m e n t s of So from 1967 to 1985 suggest a slow oscillation in absolute value which could be part of a 22-year modulation with a peak-to-peak amplitude of about 0.4 per cent coincident with the magnetic cycle of the sun. Due to the missing data between 1971 and 1976, however, it is not clear whether the trend between 1969 and 1980 was continuous or had a dip during the minimum. Even if the l a t t e r would be the case t h e lower d a t a in 1969 could still be explained by the fact that the activity maximum in 1969 was only about two thirds of the s t r e n g t h of the one in 1980.
5.
Conclusions
The p o w e r of the i r r a d i a n c e variability s p e c t r u m from 100 nHz (110 days) to 80 ~Hz (3.5 hours) can be divided into major domains with the following characteristics:
about three
-From 100 nHz to 2 I/Hz (5.8 to 110 days) the spectrum is dominated by solar activity the power of which changes during the course of the solar cycle by up to one order of magnitude. Moreover, the spectrum is characterized by prominent peaks at periods of 51,4, 23.5, 7.0 a n d 4.8 days, The v a r i a n c e in this r a n g e a m o u n t s to 172000 and 24400 ppm 2 for 1980 and 1984 respectively.
- From 2 to 15 ~ z (18.5 hours to 5.8 days) the spectrum follows a I/~ 2 law, which may be partly due to internal gravity modes. The v a r i a n c e in this r a n g e is 1915 a n d 908 p p m 2 for 1980 a n d 1984 respectively. -From 15 ~Hz to 80 ~Hz (18.5 to 3.5 hours) the spectrum follows a I/~ law, which m a y be partly due to instrumental noise. The variance in this range is 1200 p p m 2 for both years. As to the long-term changes, trends of the order of 0.02 per cent per year do exist. The question whether the up and down trends with a
15
peak around 1980 belong to an o s c i l l a t o r y m o d u l a t i o n of t h e solar o u t p u t with a period of 11 or 22 years can only be a n s w e r e d in the future. The decrease of So from high to low activity could be due to the s a m e m e c h a n i s m in the sun which produced the "little ice age" in the 17th century in Europe, when the solar activity was very low over a period of m a n y solar cycles (e.g. E d d y , 1977). The question of the 11 and/or 22 years modulation of the solar o u t p u t is also very i m p o r t a n t in the c o n t e x t of the r e s u l t s of a n a l y s i s of a n c i e n t v a r v e s (e.g. Williams & Sonett, 1985: Sonett & Trebisky, 1986) which indicate also a modulation of the climate with a major period of 11 years and a minor one of 22 years. M o r e o v e r the r a t i o of the 11/22 y e a r s m o d u l a t i o n amplitude seems to decrease with time. As the a n s w e r is not only important for the interpretation of changes of the earth climate but also for the u n d e r s t a n d i n g of the sun itself, monitoring of the solar "constant" has to be continued.
,4cknowladgemen~% I t h a n k R.C.Willson, Jet Propulsion Laboratory, Pasadena, U.S.A,, for providing unpublished ACRIM d a t a and for m a n y helpful discussions. A c k n o w l e d g e m e n t s are extended to the Swiss National Science Foundation for their continuous s u p p o r t of this work at PHOD/WRC.
REFERENCES
Abbot, C. 1942: Revised Results of Solar C o n s t a n t Observing 1923 to 1939, /Inn. Sm.z'thBon, Ast.rophy~%Obsez'v., 6, 83, Brusa, R.W. 1983: Solar Radiometry, Dissartation ETH No, 7181, Zurich. Brusa, R.W. & FrShlich, C. 1986: Absolute R a d i o m e t e r s (PMO6) and their Experimental Characterization, 2ppl, Opt,, 25, 4173. Chapman, G.A. 1984: On t h e E n e r g y B a l a n c e of Solar A c t i v e R e g i o n s ,
Nature,
308,
252.
Delache P., Laclare, F. & Sadsaoud, H. 1985: Long Period Oscillations in Solar Diameter M e a s u r e m e n t s , Nature, 317, 416. Drummond, A.J., Hickey, J.R., Scholes, W.J. & Laue, E.G. 1968: New Value of the Solar C o n s t a n t of Radiation, Nature, 218, 259. Eddy, J.A. 1977: Climate and the C h a n g i n g Sun, Clim, Change, I, 173. Foukal, P., Mack, P.E. & Vernazza, J.E. 1977: Effect of S u n s p o t s and Faculae on the Solar C o n s t a n t , 2Btroph, J., 215, 952. Foukal, P., Fowler, L.A. & L i v s h i t s , H. 1983: A Thermal Model of As~roph, J,, 267, 863. S u n s p o t Influence on Solar Luminosity, Foukal, P. & Lean, J. 1986: The Influence of F a c u l a e on Total Solar Irradiance and Luminosity, Astroph, J., 302, 826. FrShlich, C. 1977: C o n t e m p o r a r y M e a s u r e m e n t s of the Solar C ° n s t a n t " in "The Solar O u t p u t a n d Its Variation, " ed. O.R.White, C o l o r a d o Associated University Press, Boulder, p.93. FrShlich, C. 1984: Solar V a r i a b i l i t y for P e r i o d s of D a y s to M o n t h s , Adv, Space Res,, 4, No.8, 117, Frohlich, C. & Delache, P. 1984: Solar G r a v i t y M o d e s f r o m A C R I H / S M M Irradiance Data, in "Solar SsismoYogy from Space'; ed. R.K.Ulrich,
16
JPL Publ.84-84, Pasadena, CA., 173. Frohlich, C. & Eddy, J.A. 1984: Observed Relation between Solar Luminosity and Radius, 2dv,~qpace Res., 4, No.8, 121. FrShlich, C. 1986: Solar Gravity Modes from ACRIM/SMM Irradiance Data, 2 d v a n c e s in HeYio a n d 2stroseismoYogy, ZAU S y m p o ~ i u m _/23, Aarhus. FrShlich, C. 1887" Variability of the Solar "Constant" on Time Scales J.Geophys.Re~% 92, D1, 796, of Minutes to Years, Geist, J. 1972: Fundamental Principles of Absolute Radiometry and the N a ~ A B u r , S~and, U.~%% Philosophy of this NBS P r o g r a m (1968-1971), Tech, Note, 5941. Hickey, J.R., Pellegrino, P., Mashhoff, R.H., House, F. & Vonder Hear, T.H. 1980: Initial Solar I r r a d i a n c e D e t e r m i n a t i o n from NIMBUS 7 Cavity Radiometer Measurements, Science, 208, 281. Hickey, J.R., Alton, B.M., Griffin, F.J., Jacobowitz, B., Pellegrino, P. & Smith, E.A, 1982: Observations of the Solar Constant and its V a r i a t i o n s E m p h a s i s on NIMBUS 7 Results, in "Proc, I ~ M A P Red, Comm. ,?rd Scientif-l'c 2ssembly, Ha/~burg l.q~¢l"; NCAR, Boulder. Hoyt, D.V. & Eddy, J.A. 1983: Solar I r r a d i a n c e Modulation by Active Regions from 1969 through 1981, Geoph, Res, L e ~ e r s , 10, 509. Hoyt, D.V. 1979: The Smithonian Astrophysical Observatory Solar Constant Program, Rev, Geophys.Space Phy~%, 17, 427. H u d s o n , H.S., Silva, S., W o o d a r d , M. & Willson, R.C. 1982: The Effect of Sunspots on Solar Irradiance, ~qolar Phys,, 76, 211. Kendall, J.M. & Berdahl, C.M. 1970: Two Blackbody Radiometers of High 2fpl, Opt,, 9, 1082. Accuracy, Kondratyev, K.Y. & Nikolsky, G.A, 1970: Solar Radiation and Solar ActiQuart, J,Roy, Meteor.Soc,, 96, 509. vity, Kondratyev, K.Y. & Nikolsky, G.A. 1979: The Stratospheric Mechanism of Solar and Anthropogenic Influences on Climate, in "qolar Terres t r i a l Influences on W e a t h e r a n d Climate", ed. B . M . M c C o r m a c & T.A.Seliga, Reidel , Dordrecht, Holland, p.317. Kosters, J.J. & Murcray, D.G. 1981: Change in the Solar Constant between 1968 and 1978, in "VarJat1"ons of- the Solar Co/~stan~", ed. S.Sofia, NASA Report CP-2191. Murcray, D.G., Kyle, T.G., Kosters, J,J, & Gast, P.R. 1969: The MeasTelurements of the Solar Constant from High Altitude Balloons, .lus, XXI, 620. Pap, J. 1986: Variation of the Solar Constant during the Solar Cycle,
2 s ~ r o p h y s . S p a c e Sci.,
127,
55.
Plamondon, J.A. 1969: The Mariner Mars 1969 Temperature Control Flux Monitor, JPL Space Science P r o g r a m S u m m a r y 3, 162. Rieger, E., Share, G.H., Forrest, D.J., K a n b a c h , G., Reppin, C. & Chupp, E.L. 1984: A 154-day Periodicity in the Occurrence of Hard Solar Flares?, Nature, 312, 623. S c h a t t e n , K.H., Miller, N., Sofia, S. & Oster, L. 1982: Solar Irradiance Modulation by Active Regions from 1969 through 1980, Geoph. Res. Letters, 9, 49. Sonett, C.P. & Trebisky, T.J. 1986: Secular Change in Solar Activity derived from Ancient Varves and the Sunspot Index, Nature, 322, 615.
17
Williams, G.E. & Sonett, C.P. 1985: Solar S i g n a t u r e in S e d i m e n t a r y Cycles from the late Precambrian Elatina Formation, Australia, Nature, 318, 523. Willson, R.C. 1973: New R a d i o m e t r i c Techniques and Solar C o n s t a n t Measurements, Solar Energy, 14, 203. Willson, R,C. 1979: Active Cavity R a d i o m e t e r Type IV, 2ppY, Opt,, 18, 179. Willson, R.C. 1981: Solar Total Irradiance O b s e r v a t i o n s by Active Cavity Radiometer, Solar Physics, 74, 217. Willson, R.C. 1984: M e a s u r e m e n t s of Solar Total Irradiance and its Variability, Space Science Rev,, 38, 203. Willson, R.C., Gulkis, S., Janssen, M., Hudson, H.S. & Chapman, G.A. 1981: Observations of Solar Irradiance Variability, Science, 211, 700. Willson, R.C., Hudson, H.S., Frohlich, C. & Brusa, R.W. 1986: Observation of a L o n g - t e r m D o w n w a r d Trend in Total Solar Irradiance, ~cience, 234, 1114, Woodard, M. 1984: Short-Period Oscillations in the Total Solar Irradiance, Ph.D. Thesis, Un.,'v. CaY2f.. at San Diego, La Jolla, CA.
CAUSES
OF I N T E R A N N U A L
VARIATIONS
P.D.
Jones
Climatic School
O V E R THE PERIOD SINCE 1861
and
P.M.
Research
Norwich
of E a s t
NR4
United
7TJ
and Global
Understanding ture
of the
variations
developments. and
Sciences
Anglia
Kingdom
Hemispheric
back
Kelly
Unit
of E n v i r o n m e n t a l
University
GLOBAL TEMPERATURE
in the
has
and
station
past
record
improved
the
into data
on
Second,
data
markedly
improving
global
or
it
record data
is
spatial
areas
been been
found
have
tempera-
two
recent
extended
previously
have
when
air by
has
were
based
rejected ocean
surface
significantly
where
which
from
the
of
land-based
areas
and corrected
factory.
Data
been
First,
time
homogeneity
Temperature
to
been
sparse
tested
for
be u n s a t i s -
incorporated,
representativeness
of
the g l o b a l
record.
As an
far
overall
new
the
the
National for that
changes
1986a). the
Some
in
series
applied
are
1985;
the
stations been
Library have
in
been
station
a series
et al.,
1985,
Bracknell,
such
on
and
and
technical
UoK.
The pos-
fluctu-
as
station et
others
details the
exthose
where
(Jones
Complete
analyses
of this
contain
correction
1986b).
All
through
assessed,
so
has b e e n
particularly
factors and
of
used.
records
problems.
homogeneity
there
unearthed
archives,
required
of t h e s e in
concerned,
non-climatic
records
available Jones
Many
from
because used,
et al.,
of
instrumentation,
station
stations
is
has
Meteorological
result
omitted
number
data
homogeneity.
ations
record
of m e t e o r o l o g i c a l
temperature
moves, to b e
in the
temperature
searches
individual sible,
land-based
increase
station
haustive of
as
al., had
of all
corrections
reports
(Bradley
19
Data
from
1951-70
usable
reference
latitude-longitude tion
network.
grid
Figure
for
the
the
last
Northern century.
includes
d a t a for
alone
perature. the
of
historical
million
covers
Marine
the
ease
at
ships
have
the
has
problem of
In
is a
cooling
many
or
located the
in
warmer
than
the e a r l i e r
in
data
base
It is 1940
generally and
intake
(Barnett,
1984).
it
tem-
land.
For
taken
by
compilation
as
COADS
(Com-
et
al.,
1985;
of
the
marine
ob-
approximately
temperature
land
the
the
in the
the
(SST).
bucket
It
that
of
was
The
water
and
to be
common
of
the m a r i n e of o b s e r v known
SST
from
the
use water
the
the of
a
for
latter
0.3 a n d 0.TD observations
observation
measurement
more
SST
very
observa-
well
using between
For m o s t each
of
supplying
Readings
In
fabric
most
to
ho-
land-based
the m e t h o d
pipes
method.
their
in w h i c h
speed
measuring
sea
shown
bucket
by
to
1987).
the
homogeneity
of
known how
measurement
are,
way
about
intake
subject Kelly,
for
recorded.
engines.
is not
and
size,
method
been
data,
than
the
never
have
assumed
in
a bucket
ship's
technique
the h e m i -
is
types
are
problems
information
in in
measurement
the
the
affecting
was
change
all
Goodess
overcome
and
cases,
lost
the
to
the
1984;
to
taken
thermometer
thermometer
like
data,
occurred
known
surface
in i n s t r u m e n t a t i o n ,
are
all
been
are,
(Barnett,
marine
globe
(Slutz
which
sea
of
complete
is
Set)
set c o n t a i n s
of
of
only
hemispheric
observations
most
data
areas of
the
use
The
of
middle curve
1979.
difficult
sea
database.
use
1854 to
Changes
tions
ing
numerous
observations
more
series.
data
problems of
nature,
This
the
Hemisphere
land
of
to
Data
most
since
sta-
estimates
1957.
area
instrumental
observations
the y e a r s
mogeneity
the
the
the
a regular
temperature
picture
necessary
Atmosphere
the
irregular
mean
f r o m the
opportunity'.
marine
1986).
servations, 63.25
of
is
overcome
Southern
true
from
onto
since
a
30% of
it
departures
Hemispheres
the
give
about
Ocean
Woodruff,
that
observations
'ships
prehensive
Southern
not
to
as
interpolated
annual
the A n t a r c t i c
areas,
so-called
were
in o r d e r
Note
may
0nly
marine
expressed
I shows
and
Meteorological spheres
stations,
p e r i o d mean,
w a s made.
prevailed after
before
that
time
20
1850 1,0 I
I
1870 I
I
1890 i
I
1910 I
I
1930 I
.... I
1950 I
I
I
1990 I....
I
I
I
I
I
I
I
I
I
I
I
I
I
1970
I
10
1.0
0.0
I
Fig.
I:
I
1
;
1878
ese
l
1898
I
l
1910
i
i
t930
I
l
1950
I
I
-1.o
19~
1990
Annual temperature estimates from land-based d a t a for the Northern and Southern Hemispheres. Data are exp r e s s e d as a n o m a l i e s ( d e g r e e s C e l s i u s ) f r o m the 19511970 r e f e r e n c e p e r i o d (see J o n e s et al., 1 9 8 6 a , c ) .
1850 1.0
1870
1890
1910
1930
I
I
l
I
I
I ........ I
I
I
I
|
l
i
I
1950
" I'"
l
1970
I
I
1990
I
8.0
Northern Hemisphere (degree~ C) -1.0
!
!
!
r
I..... i
1,0
|
0,0
Southern Hemisphere (degrees C)
I 1850
Fig.
I 1878
I
I 1890
I
I 1910
I
I 1930
I
I 1950
I
I 1970
I
-1.0 1990
2: A n n u a l temperature estimates f r o m SST o b s e r v a t i o n s for the Northern and Southern Hemispheres. Data are expressed as a n o m a l i e s (degrees Celsius) f r o m the 19501979 r e f e r e n o e period (see Jones et al., 1986d). Upd a t e s o f the S S T s e r i e s for the f i n a l six years have been made from adjusted Climate Analysis Center a n a l y s e s ( R e y n o l d s a n d G e m m i l l , 1984).
21
Large-scale on
board
Hemispheric compared
the
have
and
with
Intuitively, be
averages
ships
ces
between
The
consistency
Once
time
the
between
of
Figure
case
2 shows
are
similar
~
0.85).
since
of
all
SST
the
data
between
the
southern
tip
Combining forward.
Figure
series
areas.
land
3
and
shows
the
MAT
or
SST
Northern
equally while
for
1.5 t i m e s
land
ent
of
sistent
by
these
effects
the
land
and
features,
in
and
ocean.
is 1986).
ocean
Agreement even better
the N o r t h e r n MAT
186]-1979
for
N H and
Hemisphere
the
Arctic
representative 45°8.
and
variations
There
is
Ocean, of
the
are p r a c t i -
and Antarctica
data
may
the
is
land and
be
except
relatively
near
ocean
(SAT)
used land
Southern
to
ocean
Hemisphere account
Figures
I,
2
the
and
for
3
warming
model levels
the
the
ocean are
ocean
is
the d i f f e r many
con-
between
trend
the
exhibited
predictions of
for
hemisphe-
portions
show
similarities
The
with
atmospheric
series
represent
and
to
consistent
straight-
The h o m o g e n i s e d
in o r d e r
Hemispheres.
increasing
( W i g l e y et al.,
for
except
and
45°8
particular,
Southern
series of
equator
Hemisphere,
weighted
Northern
be
Northern
only
Hemispheres.
weighted
area
over the
is
combined
and Southern for
For
approach. comparison
of S o u t h A m e r i c a .
the
the N o r t h e r n ric
available
(r 2
oceans
curve
oceans between
series.
estimates
series.
hemispheric
for
northern
cally
no
anomalies The
SaT
southern
MAT
this
should
should
differen-
two
similar
SST
been data.
comparison.
curve
Hemisphere
the
the
a
MAT estimates
have
systematic
justifies
corrected,
1854.
the
COADS
(1986d).
land-based
correct
the
al.
two data sets
between
correct
and
to t h o s e
of
Southern
to
SST
Whilst
representative the
been
on
any
to
hemispheres
has
annual mean
Hemisphere
very
used
(MAT) m e a s u r e d et
reliable
Thus,
of the M A T / l a n d
Southern
SH:
be
based
differences
used
hemispheric
in the
may
Jones
f r o m the
1985).
between
be
by
more
estimates
the
series
may
on the
et al.
series
and
MAT
technique
than
the
temperature
temperatures
based
(Wigley
air
homogenised
regional
those
hemispheric
same
through
of m a r i n e
been
greenhouse
of
the
gases
22
1850 l.O
t
1878 I
I
1990 t'
I
1910 I'
I
I
I
l
I
|
I
J
1930 I '"
I
1950 I
1970 I
....
1990 I
0.0
-1.0
Northern I
HemL~here (degrees C) 1 I i I (
1.0
0.0
Southern Hemisphere (degrees C)
i
l
1850
Fig.
3:
forcing
made
(KSppen,1873), the
scales.
the
two
have
activity
first
1
I
191El
'J
1930
with
the
potential
considered
to
global
mean
Wigley
et
for
commonly
example,
E1
I
t
t
195~
-I.B
I.......
1970
the
1985,
effects
ENSO phenomenon,
for of
global
most on
solar
K8ppen,
!99e
I
of
on h e m i s p h e r i c
to
mechanisms
These
(ENSO)
Here,
factors,
temperature.
forcing volcanic to-
phenomenon
dioxide,
of
year
time-
factors,
carbon
review).
these
or
and
were
to ex-
longer
activity
causes 100
proposed and
Oscillation
probable the
temperature
been
1914).
increasing
a recent two
have
considered in
of
mean
year-to-year
Ni~o/Southern
effects be
on
variations
temperature
al.,
the
most
of
mechanisms
variations
been
(see,
estimates
causal
observed
The
factors
compare
I
factors
since
and
J
I$~'0
1878
Ever
gether
I
Annual temperature estimates from a combination of land-based data and SST observations for the Northern and Southern Hemispheres. Data are expressed as anomalies (degrees Celsius) f r o m the 1950-1979 reference period (see Jones et al., 1986d).
Possible
plain
i
variations
time
scale
we
consider
volcanoes
and
are in (see and the
23
Volcanic
effects
Explosive
volcanoes
mosphere.
Once
to
two y e a r s .
formed
into
aerosol ing
Over
in
the
from that
to cool
the
The
the
case
precipitation, immediate
ash
sun
and
the
of
a
vicinity
I: S e l e c t e d
gas
surface
(Lamb,
Volcanic
aerosols
are t r a n s -
as
secondary
scatter
Earth,
by one
or
T h e net
effect
should
the
aerosols
Volcanoes it is affect
for
amount
incom-
of
to
of the v o l c a n o
gases
known
of
are
which
solar
only
inject
readily washed
out by
the
weather
in
Events
Year
Month
1902
5
~ 14 N
61.2W
4
Ksudach
1907
3
51.8N
157.5E
5
Latitude
Longitude
VEI*
Novarupta
1912
6
58.3N
155.2W
6
Bezymianni
1956
3
57.1N
160.7E
5
Krakatau
1883
8
6.1S
I05.4E
6
Tarawera
1886
6
38.2S
176.5E
5
Azul
1932
4
35.78
70.8W
5
Agung
1963
3
8.3S
115.5E
4
Two
Soufriere) 4.
The
used
in
eruption this
scured by was
the
included
See K e l l y
Index
volcanic occurred
eruptions during
later
that
because eruptions of
(1984)
its for
the
1902.
year
earlier because
in
May
analysis
a n d Sear
the
a f e w days.
Pelee/Soufriere
NOTE:
two
resident
Event
*Volcanic Explosivity
at-
for up
the
where
likely
upper
reduce
volcanic
1970).
the
can remain
sulphur
eruption.
the
into it
process
sulphate
large
only
a
thus
troposphere,
are
and
months, in
and
surface whilst
i n t o the
ash
the s t r a t o s p h e r e
initial
reaches
the
inject
aerosols
stratosphere
material
Table
the
formation.
percent
in
can
reaches
sulphate
energy
radiation
be
ash
in
the
effects
Dust
further
(VEI
are
year. Veil
(Pelee
eruptions
Guatemala
in t h a t
high
Caribbean
Both
=
were 5)
Index
details.
=
w a s not
l i k e l y to The Agung
and VEI
be
ob-
eruption
(Lamb,
1970).
24
In
order
to
is n e c e s s a r y Various of
determine
to a s s e s s
workers
have
historical
Lamb,
1970;
tion tainty
and
of
Nevertheless, tions
most
careful
it
is
synthesis eight
were
likely
tion
drew
(1981) 1
1981).
possible
of
published
heavily
on
used
lists
had
Kelly
and
is the
uncer-
availabe. erup-
basis Sear
effects. of
of
of
which
The
Lamb
the
a
(1984) selec-
Simkin
from
date
of erup-
1881-1980
catalogue
the
(e.g.
with
and
periode
information
volcanoes
basis
on
historical
On
climatic
geological
supplementary
these
the
significant
the
the
material,
the
so
that
climate.
during
on
fraught
it
eruptions.
assessment
is
identify
climate,
and
information
affected
volcanoes
catalogues
evidence,
effect
to
have
on
volcanic
Inevitably,
limited
to
to h a v e
and
eruption
climate
the
likely
identified
Table
et al.
influence
of p a s t
geological
potential
because
volcanic
compiled
accounts,
Simkin
size
the
the m a g n i t u d e
et
al.
(1970).
major
erup-
tion.
To
examine
the
temperature, 1962;
see
carried
out.
months as
a
also
First,
after
each
dual
The
and
analysed
the N o r t h e r n used
in
results. 1861
and
One-tailed cause
the
500 1980
land
assess
were
have
warming. line.
the
of
used
and
Hemisphere.
been The
ocean
determine used
as
A
indivifea-
Hemisphere We
series
for
approach
was
of
the
events
between
significance
levels.
volcanoes
significance
level.
significance
5%
36
eruptions).
Carlo
chosen
the
common
temperature
randomly
60
b y month,
four
the
was
the
for
(Northern
Monte
statistical
four to
4a
Hemisphere A
Pollak,
temperature
emphasises
Figure
(Southern
combined
to
dashed
4b
in
month
the
mean
1987) for
temperature
averaging
in t h i s w a y
shown
Southern
analyses
tests
surface
horizontal
Figure
and
order
are
expressed,
by
and
et al.,
estimates
prevailing
formed
hemispheric
(Conrad
temperature
the
on
a n d Sear
hemispheric
then
Averaging
analysis
1984,
date were
mean
event, was
results
eruptions) have
the
eruptions
epoch ~ear,
hemispheric the
response
responses.
tures.
and
eruption
from
before
composite
of t h e s e
superposed Kelly
departures
months
effect
level
are is
unlikely plotted
as
to a
25
-38 I
e,5
-2e I
-10
0
40 I '
10
20
30
I
I
I
....
.
~X
I
50
60
I
Ii
~) _~
0.0
.... . ~ . ~
n~
I
"i ~ ,~~ +i
•
-e.5
I
I
Ho~thernHemisphere (degreesC>
I
I
I
S..~.~...S~..~...?..+'Y~.
'
I -3e e,5
-30 I
I -20
I -10
-20 I
-10 I
I
""
°
I
I
0.5
.'. . . . . . . . . . . ~ . .
+'+
Southern Hemisphere (de~ree~ C) 'I I I l I -e.5 le 20 30 48 5~ &O
0 8
18 I
20 I
38 I
48 I
50 I
b)
. . . . .
Northern Hemi~phePe (degree~ C) -0.5
1
I
. _
I
I
I
1
1
1
IL.. .I~'&!._L~_ pnn.~
_ _ _
. . . . . . . . .
I -:so
4:
Figure
The
about
of Northern
temperature 0.3°C cooling
effect
of
is d e l a y e d
Hemisphere the
event.
So~thern Hemisphere (degree~ C> l I I 1 I I -e.s le 28 3e 4e 5~ 6e
o
with to
to
have
a
month
Hemisphere
the
Hemisphere
Hemisphere
extremely within
prior
Southern
appear
Southern
is
occurs
slight
tions
I -1o
maximum
few
signifloantly,
of monthly hemispheric SST estimate). (a) N o r (b) S o u t h e r n Hemisphere
eruptions
rapid
-
the
months
zero
on N o r t h e r n
maximum
of
the
eruption.
can
be
considered
on
southern
effects,
about
effect
0.2°C.
eruptions
by
0.15°C,
about
noise.
Northern
cool some
two
of The The
temperatures
on t e m p e r a t u r e s
southern
Hemi-
response
eruptions
a negligible but
0.°
. . . . . . . . . .
Superposed epoch analysis temperature data (land and thern Hemisphere eruptions; eruptions.
effect
sphere
I -2e
6.5
the
erupin the
Northern
years
after
26
Whether tion
of
these
this
tribution,
or
from
each
hemisphere
that
the
to
greater
rapidly The
than
oceans
the
land
of
the
magnitude
Northern
Southern
of
the
Hemisphere
pollutants Southern
were
less
eruptions
substantial known
to
after
the
have
land
is
to
in
the
were
in
the
northern
affected
the
characteristics
(1987)
has
tures
is
curve
The
that
events
of g l o b a l
in t h e
tropical
warming
Southern
characterised
by
The
Oscillation
tremes global
Walker
of the
Oscillation The the
most
of
Southern
are
the
to
the
Southern
in
and
northern seen
and is
Europa
most
of
the
this
may
response.
on be
zone
seen
summer the
1969;
a
the
are
Bradley tempera-
in
and
the
upper
known
to be
The
Oscillation
Carpenter,
one
extreme
1932;
in
relevant
the of
and
major southern
1957). to
events) Taken as
WMO,
indices.
Pacific,
also
together,
the
in
the the
positive may
of
1985),
shifts
complementary (cold
referred
and
with
Bliss,
measure
central
now
Rasmussen
associated
(Berlage,
exbe
of
these
E1Nifio/Southern
phenomenon.
commonly in
Ocean
departures is
collectively
(ENS0)
difference
are
Circulation.
significance
variations
Pacific
Walker
negative
distribution
so-called
the
and
Hemisphere
were
or
can
reduce of
Phenomenon
(Bjerknes, events
Oscillation
Southern
the
Finally,
effect
This
Oscillation
significance
These
pressure
volcanic
latitudes
equatorial
of
more
4a.
E1Niflo/Southern
1982).
the
seasonally-dependent.
in F i g u r e
E1Niflo
the
noted
spring
or
most
of
due
inertia.
delay
Northern
1970).
also
respond
in is
is
transported
sunsets
(Lamb,
to
thermal
Most
the
eruptions have
been
dis-
ocean
explanation
to
southern to
and
However,
higher
have
Remarkable
eruption
able
quantities.
loca-
Hemisphere
tend
in
land
lower
effect.
the
seasonal
likely
their
of m a t e r i a l
occurred.
Krakatau
Northern
were
large
were
transport
most
the
Hemisphere
likely in
of
The
to
volcanic
eruptions
Hemisphere
Hemisphere
The
owing
from
their
percentages
in
area.
areas
result
eruptions,
certain.
response
ocean
response
of
relative
not
rapid
in
sample
the
is
more
the
differences
particular
used monthly
index mean
of
the sea
Southern level
Oscillation
pressure
is
between
27
Tahiti,
Society
Oscillation
Index
Ropelewski back
to
Darwin this
Islands,
and
1866
(SOI)
Jones
based
I
has
Australia.
recently Here,
been we
Chile,
in p l a c e
in F i g u r e
1880 I
190B I
1
I
of
This
extended
use
on d a t a f r o m D j a k a r t a ,
are p l o t t e d
18~
Darwin,
(1987).
and Santiago,
index
and
a
Southern
to
1882
further
Indonesia, Tahiti.
by
extension in p l a c e
Annual
of
values
of
5.
1920 I ' l
1940 I
I 1929
I 1946
I
1760 l
I
1988 I
8.0
-3,8
Figure
E1
is
extremely America. Peruvian (1987)
I
I 1900
defined
cold water Using
I
Quinn
SST and
relationship
temperature and,
the
is
second,
et
I
I
I 19'68
23
al.
warm
1980.
between examined through
occurrence
coast
variation
identified 1880
by
off the
the
coastal
between
relation sis.
best
have
The mean
I ! 88e
I
I 1 ~8~
5: A n n u a l ( J u l y - J u n e , d a t e d b y the J a n u a r y ) e x t e n d e d S o u t h e r n O s c i l l a t i o n Index.
Ni~o
events
I
of
and
The y e a r s the
ENS0
in two the u s e
and Ecuador catalogue
the
events
SOI,
(El are
listed
first,
of
warm
or
in S o u t h E1
Bradley
Ni~os)
indicators ways:
of the
of e x t r e m e l y
Peru
(1978)
values
and
et 20
in T a b l e
Ni~o al. cold 2.
and hemispheric by
of s u p e r p o s e d
direct epoch
cor-
analy-
28
Table
Warm
2: E N S O W a r m
and Cold Events
(El Niflo) y e a r s :
1884
1888
1891
1896
1899
1902
1904
1911
1913
1918
1923
1925
1930
1932
1939
1951
1953
1957
1963
1965
1969
1972
1976
1886
1889
1892
1898
1903
1906
1908
1916
1920
1924
1928
1931
1938
1942
1949
1954
1964
1970
1973
1975
Cold years:
a) D i r e c t
Correlation
Coefficients SOI
and
perature
The
formed
over
stability plained
ture
two
the
some
should
spring
This
by
have any
SOI.
increased
immediate
when
months.
The
each
global
effect
the
25
and
of
m a y be
should of
an SOI
0.15°C, of
the
accounted
six months
with by
ex-
tempera-
30%
is t h e n
departure
eruption
12
the
was
the
temperatures
temperatures of
check
variance
about
event,
to
was per-
period.
series
+ve/-ve
1982/3
to
leads
Between
up
analysis
more
the SOI
of
the tem-
intervals
The
earlier
Global
for
the e x t e n d e d with
1926-1984,
temperature
value.
0.05°C
particular
and
lag relationship
forecasting
lowered/raised this
six
6.
Slightly
is s t r o n g e s t
between
averages
monthly
d u r i n g the
the h e m i s p h e r i c SOI.
by
in F i g u r e
relationships.
about
computed
1867-1925
relationship
by
of
S0I
shown
periods,
the
the
series
mask
were
temperature
the
are
relationship
for b y
of
lagging
results
of
by
variance
of
determination
series
months.
The
of
12-month hemispheric
E1
be
one u n i t of
N
-3,
enough
to
Chichon
in
1982.
b) Sup,erposed epoch method Superposed volcanic events separate
epoch case.
listed
analysis The
in
analyses.
was
Januarys
Table The
2
applied
in the
of
warm
were
results
the used are
as shown
same (El
the in
key
way
Ni~o)
as
months
Figures
in the
and
7a
in
cold two
(warm)
29
and
7b
using
(cold).
the
priate
number
here,
Two-tailed
Monte
Carlo
of
compared
events. to
significance
technique
the
The
by
levels
randomly
significance
volcanic
case,
were
assessed
selecting levels
the
are
because
appro-
much
there
lower
are
more
events.
0.5 I N ° r t h e r n
#
Hemisphere
I
0
1
2
3
0.5
]Southern
°'°
0
O.5
I
R~
I
6
7 8
9 1'0 1"1 1"2
Hemisphere
~ ~ ~, 5
Globe
0.25
~, g
~
~ ~
§ 1'01'i 1'2
~-
-1925
---
"~"
~"
0.0
0 1 2 3 4 5 6 7' 8 9 10 11 12 L a g ( m o n t h ) of the 1 2 - m o n t h l y t e m p e r a t u r e aeries behind the 1 2 - m o n t h l y SOl Figure
Most
6:
warm
of t h e
year
during
the
cold
Lagged coefficients of determination between the Southern Oscillation Index and hemispheric temperature series (land and SST estimate).
(El with
Niflo) the
following
periods.
temperature
uary
the
of
lags of
of
Figure
when
six months
the
effect
year
of
7
or
year.
one
behind the
on t e m p e r a t u r e
tend here
The
the
the
10
to
results
considers
warm
The
that
some
to
that
SOI which,
or
of t h e
cold warm
commence
showing
July.
shows
occurs
selected 6
used
June
Figure
spheric
events
SOI
event and
the
timing
around greatest
is s i m i l a r
maximum 16
months
are
turn,
cold
some
lags six
events
is
the
hemi-
the with
temperature
by
turn
for on
after
compatible
the in
effect
the
response
Janthose
response
the
January
months.
The
similar:
a
30
warming/cooling January low
key
temperatures
Bradley
of
0.1
date.
et al.,
-8.25
to 0 . 2 ° C
Note
the
during
some
apparent
the
10
to
16
precursor
previous
year
months of
(Figure
after
a warm 7a,
the
event:
see
also
1987).
0
-38
-28
-18
10
20
30
40
5~
60
I
l
I
I
i
I
I
,
~
i
i
I
-30
-20
-10
IO
2~
30
40
50
60
-30
-20
-10
10
20
30
40
50
6~
t
I
I
i
Northern Hemisphere (degrees C) I t I I I I
0.25
So~thern Hemisphere (degrees C)
!
I
O
I
-0.
b) O.BO ................
I
,'},
n.,~
I
I
,
~...~.....~-,.,~-
II. . . . . . . . . . . . .
7:
(desPees C) I I
_rl..f
,-,~thern
Figure
misphere I
I
I
I
l
-3O
-28
-I0
Superposed epoch temperature data (El N i ~ o ) events;
Hemisphere (degrees C)
!
!
!
le
28
3~
,
|
I
48
58
69
analysis of monthly hemispheric ( l a n d a n d SST e s t i m a t e ) . (a) W a r m (b) c o l d events.
31
Conclusions
Large
explosive
volcanic
eruptions
and
the
ENSO
phenomenon
have 0
been on
shown
to
have
hemispheric
lived.
The
months
and
similar
temperature.
duration
mediately
of the
it
occurs
after
the
at
The
are
responsible
annual,
high
frequency
of the
are
effect
time
or
for
order
effects
maximum some
volcano
factors
ture
effects,
warm
or
between
variability
50 in
of the
cold and
the
to 0.2C,
relatively
is
within
of 0.1
the
short-
order
of 6
two
years
im-
event.
These
two
50%
of
the
hemispheric
inter-
tempera-
records.
Ackngwledgements
This ergy
work
was
under
acknowledge
supported contract
by
the
number
the c o m m e n t s
United
States
Department
DE-FG02-85ER60316.
of Ms.
C.M.
Goodess
The
of
En-
authors
on an e a r l i e r
draft
of the m a n u s c r i p t .
References
Barnett,
T.P.,
over
Berlage,
the
H.P.,
1984:
Long
ocean.
Monthly Weather
1957:
Fluctuations
circulation prognostic
Bradley,
R.S.,
Northern Climatic
of
more
value.
1987:
The
trends
than
explosive
in
one
112,
the
general
year;
Verh
temperature
303-312.
atmospheric
their
nature
and
69.
volcanic
continental
(submitted).
surface
Review
in
KNMI Meded.
Hemisphere Change
term
eruption
temperature
record
in
records.
32 Bradley,
R.S.,
Kelly,
Goodess,
C.,
Northern
Bradley,
Carbon
1987:
ENS0
V.
and
Harvard C.
Tech.
Division,
and
the
Dept.
Signal
in
Nature
and
and
for
the
Rep.
No.
Washington,
of Energy,
P.D.,
Raper,
Goodess,
C.,
P.M.,
1985:
data
1984.
Tech.
Dioxide
and
set
for
Rep.
data.
Tech.
Research
A
A
grid
Northern
the
Rep.,
U.S.
B.S.G.,
Kelly,
point
P.M.
surface
Dept.
of
Washington,
and air
Hemisphere,
U.S.
Division,
review for
Cherry,
H.F.,
the
air
Division.
B.D.,
Diaz,
TR022.
Research
marine
implications DoE
Santer, R.8.,
and
Climatology,
Historical
averages.
S.C.B.,
in
temperature
problem
Bradley,
T.M.L., DoE
1987:
surface
J.K.,
temperature
Methods
C a r b o n Dioxide
temperature Carbon
1962:
global
Eischeid,
(submitted).
459 pp.
sea
and
continental
L.D.,
quality of
G.
Press,
Kelly,
data
Wigley,
Kiladis,
records.
Pollak,
calculation
Jones,
H.F.
bank
DoE
Research
H.F.,
University
temperature of
Diaz,
data
1851-1980,
Dioxide
Diaz,
precipitation
Goodess,
P.D.,
climatic
335 pp.
R.S.,
Conrad,
Jones, A
Hemisphere,
TR017, D.C.
P.M., 1985:
1851Energy,
D.C.
251
PP. Jones,
P.D., P.M.
Raper, and
surface
S.C°B.,
Wigley,
air
Bradley,
T.M.L.,
temperatures:
and Applie d Meteorology, Jones,
P.D.,
Raper,
Wigley,
S.C.B.,
T.M.L., data
1984.
Teeh.
Carbon
DoE
Dioxide
1986a~
for
Rep.
Northern
Kelly,
Hemisphere
Journal
A the
B.S.G., grid
of
Climate
Goodess,
point
Southern
TR027.
Research
H.F°,
161-179.
Cherry,
set
Diaz,
1851-1984. 25,
1986b:
temperature
R.S.,
U.S.
Division,
C.
surface
Hemisphere, Dept.
of
Washington,
and air
1851Energy,
D.C.
73
PP. Jones,
P.D.,
Raper,
Hemisphere of Climate
S.C.B.
surface
and air
and A p p l i e d
Wigley,
T.M.L.,
temperatures: Meteorology,
25,
1986c:
1851-1984. 1213-1230.
Southern Journal
33
Jones,
P.D.,
Wigley,
temperature 322, Kelly,
P.M.
and
W.,
variations
Sear,
1873:
schaft
Gber
W.,
1914:
H.H.,
Global
1984.
Nature,
266A,
Volcanic and
W.H.,
Zopf,
D.0.,
Historical
Oscillation, Bulletin, E.M.
fields
Reynolds,
the
31,
und
Vulkan-
305-328.
atmosphere,
with
meteorological
Royal
Society
sea
and
T.H., the
of
a
signiLondon,
Yang,
R.T.W.,
the
Southern
of
droughts.
surface
P.M.,
Fisheries
P.D., Index.
eruptions.
surface
Southern
23,
Oscillation/El
An
objective analysis.
global Tropical
4-5.
1987: Monthly
and
surface
Nature,
in wind
354-384.
1984:
P.D.
global
Variations
and
temperature
Jones, of
1982:
110,
W.H.,
Newsletter, Jones,
response
volcanic
Kuo
temperature
Review,
Gemmill,
Oscillation
Kelly,
the
and
statistics
Indonesian
with
Weather
and
C.F.
Southern press).
and
Carpenter,
associated
mean
K.S.
and
surface
0cean-Atmosphere
major
Ni~o
and
Monthly
monthly
C.B.,
the its
der Gesell-
663-678.
sea
R.W.
Ropelewski,
in
Short,
trends
E1
76,
tropical Nido.
Periode
Sonnenflecken
of
of
Witterung,
425-533.
1978:
Rasmussen,
Trans.
der
257-267.
Zeitschrift,
dust
explosive
0sterreichisohen
8, 241-248;
assessment
Phil.
of
11-j~hrige der
Meteorologische
impact
740-743. Perioden
Lufttemperaturen,
1970:
ficance.
On
1986d:
and
Climatic 311,
die
Zeitschrift
chronology
Sear,
P.B.,
1861
mehrj~hrige
fGr Meteorologie,
ausbrGche.
Quinn,
1984:
Nature,
~ber
insbesondere
Lamb,
Wright,
between
C.B.,
eruptions.
Temperatur.
KSppen,
and
430-434.
volcanic KSppen,
T.M.L.
An
extension
Weather
Goodess, air
of
the
Review
(in
C.M.,
1987:
temperatures
(submitted).
to
34
Simkin,
T.
et
Ross, Slutz,
al.,
R.J.,
Lubker,
Jenne, J.D.,
Walker,
R.L.,
S.J.,
I
Boulder,
Colorado.
G.T.
NOAA,
and
the
In:
B.
(Eds.).
1982-84.
U.S. pp.
and world
change
S.D.,
Boulder, Memorandum
J.
SCOPE
J~ger,
Set
Colorado,
of
Analysis
MacCracken
P.M.,
scenarios
of
and
F.M.
increasin~
Carbon
pp.
D85s
Data
the
Dioxide
Empirical detection
active
and
R.A.
Climatic
gases. Warriok
Change
and
271-322.
1985: of
1986:
and
radiatively
B.R.
the
The climate
Global
Climate
system
during
Programme.
Proceedings
January
ERL ESG-23,
1985:
Energy
Effect,
review
Editor,
V Memoirs
53-84.
effect
of
Organization, Climate
M.C.
by
29, Wiley,
critical
1986:
Data
Laboratories,
P.D.,
Kelly,
induced
Greenhouse
World
Elms,
and
55-90.
P.D.
A
Jones,
Dept.
warm
The
P.M.
Weather
IV (36),
the
Jones,
Bolin,
S.D.,
Atmosphere
World
In:
studies:
Meteorological System:
and
record.
Division,
E o0systems.
Woodruff,
Hutchinson
Woodruff,
Research
Society
Detecting
C02~
climatic
Steurer, Ocean
1932:
J.K.
(Eds.),
T.M.L., Climate
E.W.,
temperature
Research
of
World.
J.D.,
Environmental
Bliss,
atmospheric
Wigley,
the
268 pp.
Angell,
Luther
D.H.,
Meteorological
T.M.L., of
of
Hiscox,
Comprehensive
Release
Wigley,
Volcanoes
Joseph,
1985:
the Royal
World
1981:
Stroudsberg.
22-24,
218 pp.
of
a
COADS
workshop,
1986.
NOAA
Technical
CLIMATIC INFORMATION FOR THE PAST HUNDRED YEARS IN WIDTH AND DENSITY OF CONIFER GROWTH RINGS
F.H.
Schweingruber
Swiss Federal
Institute of F o r e s t r y Research
CH-8903 Birmensdorf
i. Introduction
Growth rings of trees growing in areas with a seasonal climate c o n t a i n climatological subject, made,
information.
e.g. Fritts
There is an extensive literature on this
(1976). However,
most d e n d r o c l i m a t o l o g i c a l
although many studies have b e e n
investigations
in the n o n - a r i d zones
of the earth h a v e met with little success and r e c o n s t r u c t i o n s on a y e a r - b y - y e a r basis h a v e seldom proved possible. All r e c o n s t r u c t i o n s
so
far a t t e m p t e d h a v e been formulated in terms of one or another moving average over a defined period,
e.g. a decade,
and h a v e almost exclus-
ively been based on measurements of ring width. r e c e n t l y given rise to new approaches,
Those limitations h a v e
which are d i s c u s s e d below.
- Greater a t t e n t i o n is b e i n g paid to the selection of sites and individual trees,
since it has been shown that site factors exert a
stronger influence on growth ring formation than regional c l i m a t i c conditions,
e s p e c i a l l y in areas with a temperate climate.
- To allow due c o n s i d e r a t i o n of the great v a r i a t i o n in growth ring anatomy,
the l o n g - e s t a b l l s h e d p r o c e d u r e of dating through pointer
years has been b r o u g h t into r e l a t i o n s h i p to c l i m a t i c and e c o l o g i c a l factors, with proper a t t e n t i o n to abrupt changes r a d i o d e n s i t o m e t r i c methods h a v e b e e n expanded;
in ring width;
tissue analysis has
36
been refined;
and isotope research related to growth rings and cli-
mate has been undertaken. Figures
la und lb show the morphological
ria in dendroclimatological
features used as crite-
research.
The examples given below illustrate how these new methods can greatly extend the present knowledge on dendroclimatology and that growth rings,
as sources of proxy data, can supply information on
climatic conditions over long periods of time and great geographical distances.
a
bl 1980 t974 o_o ~ t961 ~ 1958 b2
1 2 maximum densities
1,0
3
4mm
0
lJ ring widths 0,8 Q
.~ 0,6 0,4
0,2b
J
Fi~. %a Growth ring sequence from a fir displaying pointer years (growth changes within one or two years) and abrupt growth reductions (persisting for several years). Those changes which can be identified and dated by the naked eye are expressions of severe changes in the physiology of the tree. Such changes are often triggered by extreme effects of the climate (summer drought, cold periods, extremely low temperatures in winter, etc.). Fi 9 . ib Parallel diagrams from a larch sequence with pointer years; bl: tomicrograph, b2: corresponding density profile. Five different ameters of such curves are selected for interpretation. Maximum sity and ring width contain by far the greatest climatological mation.
phopardeninfor-
37
2. Temperature reconstructions
Studies by Parker and Hennoch Hughes et al.
from conifers
(1971),
in Europe
Schweingruber et al.
(1978) and
(1984) have shown that the maximum density of different
species of conifer growing on cold and wet sites in the Alps and in Scotland generally contain information on temperature during the summer months from July to September. density chronologies
Response functions
for maximum
from northern Scandinavia provide data on tem-
perature during July and August,
while those from the mountain ranges
of Central and Southern Europe supply information on conditions during July, August and September.
Ring width, on the other hand,
clearly related to weather conditions,
is less
since it is far more strongly
influenced by local site conditions prevailing d u r i n g the relevant vegetation period and the preceding year. Maximum density integrates climatological
information to a greater degree than ring width, be-
cause it is mainly an expression of the cell wall thickness of the latewood cells.
Since these survive for 2 to 3 months,
upper and northern timberlines, growth factors,
even at the
they are able to reflect limiting
in this case temperature.
Ring width,
an expression of the performance of the cambium,
in contrast,
is
whose main activity
is limited to two to four weeks in early summer. These considerations and observations led to the construction of an ecologically uniform sample net. In order to maximize the reflection of temperature
in the growth-ring patterns,
samples were taken
from normally grown trees on the coldest and the wettest sites in both the upper and the northern timberline zones, ologies
i01 local indexed chron-
(2 cores from each of 12 trees per site) were used to con-
struct 22 regional chronologies.
The growth-ring data (maximum den-
sity) were calibrated by comparing maps of annual anomaly patterns of densities or their index values, with maps of the corresponding perature anomalies
(= departures
tem-
form the the long-term average).
The
reference period for maximum density was the overall temporal range of the chronologies,
that for summer temperature the reference period
1881-1980 considered by Jones et al.
(1982).
Visual comparison of the maps revealed similar to very similar patterns
in 4/5 of the cases
(Fig. 2, Appendix).
Discrepancies can be
explained through gaps in the meteorological data and through differences in growth limiting factors in various regions of Europe, instance,
for
the growth period is limited to July and August in northern
38
Scandinavia but extends from July to October in southern Italy. Furthermore,
the possibility cannot be excluded that in years with ex-
treme conditions,
e.g. 1948, growth was not limited by temperature
alone on some sites. Similar growth maps were constructed on the basis of ring width, but it has not yet proved possible to decode the climatological mation they contain in year-by-year
infor-
terms.
This study shows that density analysis permits very reliable reconstruction of temperature patterns sphere
(boreal zone, upper timberline
for the whole Northern Hemiin mountains).
Reconstruction of
precipitation over fairly large areas is equally possible. dition, however,
The con-
is that the samples be uniformly taken the lower
timberline zone, where precipitation is essentially the only growthlimiting factor (Fig. 3; Fritts 1974, Stockton et al. 1981). Networks comprising trees on differing sites are unlikely to provide uniform climatological
signals.
Consequently,
year-by-year
reconstructions
cannot be made on the basis of data from such networks, structions
and recon-
for longer periods are always accompanied by broad,
incalculable deviations
3. Limitations of dendroclimatolo~ical
The influence of growth-limiting
research
factors varies greatly between spe-
cies and between sites, especially on non-extreme sites. study such relationships,
often
(Shiyatov and Mazepa 1986, Fritts 1974).
Lingg
In order to
(1986) investigated the differences
between spruce and fir over the past 80 years along altitudinal transects in the Valais
(Fig. 4). Ecophysiological behaviour differs quite
considerably from species to species.
In firs both maximum density and
ring width in trees growing on low and high altitude sites correlate with each other.
Only in trees of the subalpine zone does maximum
density fail to reflect differences between site or species. probable that too little attention
is being paid to the difference
ecophysiological behaviour between species dendroclimatological
networks.
It seems
in the construction of
in
39
0 C3
-0,5
°°i/la --I
00 -~ F i 1600
I
~
,
I
i
A i
i
1650
~
I
'
1700
i
L
i
f
~
1750
i
i
;
I
i
1800 Year
~
i
i
~
i m--~T'7
1850
1
i
1900
,
i
,
I
1950
Fi 9 . 3 R e c o n s t r u c t e d fluctuations in temperature and p r e c i p i t a t i o n based on new ring width chronologies from the Great Plains, USA. Relationships b e t w e e n ring widths and recorded m e t e o r o l o g i c a l parameters are calib r a t e d for the period 1900-1970. (Circles: m e t e o r o l o g i c a l measurements, lines: reconstructions, smoothed with a low pass filter). After Fritts (1983).
Climatic differences,
varying from year to year,
are strongly
modified by local site factors and affect ring formation accordingly. In a r a d i o d e n s i t o m e t r i c a l
study Kienast 1985 clearly shows the re-
l a t i o n s h i p b e t w e e n different growth ring parameters and relief. The influence of site factors on growth ring formation in the mountainous
areas of the temperate zone is summarized below
- Where the regional climate is cold and moist, altitude sites is severely limited, altitude sites is o p t i m u m
-
(Fig.
5).
growth on moist, h i g h
while growth on shallow-soil
low
(Type A).
Where regional climate is warm and dry the situation is reversed: growth is o p t i m u m on moist, h i g h altitude sites but minimal on shallow-soil low altitude sites
(Type B).
- Where regional climate is cold and dry, growth is o p t i m u m on b o t h dry, h i g h a l t i t u d e sites and moist,
low sites, but limited on moist,
h i g h altitude sites by the low summer temperatures and on shallow, low altitude sites by low p r e c i p i t a t i o n
(Type C).
40
fir
fir/spruce
spruce
E EEEEE
EEEEEE
E EEEEE
uu
~el~e
840mN 1230mSW
~
1740 m N
O ~
®
(~)@~ID
1850mSSE ® O e ~
a
® ®
ice® eee eee eee e•
®iO
•
D~-
w
•
•Oe
!
e•e
classes of gleichl~ufigkeit O 59,2-63,0 p=0,05 ® 6 3 , 1 - 6 7 , 3 p =0,01 6 7 , 4 - 7 1 , 9 p =0,001 7 2 , 0 - 76,0 76,1 - 79,9 • > 80
ring width 840 m N 1230 m SW 1240 m S 1510 m N 1740 m N 1850 m SSE
Fig.
ee®®
I
'o~ o
4
Relationships (Gleichl~ufigkeit) of maximum density and ring width between firs (Abies alba; le~t), spruces (Picea abies; center), and (right) between firs (vertical) and spruce (horizontal), along an altitudinal profile in the Valais, Switzerland for the period 19001980. Firs and spruces are from common sites; cores were taken from 12 firs and 12 spruces at each site. After Lingg (1986). The difference in behaviour between the two species and the ring parameters is evident. Firs display greater similarity over a fairly wide spectrum than spruces. This may be mainly due to the different types of root systems. While the deep root system of fir allows an efficient water supply throughout the year, the shallower, more superficial one of spruce often leads to growth inhibition through low precipitation, p a r t i c u l a r l y on low altitude sites. Fir:
maximum density: ring width
trees on all sites behave similarly. trees on sites above 840 m behave similarly, the low altitude site (840 m) behaves differently.
Spruce:
maximum density:
the trees on the lowest two sites (840 and 1230 m), the two sites at 1230 m and 1240 m, and the three highest sites display relationships among themselves. the trees of the four lowest (840-1510 m) and two highest sites display relationships among themselves.
ring width
Fir/spruce r e l a t i o n s h i p maximum density: ring width
the two lowest sites differ clearly from all the higher ones. the pattern is similar to that of spruce alone.
41
climatic conditions cold-moist
warm-dry
cold-dry
Type A
Type B
Type C
dry--moist
dry~mdist
E
t
g°
i __
cD
dry--moist
Fi~.
5
Idealized dendro-ecological diagram, showing the relationships b e t w e e n different altitudes (ordinates) and site moisture (abscissae) during years with differing weather patterns. Optimum ring growth is represented by black shading, minimum growth by white. After Kienast (1985).
year dendroecologicaI diagram and type maximum density
1910
1911
1912 maximum density or ring width
A
m very high m C
C
high
[]moderately high moderately low
ring width ]low ] weather temperature
very tow
oC 20 o
I II III IV precipitation quarters and months
I II III IV
Temperatureand precipitation abovethe long-term mean (1901-1971) m
Fi@.
Temperatureand precipitation below the long-term mean
6
Temporal series of dendro-ecological diagrams for maximum density and ring width in trees growing in the Valais (Switzerland) and the nearby meteorological station at Sitten. Arrangement of sites (squares) as in Figure 5 (vertical: altitude, horizontal: dry-moist). Growth at high altitude sites was considerably reduced during the cold years 1910 and 1912, while the dry conditions of 1911 impaired growth on the lower sites. After Kienast (1985).
42
Fig. to year
6, however,
shows that the limiting factors vary from year
in their effect on the specific ring p a r a m e t e r
the weather pattern.
That means that it is p o s s i b l e to extract
mation on d i f f e r e n t climatic components ring sequence,
p r o v i d e d that a number of site c h r o n o l o g i e s
e m p l o y e d in d e n d r o c h r o n o l o g y
infor-
from one and the same growth
ered in r e l a t i o n to each other. Unfortunately,
functions)
according to
are consid-
the p r o c e d u r e n o r m a l l y
for the analysis of time series
does not permit the limiting ecological
(response
factors to be deco-
ded on a y e a r - b y - y e a r basis.
% 1885 93 60'
94 191t
30. o,
t2
21
22
42
44
48
49
50
62
65
68
74
76
m
_n l.l_n..n N-m._l.N.m.mn
1400-1500
203
1300-1400
12
12001300
182
,_.,.Inn.n..l_.,
1100-1200
140
mmm,_ln,,,,m..,N_.mll
1000--t100
79
900-1000
66
800-~ 900
112
700°° 800
48
400-700
48
. ,
.llmmmll
n,n,.nl.,n,nl..nm n,.,,.,n Innn nNNN-
30-
0
144
t500-1600 i186
:":'"'":I;:::]
6c
height&sJ. ~max 1600-1900
I.,n,l.l..i I• l n t . t . ! 70%
. n . ,
nno;,
nm.liimn_llm
,,°Z
A l t i t u d i n a l d i s t r i b u t i o n of pointer years in spruce in the Valais, Switzerland, since 1885. The height of the columns represents the p e r c e n t a g e of spruces at p a r t i c u l a r altitudes with pointer years (Nmax.: m a x i m u m number of spruces investigated). It is n o t i c e a b l e that pointer years are only formed on h i g h a l t i t u d e sites during cold years, e.g., 1912. In n o n - e x t r e m e dry years they are m a i n l y formed on sites at lower altitudes, e.g., 1942, 1944. After Kontic et al. (1986).
43
A l t i t u d i n a l r e l a t i o n s h i p s have been d e s c r i b e d by Kontic et al. (1986),
the i n t e r p r e t a t i o n being mainly b a s e d on pointer years,
that
is, growth rings which,
for the m a j o r i t y of trees on a p h y t o s o c i o l o g i -
cally h o m o g e n e o u s
are m a r k e d l y wider or narrower than the pre-
site,
ceding or subsequent ones
(Fig. 7).
In the upper timberline zone,
tree growth is mainly limited b y
low temperature during the vegetation period, 1965,
for example.
as h a p p e n e d
At lower elevations the limiting factor is p r e c i p i -
tation during summer,
witness
1921,
1942,
1944,
1949,
1976 etc. On
medium altitude sites in the temperate zone the limiting greatly,
in 1912 and
factors vary
indeed to such an extent that it is not yet even possible
to
explain the o c c u r r e n c e of pointer years coinciding over wide areas. These three cases show the overriding n e c e s s i t y of extreme care in selecting sites for d e n d r o e c o l o g i c a l
% 70
studies.
spruce, canton of Solothurn n = 480 lOO E
"3
~ 6o
8o 60 ,~
z _c 40-
so20-
1900
ao g
ii
i
10
20
30
40
50
60
70
1980
Fi 9 . 8 S u m m a t i o n diagram for spruces with growth reduction in the canton of Solothurn, Switzerland. The d i a g r a m shows the p e r c e n t a g e of trees whose growth has b e e n reduced since 1910 in c o m p a r i s o n to the p r e c e d ing period (black: over 71%, hatched: 56-70%, white: 40-55% growth reduction). The fluctuations are evident. The phase of growth reduction b e t w e e n 1945 and 1954 is conspicuous.
44
Abrupt growth changes were long regarded in dendrochronology disturbances, disease,
as
since they often reflect individual changes such as
injury to the photosynthetically
active crown, or alteration
in the vertical position of the tree. Recent studies, however, have shown that abrupt growth changes persisting for more than three years incorporate climatic signals. A supra-regional
study of growth patterns over the present cen-
tury in several thousand conifers of different
species in Switzerland
revealed a certain trend towards a periodicity of 11-16 years, seems to be mainly governed by deficits (Figs.
which
in summer precipitation
8, 9). It is quite obvious that some of these growth changes
are due to local pollution,
disease,
or impairment of the site through
soil compaction or sinking of the ground water level. dendrochronological
It is a task for
research to clarify the origin of these irregu-
larities. a) duration of growth reduction
1910 Valais
spruce
1920 t
1930
1940
I
Solothurn Mittelland Aargau Fricktal
I
1980 n I I ] 327 I
I I
197
,,I I
i
,,,
202
t I
500
t
634
F............ ~'i~ ~................ t 4
Valais Solothurn
480
I I
I
Chur
fir
1970
1960
1950
p...~
!
Aergeu
I Chur.910~10 VaEaie c duration of grovvth-b) summagon diagram ._ ,g 14 I reductions V
,[
[
464
I I
620
II
494
I
Pine
,.V.16
Vll
621
rV__ 12
5539
.u / / / / ~ . .
--~g d) precipitation deficits
o
o_ & 3oo
""
"
" ~ _
l!i year
1910
1920
1930
1940
1950
// ',
if !ll,li I 1960
1970
1980
Fig. 9 Duration of growth reduction phases in different conifers (5539 trees) growing in different regions of Switzerland in relation to precipitation deficits in the months May-August as measured at the meteorological stations Rheinfelden, Olten and Aarau. The fluctuations are closely related to periods with low precipitation. After Schweingruber et al. (1986).
45
:~:;::i::i::!iiii:;::iiii~if
t
~
I
J
I
'
[
J
I
i ,--~'--~
~
~
~
~
o
E o~ LU i~ o
~
~
~
o
T
°
46
,,iF k ii/p ii/p ..
47
.......... i~i~i~i~i~i~i~i~ ~i~i~i~i~i~i~i~i~ili~i~i~i~i~ili ........................
48
~l~~ ~~
49
IW
50
51
#
52
f~
53
54
iiIF
55
References
Fritts, H.C., 1974: Relationships of ring widths in arid-site conifers to variations in monthly temperature and precipitation. Ecological Monographies 44, 411-440. Fritts, H.C., 1976: Tree rings and climate. Francisco. Academic Press, 567 pp.
London,
New York,
San
Fritts, H.C., 1983: Tree-ring dating and reconstructed variations in Central Plains climate. Transactions of the Nebrasca Academy of Sciences, XL: 37-41. Hughes, M.K., Schweingruber F.H., Cartwright, D., Kelly, P.M., 1984: July-August temperature at Edinburgh between 1721 and 1975 from tree-ring density and width data. Nature 308, 341-344. Jones, P.D., Wigley, T.M.L., Kelly, P.M., 1982: Variations in surface air temperature: Part i. Northern hemisphere 1881-1980. Monthly Weather Review ii0, 59-70. Kienast, F., 1985: Dendro~kologische Untersuchungen an H~henprofilen aus verschiedenen Klimabereichen. Ph.D. thesis, Univ. ZUrich, 129 pp. Kontic, R., Niederer, M., Nippel, C.A., Winkler-Seifert, A. 1986: Jahrringanalysen an Nadelb~umen zur Darstellung und Interpretation von Waldsch~den (Wallis, Schweiz). Eidgen~ssische Anstalt fur das forstliche Versuchswesen, Berichte 283, 1-46. Lingg, W., 1986: Dendro~kologische Studie an Fichte (Picea abies) und Weisstanne (Abies alba) im subkontinentalen Klimagebiet (Wallis, Schweiz) Eidgen~ssische Anstalt fur das forstliche Versuchswesen. Berichte. 287, 1-81. Parker, M.L., Henoch, W.S.E., 1971: The use of Engelmann Spruce latewood density for dendrochronological purposes. Canadian J. of Forest Res. l, 90-98. Schweingruber, F.H., Fritts, H.C., Br~ker, O.U., Drew, L.G., Schaer, E., 1978: The X-ray technique as applied to dendroclimatology. Tree-Ring Bull. 38, 61-91. Schweingruber, F.H., Albrecht, H., Beck, M., Hessel, J., Joos, K., Keller, D., Kontic, R., Lange, K., Nippel, C., Spinnler, A., Steiner, B., Winkler, A., 1986: Abrupte Zuwachsschwankungen in Jahrringabfolgen als ~kologische Indikatoren. Dendrochronologia, 4, 125-183. Shiyatov, S.G., Mazepa, V.S., 1986: Natural fluctuations of climate in the eastern regions of the USSR based on tree-ring series. Paper presented at the workshop on Regional Resource management. September 1985, Albena, Bulgaria. Collaborative Paper Internat. Inst. for Applied System Analysis. A-2361 Laxenburg, Austria. Vol. I: 47-73.
56
Stockton, Ch.W., Mitchell, L.M., Meko, D.M., 1981: Tree-ring evidence of a relationship between drought occurrence in the western United States and the Hale Sunspot Cycle. In: LAWSON, M.P., BAKER, M.E., (eds.). The Great Plains. Perspectives and Prospects. University of Nebraska Press, Lincoln and London, 83-110.
VARIATIONS IN THE SPRING-SUMMER CLIMATE OF CENTRAL EUROPE FROM THE HIGH MIDDLE AGES TO 1850
C h r i s t i a n Pfister U n i v e r s i t y of Berne D e p a r t m e n t of History Engehaldenstrasse
4
3012 B e r n e / S w i t z e r l a n d
Does
I.
the climate of the High Middle Ages
include elements for
a
w a r m i n g scenario?
Warm periods in the past may provide elements for a s s e s s i n g the
clima-
tic and human c o n s e q u e n c e s of the global w a r m i n g w h i c h is p r e d i c t e d for the
next century,
if the present trend in c o n c e n t r a t i o n of g r e e n h o u s e
gases in the a t m o s p h e r e continues sea
ice around G r e e n l a n d would
(WMO,
1986).
It is assumed that
the
retreat towards its northern coast
in
the early stage of a w a r m i n g period and then c o m p l e t e l y d i s a p p e a r in
a
later stage. The Arctic Ocean w o u l d become ice free while the c o n t i n e n tal
ice-dome at the Antarctic would persist.
during the Late T e r t i a r y for the last time.
Such a situation existed Flohn
(1984:
7, 265) con-
cludes
from
coasts
of the M e d i t e r r a n e a n together with the Alps
Europe
(up to latitudes 48 - 50 N) might obtain a w a r m - t e m p e r a t e
mate
with
the climatic evidence of this period that
some r e d u c t i o n of summer rains,
season droughts, 2
months
from
the
periods Central
i.e.
the
and
northern
south-central
with
frequent
cliwarm
while the v e g e t a t i o n period w o u l d be increased by I -
(Flohn,
1984:
9).
botanical
evidence
over
past
the
Europe
was
summer
temperatures
present
average.
This
On
the
other
available
700'000
from
years
not
mediterranean
may
have
suggests
been a warm
hand
the warm
that at
2
-
Frenzel
the that
3
interglacial vegetation
time
degrees
and moist
concluded
summer
in
although above
the
climate.
What do we know about the w a r m period in the High Middle Ages?
AD
985
Norse colonists from Iceland settled in Greenland around modern Narsaq, J u i i a n e h a a b and G o d t h a a b d i s t r i c t s
(Mc Govern,
1981:
407).
The colo-
58
nists
were
able to bury their dead deep in soil that has
p e r m a n e n t l y frozen. the
the present normal
(Lamb,
1982:
165 f).
coasts of Iceland only on the average for a 1945). ding
farther
Bohemia,
north,
1986),
settlements country;
(Koch,
1982:
170),
reports on grape h a r v e s t s
vines
(Ale-
were grown on a l t i t u d e s of 600 to 700 m in (Scherer,
1874).
wheat 1984:
was
the
In Norway, also,
36);
grown almost to the latitude of the Polar
farm hill
Circle
in the Alps p a s t u r e s could be grazed up to 2800
1976). A c c o r d i n g to
Lamb
a
from
were s p r e a d i n g up to 200 m h i g h e r than before on the
(R~thlisberger,
m
(1984: 37) m i d s u m m e r s during
"Little Optimum" were p r o b a b l y between 0.7 and 1.0 ° C w a r m e r than twentieth-century
Central Europe
A
few weeks per year
the
m e d i e v a l v i n e y a r d s in England are known up to
p r e a l p i n e v a l l e y of T o g g e n b u r g
this
warmer
T h u r i n g i a and Belgium are included in m e d i e v a l sources
xandre,
the
or more,
Drift ice reached
In Central and W e s t e r n Europe c u l t i v a t i o n of the vine was sprea-
latitude of 53 ° N (Lamb,
(Lamb,
been
In the m i l d e s t period in the early twelfth century
w a t e r in the fjords was at least sometimes 4 ° C,
than
since
(Lamb,
average in England and 1.0 - 1.4 ° C 1982:
warmer
170).
d e t a i l e d a n a l y s i s of the c l i m a t e in the Middle Ages m i g h t
allow
us
anomalies
in
to learn more on the seasonal w e a t h e r p a t t e r n s
therefore
and
that might be c o n n e c t e d with the w a r m i n g trend.
on
This
the know-
ledge may be helpful for a s s e s s i n g the economic and societal impacts of a
w a r m i n g in the future.
spring-summer this
In the f o l l o w i n g the w e a t h e r patterns
p e r i o d between 1270 and 1425 will be
data will be c o m p a r e d with the known v a r i a t i o n s
the end of the s o - c a l l e d
in climate
M a n - m a d e data and their l i m i t a t i o n s
For
the 350 years before the c r e a t i o n of the n a t i o n a l w e a t h e r Switzerland
the m o n t h l y patterns of weather and climate
described
and q u a n t i f i e d based upon a body of data
man-made.
It
surements,
until
c o m p r i s e s e x p l i c i t w e a t h e r data
that
service could
are
and proxy-data,
weekly
i.e. a v a r i e t y of infor-
w h i c h r e f l e c t s the c o m b i n e d effect of several
d u r i n g a period of several months
be
mostly
(early i n s t r u m e n t a l mea-
q u a n t i t a t i v e and q u a l i t a t i v e d e s c r i p t i o n s of daily,
and m o n t h l y w e a t h e r patterns) mation
and
"Little Ice Age".
2.
in
in the
investigated,
weather
factors
(e.g. o b s e r v a t i o n s on the freezing of
59
lakes
and the ripening of grapes and measurements of maximum tree-ring
density on logs from the upper timberline). all
types
1985 a) types
of
has of
evidence in the CLIMHIST weather allowed
data,
same has
to
data
to compare and to mutually check
to refine the interpretation and
indices for temperature and precipitation
Prior
The synchronous display of
(Pfister,
to
bank
(Pfister,
the
different
derive
monthly
1984).
the early sixteenth century man-made sources become
at
the
time less abundant and less rich in meteorological entries. two consequences:
creases,
and
This
the time resolution of the reconstruction
de-
the spatial dimension of the analysis must be increased.
The data are scattered within a large area,
which begs the problem
of
interpolation in space and reduces the reliability of the estimates, particular for precipitation. neous
proxy-data
patterns
that
are required for
estimating
the
temperature
of the vegetative period are more difficult to obtain.
sionally
phenological
Ages
order to determine and compare temperature patterns
in
standing years:
in
Also, continuous quantitative and homoge-
Occa-
observations have also been made in the
a friar of the order of St.
Dominic,
Middle in
out-
who was born in
1221 and lived in Basel and in Colmar, has included phenological observations in his Annales Basilienses et Colmarienses. earliest springs of the present millenium, first
rye
March
19th,
ears appeared around January 8th, the vine got leaves on April Ist,
sold on May 17th, date
In 1283, one of the
he wrote for instance:
the
the rye was in bloom the first new rye
the peas could be harvested from June 8th,
strawberries and cherries were ripe (Annales,
1861).
the same But
observations were not systematically carried on for some years, Thus
on was
these such as
those
made in the eighteenth and nineteenth century.
allow
quantifying roughly the thermic character of climatic anomalies.
they
only
Grape
harvest dates are available from the mid fourteenth century when
several chroniclers and annalists began to keep track of the date the
wine
1971:
50).
harvest
was fixed by public proclamation
However
(Le
Roy
Ladurie,
the records are often incomplete for many
Measurements of the maximum density of tree-rings at the upper line
are
Lauenen a
the only continuous evidence for this time.
(Bernese Oberland)
originates
The
in 1269 (Schweingruber,
when
years. timber-
series
of
1978). In
near future it will be extended back to the year 1000 (communication
by Dr. Schweingruber).
60
3.
Guidelines for the spatial extrapolation of data
Given
the
Middle
insufficient density of man-made and natural
Ages,
data
in
the
it is essential to assess which biases might occur
from
extrapolations within large areas. For this reason the spatial patterns of
temperature variation in Europe must be known for the present
tury. to
change
over time with the changing climate.
analysis
of spatial correlations
ning
instrumental
of
large
measurements.
This type of
the
analysis
will
machine
be upon
readable
1985).
the present context the spatial correlations of temperatures in the
vegetative
period
(April to September)
1901-60 and 1851-1900
If
reason
Institute of Berne based
number of long series readily available in
form (US Dept of Energy,
In
For this
should be extended back to the begin-
attempted for Europe at the Geographical a
cen-
Moreover we need to know to what extent those patterns are bound
Zurich
and in summer are provided
for
(table 1).
is chosen as a reference station the covariance of
temper-
ature patterns in the vegetative period is very high (R 2 of 65%) up
to
the shores of the Atlantic over a distance of almost 800 kilometers and still Alps
remarkable (Vienna).
August) ment
is somewhat weaker in most cases.
with
series
across the Alps to Northern Italy and to the Eastern
The covariation between the summer temperatures
of
These results are in
the significant correlations that have been vine harvest dates over distances of 800
between Geneva and Vienna)
(Flohn,
found
agreebetween
kilometers
The use and misuse of historical sources
The
meteorological
evidence contained in the chronicles and annals
the Middle Ages has been included in large compilations.
seeking to reconstruct past climates.
Historians
have drawn on the results of these reconstructions.
of
At first sight
compendia seem to provide a convenient ready-made data bank
it is therefore not surprising that they have been much used by tists
(e.g.
1985: 96).
4.
these
(June-
in their
and
scienturn
61
cO 0
0
c~
o'~
co
0
~
t~
q
-4
~
o
o ch I
cO o o
co
un
~
E
0 r~
N o c~
•
tlh ~0 ko
co
0 0 0
r0
~ 0
0 .,~
03 u5
r~ ¢J
0 o~
~ cO
I'~ ~D
~D
CO
O0
q
<
q
q
<
<
<
<
U'3
CO
P'-
u~
r,q
~
r~
~D
r~
.~1
-~
r~
,-'-t 4J 0
o
O0 O0
Oh ~
O0 ~
If) ~
I~ '~
CO
C~
~-~
~
~-~
~xl
fxl
~
IZ~
0
4a
~ 0
~
4~
oo
oJ
4~
0
~,'~
f9
~4
~4
4J
-
II)
c~
tn
0
O
.,4
-,-t
°°
~-~
0 o~ u~
© _Q E~
~
O
U
.,-4
62
Only
a decade ago it was d i s c o v e r e d
mation
about
that d o c u m e n t a r y sources of infor-
past climate are not equally reliable,
and indeed
m a t e r i a l which purports to record h i s t o r i c a l events is gravely ding. As far as the M i d d l e Ages are concerned, have
been
are
acceptance
of
original
(1978).
inaccurate or u n c e r t a i n d a t i n g of accounts
w h i c h are d i s t o r t i o n s
observations,
mislea-
the current c o m p i l a t i o n s
a n a l y s e d in detail by Bell and Ogilvie
weaknesses
Their
particular
or
misdating.
amplifications
i n c l u s i o n of events for w h i c h there is no
The c o n s e q u e n c e s can be
is
falls
misdated,
far-reaching.
which can easily occur,
into two c a l e n d a r years,
of
the
medieval
(if
at
all)
(Bell,
Ogilvie,
(1890,
1892,
did
1978).
not include 30% of the r e l i a b l e
a v a i l a b l e in print compilations
have
(Ingram et al.,
(Alexandre, been
1981:
1986).
repeatedly
192; Pfister,
To take the well
On the
evidence
demonstrated 1984:40
in
the
for
other
that
Though the w e a k n e s s e s of
was these
last
years
f.) they are still uncri-
tically used as data sources for c l i m a t i c r e c o n s t r u c t i o n w h i c h is inacceptable.
the
between
1897) as an example:
period half of the records are worthless.
Amberg
later
and t h e r e f o r e they contain a m i s h m a s h
known Swiss c o m p i l a t i o n by A m b e r g
hand
season
Most fundamentally,
works do not d i s t i n g u i s h a d e q u a t l y
of v a l u a b l e and w o r t h l e s s data
re-
If for example a cold
this event may be i n c l u d e d in a
reliable and u n r e l i a b l e sources,
of
through
given that this
c o m p i l a t i o n in an a r t i f i c i a l l y m u l t i p l i e d way. majority
main
events,
liable e v i d e n c e w h a t e v e r and spurious m u l t i p l i c a t i o n of events
winter
much
(Burga,
1985),
It is not n e c e s s a r y to comment on the value
of
s o p h i s t i c a t e d s t a t i s t i c s that are based upon data from non c o n t e m p o r a r y sources
(Pavese,
Gregori,
1985).
H i s t o r i a n s have been more cautious in s e l e c t i n g their sources. (1968)
has
drawn
from c h r o n i c l e s
in order to i n v e s t i g a t e
Schmitz
the
links
b e t w e e n m e t e o r o l o g i c a l v a r i a b l e s and the prices for grain and wine from 800 to 1350. standard Baden
Buszello
(1982) has i l l u s t r a t e d
of living of the "common man"
and Alsace and their m e t e o r o l o g i c a l
chronicles.
A
model
compilation 1425
and
to
He
the
(1986) who has brought t o g e t h e r a new critical
Silesia and N o r t h e r n
Italy,
but
to
excluding
has taken care to assess the r e l i a b i l i t y of every source
check every bit of information.
discarded.
the
contemporary
is the recent work of
of c l i m a t i c e v i d e n c e for W e s t e r n and Central Europe up
(including Bohemia,
England).
in
Switzerland,
causes from 35
of critical a w a r e n e s s
B e l g i a n Pierre A l e x a n d r e
the f l u c t u a t i o n s
in late m e d i e v a l
All u n r e l i a b l e
A l e x a n d r e only r e t a i n e d first hand o b s e r v a t i o n s
porary chroniclers.
records
were
from contem-
This is in itself an enormous task given the
fact
63
that
most
medieval sources only contain fragments
information.
As
of
meteorological
far as man-made data are concerned this evidence will
provide the basis for the following reconstruction.
5.
The representativity of tree-ring and grape harvest data
Tree-ring
data from humid Western and Central Europe do not allow very
convincing
climatic
climatic memory
reconstructions,
(Hughes et al.,
1982).
mainly because of
expected from trees at the alpine timberline, the
short
progress wood.
vegetative has
their
long
Representative results can
period controls the
be
where the temperature of growth
rate.
been made through the Roentgen density
Significant
measurements
of
Maximum density of the late wood is the single tree-ring charac-
teristic
most highly related to climatic data.
(Bernese Oberland)
A series from
Lauenen
originating in 1269 has been set up by Schweingruber
(1978, 1979). Because in some years tree-ring density data are the only evidence summer
available, (Pfister,
original
their
1984)
values
covariation with the thermal
had to be determined.
For this
were grouped into seven classes. It
stand out in the record density they
data
(Pfister,
(1616,
for the
turned
densities were very low in most of the very cold summers, of the hottest summers in the last 450 years
indices purpose out
that
whereas some
1719, 1947) do not
1985 c). This suggests that tree-ring
should be used cautiously as climatic indicators
can be cross-checked with man-made observations or grape
unless harvest
dates.
One
of
the
gradually
most representative
been
area-averaged
series of
grape
built up by Le Roy Ladurie and 103
series
of wine harvest
data
harvest
Baulant for
dates
(1980);
has they
eastern/central
France, western Switzerland and a few villages from southwestern Germany.
The
final series originating in 1484 was tested with the Parisian
temperature series for the period 1797-1879. lation
is
phenological period
The coefficient of corre-
.86 which ought to reassure anyone to sources
(Le Roy Ladurie,
Baulant,
the 1980:
reliability 263).
before 1484 wine harvest data for Dijon are contained
work of Angot
(1883) from 1366 ,
For the in
the
who, in his turn, relied upon Lavalle
(1855). The climatologist Jean-Pierre Legrand evidence after 1400,
of
(1979 a, b) has used this
when it is almost complete,
in his careful inves-
64
tigation
on
580 years.
temperature anomalies and sunspot activity over the
last
New data for the fourteenth century have been discovered by
historians who became
sensitive to this type of evidence after reading
Le Roy Ladurie's History of Climate
(1967,
1971).
Rotelli
(1973)
has
included
several series in his work on agrarian history of the Piemon-
te.
longest and the most complete
The
(Moncalieri)
from 1331 to 1424. A series from Beaune to
been
1976) made it possible
Another series from the plain of Albenga
set up for 1364-1796
Anjou in form of a small graph
(Mazzei). (Le Men~,
they were too difficult to check. C~te
d'Or
values
Data for
1982) were discarded,
series was computed from Dijon
and
Beaune;
Dijon
some
(see
missing
Appendix).
Albenga
Dijon
.73 (N=30)** .68 (N=22)*
.59 (N=18)*
Moncalieri
.30 (N=32)*
.38 (N=42)*
Significance:
*~.05 ns
r ~
**~.00
.23 (N=26) ns
N: paired observations
not significant
0.7 the correlation between the two series from the C~te d'
almost
Ladurie, Albenga
because
the covariance between the four series was determined:
Albenga
is
century
For the period after 1371 a regional
Beaune
With
late
(N Italy)
fifteenth
were interpolated using Albenga and Moncalieri
Previously
Or
period
bridge the frequent gaps contained in the Dijon series in the
fourteenth century. has
(Dubois,
covers the
at the same level as in the
Baulant,
1980:
App.
II).
later
centuries
(Le
Roy
Remarkable also is the result of
(across the Alps) while the covariance of Moncalieri with
the
C~te d'Or is weak and not even significant with Albenga.
The
value
critically maximum
of
the C~te d'Or series for
same
out order
reconstruction
In order to test the stability
the series were split into two shorter
periods.
that the correlations between the two series were of magnitude as between the Lauenen series and
averaged series of wine harvest dates from Western Europe 98):
was
assessed in cross correlating the series with the tree-ring
density series of Lauenen.
the correlations, turned
climatic
the
in
of It the
area-
(Flohn, 1985:
65
1370-1399:
-.65
(N= 29)
1400-1499:
-.43
(N= 94)
This
result suggests that the CSte d'Or series can be used as a
valid
climatic indicator.
6.
Outstanding anomalies
6.1. Definition and interpretation
In
table
2 years with strong temperature anomalies in the
vegetative
period are listed from 1270 to 1524 as far as they appear in the
tree-
ring
both.
density record,
in the series of wine harvest dates or in
The two data sets complement each other: reflect Flohn,
temperatures 1985:
95 f.),
July through September),
heat-waves
in June,
(Legrand,
appear 78).
but also
in
April-May.
1979 c:
43),
as occurred e.g.
However of
in 1616 and in 1976, (Schweingruber,
the weather patterns in the Alps may be somewhat
than those in the lowlands. A more complete list of lies
86;
that may precipitately advance the maturation
to have little effect on maximum density Also
1984:
maximum densities those in August and September
(sometimes grapes
grape harvest dates primarily
in the spring-summer period (Pfister,
1978:
different
temperature anoma-
may therefore be derived from a comparison of the two records and
a cross-checking with additional unsystematic phenological observations and weather descriptions.
Years in which the grape harvest began prior
to September 10th or later than October 20th are considered whereas
for the tree-ring densities the limits of warm and cold anoma-
lies are set to 1090 g/cm 3 and to 880 g/cm 3 respectively. 1090 g/cm 3 corresponds August
anomalous,
and September whereas densities below 880 g/cm 3
summers
that
(Pfister,
1985 c: 187).
A value above
in most cases to extremely high temperatures
were colder than the chilliest of the
point to present
in late
century
66 Table 2
Temperature
anomalies
in the warm season 1270-1524
in
Central and Western Europe
year
1270
tree-ring
wine-harvest
density
dates
unsystematic
phenological
observations
first ripe grapes of early
1170(+3)
burgundy July 13 (Alsace) 1273
1180(+3)
1274
989(0)
1282
1028(0)
Nov 18 (Basel) new wine Aug 22
1287
1124(+3)
end around Sept
1300
1138(+3)
(Strasbourg) 22(Ribeauville) 1302
873(-3)
1304
1090(+2)
1315
829(-3)
Oct 23(Limoges) first ripe grapes of early burgundy July Ist (Alsace) Nov 9 (Quimperl~) Nov 19 (Vienna)
1319
1123(+3)
1330
1075(+I)
1331
1130(+3)
NOV 9(Maillezais) beg of Sept(Liege) Aug
1333
1169(+3)
1335
723(-3)
1336
1138(+3)
1345
831(-3)
(Paris) -
cherries ripe at beg of May (Maillezais) slow maturation
of vine
(Paris) Aug(Liege)
very high sugar content (Z~rich) slow maturation (Paris,
1346
858(-3)
of vine
Torino)
vine still in bloom on Aug 2nd (Lindau)
1347
909(-2)
Nov 9(Krems)
vine still in bloom on Sept Ist (Lindau)
1350
822(-3)
-
1359
724(-3)
-
1361
1010(+2)
1366
778(-3)
Sept 9(Constance) Oct 17(Dijon)
slow maturation (Mainz)
of vine
67 slow maturation of vine
1370
870(-3)
1382
1111(+2)
Sep 12(Dijon)
1383
1095(+2)
Aug (Rouen)Sept 5
(Mainz)
(Bordeaux) 1384
1019(0)
Sep 9 (Beaune)
1385
1063(+I)
Sep 8 (Beaune)
1393
1086(+2)
1400
1062(+I)
Sep 9 (Dijon)
1420
1003(0)
Aug 23(Paris)
vine bloom begins on May Ist (St.Galler Rheintal)
Sep 9 (Dijon) Sep 2 (Beaune) Sep 16(Dijon)
new rye on May 15 (Mainz)
Aug 25(Dijon) Aug 29(Beaune) 1422
1006(0)
Aug 28(Dijon)
1436
1028(0)
Oct 27 (Dijon)
Sep 11(Beaune) 1448
-
Oct 21 (Dijon)
1456
841 (-3)
Oct 4 (Dijon)
1465
787(-3)
Oct 12(Dijon)
1473
1129(+3)
Aug 30(Dijon)
1480
890(-2)
Oct 10(Dijon)
1481
952(-I)
1488
1052(+I)
Oct 18(W Europe)
1491
884(-2)
Oct 21(W Europe)
1505
983(0)
Oct 14(W Europe)
1511
958(-I)
Oct 15(W Europe)
early rye harvest(Winterthur)
Oct 19(Lausanne)
Sources:
Oct 18(Dijon)
tree-ring data Lauenen: the series has kindly been provided by Dr. Schweingruber wine harvest dates and phenological observations up to 1426 (Alexandre, 1986) wine harvest series of Dijon (Angot, 1883) wine harvest series Baulant, 1980)
'W Europe'
(Le Roy Ladurie,
Data: Alexandre, 1986
~
/
~
_" I ~
Early ripe O %7 _ /
~
h ~ _~ ~
~
.
Wine harvest beg Sep 13
March to May very dry Cherries ripe Ap 16
- I j fruit trees full bloom Ap 11 arvest beg Aug 25 ~ r a p e s ripe Aug 16
~
~ ~
Warm at end of winter ~tb~sf~o~Df~3Mar29
Warm at end of w i n t e r fruit trees bloom Ap 13
ripe beg Ma
J
9
Graph i: The early spring-summer 1420 in Central Europe
k X
Os
69
6.2. Examples of warm anomalies
In
1420 wine harvest in Western and Central Europe began at the end of
August,
even on altitudes of 500 to 700 m (Bern,
the earliest date ever recorded. this
year outstanding,
Toggenburg).
Because the contemporaries
from
considered
it was described in most chronicles,
those in which meteorological observations were marginal. extended
southern
Thuringia to the Po valley and
France to the Vienna basin (graph 1). are available at present.
This is
even
in
The
anomaly
from
Central
For the adjacent regions no data
In order to explain the weather patterns
of
this year the phenological evidence is compared to the pattern observed in 1540 (Pfister,
1984, 1985 a) and to comparable phenological extremes
documented with thermometrical measurements
In
1420
the
(table 3).
warm phase started in February.
In March
summer
already.
The vine bloom was two weeks earlier than in 1893 - the
advanced
year
Lichtensteig 100
within the instrumental period.
m (Becker,
1969 :
i.e.
almost
first sold
in the last days of May
around Basel
(260
probably from early burgundy
m)
Baulant,
1980).
The
grapes,
The wine harvest was advanced
compared with the long term average for Western Europe
1893).
per
a month before the mean date of the present century.
at the beginning of August.
Ladurie,
for
days
142) it has been estimated that the end of the
new wine ("Sauser"),
month
most
Based on the date
(600-700 m) and according to a gradient of 3.6
bloom may have occurred
began
was by
(Le
In 1540 the heat-wave began in April
a Roy
(as in
It is reported that the development of the vegetation was slowed
down by drought. The meager evidence available for 1270, 1304, 1331 and 1336
(cp.
table 2) suggests that phenological patterns may have
comparable to those observed in 1420 and in 1540: early
burgundy
whereas and
grapes
been
in 1270 and 1304 the
were ripe at about the same time as
in 1331 the ripening of the first cherries in
in
Western
1540, France
the beginning of the wine harvest in Paris coincided roughly
with
the corresponding phenophases in 1420.
The
comparison of the phenophases in 1420 and 1540 with the correspon-
ding extremes documented with thermometrical evidence suggests that 1420
all months from February to August
in
(in 1540 from April to August)
may have been 2 to 3 degrees above the 1901-60 average.
70
;
~I~
co
,,,.,-I v
o
v
v
~
t~
0 .,-I +
+
+
X
+
,<
c~
,.-4 ~
|
A
~
U'3
m 0
V
co
~
.,-I C
,.--4
v
r,.l
,-4
,--I ,-t
P,I P,l
¢N
,-4
,-t
,-4
~
S
S
..l.J
g r-i 1.4 c0 m .,-4 4-4
-,.-I
ffl
v
0
•~
I
I
v
v
v
v
O
v
!
0
m
I~ 0
.,--I ,'~
~
'4-I 0
~0
0 4-)
I
I
r~
°1
QI
Ill
P,I
~.~
~
~
u'3
~ ~..~
'0
0
u} Q)
if)
u')
W
-~
U
71
Precipitation
patterns may only be got for 1540.
who was antistes in Zurich, tation
during
After
Heinrich
recorded a total of six days with precipi-
the 26 weeks from mid March to the
end
of
September.
two rainy days at the beginning of October the weather turned to
warm and dry again. near Schaffhausen. mediterranean
On New Year boys were still swimming in the
summer
months.
of 1304 may have been similar.
According to
but
Annual
1984: 138). the
Colmarienses flour became scarce because many mills fell dry; abundant,
Rhine
The record of this outstanding year suggests that a
type of climate persisted for about ten
precipitation may not have exceeded 300 to 400 mm (Pfister, The
Bullinger,
Annales wine was
the casks couldn't be loaded on boats because the level
of the Rhine was too low (Annales,
1861: 231).
The most outstanding spring-summer period since 1269 probably is because
1473,
in this year a very early grape harvest coincided with a tree-
ring density that is close to the maximum recorded is an abundance of observations
for this anomaly,
(cp table 2).
There
but the documentation
is not available yet.
6.3. The ice-age summers of the 1340's
From
the climatic history of the last centuries it is well known
cold summers have a certain tendency to cluster similar pattern stands out in the 1340's:
in
bloom
at
Lindau after August 2nd,
a
in 1346 the vine was still retardation
of
and 1816).
In 1345 and 1346
maximum tree-ring
among
the twenty lowest contained in the Lauenen series.
vine was still in bloom at the beginning of September.
densities
are
In 1347 the
This points to a
anomaly in July and August that is unique in the last six
ries (Pfister,
7.
vegetative
that may be compared only to the two coldest summers since 1500
(1628
cold
that
1812 to 1817). A
for 1345 a slow maturation of
the vine is reported from Paris and Torino,
growth
(e.g.
centu-
1985 c: 192).
Climatic trends in spring and summer from the High Middle Ages to 1850
For
the warm period of the High Middle Ages continuous proxy-data
are
72
not yet available. in
1018
A historical custom introduced by the order of Cluny
nevertheless allows estimating the average date of the
grape
harvest in Northern and Central France. At the mass of the Transfiguration
(Aug.
6) the new wine was dedicated at the altar
presented to every friar.
the
Gregorian
afterwards
When the maturation of the vine was delayed,
the juice of some soft grapes was taken instead.
If the correction for
rule is made the average date of this
around August 13. This
and
celebration
was
roughly corresponds to the earliest date of the
wine harvest ever recorded
(August 13, 1893 in the region of Bordeaux).
In this year the mean temperature above the long term average
from April to August was 2.6 degrees
(Legrand,
1979 b:
42 f.).
But presumably
spring-summer temperatures in the High Middle Ages were somewhat lower. Legrand admits that we do not know whether the grapes for the first new wine were grown on an espalier sheltered from the cold. Also it must be assumed
that
grapes.
In
the
wine was made from an
Switzerland
the
August 10 in very warm summers
early
variety
early burgundy grapes were (Pfister,
1984:
of ripe
on August 15 (table 3),
vest was opened.
around
84). In 1420 the first
"new wine" from these grapes was drunk at the beginning of 1540
burgundy
August,
about twenty days before the
If this delay was the same in ordinary years, date would be around September Ist in the
in
wine harthe mean
grape
harvest
Ages,
which is a few days earlier than in the warmest summers documen-
ted with thermometric measurement.
High
Middle
From a regression approach comparing
the decennial means of wine harvest dates and temperatures from 1370 to 1850 it has been estimated that an opening of the harvest on Ist
September
corresponds to a mean temperature from April to September that
is
1.7 (+ - 0.2) degrees above the average for 1901-60.
From
the spreading of the vine and the cereals to higher altitudes and
latitudes
in
the High Middle Ages it has been primarily concluded
temperature patterns in midsummer. crops
is
1984:
86)
However,
mainly promoted by temperatures in May this
and
June
(Pfister,
evidence is rather conclusive for conditions
spring and early summer
(Legrand,
to
as the maturation of both
1979 b: 43).
in
late
An advance in the mean
date of the grape harvest also suggests an earlier date of snow-melt in the
Alps
since phenophases of the vine are
significantly
with the melting dates at different levels of altitude b: 168 f.).
correlated
(Pfister,
1985
73
In
the
High
Lauenen series of tree-ring densities the warm period
Middle
Ages
can be
d o c u m e n t e d by a s s e s s i n g
the
of
the
frequency
of
p o s i t i v e a n o m a l i e s e x c e e d i n g 1089 g/cm 3.
1269 - 1 2 9 9 : 1 3
%
1300 - 1 3 3 9 : 1 8
%
1340 - 1399:
5 %
1400 - 1499:
1%
1500 - 1599:
2 %
1600 - 1699:
0 %
1700 - 1799:
5 %
1800 - 1979:
0 %
Within the entire series they account for 3% of the cases. years
1774,
1777,
cal measurement, above
1779,
1781,
the only ones d o c u m e n t e d by thermometri-
temperatures in August and S e p t e m b e r were 2.0 degrees
the 1901-60 average.
From 1269 to
1339 p o s i t i v e anomalies
curred more than once every decade on average; were part of the "normal" climatic pattern; drops to a level of 5%, occurrence have
not
after 1340 their frequency
after 1400 they became very rare. Not a single
recorded any more.
century
marks a climatic watershed.
maximum
densities
U n d o u b t e d l y the
Since
1781
early
fourteenth
The frequent occurrence
is
connected.
1420,
according
high a
harvest
It may be h y p o t h e s i z e d that summers which were outstanto the standards of later periods,
such as
those
of
1473 or 1540 were within the normal range of fluctuations during
the High Middle Ages. warm
of
they
before 1330 can be interpreted in the context of
warm climate to which an advance in the b e g i n n i n g of the grape
ding
oc-
this suggests that they
is m e a s u r e d for the seventeenth century. been
In the four
This could explain why extreme anomalies in
the
period such as the summer of 1331 are only briefly d e s c r i b e d in a
few chronicles whereas similar events after the m i d - f o u r t e e n t h have evoked e x t e n s i v e comments in a m u l t i t u d e of sources. on p r e c i p i t a t i o n does not contradict this hypothesis.
century
The evidence
From 1200 to 1310
only two decades had a moderate excess of wet summer months
(graph 2).
74
Graph
2 Precipitation Difference
patterns
in S u m m e r
of u n m i s t a k a b l y
(J, J, A)
rainy
from
and dry months
1150
to 1420
per decade
Excess of +10987_ 6_ 5_ 4_ 3_
I
rainy months
2_ 1. O_ 1_ 2_
i V/, i V/J I V/, I V//
3_ 4_
-
H
I
v~ i V/J
dry months
51 8'
L
l
1150
I
I
i
1200
1250
1300
I
1350
1400
The year indicates the beginning of the decade
Source: Alexandre,
The
frequency
Lauenen
series
of
the very
by assessing
cold the
summers
c a n be
frequency
880 g / c m 3 .
1269
- 1299:
1340
- 1379:18
1430
documented
of n e g a t i v e
0 %
1300
- 1339:
5 %
%
1380
- 1429:
0 %
- 1499:
3 %
1570
- 1599:17
1600
- 1699:
4 %
1700
- 1810:
0 %
1811
- 1860:10
%
1861
- 1979:
0 %
%
1986
from
anomalies
the below
75
The evidence suggests that the shift from the warm climate of the
High
Middle Ages to the full brunt of the "Little Ice Age" did not take much more
than two decades.
The end of the transitory period in the 1330's
stands out by an extreme variability:
five late summers out of ten were
either much too cold or much too warm: three of them (1331, 1333, 1336) are very close to the highest tree-ring density recorded, value
of
series.
The
compared period
whereas
723 g/cm 3 measured for 1335 is the lowest within the
to
frequency of very cold summers from 1340 to 1379
may
the final decades of the sixteenth century
to
1811 to 1860,
advances in the Alps.
and
the
entire be the
which were at the origin of far reaching glacier For the Aletsch glacier an advance of a
similar
magnitude in the late fourteenth century has been recently demonstrated by Holzhauser
In
(1984).
order to compare the spring-summer climate in the Late Middle
with the
that of more recent periods the series of wine harvest dates CSte
Europe means
d'Or has been joined to the series for Western
(Le Roy Ladurie, of
grape
Baulant,
harvest dates
1980).
and
The curve of the
(graph 3) is based upon the
Ages for
Central decennial
CSte
d'Or
series until 1500 and on the other series for the following period. The differences
between
both
series
in the
overlapping
period
m a y be
neglected.
Within the series two levels may be clearly distinguished: 1430
to the long term average; curve
from 1380 to
the beginning of the wine harvest was advanced six days
from about 1450 to the late 19th century the
fluctuates around a lower level and displays the known
of glacier history: the 1770's,
the late sixteenth century,
the 1820's.
compared
advances
the 1690's, the 1740's,
The small trough in the 1490's may have
ceded a minor advance of the Lower Grindelwald Glacier, which, was very close to the valley
(Pfister,
in preparation).
pre-
in 1535,
The advanced
development of the vegetation from 1380 to 1430 may be explained by the complete
absence
of ice age summers and by an enhanced
warm spells in the spring-summer period and the 1420's temperatures
(e.g. 1382-1385).
frequency
of
In the 1380's
from April to September may have been
degrees above those of our century. A rapid and prolonged of the alpine glaciers may be hypothesized for this period.
0.5
melting back
76
0 0
0
/
0 0 CO
0 I
~:o co
0
0 r~ r¢l
~ ,-.-4
0
m ~
0 0
..~ 4A i>o
t/]
,-,1 I o o o00 u-~c0
rclo 0
II ~o
0
o
0
oo o c~u'J
U 0 0 0"1
113 0 ©
o
S
....
!,4
0 o
O~
77
APPENDIX GRAPE HARVEST DATES ORIGINAL AND INTERPOLATED SERIES YEAR
1370 1371 1372 1373 1374 1375 1576 1377 1378 1579 1380 1381 1382 1385 1384 1385 1386 1387 1388 1389 1590 1591 1592 1395 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1435 1454 1435 1436 1437 1438 1439 1440 1441 1442 14~3 1444 1445 1446 1447
ALBENGA
20 22
2~ 5 19 24 24 15 27 20 25 50 17 22 21
2i 2~ 13 23
2i 25 23 23 2O 23 24
MONCALIER1
BEAUNE
2i
16 25 29 29
2fi 2i z~
12
2o
FROM
DIJON
2~
20 12
3i 25 24
3i
16 38
2~ 14 13 21 28 26 24 25 27 17 10 38 17 10 27 20 24 20 9 17 -3 23 10 22 10
2i
25 26 10 11 20 19
2~
28 18 13 26 22 56
: 'A' FROM ALBENGA 'M' FROM MONCALIERI
COTE-D'OR
2~
2i
38
IST
33 30 24 37 18 29 35 22 31 22 14 25 -6 22 -3 25 11 16 14 25 36 24 15 19 18 12 1 25 56 28
2~
22 21 21 15 5 6 7 18 20 28 24
SEPT.
28M 26M 29 22 33 22 18 22A 25 24 23 21 13 7 8 8 20 27 27 24 20A 19 36 9 38 21A 27M 22 25 26 10 13 17 20 30 28 26 27 3O 24 17 38 18 20 30 21 28 21 12 21 -5 23 4 23 II 16 19 25 36 24 15 19 18 12 i 25 56 28
20 3i 20 16
2~ 23
26
INTERPOLATION
AVIGNON
25 23 26 18 26 32 28 20 26 26 23 30 21 54 19 15 29 29 29 24 29 29 22 14 32
(DAYS
28 18 15 26 22 56
= I) N.-EUROPE
78
APPENDIX GRAPE HARVEST DATES ORIGINAL AND IHTERPOLATED SERIES YEAR
ALBENGA
MONCALtERI
1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1~73 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 I~99 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 INTERPOLATION : ' V '
AVIGNON
21 13
28
21
2i 14 29 21 12 21
2~ lO 17
3~ 2~ 2~ 7 16 29 1
FROM AVIGNON
7 28 18 13 5 33 20 35 28
BEAUNE
COTE-D' OR N.-EUROPE
DIJON 50 29 25 39 24 27 34 31 33 14 18 36
li 3i 14 41 27 27 32 2O 37 II 25 -2 39 31 28 4t 19 16 39 47 16 15 20 43 20 22 42 31 25 45
38
50 29 25 39 24 27 34 31 33 14 18 36 22V 16 9V 35 14 41 27 27 32 2O 37 11 23 -2 39 31 28 19 16 59 47 16 15 20 43 20 22 42 31 25 45
31 37 2O 26 47 27 27 50
38
3~
43 28 21 3O 2O 30 ~4 2~
18 12 42 4i 26 28 14 19 29 28 14 43 28 21 3O 2O 3O ~4 2~
3i
3~
18 12 42
26 28 14 19 29 28
35 12 26 32 ~0 35
g
35 12 26 32 4O 35
g
-5
-5
2i
2i
18 12 4O 31 26 28 14 I9 26 17 17 43 29 19 32 25 3O 44 24 2B 29 31 11 22 28 37 23 23 2~ 17 14 2O
79
APPEHDIX GRAPE H A R V E S T DATES O R I G I N A L AND I N T E R P O L A T E D S E R I E S CORRELATION
AHD D I F F E R E H C E OF M E A N S IN THE C O M M O N YEARS ALBENGA
ALBENGA
BEAUNE 22
-
0.677
- 5 . 4 6
BEAUHE
22 0,677 3.46
DIJON
18 0 . 5 7 9 -1.50
-
30 0 . 7 3 4 -2.75
DIJON
MONCALIERI
18 0,579 1.50
26 0.230 3.27
30 0,734 2.75
32 0.300 4.78
-
42 0 . 3 7 9 -2.35
MOHACALIERI 26 0,230 -5.27
52 0.300 -4.78
42 0.579 2,35
AVIGNON
-
8 0,174 8.63
5
41 0,825 -0.36
23 0.354 -10.22
i
WEST-EUROP
IST LINE 2NO LINE
-
-
AVIGNON
i
-
8 0.174 -8.63 5
: H U M B E R OF C O M M O N VALUES, C O E F F I C I E N T OF C O R R E L A T I O H = D I F F E R E N C E OF MEANS : T O P - S E R I E S MINUS L E F T - S E R I E S
W.-EUROPE
41 0.825 0.36 23 0.354 10.22
80
References Acknowledgments are due to Prof. J o h n Post, B o s t o n , for r e a d i n g the m a n u s c r i p t a n d for m a k i n g h e l p f u l suggestions, to A n d r e a s Lauterburg a n d H a n n e s S c h G l e for p r o v i d i n g invaluable h e l p in programming and to Dr. P. A l e x a n d r e , Embourg (Belgium), for s e n d i n g a d d i t i o n a l data. Alexandre, P. 1986: Le Climat en Europe au Moyen-Age. Contribution ~ l'histoire des v a r i a t i o n s climatiques de 1000 ~ 1425, d'~pres les sources narratives de l ' E u r o p e occidentale. Paris (Ed. de l ' E c o l e des H a u t e s E t u d e s en Sciences Sociales). A m b e r g , B. 1890, 1892, 1897: B e i t r ~ g e zur C h r o n i k der W i t t e r u n g und verwandter Naturerscheinungen mit besonderer RGcksicht a u f das G e b i e t der R e u s s u n d der a n g r e n z e n d e n G e b i e t e der A a r e u n d des R h e i n e s . In: J a h r e s b e r . HSh. L e h r a n s t . Luzern. A n g o t , A. 1885: E t u d e sur les V e n d a n g e s en B u r e a u C e n t r a l M ~ t ~ o r o l o g i q u e de F r a n c e ,
France. Annales A n n i e 1883.
du
Annales, 1861: A n n a l e s B a s i l i e n s e s et C o l m a r i e n s e s ed. b y Ph. Jaffe. M o n u m e n t a G e r m a n i a e H i s t o r i c a , scriptores, 17, 189231. ( q u o t e d b y A l e x a n d r e , 1986). Becker, N. 1969: Beitrag zum Menge-G0te-Problem W e i n b a u . In: W e i n w i s s e n s c h a f t , 24, 172-190. Bell,
im
deutschen
W., O g i l v i e , A.E. 1978: W e a t h e r c o m p i l a t i o n s as a s o u r c e of d a t a for t h e r e c o n s t r u c t i o n of E u r o p e a n c l i m a t e d u r i n g the m e d i e v a l p e r i o d . In: C l i m a t i c C h a n g e I/4, 3 3 1 - 3 4 8 .
Burga, C.A. 1985: Naturohronlken.
Pal~oklimatische Auswertung yon B0ndner In: G e o g r a p h i c a H e l v e t i c a 40/4, 196-204.
B u s z e l l o , H. 1982: " W o h l f e i l e " u n d " T h e u r u n g " a m 0 b e r r h e i n 15401525 im S p i e g e l zeitgen8ssischer erz~hlender Q u e l l e n . In: Bauer Reich und Reformation. Festschrift fGr G G n t h e r F r a n z z u m 80. G e b u r t s t a g , hg. y o n P. B l i c k l e , 18-42, S t u t t g a r t . Dansgaard, W. 1984: Past c l i m a t e s and their f u t u r e . In: F l o h n , F a n t e c h i , 2 0 8 - 2 4 8 . Dubois, H. vall@e 1430).
relevance
to
the
1976: L e s f o i r e s de C h a l o n et le c o m m e r c e dans la de la S a o n e ~ la f i n d u M o y e n - a g e ( v e r s 1280 - v e r s Paris.
F l o h n , H., F a n t e c h i , R. (ed.) 1984: The C l i m a t e of E u r o p e : Past, Present and Future. Natural and Man-Induced Climatic Changes: A European Perspective. Dordrecht (Reidel). Flohn, H. 1984: Ice-free Arctic Flohn, Fantechi, 248-268.
and
glaciated
Antarctic.
In:
F l o h n , H. 1985: A c r i t i c a l a s s e s s m e n t of p r o x y d a t a for c l i m a t i c reconstruction. In: The C l i m a t i c Scene, ed. b y M.J. T o o l e y & G°M. S h e a i l , 9 3 - 1 0 3 , L o n d o n ( A l l e n & U n w i n ) .
81
Holzhauser, H.P. 1984: Zur G e s c h i c h t e der A l e t s c h g l e t s c h e r und des F i e s c h e r g l e t s c h e r s . Physische Geographie, 13. Z G r i c h (Geographisches Institut). Hughes, M.K., Kelly, P.W., Pilcher, (eds): C l i m a t e f r o m tree rings. Press; q u o t e d b y Flohn, 1985).
J.R., La Cambridge
Marche Jr, C.V. ( C a m b r i d g e Univ.
Ingram, M., Underhill, D., Farmer G. 1981: The use of documentary sources for the s t u d y of p a s t c l i m a t e s . In: Climate and History (Hrsg. T.M.L. Wigley et al., Cambridge), 180-913. Koch,
L. 1945: Dansgaard,
Meddelser 1984).
Lamb,
H. 1982: (Methuen).
Climate,
Lamb,
H. 1984: Climate Climatic Fluctuation 64.
om
Gronland,
history
and
in the L a s t a n d Change.
130, the
Nr.
modern
3
(quoted
world.
in
London
Thousand Years: Natural In: Flohn, Fantechi, 25-
L a v a l l e , M.J. 1855: H i s t o i r e et S t a t i s t i q u e grands vins de la C6te d'Or. Avec Gamier. D i j o n (Picard).
de le
Legrand, J.P. 1979a: Les fluctuations exceptionnelles durant les Saisons Estivales depuis le M o y e n - A g e . In: L a series, 16, 167-181; 18, 131-141.
la v i g n e concours
et des de J.
M@t@orologiques Printani@res et M@t@orologie, 6th
Legrand, J.P. 1979b: L'expression de la v i g n e au t r a v e r s du c l i m a t d e p u i s le m o y e n - a g e . R e v u e F r a n g a i s e d ' O e n o l o g i e 75, 23-52. Le
Men@, M. 1982: Les c a m p a g n e s A n g e v i n e s a la fin du M o y e n - A g e (vers 1350 - v e r s 1530). E t u d e @ c o n o m i q u e , N a n t e s .
Le
R o y L a d u r i e , E., Baulant, M. 1980: G r a p e h a r v e s t s f r o m the fifteenth through the n i n e t e e n t h centuries. In: J. of Interdisciplinary H i s t o r y 10/4, 8 3 9 - 8 4 9 .
Le
Roy Ladurie, E. 1971: Times of Feast, T i m e s of Famine. A history of c l i m a t e since the year 1000. L o n d o n (Allen & Unwin). (Translation of the 1967 F r e n c h e d i t i o n : L ' h i s t o i r e du c l i m a t d e p u i s l'an mil. Paris).
M a z z e i , M.: Le v a r i a z i o n i del c l i m a n e l l a p i a n a di A l b e n g a in eta' m o d e r n a . T h e s i s at the U n i v e r s i t y of Genova, F a c u l t y of L e t t e r s . ( S o u r c e a n d d a t a f r o m A l e x a n d r e , 1986). M c G o v e r n , T.H. 1981: E c o n o m i c s of e x t i n c t i o n in N o r s e G r e e n l a n d . In: W i g l e y et. al., C l i m a t e and History. Studies in p a s t c l i m a t e s a n d their impact on man, C a m b r i d g e , 4 0 3 - 4 3 3 .
82
Pavese, M.P., Gregori, G.P. 1985: A n a n a l y s i s of s i x c e n t u r i e s (XII t h r o u g h X V I I c e n t u r y A . D . ) of c l l m a t i o records from the Upper Pc V a l l e y . In: W. SohrSder (ed.), H i s t o r i c a l e v e n t s a n d p e o p l e in g e o s c i e n c e s , F r a n k f u r t , 1 8 5 - 2 2 0 . Pfister, Ch. 1984: K l i m a g e s c h i c h t e der S c h w e i z 1525-1860. Das K l i m a der S c h w e i z y o n 1 5 2 5 - 1 8 6 0 u n d s e i n e B e d e u t u n g in der Geschichte von BevSlkerung und Landwirtschaft. Vol I. Bern (Haupt). P f i s t e r , Ch. 1985a: C L I M H I S T - a w e a t h e r d a t a Europe 1525 to 1863. May be ordered H a l l e r s t r a s s e 50, C H - 3 0 1 2 Berne.
b a n k for C e n t r a l from METEOTEST,
Pfister, Ch. 1985b: Snow cover, snow lines Central Europe s i n c e the 16th century. S c e n e , ed. b y M.J. T o o l e y a n d G.M. S h e a i l ,
and glaciers in In: The C l i m a t i c L o n d o n , 154-174.
Pfister, Ch. 1985c: Ver~nderungen der Sommerwitterung im sGdlichen Mitteleuropa yon 1270-1400 als Auftakt zum Gletscherhochstand der N e u z e i t . In: G e o g r a p h i c a Helvetica 40/4, 1 8 6 - 1 9 5 . Rotelli, C. 1973: U n a c a m p a g n a medievale. 8toria agraria Piemonte frail 1250 e i l 1450. T o r i n o ( E i n a u d i ) o
del
R S t h l i s b e r g e r , F. 1976: G l e t s c h e r - u n d K l i m a s c h w a n k u n g e n im R a u m Zermatt, Ferp~cle und Arolla. In: 8000 Jahre Walliser Gletschergeschichte. Ein Beitrag zur Erfors0hung des Kllmaverlaufs in der N a c h e i s z e i t . D i e A l p e n 5 2 / 3 - 4 , 58-152. Scherer,
G.S.
1874:
Kleine
Toggenburger
Chronik,
St.
Gallen.
S c h m i t z , H.J. 1968: F a k t o r e n der P r e i s b i l d u n g fGr G e t r e i d e W e l n in der Z e i t y o n 8 0 0 b i s 1350. S t u t t g a r t ( F i s c h e r ) .
und
Schweingruber, F.H., F r i t t s , H.C., B r ~ k e r , U°, Drew, G., Sch~r, E. 1978: The X-ray technique as applied to dendroclimatology. In: T r e e - R i n g B u l l e t i n 38, 61-91. Schweingruber, F.H., Br~ker, 0.U., Sch~r, E. 1979: Dendroc l i m a t i c s t u d i e s on c o n i f e r s f r o m c e n t r a l E u r o p e a n d Great B r i t a i n . In: B o r e a s 8, 4 2 7 - 4 5 2 . U.S. WMO
Department hemisphere
of E n e r g y 1985: A c l i m a t e d a t a - b a n k l a n d a r e a s 1 8 5 1 - 1 9 8 0 . T R O No 17.
for
northern
1986: R e p o r t of the I n t e r n a t i o n a l C o n f e r e n c e on the A s s e s s m e n t of the R o l e of C a r b o n D i o x i d e a n d of O t h e r G r e e n h o u s e Gases in Climate Variations and Associated Impacts. V i l l a c h , A u s t r i a , 9 - 1 5 O c t o b e r 1985, W M O No. 661.
NORWEGIAN SEA DEEP WATER VARIATIONS OVER THE LAST CLIMATIC CYCLE: PALEO-OCEANOGRAPHICAL IMPLICATIONS
J. C. Duplessy , L. Labeyrie Centre des Faibles Radioactivit~s Laboratoire mixte CNRS-CEA F-91190 Gif sur Yvette, France and P.L. Blanc C.E.A./I .P. S .N./D.P .T. CEN/FAR B.P. 6-92265 Fontenay aux Roses Cedex,
i.
Introduction
The N o r w e g i a n world
ocean
formed flows
there out
by
over
few
the
heavy
because
analysis,
because
organisms
and
temperature
isotopic
place the micro-organisms ratios
are e x p r e s s e d
sample
from that
all
ocean
water
180/160 ratio) ice on the
years. live
in marine
water
continent.
the
of
In order
type
until
is
water
portion
deep-water
bottom). the
surface
of in
reaching
The
formed
1955;
At any given
dependent
from the
water
time,
ocean
core
micro-
both
1978).
the
180/160 in a
180/160 ratio light
and d e p o s i t e d sea
by
on
isotopically
past
of the
sea
180/160 ratio
the
and
at the time and
Duplessy, of the
more
to estimate
occurrence
carbonate is
of
waters)
in deep
deviation the
at a rate shells
parameter
of the ocean
(Emiliani,
deposited contains
in
sediments
higher
is removed
live
important
as the relative be
which
deep
a major
important
material
that
composition
in a standard. must
This
ratio
lived
This
all oceanic basins
on the
is an
180/160
deposited
and
winter. forms
of the modern water
1971).
(animals
180,
the
areas dense
floor are made of material
(that
isotope,
and
the most
it penetrates
deep
during
strait
(NADW),
thousand
foraminifera
of
cooling
(Reid and Lynn,
per
critical
volume
Denmark-Faeroe
foraminifera
oxygen
of the most
large
surface
from the ocean cm
planktonic benthic
the
Deep Water
the North Pacific Cores
is one
of
intense
the world ocean,
a
Sea
because
North A t l a n t i c
of
France
of (low
as glacier
surface
temperature
from isotopic analysis, it is thus n e c e s s a r y to estimate independently the t i m e - v a r i a t i o n s of the global ocean 180/160 ratio. This is the goal of the present paper. Micropaleontological have
shown
that
the
and
conditions
isotopic
analysis
prevailing
at the
of N o r w e g i a n Norwegian
Sea
Sea
cores
surface
84
have not been
remained
covered
constant
with
ice
As a consequence, not
active
glacial that
most
the m e c h a n i s m
and the N o r w e g i a n
ocean
the
over the last climatic
during
(Duplessy
amplitude
of
of the
last
by which
al.,
the
150
a source
1975). kyr
These
sea has
(Kellogg,
of deep water
isotope
also
for the
demonstrated
record
of
benthic
f o r a m i n i f e r a was smaller than in the other oceanic basins and glacial 2180 value was similar to that of b e n t h i c f o r a m i n i f e r a deep
Atlantic
hydrological Water
was
et al.
Ocean,
indicating
characteristics.
assumed
(1975)
to have
suggested
remained
that
the
Water could have been warmer However, Norwegian about
the
107
latitude
kyr
ago
1985)
Water
noticeably
was
necessary Norwegian
to
contradicts warmer
re-assess
deep
Norwegian
water
during
Sea Deep
formed
isotope
1985)
and
in
substage
that
Boyle and Keigwin,
hypothesis
during
the
Deep
Duplessy
the
the 5d, high
of cold deep water during the whole
1987;
the
similar
temperature,
of the
Shackleton,
1980;
had
last ice age than today.
that
was a source
et al,
Fairbanks,
and
waters
that the from the
in which the Pacific
temperature
interglaciation
(Duplessy
(Duplessy
deep
at a constant
evidence
last
N o r t h Atlantic
glaciation
both
during the
recent
Sea after the
that
By using a model
1980).
is formed today was
authors
oxygen
and this
age
deep water
Sea was not
et
cycle
ice
the
that
1982;
Norwegian
last
ice
age.
of
the
evolution
reconstruction
Sea Deep Water during the last climatic
Mix and Sea
Deep
is
thus
It
of
the
cycle and to determine
when it was cut off from the world ocean circulation.
2.
Strategy
The
isotopic
difficult
for
considerably sediment and
record. Second, shells
in
(1959)
showed
late
and
core
compare
all
reasons. time
(Kellogg several
et
al.,
cores
and
that
in
in the
the the
are
1978). order
It to
entire
fauna
fauna
is thus
southeastern
must
a
Norwegian
a
in
a to
complete
rest
on
vary single
analyze climatic
Holtedahl
downslope
Streeter
made
foraminiferal
problem.
displaced Sea.
is
rates
necessary
severe
been
Sea
recorded
generate
is
has
Norwegian sedimentation
not
reconstruction
displaced
an
of
First,
intervals
a paleoclimatic
place
glacial
paleoceanography
several
et
al.
in the (1982)
have shown that the p e r c e n t a g e of d i s p l a c e d s p e c i m e n s is high during glacial c o n d i t i o n s and related this d i s p l a c e m e n t to ice rafting, which can
be
easily
sediment.
recognized
by
the
During the last glacial
high
maximum,
percentage
of
these authors
quartz
in
the
showed that the
sediment is b a r r e n from benthic (i.e. deep water) foraminifera in place, w h e r e a s the a b u n d a n c e of d i s p l a c e d b e n t h i c fauna may be high. Third, p o r c e l a n e o u s species, n o t i c e a b l y Pyrgo , are abundant in portions of the cores.
However,
repeated
analysis
of
the
isotopic
composition
of
85
specimens
of
deviations
up to
this
(Duplessy et al.,
genus
have
shown
1 per mil toward
an
either
1984). As Pyrgo
unexplained light
dispersion
or heavy
was u s e d by D u p l e s s y
with
isotope
value
et al.
(1975)
to g e n e r a t e the i s o t o p e r e c o r d of core K-II, it m a y c o n t r i b u t e to i n a c c u r a c i e s in this record. The most a b u n d a n t h y a l i n e b e n t h i c species are C i b i c i d e s w u e l l e r s t o r f i and O r i d o r s a l i s t e n e r in the H o l o c e n e sediment (Belanger and Streeter, 1980). Streeter et al. (1982) showed that Cibicides is dominant during Holocene, isotope stages 4 and 5 and is absent d u r i n g isotope stages 2, 6 and most of stage 3. By contrast, Oridorsalis becomes dominant when Cibicides is rare or a b s e n t , n o t i c e a b l y d u r i n g glacial conditions. We t h e r e f o r e u s e d only these two species in order to generate a benthic 2180 record extending to the last interglaciation. In this paper, we present analyses of five sediment cores from the Norwegian Sea (Fig. I). Since none of them yielded a complete record of the conditions prevailing during the last climatic cycle, we developed a detailed stratigraphic framework in order to express the 2180 variations r e l a t i v e to a common time scale and we g e n e r a t e d a s t a c k e d b e n t h i c isotope r e c o r d d e s c r i b i n g the hydrological e v o l u t i o n of the N o r w e g i a n Sea Deep Water. We then p r e s e n t an e x a m p l e i l l u s t r a t i n g how the difference between this record and that of either benthic or planktonic foraminifera from the major oceanic basins may be interpreted in terms of temperature variations for the ocean water in the past.
3.
Isotopic
Calibration
of
the
Benthic
Species
Since b e n t h i c fauna vary in response to e n v i r o n m e n t a l
and climatic
changes, g e n e r a l l y any s i n g l e b e n t h i c f o r a m i n i f e r a l s p e c i e s is not a b u n d a n t e n o u g h for analysis t h r o u g h o u t the length of a deep sea core. In o r d e r to get a c o m p l e t e i s o t o p i c record, one r e f e r e n c e species is chosen and the other species are c o m p a r e d to that r e f e r e n c e species. F o l l o w i n g the r e c o m m e n d a t i o n of D u p l e s s y et al. (1984), we a d j u s t e d in this p a p e r 2180 values to Uvigerina , b e c a u s e this a d j u s t m e n t p r o v i d e s the oxygen isotopic composition of calcite in equilibrium with sea water at the time of~dep0sition. The c a l i b r a t i o n of C i b i c i d e s versus U v i g e r i n a for 2180 is now well e s t a b l i s h e d by thousands of analyses (Duplessy et al., 1970; 1984; S h a c k l e t o n and Opdyke, 1973; S h a c k l e t o n and Cita, 1979; B l a n c and Duplessy, 1982) and we applied the classical adjustment of +0.64. By contrast, Oridorsalis tenet was very poorly c a l i b r a t e d for both oxygen and carbon. We m a d e this c a l i b r a t i o n by c o m p a r i n g the 2180 and 813C values of O. tener and C. wuellerstorfi d u r i n g isotope stages I, 4, and 5, when both species were abundant and clearly coexisted. These data (Table 1 in Appendix] show that it is r e a s o n a b l e to adjust the 2180 of
86
30 °
15 °
0°
t5 °
30 ~
,/" /
J
75 ~
t I
/ 7
70 °
60 °
/
J
50 ° 30 °
Figure
15 °
1
:
Core
CH 77-07 K-f1 V 27-60 V 27-86 V 28-38
15°
0o
30 °
location. 66o36 ' 71047 ' 72°11' 66036 ' 69023 '
N N N N N
10031 ' 1°36 ' 8°35' l°07 ' 4024 '
W W E E W
1487 2900 2525 2900 3411
m m m m m
87
O.
tenet
by adding +0.36 to the measured values.
Conversely,
the ~13C
difference between O. t e n e t and C. w u e l l e r s t o r f i is not constant with time and our data indicate that it is not possible to estimate past ~13C of ~CO 2 dissolved in sea water by analyzing O.tener. Such a difficulty is not unusual and has already been observed for other benthic species, (Zahn et al., 1986). We therefore did not attempt any such as Uvigerina adjustment of the measured ~13C values of O.tener and used only the 813C values of C. w u e l l e r s t o r f i in order to estimate the carbon composition of ~ C O 2 dissolved in the Norwegian Sea deep water.
4.
Displaced
Foraminiferal
Shells
in
Glacial
isotopic
Sediment
Evidence for the presence of displaced shells, probably transported with ice rafted material, is abundant (Holtedahl, 1959; Belanger and Streeter, 1980; Streeter et al., 1982). We observed that in large sediment samples from core K-II, small amounts of C. w u e l l e r s t o r f i shells may be found in glacial sediment and the question arises whether these specimens are in place or displaced. We therefore made a detailed and O. comparison of the 2180 and ~13C variations of C. w u e l l e r s t o r f i tener in this core. Fig. 2 shows that whereas O. t e n e r exhibits isotopic variations roughly correlated with those of N. p a c h y d e r m a , C. wuellerstorfi had a constant isotopic composition during the whole isotope stages 2 and 3, characteristics of interglacial climate. By contrast, during isotope stages i, 4 and 5, when both species are abundant, they provide a similar record. We thus believe that the few C. wuellerstorfi shells found in glacial sediment are displaced specimens. We therefore analyzed C. w u e l l e r s t o r f i in the other cores only during stages I, 4, and 5, when this species is abundant. This strategy minimizes the perturbation effects of both bioturbation and transport (Bard et al., 1987).
5.
The
Planktonic
and
Benthic
Isotope
Record
and The oxygen and carbon isotope ratios of N. pachyderma, O . t e n e r in the five cores are r e p o r t e d in Table i. N. pachyderma left coiling has been used to derive the planktonic record, because it is adapted to very low temperature and is the most abundant planktonic (i.e. surface water) foraminifer in the Norwegian Sea. This species develops mostly during spring to summer. Its 2180 is in isotopic equilibrium for the hydrological conditions corresponding to a mean depth habitat of 80-100 m in the Norwegian Sea (Kellogg et al., 1978). C.wuellerstorfi
88 2,50
a18o
K-11
3,00
0
A+
3,50 4,00
0
o
0
//
~ oo
Z
o I~ llo#ol\?kro'o, ~1;o,~1 \ o O . , l l \ l xl 8 ~"
I
4,50
N. pachyderma 0-0
~" %
O Jr "B~
Q~o e ' ~ o '--
0 •
'...,,
, ~•
-\
,
I
\.o.~,~
9" J
~%"\
!
I
-
%
l
"'-'o . ,oo°r
\
~'t," ~ / \ e, L , {.o. I V%L /
/
l ~
qro
~, ."r"m
v.:./-." %
5,00
l~Jt
.
o
\
e
5,50 0
I
:
:
:
50
100
150
200
250
a 18 0
3,00
3,50
:
t •
; Depth ................(cm) 300
350
K- 11
I. ~
di+spiaced shetls
..
.
,,-
," 4 ",,~,eb% ,~ .~•' ~ %
4,00
•
&%
~ I
..
,-J~1~ .
•
•
.
6,
?,"
•
./~%" l
It / ~ "" .,,- "V
.
r~ 4,50
•~ Cibicides Depth (cm)
5,00 0
50
1O0
t 50
200
250
300
350
2 : O x y g e n i s o t o p e r e c o r d of N. p a c h y d e r m a , O. t e n e t (upper curves) and C. wuellerstorfi (lower curve) in core K-II. shells C o n s t a n t i s o t o p i c ratios of the rare C. w u e l l e r s t o r f i found in g l a c i a l sediment are i n t e r p r e t e d as an indication that t h e s e s h e l l s h a v e b e e n t r a n s p o r t e d a n d are not i n d i c a t i v e of local c o n d i t i o n s at the t i m e of d e p o s i t i o n of the sediment. N, pachyderma lives in surface w a t e r . O . t e n e r and C. wuellerstorfi live at depth on the bottom.
Figure
89
N. pachyderma occurs t h r o u g h o u t n e a r l y the whole of the last climatic cycle (Fig.3-7). The 2180 b e n t h i c r e c o r d has b e e n e s t a b l i s h e d b y analyzing monospecific samples. The analyses are plotted (Fig. 3-7) with correction of +0.64 for C. wuellerstorfi and +0.36 for O. tener, to take a c c o u n t of t h e i r r e s p e c t i v e d e p a r t u r e from i s o t o p i c e q u i l i b r i u m . When several m e a s u r e m e n t s were made at the same level, only their mean value was plotted. in N o r w e g i a n Sea sediment cores The 2180 records of N. pachyderma reflect the w o r l d - w i d e 2180 signal (Shackleton and Opdyke, 1973) modified s o m e w h a t by t e m p e r a t u r e and s a l i n i t y e f f e c t s p e c u l i a r to this area (Kellogg et al., 1978). The records of cores K-If, V 27-60, V 27-86 and V 28-38 e x t e n d until the b o u n d a r y between isotope stages 6 and 5 and show a great similarity. Their amplitude is larger than 2 per mil and they all exhibit a sharp peak c o r r e s p o n d i n g to isotope substage 5e. By contrast, the benthic record exhibits a much smaller amplitude, which is close to 1 per mil. The most striking difference between the p l a n k t o n i c and the b e n t h i c records is observed at the transition b e t w e e n substages 5e and 5d, which marks the b e g i n n i n g of glaciation: The mean amplitude of 2180 change between the peak of 5e and the maximum value of 5d is 0.75 per mil for the benthics whereas it is 1.6 per mil for the planktonic foraminifera. This d i f f e r e n c e between the p l a n k t o n i c and the b e n t h i c s i g n a l c a n n o t be e x p l a i n e d by b i o t u r b a t i o n , because the benthic foraminifera are abundant during the whole stage 5. It therefore implies that the summer surface waters and the b o t t o m waters evolved in very different ways during the beginning of glacial conditions. The d e t a i l e d records of core v 28-38 and V 27-60 also show that the 2180 increase of N. p a c h y d e r m a p r e c e e d s that of the b e n t h i c s . This s u g g e s t s that a c o o l i n g of surface w a t e r s by 3-4°C o c c u r r e d d u r i n g isotope s u b s t a g e 5e, b e f o r e the d e v e l o p m e n t of a s i g n i f i c a n t amount of ice o v e r the c o n t i n e n t s . This c o o l i n g p h a s e is f o l l o w e d by a small isotopically light peak, which does not coincides with isotopic substage 5c, as d e f i n e d in the benthic record, but preceeds it by more than i0 cm in all the records. This i l l u s t r a t e s that the d e t a i l s of the isotopic in the high latitudes are not easily correlated record of N° pachyderma with those of lower latitudes. and Figures 3-7 also display the ~13C variations of N. pachyderma C. w u e l l e r s t o r f i p l o t t e d a g a i n s t d e p t h in the five cores. A common feature to all the C. wuellerstorfi records is the p r e s e n c e of high ~13C values during the whole of stage 5 and stage 4. These values are more positive than those m e a s u r e d in benthic records from the North Atlantic (Shackleton, 1977; Duplessy, 1982; S a r n t h e i n et al., 1984; Mix and Fairbanks, 1985; Zahn et aL, 1986) or the o t h e r o c e a n i c b a s i n s (Shackieton et al., 1984) and are indicative of recent contact with the atmosphere (Duplessy and Shackleton, 1985). Gas e x c h a n g e with the atmosphere was possible at that time, since Belanger (1982) observed the p r e s e n c e of c o c c o l i t h s in N o r w e g i a n Sea b o t t o m s e d i m e n t s t h r o u g h o u t stage 5 and into stage 4. The presence of coccoliths implies that summer
90
0t80
K-tl
%
3,00
o
IJ . ,4 °
3,50 o
u
4,00
4,50
io °%oo ~'o.OP
''
5,00
L:
keJ
=
•
o o
oo-
r~, r...':
r;.! ~ 1 1 " / " ~ I I ~ ,, O\z • O'd~..,.,.Ag.# " "eoe
-"
"
tqenthics
5,50
0
Depth (cm) I
!
50
100
6,00
1,80
T
,,
I
;
:
I
200
250
300
350
,:
150
K-11
13 C
1,80
1 !I/~,',,
• ~\ 1,40
~
CJbioide~
• t't~.'r,
""
I/'v:-,
1,00 0,80 _
0,60 0,40 0,20
0,00
,o%
~,oa~ 1o o
t
o
ot\
_
I~00,,. /
/ 1
co~
c % od "oti"R
p,~Sq!
s'
VV~
|00 o oJ~Dof
~*"0
0
,-~
~'0~ ~. N. pachyderma
~
°o
o b%
-~ v~O
T
o
• oo
t Depth (CM)
O
-0,20 0
50
100
150
200
.250
300
350
F i g u r e 3 : Oxygen and Carbon isotope record of N. p a c h y d e r m a , Carbon isotope record of C. w u e l l e r s t o r f i and Oxygen isotope lives in record of benthic calcite in core K-If. N. p a c h y d e r m a surface water. C. w u e l l e r s t o r f i lives at depth on the bottom.
91
O 180
2,50
V 27-80
o
oJo% i ~i°
6<6!I. "W'o
~.o,o o 3,00
o
t
3,50
~0
Ao
0
1/1
4,00
/t •
O
o
P 1 o,~'od ~ -'\ q ,, Oo P ;oio,O~o. c6r~c. ,, -x d ~ o o u ;o ~ ~ "~'e \_ooo o,, o'o" 9 Jill -.'. , \ 2
o
: • t'; n ' ~ s
4,50
,°oo °
/
•~.:.. -- "O ~-.¢"--'~._
5,00
u
~
;---_.i--"
Tr~i,..16
% ~,,,
\
\=~.t~o=
;.*-.'~;-"
\; ""
5,50
Depth (cm) 6,00
i
I
I
I
l
100
200
300
400
500
13 C
1,80
V 27-60
I
I
600
700
•
1,60 1,40
J,
r;,tt
I . L~
o=c=e=
,,# %,}.~. •
1,20
1,00
+
,o
•
"l,l
0,80 • ,~;)\
0,60
0./%
0,40
o
0,20 0,00
o
o/[
\,,
O
_
nu
I~o
/
\ Io
off fo ~i o_t ~ .. ~,~o.m~ oOJ o ~° / _o o. f,. ! I ,o, ~ fo o4 d "o, .o _
,~ ~
OTc,o" \ ~.6 \ _..ocoo o -oo~
/+o\
~o6 \/I ~
Io [ I
211
~.o ,~
o
-Lo
L to/ eo
r "o%
o
-0,20 ;
;
~....
100
200
300
',
:
400
500
I Depth (cm)
-0,40 0
600
I
700
F i g u r e 4 : O x y g e n a n d C a r b o n i s o t o p e r e c o r d of N. p a c h y d e r m a , C a r b o n i s o t o p e r e c o r d of C. w u e l l e r s t o r f i and Oxygen isotope r e c o r d of b e n t h i c c a l c i t e in core V 27-60. N. p a c h y d e r m a lives in s u r f a c e water. C. wuellerstorfi lives at d e p t h on the bottom.
92
,:918 O
V 27-86 N. pachyderma
3,0O
o o
0
.,.o
4,50
I
o
,o \ ,.o--o
,'oJ
"-,-, %oo,O / o \
5,00
,
°~o
\/ o.
o
/,,. o
~.'
~'0", IX;-"
/
enthics
5,50
\./
6,00
[
;
I
50
1O0
I
150
iDepth(cm)
I
200
1 350
300
250
V 27-86
1,80 I a 13 C 1,60
•
pO...O
1,40
•
;V\
::/'...,.,..
Cibicides
~e
1,2o 1,00 0.80
O
0,40
T
/%
o
0
oo \O" o /
O
O-o
°?% /\ /\ o,, / \ / ~ 0,20 -~ " - ' o , , j_ %,,/o~o,, o \ 1 ~ / ' ~ 0,00 -0,20
o
N, pachyderma
0,60
o~
o /
dX%
X
I O
Xo!
'o
oo 'oo o'
o~
11
o
\~
Depth (cm)
o
|
:
i
;
!
:
I
I
0
50
100
150
200
-250
300
350
F i g u r e 5 : O x y g e n and C a r b o n isotope r e c o r d of N. p a c h y d e r m a , C a r b o n i s o t o p e r e c o r d of C. w u e l l e r s t o r f i and O x y g e n isotope r e c o r d of b e n t h i c calcite in core V 27-86. N. p a c h y d e r m a lives in surface water. C. w u e l l e r s t o r f i lives at depth on the bottom.
93 V 28-38
2,50 '8 180
o
o
¢
oo\/\
3,00
o
o
N. pachyderms
3,50
/\
/ !c~l i
o•
4,00 u\ I~ 00 00 R ,0 cO%£ crdl ]~ 1o _o o k 1-5!,,.6cc'-~'
Senthics 4,50
,. • ,•.
"~,,ep"e.%-" • h i
5,00
°11
"~_1,0,
oc,
~.e, ,.~.
l, o ell
I,e,et,,,~, e'~!'e_ l* e Lo
~'%" D,nu,
,_
~• Depth (cm)
5,50
l
I
1
i
I
100
200
300
400
500
V 28-38
o~ 13 C
1,8
I
60O
1,6 1,4
C;bicides ~r "~ ,
O~,~O
1,2
I•I
•
1
% o
o m oo~° N paehyderma 00,0 ,/~ 0 0 YII\~I i"
0,8
~i~i,ol/
"
0,6
o,.
o o
0,4 (?,2
%~_5~ .
o "oo/
~
o o 0
1
t|l
o"~o
o
• I
o• Io~8 o. o
Co
0o
,~
.co
0
100
o6~
~o
0 ~
o°
{ o
,ooo ,
!1
{/~
~0 '"I...........
o
¢oo1~1 o%O;~ ,o °1 /
0 -0,2
~.oo o;.l n
{I
Depth (era)
0
:
:
!
l
I
200
300
400
500
600
F i g u r e 6 : O x y g e n a n d C a r b o n i s o t o p e r e c o r d of N. p a c h y d e r m s , C a r b o n i s o t o p e r e c o r d of C. w u e l l e r s t o r f i and Oxygen isotope r e c o r d of b e n t h i c c a l c i t e in core V 28-38. N. p a c h y d e r m s lives in s u r f a c e water.C, wuellerstorfi lives at d e p t h on the bottom.
94
(918 o
CH 77-07
3,50
4,00
/.....-,: .\.. -/ oo-o.o.o,0,\{,.,.
4,50
•
5,00 ~'2 Benthic s O
I 50
5,50
1,50
I" 100
: 200
150
(9 13 C
I 250
l 300
Depth (cm) I I 350 400
CH 77-07
o 1,00
0,50
~'% o%
~.....
A
\p-.... "~P
0,00
,f•'~ W
• -0,50
.O"O'O X
T
50
100
-0-
"o.0.•.o
I.-"
"~o~•
/
/e'O
N. pachyderma
X ? •
;
"-.. •
! ....... 150
:
:
]
200
250
300
Depth (cm) , ] i 350 400
F i g u r e 7: O x y g e n and C a r b o n i s o t o p e r e c o r d of N. p a c h y d e r m a , C a r b o n i s o t o p e r e c o r d of C. w u e l l e r s t o r f i and O x y g e n isotope record of benthic calcite in core CH 77-07. N. p a c h y d e r m a lives in surface water. C. wue!lerstorfi lives at depth on the bottom.
95
temperatures
in the photic zone at that time were warmer than 2-3°C, as
coccospheres
do not
live today at lower temperatures.
We thus
that the Norwegian Sea was free of ice, at least during summer
conclude and that
it was an active site of deep water formation d u r i n g this time (from about 128 to 62 Kyr B.P.). The active convection in the Norwegian Sea is also supported by the high degree of correlation between the ~13C records of the p l a n k t o n i c and benthic foraminifera (r = 0.80 for core V 28-38, 0.73 for core V 27-60, 0.62 for core K-If and 0.40 for core V 27-86, which has a poor resolution). sharply d e c r e a s e s by the end of The abundance of C. w u e i l e r s t o r f i stage 4 (in cores V 27-60, K-II, V 27-86) or at the early b e g i n n i n g of stage 3 (in core V 28-38), w h e r e a s O . t e n e r is still p r e s e n t . U n d e r modern conditions, C. wuellerstorfi is the most abundant hyaline species in N o r w e g i a n Sea s e d i m e n t s at a w a t e r d e p t h b e t w e e n 1250 and 2900 m ( B e l a n g e r and Streeter, 1980; M a c k e n s e n et al., 1985) and O. t e n e t b e c o m e s d o m i n a n t b e l o w 2900 m. F r o m our data alone, we c a n n o t tell w h e t h e r C. w u e l l e r s t o r f i was d r a s t i c a l l y r e d u c e d in a b u n d a n c e during glacial conditions or whether the faunal b o u n d a r y which is at present at 2900 m m i g r a t e d upward. However, we b e l i e v e that this species p r o b a b l y d i s a p p e a r e d almost e n t i r e l y from the N o r w e g i a n Sea during the remaining of the glaciation, b e c a u s e its absence in g l a c i a l s e d i m e n t s has also b e e n o b s e r v e d in s h a l l o w e r 1984).
cores
(Streeter et al,
1982;
Sejrup et al.,
S t r e e t e r et al. (1982) i n t e r p r e t e d the change in d o m i n a n c e from C. to I O . t e n e r as an a b r u p t shift to an i c e - c o v e r e d Norwegian Sea. D e e p w a t e r f o r m a t i o n w o u l d not be p o s s i b l e in t h e p r e s e n c e of a p e r m a n e n t ice cover and we b e l i e v e that the benthic fauna wuellerstorfi
shift c o i n c i d e s w i t h the d e v e l o p m e n t of h y d r o l o g i c conditions which did not p e r m i t d e e p w a t e r formation. This h y p o t h e s i s is s u p p o r t e d by the v e r y n e g a t i v e ~13C values measured in O. tenet from g l a c i a l sediments, that are b e s t e x p l a i n e d by the p r e s e n c e of a d e e p w a t e r - m a s s p o o r l y o x y g e n a t e d at that time. W i t h this hypothesis,
the r e - a p p e a r a n c e
of C.
in all the cores by the end of the s e c o n d step of the d e g l a c i a t i o n (Termination IB, Duplessy et al., 1981). dates the renewal of the deep w a t e r v e n t i l a t i o n . Sejrup et al. (1984) r e p o r t e d that C. wuellerstorfi r e i n h a b i t e d the N o r w e g i a n Sea at s h a l l o w e r d e p t h within the first p a r t of the d e g l a c i a t i o n (Termination IA), about 5,000 years earlier. This suggests that the convection started d u r i n g the middle of the d e g l a c i a t i o n , but that the w a t e r m i x i n g b e c a m e s t r o n g e n o u g h to r e n e w the w h o l e N o r w e g i a n Sea deep w a t e r m a s s only at the end of the m e l t i n g phase of the continental ice sheet. wuellerstorfi
6.
Absolute
Chronology
We can i d e n t i f y several i s o t o p i c d a t u m levels in the ~180 records of N. pachyderma and of the benthic species, which are widely recognized
96 and dated outside the N o r w e g i a n Sea: Termination I, which has been dated b y 14C A c c e l e r a t o r m a s s - s p e c t r o m e t r y (Duplessy et al., 1986; B a r d et ai.,1987).
The 4/5a and 5e/6 boundaries can be clearly identified in all
the cores and their age has been a s s i g n e d within the framework of the astronomical theory of the Pleistocene climate (Imbrie et al., 1984). O t h e r i s o t o p i c b o u n d a r i e s are not e a s i l y recognized, n o t i c e a b l y within stages 3 and 5. We refined the c o r r e l a t i o n b e t w e e n the 5 cores t h r o u g h t h e i r ~13C records of N. pachyderma, w h i c h p r e s e n t a close similarity over the last 140,000 years in the high latitude of the North A t l a n t i c and in the N o r w e g i a n Sea (Labeyrie and Duplessy, 1985). A final a d j u s t m e n t has b e e n m a d e by m a t c h i n g the e v e n t s r e c o g n i z e d in the 2180 b e n t h i c records, b e c a u s e the N o r w e g i a n Sea Deep W a t e r must have had rather u n i f o r m properties within this single basin. The c h r o n o l o g y applied to this s t r a t i g r a p h i c framework is obtained by piecing that of Duplessy et al. (1986) for the last 25,000 years with that of Paterne et al. (1986) for isotope stage 3 and with the SPECMAP t i m e - s c a l e for stages 4 and 5 (Imbrie et al., 1984). The estimated ages of the levels taken as reference in each core are reported in Table 2. The ~13C record of N. pachyderma e x h i b i t s isotopically light peaks, w h i c h r o u g h l y c o r r e l a t e w i t h the o x y g e n i s o t o p e s t a g e s 2 and 4. We e s t i m a t e d the age of the 4 m a j o r transitions, b e c a u s e they may be used as additional stratigraphic markers. Results (Table 3) show that whereas the limits of the p e a k a s s o c i a t e d with isotopic stage 2 are a p p r o x i m a t e l y in phase with the oxygen isotope record, those a s s o c i a t e d w i t h isotopic stage 4 lag the 2180 record by about 5,000 years.
7.
Stacked
Benthic
Oxygen
Isotope
Record
from
the
Norwegian
Sea
For e a c h core, we u s e d the n o r m a l i z e d 2180 values of the benthic foraminifera and the reference levels reported in Table 2. We calculated the age of each level in the core by linear i n t e r p o l a t i o n b e t w e e n the upper and lower reference levels. We put t o g e t h e r the data of the five cores in order to obtain a matrix with two columns, one for the age and the other for the benthic calcite 2180 value. We then estimated the 2180 value each 1/3 kyear by cubic spline i n t e r p o l a t i o n and f i l t e r e d the r e s u l t i n g record by a 15 point least square quadratic filter (Savitzky and Golay, 1964). Finally, we c a l c u l a t e d from the filtered record the b e n t h i c c a l c i t e 2180 value each t h o u s a n d years. These 2180 values are p l o t t e d against time in Fig. 8. The m a j o r trends of the N o r w e g i a n Sea 2180 b e n t h i c record can be easily recognized, in p a r t i c u l a r the low a m p l i t u d e of the 5e/5d transition, w h i c h is m u c h smaller than in the Atlantic, Indian and Pacific oceans. The oscillations corresponding to isotopic substages 5a5d are well marked, but the 2180 v a l u e s c o r r e s p o n d i n g to i s o t o p i c
97 Table 2: Age assigned to reference levels. Core CH 77-07 Depth Age (cm) (kyr) 0 I00
0 9,8
Core K-11 Depth Age (cm) (kyr) 0 15 20 30 59 90 100 113 130 154 171 186 190 193 t98 204 218 242 255 270 272 274 285 300 303
5 10,5 13 15 25 32,7 33,5 37 40,7 44,8 50,3 51,8 56,1 57,4 60,3 65 71 80 87 99 100,5 117 122 127 135
Core V27-60 Depth Age (cm) (kyr) 0 70 140 160 190 240 280 320 360 375 385 403 405 420 440 450 461 475 485 500 515 536 560 640
0 6 10,5 13 15 25 37 40,7 50,3 56,1 60,3 65 68 71 80 87 99 108 112 116 119 122 127 135
Core V27-86
Core V28-38
Depth (cm)
Age (kyr)
Depth (cm)
Age (kyr)
0 9 30 50 80 120 130 169 215 227,5 243 260 270 280 300 325 340
3 6 10,5 15 25 37 40,7 48,8 62,3 65 80 87 99 112 118 122 127
0 69 112 164 204 222 228 236 253 302 316 328 342 376 391 408 414 436 450 481 502 517 528 532 541 555 578
0 6 10,5 15 25 32,7 33,5 37 40,7 45 47,9 48,8 51,8 56,1 57,4 59,3 65 71 80 87 99 107 112 113,3 118 122 128
Table 3: Age of the 4 major transitions in the carbon record N. pachyderma in Norwegian Sea sediment cores. Transition
1/2 2/3 3/4 4/5
K-11 V27-60 V27-86 V28-38
13,6 25 50,7 61,9
11,8 28 59 64
15 25 56 70
14 26,7 58,4 66,1
mean Age St. dev. (kyr) (kyr) 13,6 26,2 56,0 65,5
1,3 t,5 3,8 3,5
98 Norwegian Sea
4,00 I
~9 18 O
Stacked Benthic Record .%
6c~
4,50 fo o o.
o
o
-
'-'
o~co o P~
5,00
toc~
~,6 o
a
,o
o
o , (co G o o o
Ib
,.o "
to.co •
oo ~
o
q
b o. %co co o , c6~o
o
o
o
~6
o o
l o
5,50
% o
Age (kyr) 6,00
'
'
'
I
20
'
'
'
t
'
'
40
'
;
'
'
60
'
1
'
'
80
'
;
'
'
100
'
!
'
'
'
120
[
140
F i g u r e 8 : O x y g e n i s o t o p e r e c o r d of b e n t h i c c a l c i t e in N o r w e g i a n Sea s e d i m e n t d u r i n g t h e last c l i m a t i c c y c l e c a l c u l a t e d by stacking the benthic records of the 5 N o r w e g i a n Sea cores.
3,50
~180
Benthic
records
~t~
Pacific
5.00
:~
~!
A
"
4 n
5,50
6,00 0
1 20
I 40
I 60
I 80
1 100
Age (kyr) I ,' '1 120 140
F i g u r e 9 : C o m p a r i s o n of the s t a c k e d o x y g e n i s o t o p e r e c o r d of b e n t h i c foraminifera from the Norwegian Sea with that of Pacific core V 19-30 analyzed by N. J. Shackleton at Cambridge.
99
substages during
5a
surface Water
water to
substage
5a
at
than
present
cold
a first
has
only
resulting
from
sheets.This (Chappell
Deep The
for
surface
are
isotopic
temperature
confirmed
substages
5a-Sd
by
low
mass
(-I°C)
at
depth
surface
we
can
the and
a
lower
is -1.7°C, assume
significant case, of
of
the
the
similar
to the N o r w e g i a n
1986; stages
Sea
Dodge et al., 2 and 3, C.
account As
Norwegian
Sea
variation
foraminiferal world
northern
is supported by the sea level
the
sinking.
temperature
the benthic
variations decay
the
Its
for
into
during
that
formed
today.
limit
taking
experience
therefore
In this
growth
and
waters
any
warmer
the
(Kellogg,
which was
very
glacial
Sea
than
registered
the
sea
be
the
period.
not
and Shackleton,
During
the
during
whole
documented
temperature
could
experience
closely
that
Norwegian
Sea Deep Water,
5 and 4.
the
the this
well
measured Since
water,
that
hypothesis
variations
low
species
water
not
stages
implies
those 5e).
Sea was a sink for cold
4,
during is
than
substage
the Norwegian
is
warming
did
the
record
surface
approximation,
Water
during
heavier
5 and
which
These
subpolar
of this
adiabatic
mil
(isotope
stages
1980),
temperature
temperature
per
temperature
cover,
As a consequence, this
Deep
low
today.
of
0.i
conditions
isotope
a
sea-ice
disappearance from
only
indicate that the Norwegian
(Kellogg,
cooler
1976).
are
during
remained
proximity
the
5c
interglacial
313C values
Cibicides
was
and
full
ocean
3180 2180,
hemisphere
record,
stacked
which
benthic
ice shows
record
1983). is absent
wuellerstorfi
and
the same r a t i o n a l e cannot be a p p l i e d since we cannot i n t e r p r e t the ~13C in terms of those of the total d i s s o l v e d CO 2. we variations of O. t e n e r shall thus Pacific
compare
(Chappell
and
are highly
115,000
yr B.P.).
interval
and
therefore
Since
the
global
record.
explanations
Sea
2180 record. sea
water
Second,
record
their
significantly
the
below
foraminifera isotopic
water circulation sinking
level
volume
surface
we
The
observed
water
isotope
V19-30
during
conclude
to
this
that
the
sharp
2180
point,
would
which
(1986),
place
in
would
correspond
today.
The
at high
the
sea water
correspond is
the Pacific
Norwegian to deep
Sea
record
which
If
sediment
water
tempe-
of the
et ai.,1987)
latitudes,
to
impossible.
for that period.
reconstruction
(Duplessy
Sea
first,
as a global
2180 increase
the
increase
in the Nor- wegian
sea water 3180 record
than
two
from 28,000
for this observation:
and Shackleton in
the
(i.e.
core
record,
freezing
in the North Atlantic
of cold
that
2 may be considered
composition warmer
of
changes.
2 is not
the
are
shows
there only sea water 318Ovariations
the Pacific
as a global
cores,
Sea record with that of
3 to 5d
record
are possible
as showed by Chappell
rature
North water
ice
case,
temperature
benthic
sea
stage
of stage
In that
may be c o n s i d e r e d the
isotopic
the
9
stages
Sea record reflects
to isotope
Norwegian
Fig.
during
the
corresponding Three
1986).
correlated
parallels
benthic N o r w e g i a n
Norwegian
which has been closely tied to sea level variations
Shackleton,
records time
the stacked benthic
Core V19-30,
deep
indicates
fed the deep
A t l a n t i c b a s i n and c o n t r i b u t e d to the slow renewal of the deep in the N o r w e g i a n Sea. Since temperature is a c o n s e r v a t i v e tracer
iO0
for deep seems
waters,
rather
the presence
unlikely
explanation,
which
shells
in the
that
found
the
of the
we
Norwegian
favor,
between
with
previous
observations
such
a barren
zone
renewal basin
and
the
(Jansen
concomitant
A
poor
is
that
levels
25,000
of
Streeter
benthic
by
the
Record
of
the
of
et
benthic 2 are
al.
the
water ice
which
in
Ocean
in agreement presence
from
the
the
deep
and
also
of the
from
in a reduction
(Belanger,
2180
of
weak
Norwegian
as a consequence
cover
and
at the peak
The
result
results
surface
in place
life
yr B.P.,
may
The
foraminifera
not
(1982).
1984)
permanent
Global
third
and Ii,000
et al.,
of food at the sediment
basin
conditions.
of b e n t h i c
foraminifera
Sejrup
in the N o r w e g i a n
rare
stage
devoid
about
oxygenation
caused
the
from
low surface p r o d u c t i v i t y
availability
8.
for
deep water
hydrological
entirely
et ai.,1983;
stratification
these
sediment Sea was
glaciation
of warm
under
the
of the
1982).
Variations:
Major
Consequences.
Our
data
remained cycle. water
demonstrate
This
permanent
formation
stages
i,
The this
solely
record
stage
is
isotope 3,
present
temperature
occurred
isotope calcite
not
complete,
because the
sequence
glaciation.
the
the
variations
benthic
2180
stages
2 and
stages
i,
in Fig. function to
2180
of time. variations
variations
during
the
5. The
i0 as departure
the
the
a
Sea
ocean.
deep
isotope
source
of
is
low
nil
single
of Pacific
are
during
when core
core
sea
stack
benthic
record
record
as the
mass
climatic
ice
V19-30
exhibits
By
a
during the of
record of core V19-30 of the
Norwegian in Table
global
Sea
4 and
ocean 2180 as a
record
experienced
cycle.
was
is incomplete.
is reported
reference
which
absent isotope
record of the variations
the isotopic
resulting
therefore
Unfortunately
foraminifera
almost
a complete
be used last
of
during
Norwegian
global
accuracy
was
record
water
Sea
last climatic
process
basin
in per mil from the modern
It m a y of
the
record of one
obtained
3 with
4 and
Norwegian
only the ocean water 2180 v a r i a t i o n s
We t h e r e f o r e
for
to
benthic
its
the global ocean water 2180 by piecing for
the
of the N o r t h A t l a n t i c from
sedimentation
reflecting
due
Norwegian
of the
because
Moreover,
sedimentary
of
over the entire
2 and 3.
2180 r e c o r d
2180 2.
waters is
in the
stages
the stage
so that
By contrast, good
deep
4 and 5 a n d to the p r o x i m i t y benthic
reflects
the
low temperature
cold
which
deep water during
during
that
at q u a s i - c o n s t a n t
corresponding
no
comparing
temperature it
to
the
b e n t h i c r e c o r d of other o c e a n i c basins, it w o u l d p r o v i d e a m e t h o d to e s t i m a t e the deep water t e m p e r a t u r e v a r i a t i o n s over the last climatic cycle
(Labeyrie et al., 1987). The r e c o r d of g l o b a l ocean
estimate
sea
surface
hydrological
2180
variations
conditions.
may
However
also the
be
used
to
problem
is
101
TABLE 4: Oxygen isotopic ratio of the mean ocean water as a function of time during the last 135,000 years
Age
kyr 0 1 2 3 4 5 6 7 8 9 I0 11 12 13 I4 I5 16 I7 18 19 2O 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45
mean sea water ~w 0,00 -0,02 -0,01 0,01 0,01 0,05 0,10 0,17 0,27 0,32 0,37 0,42 0,48 0,48 0,88 0,91 1,05 1,01 0,94 0,95 1,01 1,00 0,95 0,80 0,70 0,51 0,56 0,59 0,60 0,65 0,67 0,65 0,63 0,68 0,63 0,55 0,48 0,55 0,61 0,67 0,67 0,64 0,58 0,63 0,64 0,6t
Age kyr 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
mean sea water Ow 0,59 0,57 0,58 0,44 0,33 0,38 0,45 0,53 0,56 0,57 0,58 0,57 0,43 0,20 0,35 0,45 0,48 0,69 0,70 0,71 0,67 0,63 0,55 0,56 0,49 0,45 0,40 0,40 0,44 0,47 0,39 0,26 0,17 0,08 0,08 0,15 0,25 0,18 0,09 0,15 0,29 0,40 0,43 0,37 0,28
Age kyr 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 1I0 111 112 113 1t4 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135
mean sea water
Ow 0,20 0,23 0,29 0,28 0,17 0,10 0,10 0,08 0,02 0,09 0,18 0,22 0,24 0,34 0,48 0,56 0,51 0,43 0,41 0,47 0,50 0,46 0,45 0,42 0,41 0,30 0,22 0,11 0,05 -0,06 -0,15 -0,14 -0,11 -0,05 -0,03 0,00 0,09 0,20 0,35 0,49 0,61 0,73 0,85 0,98 1,10
102
Global Sea Water
3 180
-0,20
Oxygen isotopic ratio variation
%
0,00
?% 0 et-d b
~O O
n,k t~
IQ
I
n
(o
O
o
,,
0,60
c.~o
0,80
o
o~ '~" / h
!
o
~'-
I
c. 9
£
o
o 0
,
l
o
0o.,
I o
Y
x4 o.,
((3 o O 0
o^ #Vo / '
q.d~'o~
~,
I
o^ z% Is
l~
q
_ o
io [ oo~_.^ 0 u(_~
j
q
0,40
;- t
o'~d~
u
o
0,20
O ,4
o
I
o
co
~ .e, co OdO~ ~-~'
'
o
I
o
| .o
o
I
o o
1,00 Age (kyr) 0
o
I
l'
l
1
I
!
I
20
40
60
80
100
120
140
1,20
F i g u r e i0 : V a r i a t i o n s of the oxygen i s o t o p i c c o m p o s i t i o n of the global ocean w a t e r d u r i n g the last c l i m a t i c cycle o b t a i n e d by p i e c i n g the s t a c k e d b e n t h i c N o r w e g i a n Sea r e c o r d of i s o t o p e stages i, 4 and 5 w i t h the b e n t h i c r e c o r d of P a c i f i c core V 19-30 for glacial isotope stages 2 and 3.
Temperature °C
V 28-38
! I
9"
ge
F e:x
.+
0
x
x~ * I%,,+++.+/
~++ ---X ~;z ~" ,,, " } x ' X x
X'~X + X
+ !
%i ,+
o~-X l•
;.,_V L,~ ,m
U )} °+ X
'~
(e.
-,',
•,
, 14
l,
•t l
((++I" z
Age (kyr)
]
I
I
l
l
20
40
60
80
100
1
I
120
140
F i g u r e II : E s t i m a t e of sea surface t e m D e r a t u r e v a r i a t i o n s in the N o r w e g i a n Sea c a l c u l a t e d from the 3180 d i f f e r e n c e b e t w e e n the p l a n k t o n i c r e c o r d of core V 28-38 and the global sea water 3180 record. B e t w e e n 9 and 65 kyr, no c o n s t r a i n t s are a v a i l a b l e for surface salinity and 3180 variations and sea surface temperature are p r o b a b l y ~ e r e s t i m a t e d as a c o n s e q u e n c e of m e l t i n g of ice d e p l e t e d in O (this p a r t of the r e c o r d is d r a w n in d a s h e d lines) .
103
more c o m p l i c a t e d than the interpretation of the b e n t h i c record b e c a u s e t h e 0180 p l a n k t o n i c r e c o r d depends on three variables: global 0180 variations,
sea
surface
temperature
(at the
season
and depth
at which
the foraminifera have lived), and local surface water 0180 changes, which may be due to local changes either in the e v a p o r a t i o n / p r e c i p i t a t i o n ratio or in the rate of exchange between surface and deep waters. If micropaleontological transfer functions can i n d e p e n d e n t r e c o r d of sea surface t e m p e r a t u r e c o m p a r i s o n of the p l a n k t o n i c 0180 record with
be u s e d to o b t a i n an (SST) variations, the the global ocean 0180
record and with the SST record should provide a m e t h o d to estimate the local sea surface 0180 values and therefore the sea surface salinity. In the Norwegian Sea, where micropaleontological transfer functions are not a c c u r a t e , e s t i m a t e s of SST c o u l d be o b t a i n e d w h e n some c o n s t r a i n t s exist on surface salinity. As an example, we compared the in core V 28-38 with the global sea water 0180 record of N. pachyderma 0180 record. In order to compare these records, we e s t i m a t e d by linear e a c h 1,000 years. We i n t e r p o l a t i o n the 0180 values of N. pachyderma a s s u m e d t h a t the 0180 d i f f e r e n c e b e t w e e n surface and deep water has r e m a i n e d n e g l i g i b l e d u r i n g the whole time the N o r w e g i a n Sea was an active source of deep water. The variations of sea surface 0180 value may thus be estimated by adding the modern value to the global sea water 0180 record. The sea surface temperature has then been c a l c u l a t e d by using the p a l e o t e m p e r a t u r e e q u a t i o n (Shackleton, 1974). Results are reported in Fig.ll. The e s t i m a t e for m o d e r n c o n d i t i o n s (5.5°C) falls right within the range of p r e s e n t l y m e a s u r e d SST (3°C during winter and 8°C during summer). The e s t i m a t e d t e m p e r a t u r e s exhibit values similar to the present ones only during the peak of isotopic substage 5e, and a major t e m p e r a t u r e drop o c c u r r e d by the end of substage 5e. During the r e m a i n d e r of stage 5, the t e m p e r a t u r e s were in the range I-3°C, as expected from the presence of coccoliths in the sediment from this time. During the glaciation, we have no constraint on the surface salinity and 0180 v a l u e s a n d our m o d e l is not v a l i d any more. The e s t i m a t e d temperatures are p r o b a b l y too high, since salinity and 0180 are expected to decrease, as a consequence of the presence of a q u a s i - p e r m a n e n t ice cover over the N o r v e g i a n Sea. High t e m p e r a t u r e e s t i m a t e s are obtained for the Holocene, when our model is valid, with temperature higher than today about 7,000 Kellogg (1976).
9.
years
ago,
in
agreement
with
the
observations
of
Conclusions
O x y g e n and c a r b o n i s o t o p e a n a l y s e s of p l a n k t o n i c a n d b e n t h i c foraminifera from five N o r w e g i a n Sea sediment cores demonstrate that the N o r w e g i a n Sea was an active area of deep water formation not only during
104
full interglacial conditions (isotopic stages 1 and 5e) but also during the early part of the glaciation (isotope stages 4 and 5a-Sd) . As sinking of surface water to the depth is linked to its cooling during winter, the Norwegian Sea deep water temperature was close to the freezing point during the time intervals 0-8 kyr B.P. and 65-128 kyr B.P.. Consequently the oxygen isotope record of this period is a record of the global sea water 2180 variations due to the growth and decay of continental ice sheets. A complete record of the global sea water 2180 variations during the last climatic cycle has been established by piecing the stacked Norwegian Sea benthic record of stages i, 4, and 5 with the benthic record of Pacific core V 19-30 for glacial stages 2 and 3.The resulting record may be used to -extract either the temperature signal present in benthic records from other oceanic basins or the variations in surface h y d r o l o g y recorded by the oxygen isotope variations of planktonic foraminifera. C.F.R.
Contribution
N ° 870.
REFERENCES Bard, E., M. Arnold, J. Duprat, J. Moyes, and J.C. Duplessy, Reconst r u c t i o n of the last deglaciation: D e c o n v o l v e d records of 2180 profiles, m i c r o p a l e o n t o l o g i c a l v a r i a t i o n s and a c c e l e r a t o r mass spectrometric 14C dating, Climate Dynamics, I, 101-112, 1987. Belanger, P.E., Paleo-oceanography of the Norwegian Sea during the past 130,000 years: coccolithophorid and foraminiferal data, Boreas, Ii, 29-36, 1982. Belanger, P°E., and S.S° Streeter, Distribution and ecology of benthic foraminifera in the Norwegian-Greenland sea,Mar. Micropal, 5, 401428, 1980. Blanc, P.L. and J.C. Duplessy, The deep water circulation during the Neogene and the impact of the Messinian salinity crisis, Deep Sea Res., 29, 1391-1414, 1983. Boyle, E.A. and L.D. Keigwin, Deep circulation of the North Atlantic over the last 200,000 years: Geochemical evidence, Science, 218, 784-787, 1982. Chappell J. and N.J. Shackleton, 324, 137-140, 1986
Oxygen isotopes and sea level, Nature,
Dodge, E.D., R.G. F a i r b a n k s , L.K. P l e i s t o c e n e sea level from raised 219, 1423-1425, 1983.
Benninger, coral reefs
and F. Maurasse, of Haiti, Science,
Change, edited Duplessy, J.C., Isotope studies, in Climatic Gribbin, Cambridge University Press, Cambridge, 46-67, 1978.
by
J.
105
Duplessy, J.C., Circulation des eaux profondes nord atlantiques au cours du dernier cycle climatique, Bull. Inst. G4ol. Bassin d'Aquitaine, 31, 379-391, 1982. Duplessy, J.C., C. Lalou, and A.C. Vinot, Differential isotopic fractionation in benthic foraminifera and p a l e o t e m p e r a t u r e reassessed, Science, 168, 250-251, 1970. Duplessy , J.C., L. Chenouard, and F. Vila, Weyl's theory of glaciation supported by isotopic study of Norwegian core K-II, Science, 188, 1208-1209, 1975. Duplessy, J.C., J. Moyes and C. Pujol, Deep water formation in the North Atlantic Ocean during the last ice age, Nature, 286, 479-482, 1980. Duplessy, J.C., G. Delibrias, J.L. Turon, C. Pujol, and J. Duprat, Deglacial warming of the northeastern Atlantic ocean: correlation with the p a l e o c l i m a t i c evolution of the european continent, Palaeogeogr., Palaeoclimat., Palaeoecol., 35, 121-144, 1980. Duplessy, J.C. and N.J. Shackleton, Response circulation to the Earth's climatic change ago, Nature, 316, 500-507, 1985.
of global deep-water 135,000-107,000 years
Duplessy, J.C., M. Arnold, P. Maurice, E. Bard, J. Duprat and J. Moyes, Direct dating of oxygen-isotope record of the last deglaciation by C-14 accelerator mass spectrometry, Nature, 320, 350-352, 1986. Duplessy, J.C., N.J. Shackleton, R.K. Matthews, W. Prell, W.F. Ruddiman, M. Caralp, and C.H. Hendy, 13C record of benthic foraminifera in the last interglacial ocean: Implications for the carbon cycle and the global deep water circulation, Quat. Res., 21, 225-243, 1984. Duplessy, J.C., N. J. Shackleton, R. Fairbanks, L. Labeyrie, D. 0ppo and N. Kallel, Deep water source variations during the last climatic cycle and their impact of the global deep water circulation, Paleoceanography (submitted), 1987. Emiliani,
C., Pleistocene
Temperatures,
Holtedahl, H., Geology and paleontology J. Sed. Petrol., 29, 16-29, 1959.
J. Geol.,
63, 538-578,
of Norwegian
1955.
sea bottom cores,
Imbrie, J., J.D. Hays, D.G. Martinson, A. Mc Intyre, A. Mix, J. J. Morley, N. Pisias, W. Prell, and N.J. Shackleton, The orbital theory of Pleistocene climate:Support from a revised chronology of the late marine 2180 record, in Milankovitch and Climat~ edited by A. Berger et al., D. Reidel, Hingam, Mass., 1984 Jansen, E., H.P. Sejrup, T. Fjaeran, M. Hald, H. Holtedahl and O. Skarbo, Late W e i c h s e l i a n p a l e o c e a n o g r a p h y of the southeastern Norwegian Sea, Norsk Geologisk Tidsskrift, 63, 117-146, 1983 Kellogg, T.B., Late Quaternary climatic changes: evidence from cores from Norwegian and Greenland Seas, Geol. Soc. Am. Memoir, 145, 77ii0, 1976. Kellogg, T.B., Paleoclimatology and paleo-oceanography of the Norwegian and Greenland seas: Glacial-interglacial contrasts, Boreas, 9, 115137, 1980.
106
Kellogg, T.B., N.J. Shackleton, and J.C. Duplessy, Planktonic foraminiferal and oxygen isotopic stratigraphy and paleoclimatology of Norwegian sea deep sea cores, Boreas, 7, 61-73, 1978. Labeyrie, L.D. and J.C. Duplessy, Changes in the oceanic C13/C12 ratio during the last 140,000 years: high latitude surface water records, Palaeogeogr., Palaeoclimatol., Palaeoecol., 50, 217-240, 1985. Labeyrie, L. D., J.C. Duplessy, and P.L. Blanc, Deep water formation and t e m p e r a t u r e variations over the last 125,000 years, Nature, 327, 477-482, 1987. Mackensen, A., H.P. Sejrup, and E. Jansen, The distribution of living benthic foraminifera on the continental slope and rise off southwest Norway, Mar. Micropal., 9, 275-306, 1985. Mix,
A. and R.G. Fairbanks, North Atlantic s u r f a c e - o c e a n control of Pleistocene deep ocean circulation, Earth Planet. Sci. Lett., 73, 231-243,1985.
Paterne, M., F. Guichard, J. Labeyrie, P.Y. Gillot and J.C. Duplessy, Tyrrhenian sea tephrochronology of the oxygen-isotope record for the past 60,000 years, Mar. Geol., 72, 259-285, 1986. Reid, J.L. and R.L. Lynn, On the influence of the Norwegian-Greenland and Weddell seas upon the bottom waters of the Indian and Pacific oceans, Deep Sea Res., 18, 1063-1088, 1971. Sarnthein, M., H. Erlenkeuser, R. von Grafenstein, and C. Schroeder, Stable-isotope stratigraphy for the last 750,000 years: "Meteor" core 13519 from the eastern equatorial Atlantic, Meteor Forschungergebnisse, Reihe C , 38, 9-24, 1984. Savitzky, A. and J. E. Golay, Smoothing and differenciation of data by simplified least squares procedure, Anal.Chem., 36, 1627-1639, 1964. Sejrup, H.P., E. Jansen, H. Herlenkeuser, and H. Holtedahl, New faunal and isotopic evidence of the late Weichselian-Holocene oceanographic changes in the Norwegian Sea, Quat. Res., 21, 74-84,1984. Shackleton, N.J., Attainment of isotopic equilibrium between ocean water and the benthonic foraminifer genus Uvigerina : isotopic changes in the ocean during the last glacial, Colloque International du Centre National de la Recherche Scientifique, N ° 219, 203-210, 1974. Shackleton, N.J., Carbon-13 in Uvigerina: Tropical rainforest history and the equatorial Pacific carbonate dissolution cycle, in The Fate of fossil fuel C02 in the Oceans, edited by N. R. Anderson and A. Malahoff, pp 401-428, Plenum, New York, 1977. Shackleton, N.J., and M.B. Cita, Oxygen and carbon isotope stratigraphy of benthic foraminifera at site 397: Detailed history of climatic change during the late Neogene, Initial Reports of the Deep Sea Drilling Project, 47, 433-445, 1979. Shackleton, N.J., J. Imbrie, and M. Hall, Oxygen and Carbon isotope record of east Pacific core V 19-30: Implications for the formation of deep water in the Late P l e i s t o c e n e North Atlantic, Earth and Planet. Sci. Lett., 65, 233-244, 1983.
107
Shackleton, N.J. and N. D. 0pdyke, Oxygen isotope and paleomagnetic stratigraphy of equatorial Pacific core V28-238: Oxygen isotope temperatures and ice-volumes on a 105 year and a 106 year scale, Quat. Res., 3, 39-55, 1973. Streeter, S.S. , P.E. Belanger, T.B. Kellogg, and J.C. Duplessy, Pleistocene p a l e o - o c e a n o g r a p h y of the N o r w e g i a n - G r e e n l a n d Benthic foraminiferal evidence, Quat. Res., 18, 72-90, 1982.
Late Sea:
Zahn, R., K. Winn, and M. Sarnthein, Benthic foraminiferal ~13C and accumulation rates of organic carbon: Uvigerina peregrina group and Cibicidoides wuellerstorfi, Paleoceanography, i, 27-42, 1986.
108
Table 1: Core K- 11 depth (cm)
N. pachyderma ~18 013
0 3 5 8 10 12 15 16 17 18 20 23 25 26 29 30 32 35 38 40 45 50 53 56 59 60 65 68 70 73 75 80 83 85 90 94 95 100 103 105 106 110 113 115 120 121 124 125 126 130
3,02
0,49
3,06
0,33
3,22
0,42
3,60 3,74
0,51 0,43
3,25 3,19 3,77 3,55
0,55 0,48 0,33 0,09
4,59
0,14
4,71
0,14
4,57 4,73 4,65
0,11 0,22 0,10
4,75 4,53 4,58 4,37 4,52 4,46 4,53 4,52 4,58
0,08 0,31 0,31 0,21 0,28 0,33 0,39 0,22 0,47
4,16 4,35
0,27 0,30
4,57 4,48
0,22 0,40
4,46 4,68 4,09
0,41 0,39 0,36
3,74 4,56
0,25 0,37
4,36
0,37
4,65
0,46
Cibicides wuellerstorfi 018 013
Oridorsalistenel Benthic ~18 ~13 ~18
3,83 3,89
1,30 1,36
4,31
-0,61
4,57 4,53
4,11
1,21
4,48
-0,84
4,80
4,01
1,25
4,76
-1,07
4,89
3,92
1,13
3,84
1,17
4,66
-1,59
4,75
3,85 3,90
1,19 1,27
4,76 4,96
-1,59 - 1,50
5,12 5,32
3,99 3,92 3,93 3,91 3,91 3,94 3,90 3,91 3,94
1,32 1,23 1,26 1,17 1,22 1,28 1,32 1,17 1,21
4,77 4,71 4,80
-1,56 -1,49 -1,79
5,13 5,07 5,16
5,00 5,01
-2,20 -2,04
5,36 5,37
4,81 4,59
-1,49 - 1,89
5,17 4,95
4,25 3,98 3,88 3,95
1,21 1,27 1,18 1,21
4,68 5,06 4,71 4,95
- 1,73 -1,60 -1,73 -1,49
5,04 5,42 5,07 5,31
4,12 3,91
1,44 1,30
5,01
-1,54
5,37
3,92 3,91
1,21 1,19
4,83
-1,53
5,19
3,88 3,89
1,21 1,28
4,96
-1,34
5,32
3,85
1,26
4,94
-1,15
5,30
3,94
1,74
4,71
-1,31
5,07
4,00 3,93
1,42 1,26
4,88
-1,37
5,24
3,82 4,07
1,26 1,30
5,08
-1,10
5,44
4,56
109
Table 1: Core K- 11 (cont.)
depth (cm) 133 ! 35 138 140 144 145 t46 148 149 I50 152 154 155 157 160 165 170 171 173 175 176 179 180 182 185 187 190 194 195 196 198 200 20I 202 204 208 210 214 215 218 220 224 228 230 234 238 240 242 244 245
N. pachyderma ~18 ~I3 4,27
0,47
4,39
0,39
4,65
0,55
4,06
0,44
4,28 4,22 4,42
0,40 0,23 0,30
0,09
4,25 4,29 4,46 4,66 4,48
-0,04 0,10 0,20 0,13 0,02
0,65
4,32
0,56
4,19
0,59
3,98
4,85
-1,20
5,21
4,04
1,46
4,74
-1,30
5,10
3,98
1,19
4,91
-1,10
5,27
4,10 4,00
1,30 1,22 5,03
-1,12
5,39
3,94 3,97
1,28 1,27
4,63
- 1,43
4,99
3,84 4,04 3,87
1,22 1,30 1,26
4,65 4,86
-1,33 -1,76
5,01 5,22
4,02 4,02
1,26 1,27
4,75
-1,42
5,11
4,00 3,95
1,39 1,33
4,80 4,79
-i,42 -1,50
5,I6 5,15
3,99 4,10 3,92 3,84 3,91
1,30 1,37 1,21 1,24 1,27
4,86 4,74 4,74 4,89 4,46
-1,35 -1,45 -1,45 -1,66 -1,33
5,22 5,10 5,10 5,25 4,82
4,03 4,22
1,28 1,50
4,90 4,60
- 1,78 - 1,79
5,26 4,96
4,50 4,52 4,65 4,41 4,42 4,31
4,86
-1,25
1,36 1,37 1,32 1,34 1,44
5,18 5,16 5,29 5,05 5,06 4,95
4,24
1,52
4,88
4,29 4,24
1,36 1,25
4,93 4,88
4,19 4,17 4,1I 3,92 3,98 4,26
1,20 1,62 1,63 1,40 1,6t 1,22
4,83 4,81 4,75 4,56 4,62 4,90
0,27
4,55
3,99
1,27
0,28
4,34
4,63
4,08
0,17
4,48
4,23
Cibicidcswuellerstorfi Oridorsalis tenel Benthic ~I8 ~13 ~18 813 ~18
0,51 0,68
110
Table 1: Core K-11 (cont.) depth (cm) 248 250 252 255 258 260 262 264 265 268 270 272 274 275 279 28O 282 284 285 288 290 292 298 300 303 305 310
N. pachyderma Cibicides wuellerstorfi 018 013 018 013 4,25
4,i8
4,29
Oridorsatis tene~ Benthic 018 013 018
3,86
1,54
4,50
4,29 4,36 4,25
1,33 1,3 t 1,40
4,93 5,00 4,89
4,08 4,11 4,09 4,05
1,46 1,35 1,03 1,41
4,26
-0,76
4,72 4,75 4,73 4,69
4,05 3,86
1,33 1,18
4,35
-0,80
4,69 4,50
3,70
1,08
3,85 3,76 3,79 3,82
1,19 1,13 1,10 1,16
3,82 3,70
1,17 1,02
0,62
0,54
0,72
4,07
0,50
3,42 3,46
0,45 0,37
3,10
0,I0
3,83 3,85 3,56 3,35
0,26 0,23 0,15 0,06
4,75 4,90
0,08 0,06
4,34 4,11
-1,21
4,49 4,44 4,43 4,46 4,46 4,34
5,24
-2,33
5,60
Table 1: Core V 27-60 depth N. pachyderma left (cm) 018 013 12 14 20 30 34,5 40 44 52 55 60 70 74 78
2,84 2,71 2,82 2,87 2,81 2,76 2,67 2,81 2,7 2,69 2,5 2,42 2,65
0,40 0,34 0,57 0,67 0,46 0,62 0,44 0,45 0,35 0,22 0,26
Cibicideswue~A~storfi 018 013 3,88 3,65 3,90 3,97 3,81 3,99 3,67 3,91 4,12 3,85 3,88 4,I2 4,17
1,24 1,15 1,28 1,24 1,3I 1,34 1,31 1,19 1,20 1,13 1,11 1,12 1,10
Oridorsalis tener 818 013 4,15 4,05 4,26 4,24 4,15 4,11 4,06
-0,83 -0,77 -0,72 -0,70 -0,63 -0,93 -0,84
4,21
-0,87
4,28 4,24
-1,1I -0,71
Benr2uc ~I8 4,52 4,35 4,53 4,61 4,48 4,55 4,37 4,55 4,67 4,49 4,58 4,68 4,81
111
Table 1: Core V 27-60 (Cont.) depth N. pachyderma left (cm) 018 013
80 85 90 99 104 110 117 120 127 137,5 140 160 170 186 190 200 210 220 230 240 249 259 270 280 29I 302 310 320 332 340 350 360 370 375 380 385 390 395 401 403 405 408 410 415 417 420 425 428
Cibicides wuetlerstorfi Oridorsalistener ~18 ~13 ~18 ~t3
2,8 2,63 2,85 2,95 2,82
0,06 0,09 0,18 0,17 0,20
2,93 3,56
0,33 0,36
3,65 4,07 3,62 4,34
0,54 0,34 -0,06 -0,02
4,17
0,95
4,84 4,72 4,62 4,59 4,56 4,43 4,57 4,74 4,53 4,39 4,43 4,59 4;5 4,66 4,47 4,31 4,13 3,88 3,98 4,45 4,38 3,2 3,94 4,67 4,61
0,05 0,02 0,04 0,00 -0,11 -0,05 0,09 0,22 0,13 0,08 0,I2 0,38 0,31 0,48 0,23 0,01 0,07 0,18 -0,09 -0,04 0,26 -0,17 -0,14 0,03 0,33
4,05 4,09
1,10 1,20
4,38
0,23
4,49 4,28
0,46 0,45
4,33 4,16
0,64 0,60
4,02 4,11 4,12 4,20 4,24
4,41 4,64 4,55 4,42 4,47 4,50 4,48 4,22 4,28 4,49
1,05 0,97 0,91 0,97 1,25
1,17 1,22 1,29 1,26 1,29 1,32 1,47 1,52 1,77
Benthic a18
4,38
-I,08
4,35 4,46
-1,32 -1,28
4,62 4,55
-1,06 -0,84
4,42
-0,83
4,81 4,81
4,71 4,85
-1,40 -1,61
5,07 5,21
4,9I
-1,70
5,27
4,76 4,50 4,92
-1134 - 1,42 -1,58
5,12 4,86 5,28
4,82
-t,99
5,18
4,84
-2,16
5,20
4,70 4,75 4,74 4,82 4,88 4,98 4,92
5,05 5,28 5,19 5,06 5,11 5,14 5,12 4,86 4,92 5,13
112
Table 1: Core V 27-60 (Cont.) depth N. pachyderma left (cm) 018 013 431 435 437 440 445 446 450 458 461 465 467 475 480 485 487,5 492 495 498 500 505 508 510 515 520 525 527,5 532 532 535 536 545 548 552 555 560 565 57O 575 580 584 591 610 620 640 650 66O 670 680
3,96 3,83
0,41 0,30
3,49
0,53
4,15 4,19
0,65 0,60
4,17 4,26
0,6I 0,64
3,96 3,94 3,82
0,44 0,37 0,34
3,81 4,28
0,33 0,69
4,34 4,38
0,66 0,74
3,73 2,81
0,52 0,20
2,84
0,17
3,i1
0,29
2,8 2,56 2,72
0,07 0,01 0,03
2,75 2,85 3,28 4,3
-0,07
3,61 4,27 4,17 4,13 4,23 3,99 4,78 4,7 4,56 4,47 4,6
0,18 -0,14 -0,I1 -0,13 -0,04 -0,15 0,11 0,14 0,04 O,10 0,02
0,13 0,28
Cibicides wueUerstorfi Oridorsalistenet 018 013 018 013 1,43 1,32 1,27 1,22 1,57 1,33 1,46 1,26 1,26 1,33 1,29 1,31 1,24 1,16 0,99 1,23 1,30 1,10 1,19 1,23 1,08 1,17 1,10 0,99 1,07 0,8"9
Benthic 018 4,94 4,57 4,66 4,37 4,89 4,36 4,83 4,50 4,44 4,80 4,80 4,99 5,04 5,01 4,93 4,98 4,98 4,91 4,99 4,62 4,71 4,79 4,44 4,45 4,58 4,34 4,44 4,29 4,32 4,03 4,52 4,39 4,40
4,30 3,93 4,02 3,73 4,25 3,72 4,19 3,86 3,80 4,16 4,16 4,35 4,40 4,37 4,29 4,34 4,34 4,27 4,35 3,98 4,12 4,15 3,80 3,81 3,94 3,70 3,80 3,73 3,68 3,46 3,88 3,75 3,76
0,87 0,82 0,87 0,94 0,88 0,89
3,91
0,69
4,55
4,10
1,19
4,74
4,29
-0,51
3,85
-0,75
3,60
5,29 5,17 5,07 5,17 5,13
-1,70 -1,86 -I,86 -1,81 -1,92
5,65 5,53 5,43 5,53 5,49
113
Table 1: Core V 27-86 Depth (cm) 1 5 9 15 19 30 40 50 55 60 70 80 100 110 120 130 140 150 160 169 210 215 220 225 230 235 240 246 250 255 260 265 270 275 280 285 289 295 300 305 310 315 321 325 330 335 340 360
N. pachyderma left 318 313 2,68 2,67 2,54 2,63 3,04 4,30 4,47 4,72 4,71 4,74 4,55 4,68 4,36 4,25 4,52 4,38 4,10 4,38 3,40 3,64 4,52 4,34 4,46 4,63 4,51 4,30 3,78 3,92 4,05 4,39 4,16 4,02 4,20 3,78 3,84 3,75 4,67 4,45 3,87 3,28 3,08 3,26 3,02 4,04 4,13 4,14 3,92
0,27 0,18 0,34 0,35 0,26
Cibicideswullerstorfi 218 313 3,95 4,07
1,39 1,36
4,08 3,94
1,31 1,21
0, I9 0,08 0,04 -0,07 O,11 0,03 0,22 0,t2 0,28 0,42 -0,04 0,38 -0, I0 0,15 0,07 -0,01 0,27 0,40 0,39 0,51 0,54 0,68 0,47 0,53 0,21 0,32 0,23 0,5 t 0,62 0,38 0,28 0,16 0,18 -0,08 0,19 0,23 0,28 0,44
Oridorsatis tenet 018 313
Benthic 218
4,25 4,21 4,30 4,27 4,43
-0,86 -0,76 -0,83 -0,87 -0,97
4,60 4,64 4,66 4,68 4,69
4,92
-2,14
5,28
5,34
-1,98
5,70
4,77
-1,94
5,13
4,65
- 1,90
5,0 I
4,74 4,85 4,76 4,60
-1,36 -1,18 -1,60 -1,59
5,10 5,21 5,12 4,96 4,53
4,53
-i,21
3,89
1,28
4,36 4,33 4,69 4,35 4,49 4,04 4,48 4,26
1,33 1,52 t,45 1,27 1,24 1,40 1,54 1,36
4,72 4,70
-I,24 - 1,02
4,61
-I,07
4,27
1,57
4,72
3,87
1,43
4,07
4,12 4,15 4,41 4,60 3,98 4,03 4,23 3,78 3,84 3,8 t
1,24 1,27 t,20 1,29 1,38 1,31 1,27 1,29 1,17 1,t5
4,76 4,79 5,05 5,24 4,62 4,67 4,87 4,42 4,48 4,45
3,98 3,84
1,2t 1,21
4,62 4,48
5,00 4,93 5,33 5,04 5,10 4,68 5,12 4,94 4,91
-1,13
4,47
114
Table 1: Core V 28-38 depth (cm) 48 58 69 8O 92 101 112 119 132 142 148 152 158 164 172 175 184 192 194 197 204 208 212 215 218 222 228 232 236 248 253 258 262 268 272 278 282 294 302 308 314 318 323 328 336 342 352 358
N. pachyderma left 318 ~13 2,94 2,9O 2,72 2,90 3,19 2,84 2,84 2,98 3,24 4,34
0,59 0,5O 0,40 O,55 0,43 0,40 0,40 0,27 0,51 0,45
4,11 4,33 4,63 4,50
0,25 0,08 0,14 0,17
4,79 4,84
0,12 0,17
4,55 4,53 4,55 4,68
0,34 0,11 0,20 0,31
4,40 4,72 4,73 4,39 4,52 4,38 4,68 4,56 4,65 4,65 4,62 4,63 4,54 4,42 4,38 4,38 4,08 4,32 4,06 3,83 3,67 3,96 4,06 4,19
0,34 0,43 0,47 0,49 0,47 0,52 0,44 0,50 0,45 0,50 0,38 0,39 0,31 0,29 0,18 0,17 0,09 0,17 0,08 0,25 0,05 0,27 0,38 0,37
Cibicideswuetlerstorfi 318 313 3,90 3,89 3,94 3,97
1,30 1,27 1,32 1,22
Oridorsalis tenet ~18 ~13
Benthic ~18
4,16 4,36 4,32 4,35 4,39
-0,54 -0,63 -0,92 -0,93 -0,69
4,53 4,62 4,63 4,66 4,75
4,45 4,36 4,66 4,57 4,71 4,54 4,47 4,60
-0,96 -0,95 -1,00 -1,13 -1,34 -1,88 -2,07 -1,20
4, 81 4,72 5,02 4,93 5,07 4,90 4,83 4,96
4,57 4,63
-1,03 -1,21
4,93 4,99
4,54
-1,20
4,90
4,29 4,58
-1,60 -0,93
4,65 4,94
4,79
-1,49
5,15
4,39 4,68 4,44 4,24 4,50 4,93 4,58
-1,08 -1,14 -0,93 -1,29 -1,10 -1,23 -1,16
4,75 5,04 4,80 4,60 4,86 5,29 4,94
4,70 4,62 4,60 4,45 4,57 4,41
-1,36 -1,28 -1,31 -1,21 -1,41 -1,52
5,06 4,98 4,96 4,81 4,93 4,77
4,82 4,80 4,64 4,48 4,69 4,52 4,63 4,70
-1,53 -1,62 -1,49 -1,23 -1,54 -1,43 -1,25 -1,37
5,18 5,16 5,00 4,84 5,05 4,88 4,99 5,06
115
Table 1: Core V 28-38 (Cont.) depth (cm) 362 368 374 378 384 39I 396 398 402 408 414 418 425 432 434 436 438 442 445 447 448 453 457 458 462 468 475 478 481 486 491 498 502 508 517 518 520 521 522 527 528 532 536 541 543 548 552 558 562 566 571 573 578
N. pachydermaleft 018 0t3 4,23 4,33 4,51 4,10 4,54 4,39
0,16 0,30 0,16 0,14 0,I9 0,12
4,58 4,00 3,73 4,59 4,68 4,48 4,24 4,24 4,40 4,13 4,30 3,91
0,28 -0,07 -0,02 0,21 0,31 0,75 0,51 0,74 0,47 0,69 0,72 0,37
3,64 3,60
0,51 0,64
4,20 4,02 3,92 3,85 4,33 4,38 3,94 4,15 4,23 4,19 4,10 4,19 4,29
0,79 0,35 0,71 0,85 0,77 0,70 0,59 0,66 0,79 0,66 0,78 0,66 0,79
4,I4 4,12 3,88 3,87 3,88 4,05 4,17 3,62 3,26 3,33 2,73 2,70 2,57 4,13 4,05 3,55
0,59 0,82 0,46 0,49 0,50 0,56 0,75 0,49 0,41 0,42 0,i9 0,10 -0,01 0,34 0,39 -0,07
CibicideswueUerstorfi 018 013
4,01
Oridorsalis tenet 018 013 4,68 4,85 4,83 4,73 4,18
-1,I0 -1,41 -I,21 -1,07 -1,63
4,64 4,87 4,79 4,93 4,89 4,94 4,65 4,57 4,35 4,5I 4,55 4,61
-I,65 -1,14 -1,44 -1,72 -1,31 -1,01 -1,24 -0,94 -0,99 -0,95 - 1,04 -1,00
4,54
-1,10
1,20
4,05 4,34
1,38 1,26
4,46 4,26 4,28 4,06 4,1I 4,25
1,44 1,50 1,68 1,62 1,63 1,51
3,88 3,45 3,88 4,16
1,05 1,24 1,45 1,53
4,17
1,35
4,17 4,28 4,29 4, I2 4,15 3,73 3,87 3,91
t,66 1,74 1,58 1,56 1,57 1,52 1,46 1,47
4,18
1",44
4,27 4,02 4,05 3,89 3,81 4,13
1,54 1,36 1,23 1,27 1,14 1,18
4,04 3,80 3,74 3,83 3,78 3,77
1,22 1,26 1,19 1,04 1,00 0,96
4,52 4,43
-1,00 -1,07
4,38 4,37 4,45 4,40 4,47 4,33 4,19 4,45 4,65 4,55 4,52
-1,I7 -t,06 -0,99 -0,85 -1,01 -0,90 -1,17 -1,18 -1,17 -1,17 -1,01
4,39 4,43 4,11
-1,41 -I,12 -1,06
4,22 4,40 4,11 3,98 4,07 4,15 4,17 4,35 4,37 4,43
-0,81 -0,87 -0,87 -0,90 -0,82 -0,76 -0,94 -1,26 -1,29 -1,20
Benthic 018 5,04 5,21 5,19 5,09 4,54 4,65 5,00 5,23 4,92 5,14 5,25 5,21 4,96 4,93 4,71 4,81 4,90 4,97 4,52 4,52 4,80 4,88 4,80 4,78 4,83 4,87 4,76 4,81 4,69 4,53 4,68 5,01 4,87 4,88 4,9t 4,66 4,72 4,66 4,46 4,77 4,58 4,72 4,46 4,36 4,45 4,47 4,47 4,71 4,73 4,79
116
Tabte 1: Core CH 77-07 depth N. pachyderma Ieft cm 018 013 0 3,78 0,51 8 10 3,74 0,54 18 2O 3,81 0,76 28 30 3,72 0,53 38 40 3,77 0,29 48 50 4,i0 0,29 58 60 3,95 0,23 68 70 3,92 -0,08 78 80 3,92 0,08 88 90 4,24 0,04 i00 4,42 0,03 105 108 110 4,27 -0,03 118 t20 4,22 -0,24 128 4,63 -0,10 130 4,82 -0,02 133 3,94 -0,03 135 4,39 -0,09 138 4,12 0,10 140 4,04 -0,01 143 4,52 0,04 145 4,70 0,06 150 4,74 -0,02 160 4,61 0,12 170 4,50 0,18 180 4,56 0,23 190 4,52 0,24 200 4,74 0,38 210 4,51 0,11 220 4,39 0,05 230 4,19 0,02 240 4,55 -0,04 250 4,32 0,03 260 4,57 -0,01 270 4,55 -0,04 280 4,44 -0,05 290 4,51 -0,02 310 4,69 0,05 330 4,46 0,07 340 4,29 -0,04 360 4,50 -0,02 380 4,I8 0,20 390 4,35 0,17
Cibicides wuellerstorfi 018 013 3,86 1,10 3,82 0,92
Oridorsalistener 018 0t3 4,20 -0,97 4,09 -0,95
Benthics 018 4,53 4,46
3,95
0,79
4,22
-0,93
4,59
3,84
0,95
4,15
-1,00
4,50
4,00
0,91
4,64
4,00
0,81
4,64
4,05
0,82
4,27
-1,96
4,66
4,1 t
0,83
4,23
-I,26
4,67
4,10
0,77
4,02
0,58
4,74 4,25
-1,45
4,64
4,66 4,65
-1,37 -1,44
5,02 5,01
4,94
-1,89
5,30
NUMERICAL MODELS OF CLIMATE
Hartmut Grassl Forschungszentrum Max-Planck-StraBe D-2054 Geesthacht
Geesthacht I
I. INTRODUCTION
The
climate
cryosphere, partment.
on earth
is the result
biosphere
and lithosphere
At a given
parameters partments
merely
insolation
results
from
of many interactions as well
the strong
observed
the drastically
between
as interactions natural
different
atmosphere,
within
variability
response
ocean,
a single
com-
of climate
time of the com-
to changed boundary conditions.
Stimulated
by extreme weather and climate conditions man started to establish a mete-
orological
network in the 18th century leading to nowadays famous time series especi-
ally for temperature. an
area
after
paleoclimatic nition:
a sufficiently evidence
climate
ultimate
The observations
weather
long
were meant as giving the climate statistics
time
(some
decades)
and these direct observations
is the
average
forecast
weather
interval
and for
all.
However,
have led to a new climate defi-
after averaging
of a few weeks
once
of
over more than the expected
including
all statistical
parame-
ters, thus also the probability for a distinct deviation from the mean. Since climate varies
on all time scales also the time span used for averaging has to be specified.
Climate defined in such a way accounts for the dominance of dynamics.
Being aware of the strong dynamics of climate it is clear that we need a continuous global observation
network. At the same time we would - given such a global observing
system - only be able to foresee periodicities neither
any change
in climate undisturbed
accounting for a high percentage
a sufficiently
strong
periodicity
below
a few hundred
of variability
are encountered.
accurate fine mesh global observing system
besides
yearly
years
another
and daily cycle tool
by man, if strong
could be derived for time scales
for getting more
insight
has not only been
created but also recently became a main research direction for climatologists: ical
models
of global
climate,
although
the ability
conditions mentioned was not at all foreseeable. to be realized: variability
The
present
inability
to forecast
Another
to separate
Since
exists nor a known
climate
'pressure'
in observations
numer-
under
the
for modellers
has
natural
climate
from man made climate change. Two main avenues have been followed in cli-
mate modelling:
118
-
simple, mainly one-dimensional
models for the understanding
of distinct processes
and basic interactions -
use
of
three-dimensional
meteorological
services,
After shortly introducing tions
in chapter
the chapters
general
atmospheric
circulation
models,
the basic applications
by
of modelling and the necessary equa-
2, the main errors and drawbacks will be named
4 and 5 are devoted to results
ural and disturbed
developed
also for climate studies.
in chapter
of numerical modelling
3, while
both for the nat-
climate system.
2. THE BASIS OF CLIMATE MODELLING
The
basic
purposes This
laws
necessary
in meteorology
dependence
applications
will
I)
be
at growing
2.1 Meteorological
For a short
boundaries
Thus,
2)
firstly
for
(12 or 24 hours)
meteorological
and secondly
for distinct
for climate
turbulent
heat
field neither
like radiation flux divergence,
fluxes
play
a
dominant
role
nor
irre-
heat of con-
changes
at
the
ocean and land surface have to be introduced.
equation
of motion of
and conservation
potential
vorticity
are
distinct vertical resolution
is necessary.
For a 5-day weather
the irreversible
as well
involved
forecast, of the pressure
in the atmosphere
the equation
balance
compartments
depend on the time scale envisaged.
Applications
term
and
shown
system
strongly
time scales.
versible processes densation
and the climate and climatology
forecast
as land surface
parameter
ever, may still be held constant
changes.
of total mass often combined the
only
processes
basic
in a
involved.
No
become an important part
The ocean surface
at its typical
laws
temperature,
how-
seasonal values - or better - at
its actual values at the start of the forecast.
Now the basic
thermodynamic
tion of state
(for the atmosphere
leading - besides
laws,
the prognostic
tion for temperature.
the radiative simply equation
transfer
equation
and the equa-
the ideal gas law) have to be included, of motion - also to a prognostic
If the water cycle is also described
water mass balance equation has to be solved.
equa-
in detail a prognostic
119
3)
For a 2 week weather of the ulbimate
forecast - still not possible
threshold
for deterministic
weather
at present,
but on this side
forecasting
- the short term
changes in the oceanic deck layer as well as surface water balance relations to
be
added,
thus
not
strongly the intrinsic
While
it was
rather
only
simple
soil-vegetation-atmosphere poorly
understood
bringing
interactions
in the
ocean
layers
but
also more
between soils and vegetation.
to add a known interaction
biological
upper
have
basic
opens
processes
have
equation
a
new
to
be
kind
under of
I) and
2) the
difficulty:
often
formulated
in mathematical
equations.
2.2 Climate Applications
All the above meteorological
applications
the main purpose of this contribution: the basic equations By enlarging
involved
answering
climate models. Nevertheless
in both long term weather forecast
the time interval
hance difficulties
were not directly
we reach
climate models.
questions
for
3) has pointed to
and climate models.
On one hand we thereby
en-
because of the need to describe at least the global atmosphere and
upper parts of the global ocean, on the other hand we no longer need an initial data set at the start being
of the computation
both the failure
cause of the large
impact
due
interactions)
to
non
linear
very near to reality.
of deterministic of minute
weather
errors
and
in the initial
the wish
The reason for the latter
forecasting
for
climate
after
a few weeks
data on the final statistics
for
(be-
result
the time
interval desired.
The
two main
streams
of climate research
with respect
to modelling
(see WMO,
1984)
presently are:
I)
Understanding models
variations
coupled to an upper ocean model.
permanent
2)
of year-to-year
by using atmospheric
general
circulation
The ocean model has to reach the deepest
thermocline.
Decadal changes of climate caused by man's activities.
This modelling task enlarges both the climate system compartments deeper
layers
equations
for
of soil,
big parts of the cryosphere
distinct
substances,
and thus the trace substance
the latter
composition
- the entire ocean,
- and the number of mass balance
describing
of the atmosphere.
the
atmospheric
chemistry
120
While task I is clearly tied to a better understanding of the natural climate system, task 2 is trying to isolate man's scales
is fully understood.
composition from
another
change, part
thereby
of
influence before the natural system on these time-
Task 2 has also to rely on the scenarios of atmospheric often
loading
'forecasting',
which
climate model is totally
results
depending
with
uncertainties
on extrapolation
of
man's energy use and thus also on political structures.
The base of climate models
as compared
to that of models of socio-economic
develop-
ment is rather clear:
most equations are known
-
-
the knowledge
of the climate
system is advanced
enough to tell us the compart-
ments to be handled for a distinct climate time interval -
empirical relations between vegetation cover and reflectivity
of the surface are
reasonably well established -
most important chemical reactions
are also known. Despite this rather optimistic
view the main error sources of climate models have to be discussed before results from different groups are presented in chapters 4 and 5.
3. ERRORS INHERENT TO CLIMATE MODELS
3.1 Numerical Errors
The basic physical equations are non linear thus analytic solutions do not exist. In turn,
the numerical solution means integration of discretized differential
in space and time.
Thus,
numerical
errors
inherent
equations
in any discretization are always
present, although more advanced numerical techniques recently have reduced the impact on model
results.
However,
a further
strong
reduction
of
this
error
is not easy,
since a finer grid or mesh which would help also means - due to numerical instabilities otherwise
imminent - a smaller
time step, soon surmounting the capacity of any
big computer. Halving the grid size increases the computer time in a three-dimensional climate model by a factor of 16.
3.2 Parameterlzation Errors
A second
even
larger
parameterization
error
error.
is
Because
also the
tied basic
to
the
laws
restricted in continuum
computer physics
capacity: are
the
a closed
121
system only if applied unaveraged lence element,
to volume elements smaller than the smallest turbu-
any volume averaging
of these equations
leads to correlation
of turbulent quantities with the need for a closure hypothesis. ses
for small
scales
scall
turbulence
with
space
scales
products
Many closure hypothe-
of a few hundred
meters
and time
of half an hour exist, the simpler ones also used in climate models.
However,
the typical grid size of a few hundred kilometers makes the parameterization worse:
all unresolved
scribed
especially While
for oceanogaphers
meteorologists
pressions),
- not only small
I also underlines
instance cell
the possible
in Figure I (taken from WOODS,
the synoptic
disturbances
development their
quantities
on
within
de-
Figure
also means detecting
of thunderstorms
on mean
eddies.
will always be needed
parameterization
of a number
effect
1985).
(mld-latitude
mesoscale
of parameterizations
For meteorologists
and describing
mean quantities
at present
- have to be de-
That this is a basic problem
cannot resolve the corresponding
that a large amount
future,
scall turbulence
at the grid points.
is clearly visible
resolve
oceanographers
the foreseeable
grid
processes
in terms of mean quantities
problem
in for
a horizontal
the grid just
knowing
at the grid.
ATMOSPHERE
Log TRUNCATION GCALE FOR GLOBAL CIRCULATtON MODEL ~O A,D.
Log E
] EDDY RESOLVING TRUNCATION SCALE FOR GLOBAL CIRCULATION MODEL 2000 A,D.
~.
l
-3
UNRESOLVED
-4
Figure
-3
-2
0 C EA N
-1
0
1
2
M(
3~
4
I: Typical spectra of motion in the ocean and atmosphere showing the energy peaks associated with eddies. Existing computers permit a truncation scale of 100 km. By the end of the century it should be possible to resolve the ocean eddies.
122
3.3 Incomplete
A third
Equations
error,
equations. by the full
only
sometimes
tied
This may be caused equations,
three-dimensional
for instance
climate
sphere contribution
to the computer
capacity,
is due to
by the wish to avoid certain processes
models
sound waves,
suffer
incomplete
also described
or just by lack of knowledge.
strongly
Most
from this error as far as the bio-
is involved.
3.4 Test of Climate Models
First of all,
present
climatic
quite well. Since meteorologists
200"
I
conditions
should
be reproduced
by any climate model
maintain a global observing system in the atmosphere
1
I
I
'*
I
I
100-
I
I
I
1
I
......... r.
O.
/J
-100 •
-
'°,,...]
-200
,
120 ° E
180 °
120~'W
.....
DWD NMC " OSU ........ T 4 0
-
-
60°W
$
0°
60°E
HUM ----- T21 ----- CCC ....... N C A R
Figure 2 : 3 0 ° - 60°N mean 500 hPa geopotential, January climatology (in geopotential meters) from six models and two observation analyses. DWD: German Met Service analyses from 1967 - 1985; NMC: US National Meteorological Center analyses from 1956 - 1966; T21 and T 4 0 : 2 versions of the European Centre for Medium Range Weather Forecast model resolving 21 or 40 waves along a latitude circle, HUM~ Hamburg University; CCC: Canadian Climate Centre; NCAR: National Center for Atmo~pherlc Research; OSU: Oregon State University.
123
the most
advanced
- because
emerged from weather test.
Six atmospheric
aging
over
of continuous
forecast models. general
circulation
30 days after having
testing
reached
models typical
two analyses of the 500 hPa pressure surface
To attribute only apply subgrid
distinct
different
scale processes.
factorily
but also contain
Nevertheless,
by all models.
used as climate January
are those which
models,
conditions,
is nearly
i.e. aver-
are compared
impossible,
or omit different
the main troughs
to
since they not
parameterizations
and ridges
Please note that all six socalled
spheric general circulation
models
(~ 5 km height).
errors to distinct models grids
- climate
Figure 2 from STORCH et al, 1986, shows such a
for
are given satis-
climate models
are atmo-
models with a very crude account of the ocean surface or
surface layers only.
4. RESULTS FOR THE NATURAL SYSTEM
The distinction
between natural and perturbed
because
mankind
surface
albedo as well as evaporation
is part of the natural
tion of the atmosphere.
Despite
climate system is somewhat
artificial,
system and has since many years changed and since
industrialization
the
also the composi-
this dilen~a we speak of climate model results
for
the natural system if no specific man-made changes have been imposed.
Although all modellers know that a climate model should include all relevant compartments and
a tested
ocean
climate
coupled
models
are
ocean-atmosphere coupled
of single components
in small
errors of momentum,
large deviations
A first
result
driven
is still not available.
is a model
climate
less realistic
by a fixed other component.
heat and substance
If atmosphere
The reason
exchange at the boundary
as the is found
leading
to
in the coupled system through new feedbacks.
step towards
persistent
the
model
surface
coupling
is shown
in Figure
temperature
anomaly.
Imposing
3 for the biggest ocean
surface
in the 20°S to 20°N area of the Pacific Ocean onto an atmospheric
general
circulation
model
the
resulting
difference
between
observed
Darwin,
(a measure for the intensity of the EI-Nino phenomenon)
observed
Australia,
and
compares well with the
(BIERCAMP et al, 1986).
A further step, a combination tion model model
pressure
the
and most
temperature
Tahiti
ocean
tested
of the Hamburg University
(emphasis on cloud formation)
confirms
(see Fig.
4 and Table
atmospheric
general circula-
with an ocean deck layer model and a sea ice
I), that at least global
variety of climate parameters compare well with observations.
and zonal means of a
124
0
-3
-6
t970
1975
1980
Figure 3: Monthly mean surface pressure difference 1983; observed (--) and calculated ( m , ) .
1
: : Jan-Climatology Jan - Simulation " 90 days
"E100-
Darwin
/ 1100
80-
-8O
60-
- 60
E
6o-
minus
Tahiti
for
1970 to
4-+ Jan-Climatology Jan -Simulation " 90 days
5040
30
%0
40-
20
2Ot 0
i
10-
~
~
~
,
,
,
~
,
i
-90-70~50-30-10 10 30 50 70 90 Latitude
0
0
I
I
I
I
-90-70-50-30-10
i
I
I
I
i
10 30 50 70 go Latitude
Figure 4: Simu&ated ( - - ) versus observed (*~-~) zonal mean cloud cover for January as a function of latitude, left part: total cover, right part: low clouds.
125
Since some modelling
groups are strongly engaged in the coupling of models we will
soon have a variety of different results which hopefully will lead into a separation of
that
part
of
anomalies
to
be
forecast
and
another
part
constituting
climate
'noise' for instance within the time domain year-to-year variability.
Table I:
Calculated and observed global mean climate variables
Simulation Experiment
Surface temperature (°C)
Precipitation (mm/day)
Cloud Cover (%)
Planetary Albedo (%)
"15.0
2.62
53.9
28.0
4.0
First 10 years coupled
14.7
2.59
53.3
27.9
4.11
Second 10 years coupled
14.6
2.49
53.2
27.8
4.67
Observation
14.9
2.74
52.1
Fixed boundary eonditions~
Sea Ice Cover (% of Earth Surface)
~ 30
4.8
5. RESULTS FOR THE PERTURBED SYSTEM
A big part of the recent research effort in climatology is due to the observed regional and the anticipated global impact of man's activities on climate. Man's impact may be threefold:
-
change of surface parameters like reflectivity, evaporation and roughness change
of
the atmospheric
composition
by emission of trace gases
and aerosol
particles -
waste heat.
Only the emission of trace gases into the atmosphere has been discussed long enough in order to give a somewhat consolidated picture; all the other man made changes are either not tackled in enough detail as for instance surface roughness change or are still not of global impact as waste heat.
126
5.1 Reaction
to an Increased Trace Gas Co~tent
Although the answer to the question: ed atmospheric model,
all
models
either
greenhouse
the
simplified
public
debate
uncoupled
one-dimensional
effect by a number
oxide
(COz),
ordered
methane
according
(CH~),
increased
temperature
greenhouse
gas? Will
nearly
There
response
also the
increased
concentration
(CFM's),
at an anticipated
increase
type
of
the
and
cloud
respond to changed atmospheric
and lower
of
cover
if stating:
(carbon di(N20)
water
Does slightly
vapour,
be changed? Surely,
the
main
How does
the
a part of these
general circulation models and
a doubling of CO2-content
will lead
1.5 and 4.5 K, with stronger warming
warming
if
What is still a matter
circulation?
atmospheric
increase between
and stronger
for very
dinitrogenoxide
increase).
concentration
clouds
has been reached
of climate
- or coupled
of the global water cycle to this stimulus.
to a mean global temperature high latitudes
models
is no reason to discard an increased green-
is answered by three-dimensional
a consensus
on the stationary
circulation
chlorofluoromethanes
to importance
is the reaction
questions
models.
based
general
of gases with observed
of debate
ocean circulation
is still
- as for
house
How will the climate system react to an increas-
effect? should be given by a time dependent coupled climate
in winter than in summer
in
in low lati-
tudes.
Since the regional by comparison
pattern
to present
is not well simulated
climate,
as shown during climate model tests
mainly because of the poor space resolution
km) of most climate models, only zonal mean changes are depicted
Because the additional al results trace
gases
greenhouse
if compared will
not
in Figure 5.
effect of a well mixed gas leads to nearly identic-
to the CO~ result, change
(- 500
the general
the combined temperature
action of the aforementioned
increase
pattern.
Recent esti-
mates of future trace gas levels based on presently measured growth rates
(RAMANATHAN
et al, 1985) have pointed to an equal importance of COs and the other trace gas group for the next 50 years.
When shall climate
the doubling
model
agricultural the
activity
1972 forecasts
ocean model's
of CO2 occur?
is needed.
for
reaction
The
future
of mankind
In order energy
to answer
consumption
has been estimated
1985, with no success.
this question and
industrial
frequently,
Therefore,
but,
more as
than a well
as
if we remember
Figure 6 displays a global
to a wide range of fossil fuel use patterns.
The main result
of this global ocean model is: The CO2 uptake of the ocean is inversely proportional to CO2 growth
rates, or in other words, halving
time needed for a distinct concentration
the growth rate more than doubles the
increase.
127
11
I
10
I
1
t
1
I
,
I
11 ° C
10
a = M a n o b e and Wetherald 1980 b : Schtesinger 1982 d = W a s h i n g t o n and Meehl 1983 clouds calculated c = W a s h i n g t o n and Meehl 1983
9 8
9 8
7 6
6
<~5
5
i--
4-
k
/
4
3-\L
3 "~- ._ . - E - J ' - A
2 -
t,.
~ / ....
90N
, v..-- \ \"
"%
I
I
I
70
50
30
z:.
2
Z:2"I
I
1 1
1ON 10S 30 Latitude
I
I
50
70
90S
Figure 5: Zonal mean temperature increase due to a doubled CO2 content as a function of latitude for different three dimensional models.
Atmospheric
C0 2 Content
j.6-~\ 1750
"ppmv/l~ \\&
growth rate - - - -
+4,3%1y
no growth . . . .
constant÷2'3%/Y
/i F'-...,, ...,... ,,oo /.,/ 1250
iil60
~.~j.~ 2 , present level
500 29:
! 2000
i 2200
t 2 400
~ 2600
I 2800
Figure 6: Atmospheric COz content as a function of time for three fossil fuel use growth rates. The strongly different percentage remaining in the atmosphere shows the importance of the ocean uptake and its relation to growth rate.
128
5.2 Atmospheric
The radiative Therefore,
transfer
relevant
is ozone,
reactions
and
ozone levels
mensional
creases
as
complicated
well
mainly
Their
ozone
as
The outstanding
the result
vertical
transport.
resolution)
result
is:
Upper
The
there
is
total
ozone reduction
increases
compensation
of
mainly
both
due
effects,
substance
simulation
increasing
growth
to the
in this
nitrous
chemistry
is
have
concentration
(their
oxide
southern
future
from one- or two-di-
ozone
rates
of
in concentration
models of atmospheric stratospheric
on chlorofluoromethane
substances
of a large number of chemical
the catalytic compounds for ozone destruction)
concentration no
being
by trace substances.
from precursor
than for C02. Up to now only results
main
depending
determined
with a number of gases
(vertical and latitudinal
struction delivering ic
to climate may be important.
horizontal
published.
is mainly
formed in the atmosphere
its concentration
in an atmosphere
thus far more
been
in the atmosphere
also gases or particles
not directly respect
Chemistry and Climate
de-
photolytic
de-
while tropospher-
emissions.
hemisphere
Globally
expecting
more
than the northern hemisphere.
5.3 Possible Effects of Aerosol Particles
For
aerosol
particles
it is more
scribe their behaviour.
-
-
difficult
Additional
They are used as condensation
varying
gases
to de-
exist for aerosol particles:
conversion due to coagulation
they settle depending on size.
No three-dimensional two-dimensional transfer al,
strongly
nuclei
they are formed through gas-to-partlcle they change their size and composition
-
than for
important processes
code
1984).
model
global
of aerosol
tropospheric
in order
transport
aerosol
to show radiation
It accounts
is known
transport
to the author.
model
coupled
However,
to a radiative
flux changes has been developed
for all the processes
mentioned,
a
uses anthropogenic
(GRASSL et particle
input and allows for sulfuric acid formation from sulfur containing gases.
Due
to the complexity
two
dimensions
shown by NEWIGER
the
of the processes
results
(1985),
should
be
involved
and also due to the restriction
interpreted
(Fig. 7), the possible
flux changes of the same order of magnitude
very
carefully.
Nevertheless,
to as
albedo change leads to net radiation
as for a doubling of CO2 but of opposite
129
o~ 3 ,<
2
0
I
I
I'
I
I
I
60
90
I
I
I
30
0
LATITUDE ['l Z
2-
.< <1
1
0 - 0.5 9O
I
!
....I.....................i .......... l'"
0
3O
I
i
6
LATITUDE ['1 Figure 7: Albedo differences caused by present amounts of anthropogenic aerosol particles during summer (upper part) and during winter for particles with 20 % soot content (---) or without soot. The shaded area shows the gas-to-particle conversion effect.
sign. The result is very sensitive to the soot content and the change in aerosol size distribution. cloudless
A big
areas
part
alone,
of
but
the
albedo
is caused
change
is not due
by the change
to aerosol
in optical
particles
parameters
in
of water
clouds, which was also taken into account.
The effect at the possible
present
anthropogenic
it compared to the trace gas effect. However, ed easily and soon.
aerosol load is too big to discard
consolidated
results will not be reach-
130
The influence of stratospheric Since they are dominated
aerosol particles
by volcanic
eruptions
on climate
is no longer questioned.
the anthropogenic
influence
should be
small.
6. STRATEGY FOR A DETECTION OF CLIMATE CHANGE
The strong
natural
contribution
variability
in climate
of climate
parameter
does not mean a stop of a possible of
for
instance
(deep ocean, of
trends.
even
a detection
of the man-made
a stop of emissions,
however,
climate change because of the long residence
chlorofluoromethanes
ice sheets),
has hampered Since
and
the
slow
there has to be searched
reaction
of
climate
time
components
for a strategy of rapid detection
a change.
l Ways to Detect a Man-Made Climate Change I Time Series of Trace Substances -paleo d a t a - present obser rations - estimates
l
for
the future
I
Global Climate Data l ~posfandpresent} I -Temperature / I - Precipitation
t - Wind
Analysis of Climate ChQnge Assumptions - Observations - areas with --~ maximum man made signal betterns - Model
for Wesentor past
ctimote
/
1
Insolation | I - Ocean Tempera- l j lure J
I -
specific
Comparison {post~d present) I
-
volcanoes
~---
orbital
| J
parameters
is
proposed a
for
man made contribution
Figure 8: Ways to detect a man made climatic change
principle
Contribution
I Global Geophysical Data | -
The strategy
~'~ 'SipQrat ion
by many scientists
combination
of
(schematic)
is schematically
observations
with
displayed
modelling.
The
in Figure 8. The observations
must
131
- besides direct global climate observations - also include paleo data both for trace substances and surface parameters as well as geophysical
data on volcanoes.
Before
model results are given for perturbations they have to be tested with present climate and/or paleo climate. After
successful tests specific anticipated perturbations have
to be applied to these models showing us and specific patterns of a
areas of maximum
relative man-made signal
result of a perturbation. These hints then should be used
to look with more intensity into observations in these areas or for these patterns.
The present uncertainty
o f the effect of mankind on climate should,
not be the reason for no measures or precautions,
in conclusion,
instead we should give some proba-
bility also to the worst case.
REFERENCES
BIERCAMP, J.; LATIF, M.; von STORCH, H.; WRIGHT, P.B.; 1986: Preparational studies for coupling an oceanic and an atmospheric GCM; Research Activities in Atmospheric and Oceanic Modelling, CAS/JSC Working Group on Numerical Experimentation, Report # 9, 8.29 - 33, WMO, Geneva. GRASSL, H.; LEVKOV, L.; NEWIGER, M.; REHKOPF, J.; 1984: Untersuchung des Einflusses anthropogen-bedingter Aerosolbildungen auf das Klima, Forschungsbericht 104 02 621, Umweltbundesamt, Berlin, 148 p. MAIER-REIMER, E.; HASSELMANN, K; 1987: Transport and storage of CO2 in the ocean - an inorganic ocean-circulation cycle model, submitted to Climate Dynamics ~r 63-90. NEWIGER, M.; 1985: EinfluB anthropogener Aerosolteilchen auf den Strahlungshaushalt der Atmosphere; Hamburger Geophyslkallsche Einzelschriften, Reihe A, Heft 73, 87 p. RAMANATHAN, V.; CICERONE, R.J.; SINGH, H.B.; KIEHL, J.T.; 1985: Trace ga~ trends and their potential role in climate change, J. Geophys. Res. 90, D3, 5547 - 5566~ ROECKNER, E.; LOWE, P.; BIERCAMP, J; 1986: Climate simulations with a simple coupled model; Research Activities in Atmospheric and Oceanic Modelling, CAS/JSC Working Group on Numerical Experimentation; Report 9, WMO, Geneva. STORCH, H. v.; ROECKNER, E; CUBASCH, U.; 1985: Intercomparison of extended range January simulations with general circulation models: statistical assessment of ensemble properties; Beitr. Phys. Atm. 58, 477 - 497. WMO, 1984: Scientific Plan for the World Climate Research Programme; WCRP Publ. Set. No 2, Geneva. WOODS, J.D.; 1985: The World Ocean Circulation Experiment, Nature 314, p. 501 - 511.
SENSITIVITY OF PRESENT-DAY CLIMATE TO ASTRONOMICAL FORCING
Ch. Tricot and A. Berger Unlversit~ Catholique de Louvain Institut d'Astronomie et de G~ophysique G. Lema~tre 2 Chemin du Cyclotron 1348
Louvain-la-Neuve
i. Introduction
Many mechanisms have been suggested to explain the glacial-interglacial during Quaternary (Berger,
cycles
1979b). The "astronomical" or "Milankovitch" theory in-
volves variations of solar radiation available at the "top of the atmosphere" (called
here
extraterrestrial
insolation)
(e.g., Milankovitch,
due
1941;
to
the
secular
Berger et al.,
perturbations
of
the
Earth's
orbit
1986).
Basically, the astronomical theory assumes that the surface air temperature
1984; Berger and Tricot,
is directly related to the insolation available at the Earth's surface for a completely transparent atmosphere and that the climate is sensitive to the changes in the distribution of that insolation among latitudes and seasons. The northern high latitudes are thought to be the most sensitive regions because of maximal continentality of these regions. Recent astronomical computations (Berger,
1984; Berger and
Pestiaux, 1984) allow to claim that these variations of solar insolations are accurately known over the whole Quaternary and provide us with a well defined external forcing over this time period. The power spectra of the insolation variations display typical characteristic peaks near 40,000, 23,000 and 19,000 years,
the first
being associated with obliquity of the Earth's axis and the latter two with precession of the longitude of the perihelion (Berger,
1977). Since the pioneer paper of
Hays et al. (1976), the astronomical theory has been given substantial support. That paper and others Imbrie,
1980)
in the
domains
frequency
(Pestiaux et al.,
show that the Earth's
1987)
and
time
(Imbrie
and
orbital parameters play an important
role in determining the succession of glacials and interglacials over the last million years.
However, there still remain difficulties in explaining how these changes in insolation could be sufficient to initiate or end glacial periods. Since a few years numerous climate models have tackled this problem but without definite conclusions. Simple climate models (e.g.,
energy balance models) cannot take into account
in a
133
correct the
way all relevant
other
hand,
give results plexity.
which
Also
seriously
more
are often
solar
work,
radiative
the
to analyse
in the
general
because
large amounts
itself.
some results
of their
able
(Lenoble,
cause
the radiative 1977).
depletion
by gases, transfer
the
climate
(called
absorption and scattering
the
received
of
solar
Finally,
at
aerosols
radiation
the
solar
In sec-
in section
the Earth's
and clouds.
in
surface
and compared
technique
SOLAR) to handle
the
atmosphere
A number of techniques
in an absorbing-scattering
Ozone
is assumed
(Joseph
atmosphere et al.,
the interactions
to
are avail-
1976) was used
is assumed. The first
Three
the second
from the pres-
similar layers
only in the upper
are
layer,
within each layer.
approximation,
the scattering
parame-
layer containing more than one scattering or absorbing compo-
to give a set of effective parameters single
scattering albedo
for the layer. The effective
and asymmetry factor are given,
respec-
tively, by
~e = ~ 6i
are
between molecular
level of the cloud bottom,
to be present
In order to be used in the delta-Eddington
optical thickness,
of
is described.
are presented
top to the top of the atmosphere.
sky.
ters for a homogeneous
be
transfer
while H20 , C02, 02 and aerosols are distributed
nent are combined
can
insolation.
from the surface to the pressure
for a clear
the changes
surface
parameterization
a three-layer model of the atmosphere
level of the cloud
the im-
than the changes in the extra-
layer is filled up with an averaged cloud and the third layer extents sure
in-
in a simple way.
For our application,
defined
changes
Indeed,
and by
are discussed.
insolation
The delta-Eddington
in our parameterization
layer extents
two
of the shortwave radiative
that
time, which
climatic
changes.
the present state of the atmosphere
and scattering
with
the
of the extraterrestrial
2. Parameterization
deal
of
On
models)
inherent com-
of computer
used in our computations
for present-day
long-term variations
factors
section
system.
forcing by examining
the atmosphere
for climate studies In
climate
circulation
system.
upon the insolation
within
in the atmosphere
into account
absorption
absorbed
insolation
to the variations
The
attenuation
more relevant
transfer
tion three,
taking
less simple
acting
(e.g.,
we focus on the astronomical
radiation
physically
terrestrial
four,
models
response of the climate
pact of the atmospNeric
thought
feedbacks
limit their practical use for the study of long-term
In the present
the
and
numerical
the use of 3-D models requires
cluding the transient
in
processes
complex
(i.I)
134
=E
e
6
sc,i
/8
(1.2)
e
(1.3)
ge = E 6sc,i gi/6sc where 6sc,i = component i, 6sc
is the optical
mi6i
thickness
due to scattering by the atmospheric
= ~e6e ,
6i' ~i and
gi are respectively
the optical
thickness,
the single scattering
albedo and the asymmetry factor of the atmospheric component i.
According considered. Rayleigh
fo Fouquart
(1986),
Parameterizations
scattering
one
single spectral
of the spectrally
and of the spectrally
(0.25-4~m)
will be
thickness
for the
averaged optical
averaged
and 03 are used in each layer. For a gaseous
interval
transmittances
for H20 , C02,02
absorber, the optical depth in a layer
j is given by
(2)
A6j = - Dj In [Trj+ 1 / Trj]
where Tr. is the transmittance due to the gaseous amount encountered by the solar J radiation from the top of the atmosphere to layer j, before entering it, Tri+ 1 is similarly defined but includes the gaseous
amount encountered
in the
layer, ~] is the cosine of the effective
zenith
angle of the solar radiation going
through the layer j.
This definition of the gaseous optical depth for a layer follows the method proposed by Braslau and Dave (1973). In our scheme, downward radiation in each layer is treated as a collimated
radiation with the same zenith angle ~s the incident
ation at the top of the atmosphere, same assumption below
the cloud
radiation
is made for the upper layer, while the downward radiation is assumed
to be diffuse,
in both cloudy and clear
the same being assumed
conditions.
For diffuse
into and
for the upward
radiation,
the angular
integration to obtain fluxes is avoided by using the diffusivity approximation radiation
is assumed
radi-
in the case of a clear sky. For cloudy sky, the
: the
to be collimated with a zenith angle chosen here equal to 53
degrees.
The
delta-Eddington
single homogeneous
approximation
gives
layer with fixed values
the reflection of
and
transmission
de, w e and ge" To combine
of a
the three
layers, we use the ascending method presented by Fouquart and Bonnel (1980, see also Fouquart,
1986).
Starting
from the
layer
near
the surface and
accounting
for the
135 +
combined reflectance of the underlying layers, the reflectance Rj+ 1 at level j+l and the
transmittance T~J of layer j are given by : +
~
,,
Rj+ 1 = Rj + R- Tj Tj / (1 - R'~R-)]
and T7 2
(3)
= T. / (i - R~R-) J J
(4)
where Rj• and T.J are the reflectance and transmittance of the isolated layer j from above and given by the delta-Eddington parameterization, R
is the combined reflectivity of the underlying layers,
R.J and T.J are the reflectance and transmittance of the isolated layer j from below and given by the delta-Eddington parameterization.
The denominator tween
the
layers
layers.
is
of Eqs This
large which will
bright surface (e.g., (4)
between
account
the
that
the
(3) and
(4) accounts
term is especially typically
be
reflectance
and and
reflections
the reflection
be-
between
the case when a thick cloud overlies
a snow covered surface).
reflectance
for the multiple
important when
transmittance transmittance
The differentiation from above of
and
below
a homogeneous
a
in Eqs (3) and takes
isolated
into layer
depend on the incidence direction of the solar radiation. For the layer near the surface, R- is equal to the surface reflectivity or surface albedo.
Assuming a downward numbering of the layers, the downward and upward fluxes at level j are given, respectively, by : +
+
FN = Ftop
j-I
k~l Tk
+ + + F. = F . R. J J 3
+ where Fro p is the downward extraterrestrial solar radiation.
Finally, partial cloudiness is treated by calculating the fluxes separately for a clear and
a completely
overcast
sky and
fluxes by the clear and cloudy fractions :
F. = (l-n) (Fj)clea r + n(Fj)cloudy
then weighting linearly
the respective
136
F'+ ( l =- n ) j
(F~)clear + n(F~)cloudy
with n the fractional cloud amount.
3. Present-day insolation pattern at the surface
This section provides the annual variation of mid-month daily insolation at the Earth's
surface for the present-day
atmospheric
undertaken with the solar parameterization tion and were performed
conditions.
The computations were
(SOLAR) introduced in the previous sec-
for each latitude from 90°N to 90°S by step of i0 ° and for
each month.
To obtain "mid-month" insolation, a constant
increment of the true solar longi-
tude % is used starting with % = 0 at vernal equinox. The mid-month values are thus defined by A% = 30 ° and they are located now around the 20th of each month. Because the length of the astronomical
seasons is secularly variable
(Berger,
1978), these
mid-month values are not related to a fixed calendar date in the past. However,
in
all the computations presented here, we will consider mid-month insolation as representative of monthly mean insolation.
Figure l.a shows firstly the distribution of the daily incident solar radiation at
the
Earth's
expressions
of
surface this
for
a
transparent
extraterrestrial
atmosphere.
insolation
depend only on the duration of sunlight,
are
the latitude,
distance between the Earth and the Sun and
The
given
classical in
Berger
analytical (1978)
and
the solar declination,
the
the value of the solar constant
(taken
here equal to 1368 W m -2 following Willson et al. (1981)). Since the Sun is presently closest to the Earth in January (during northern hemisphere winter),
the distri-
bution of solar energy is slightly asymmetrical and the maximum radiation received in the southern hemisphere is greater than that received in the northern hemisphere. The maximum insolation occurs at summer or winter solstice at either pole owing to the
long
solar day
(24 hours).
The
annual
variations
are maximum
at poles
and
decrease regularly to the equator.
To estimate the atmospheric attenuation and compute the solar energy incident at the Earth's
surface,
our computations,
some atmospheric and surface characteristics must be known.
we have used
the
zonally and monthly
In
averaged fields of surface
temperature and humidity published by Oort (1983). The vertical profiles are deduced from the surface values with Rennick's for temperature
and humidity,
(1977) and Smith's
respectively.
An
effective
with monthly cloud amount from Berlyand and Strokina
(1966) parameterizations single
cloud
is assumed
(1980). For each latitude,
the
137
J
F
M
A
M
J
J
A
S
0
N
D
20-
,7,
200
oi
4060-
-80
80-
J
F
M
l
M
-aJ
LU
J
J
l
S
0
A
S
0
N
0
MONTH F
M
A
M
J
J
N
D
8
B0
6
60
4
40
2
20
F--
0
,,~ -Z ,-t -4
2040-
-5
60-
-8
80-
J
F
M
A
M
-b-
J
J
J
A
S
0
hi
D
A
S
0
N
0
N
D
MONTH
F
M
A
M
J
J
81 61
LLI
41
2~ I-
1,-,,~ -2 ,,.,.I -4 -6 -8
J
F nC~
M
A
M
J
J
A
S
0
MONTH
Figure i. The daily variation_~f present insolation as a function day of year in units of Wm (a) at the top of the atmosphere, the surface, (c) absorbed at the surface.
of latitude and (b) incident at
138
cloud position and optical thickness are kept fixed through the year. The values of these parameters are adopted from Chou et al. (1981), except that their values for cloud optical thickness were diminished by approximately 2 between 20°N and 20°S to reduce the differences between computed and observed planetary albedos. For cloud, the asymmetry factor g is taken equal to 0.85 and the single scattering albedo is related
to the
fixed optical
thickness
following Fouquart
and Bonnel
total ozone amount is computed following the parameterization (1979)
and
the CO 2 concentration
models were considered
: maritime,
is
fixed
to 330
The
given by Van Heuklon
ppmv. Three
continental and unperturbed
(1980).
kinds
of aerosol
stratospheric,
fol-
lowing WMO (1986). D. Tanre (personal communication) provided us with the spectrally averaged scattering parameters 0.55 ~ m were taken
for each aerosol model and the optical thickness at
from a WCRP report
profile for aerosols
(WMO,
1981). The
is calculated by weighting
zonally averaged vertical
the vertical
profiles above ocean
and continent (except at 80=S and 90°S, where the aerosols were considered as mar i time aerosols
in the planetary
boundary
layer).
Finally,
the surface
albedos were
interpolated from data given by Robock (1980).
Contrary
to
the case of a transparent
atmosphere,
daily
insolation W at the
Earth's surface, when the atmospheric attenuation is taken into account, requires a daily integral to be solved numerically. This integral can be written as follows :
H0
r)2 I
w = so ( T-
-HO
Tr(cos z) cos z dH
(5)
where S O is the solar constant,
H 0 is the hour angle of the sunrise, r
m
is the mean distance from the Earth to the Sun,
r is the actual distance between the Earth and the Sun,
z
is the solar zenith angle,
Tr(cos z) is the atmospheric transmission,
function of cos z and computed by
the solar model.
To save computer time, the daily insolation given by Eq.(5) can be approximated by using a daily mean atmospheric transmission which allows Eq.(5) to be written : r m )2 W = SO ( ~ cos z Tr
(6)
139
H0 with cos z = I
H0 cos z dH / ;
-H 0
dH -H 0
and Tr is computed using an effective daily mean cosine of the zenith angle, (cos
z)
.
Our SOLAR parameterization was compared with a version of the more sophisticated SUNRAY model developed by Fouquart and Bonnel (1980). Both solar models use the same Rayleigh parameterization and gaseous transmittances. Multiple scattering is solved in the two models by the delta-Eddington
scheme. However,
in SUNRAY,
the interac-
tions between scattering and molecular band absorption are taken into account by an extension of the photon path length distribution method, which deals with the scattering and absorption processes separately. SUNRAY also uses a higher vertical resolution (between i0 and 15 layers).
In order to illustrate this comparison, Table 1 gives, for some selected latitudes, the mid-month daily insolation absorbed at the Earth's surface for January and July as computed with Eq.(5) using SUNRAY (row A, daily integration) and SOLAR (row B, daily integration) and with Eq.(6) using SOLAR (row C). The comparison of rows A and B shows that SOLAR overestimates tion at the surface.
slightly and systematically
The closest agreement
between
the daily insola-
rows A and C was
obtained by
using (cos z)* given by Eq.(7), except for the highest latitudes where the differences are similar to these between rows A and B :
(cos
z)
= 0.75
cos
z + 0.25
cosmax
(7)
where cosmax is the cosine of the daily minimum value of z, i.e., its value at solar noon, given by cos (~-~) with ~ the latitude and 6 the solar declination.
Similar
results were obtained for the other months.
Figures l.b and
l.c give the distribution
of, respectively,
the present daily
incident and absorbed solar radiation at the surface computed with SOLAR and using Eq.(6) with (cos z)
given by Eq.(7). The difference between these two figures
is
due to the surface albedo which defines the fraction of the incident solar radiation reflected by the surface. The impact of this factor is particularly visible in summer high latitudes. Furthermore, the difference between the extraterrestrial insolation pattern (shown in Fig. l.a) and the pattern for the incident solar radiation at the surface (Fig. l.b) is related to the atmospheric
transmission
which decreases
from low to high latitudes mainly as a result of the increase with latitude of the
140
averaged tion
solar zenith
angle
at the surface
maximum
at the
the maximal
and of cloudiness.
is located
summer
pole
insolation
now
in the
In consequence,
tropics
in the northern
during
hemisphere.
is yet above Antarctica
the maximal
summer with
insola-
a secondary
In the southern hemisphere,
during summer with a secondary maxi-
mum in the tropics.
Table i. Mid-month daily insolation absorbed at the Earth's surface for present January and July at some selected latitudes. Results of row A are obtained with a version of SUNRAY (Fouquart and Bonnel, 1980) integrated through the day. Results of row B are obtained with the present SOLAR parameterization integrated through the day and results of row C are given by SOLAR using the effective daily averaged_~osine of the zenith angle as defined by Eq.(7). The insolation is given In Wm
Latitude
A
B
O
SUNRAY
SOLAR
SOLAR with (cos z)
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
January
.
.
.
.
.
.
.
July
.
.
.
.
.
.
.
.
.
.
.
.
.
.
January
July
January
July
70
0
145
0
149
0
141
50
28
201
29
205
29
201
30
116
246
118
250
117
248
0
221
211
224
214
221
211
-30
270
113
274
115
271
113
-50
229
34
232
35
227
35
-70
121
0
126
0
119
0
To test we have solar al.,
further
compared
radiation 1982;
the reliability
these
results
as computed
Ou and Liou,
by some other
1984) and with
deduced from satellite observations These comparisons,
of our computations
to the zonal
not presented
the
zonal
for present-day of the
(Stephens averages
annual et al.,
conditions,
mean 1981;
of planetary
absorbed Peng et
albedos
as
et al., 1981; Hartmann et al., 1986).
show that our model calculations
rally well with all these other theoretical
4. Past insolations
authors
(Stephens
here,
averages
agree
gene-
the changes in the absorbed
solar
studies and observations.
at the Earth's surface
In order to estimate
past climatic variations,
radiation by the atmosphere
and at the surface can be thought
physically more rele-
141
vant than the changes
in the extraterrestrial
these variations
the last million years requires
over
and the atmospheric constituents
radiation.
An accurate
the surface
estimate of
characteristics
to be known over this period. A well known example
concerns the surface albedo, which is directly related to the ice sheet growths and decays and the sea ice extent, mainly around the Antarctica. Qualitative estimates of the changes of some parameters begin to be available for some particular regions and times of the Quaternary period (e.g., Jouzel et al., 1982; Lorius et al., 1984; Kutzbach and Street-Perrott,
1985). However,
continuous and global time series re-
cording the variations of these parameters are far for being yet available. Nevertheless, some sensitivity studies could be undertaken as a first step to the simulation of past insolations and climates.
In this section, the variations of the absorbed and incident solar radiation at the Earth's surface are computed for a period extending from 200 kyr BP (Before Present) to present, keeping all atmospheric and surface characteristics equal to their present-day values.
Thus
such computations
must be viewed
only as a
sensitivity
study of the present-day climate to the well known variations of the astronomical forcing. A similar work was undertaken by Ohmura et al. (1984) but using a simplified
form
of the global atmospheric
transmission
and
considering
only the solar
radiation incident (not absorbed) at the surface.
According
to
the
Milankovitch
astronomical
theory
of paleoclimates
(Milanko-
vitch, 1941), ice ages are initiated when cool summers in high latitudes can lead to the persistence of the snow field over the continental areas all years long. Therefore, we will only discuss in this study the variations of insolation for July (summer for northern hemisphere) and January (summer for southern hemisphere). For this application, the approximation given by Eq.(7) has been further validated at 125 kyr BP and 115 kyr BP again by comparing for these periods the results given by SUNRAY and SOLAR. During these two particular periods, July experienced the largest positive (at 125 kyr BP) and the largest negative (at 115 kyr BP) deviations from present-day conditions
for
the northern hemisphere
over
the last 200 kyrs
(Berger,
1979a). Table 2 gives the mid-month absorbed solar radiation at the Earth's surface for some particular latitudes
and for January
and July as computed by SOLAR with
Eq.(7) and by SUNRAY with a daily integration. As for present-day computations, both models agree very well which reinforces our confidence to use our parameterization for the whole period considered here.
142
Table 2. Mid-month daily insolation absorbed at the Earth's surface in January and July for 125 kyr BP and 115 ky= BP at some selected latitudes. SUNRAY is based on the faodel of Fouquart and Bonnel (1980) integrated through the day while SOLAR is the present parameterization used with the effective daily averaged cosine of zenith angle as defined by Eq.(7). The insolation is given in Wm- . 125 KBP .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Latitude
.
.
.
.
.
.
.
.
.
.
.
115 KBP .
SUNRAY January
70
.
0
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
SOLAR
July
165
January
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
SUNRAY
July
0
161
January
0
.
SOLAR
July
132
January
0
July
131
50
25
227
25
226
32
188
33
187
30
103
277
103
278
126
233
127
234
0
198
236
197
236
235
203
234
203
-30
243
125
243
125
282
iii
283
Iii
-50
206
37
204
38
235
35
230
35
-70
ii0
0
108
0
122
0
121
0
Figures 2.a and 2.b show the long-term variations of the deviations from today values of solar radiation for January. Each figure represents i00 kyrs. Figures 3.a and 3.b show similar variations for July. In these figures, the top panel represents the extraterrestrial solar energy, the middle one the incident solar energy at the Earth's surface and the bottom one the absorbed solar energy at the Earth's surface. As discussed by Berger (1979a), it is especially interesting to note the large positive deviations for the July insolation at the top of the atmosphere in high latitudes for i0 kyr BP (40 to 50 Wm-2), for 85 kyr BP (40 Wm-2), for 103 kyr BP (50 Wm -2) and for 125 kyr BP (60 Wm-2). All these deviations represent more than i0 per cent of the insolation received today at these latitudes and their timing corresponds closely to the timing of the warm and cold stages of the last 130 kyrs.
The variations of the incident
solar radiation at the surface (middle panels)
show similar pattern but the absolute value of the variations are reduced by a factor 2 at low latitudes and up to around 4 at high latitudes.
In fact,
the atmo-
spheric attenuation reduces the absolute values of the variations as expected, but not the relative ones. For the variations of the absorbed solar radiation at the surface (bottom panels),
the largest deviations are no longer found in high latitu-
des but in tropical and middle latitudes. The latitude reduces
further the absolute values
increase of the surface albedo with of the insolation and therefore its
absolute variations. Because of the high summer surface albedo for latitudes southward of 650S, this feature for the absorbed insolation is especially well pronounced in Fig. 2.a and 2.b.
143
I
l
MID-MONTH
oo "
i
I
I
I
)
INSOLATION
JANUARY
,*8 7u
s o ,-
eo 4S ao IS =
ao Is. o. -~o
:l:O;hlltt 11. il i ) ~/ ,, I I ; / . ' \ ~ , ! l l , - , l l l . ~-,~t'.illW/I.';/,,'~ ! i ~ ? , l t ~ \ \ ' , ~ ' / M , ~ -",~.'// / I I U ' / tl!!lll\tb,d I.!'.l\~/J~,iii' ii ~:/."~l~t'>hh I l l ,-,,'(if l l;/,V.'.:.'.'f, lll \\\~,,o ~',',\\~"A:,?;! ,!i
'
--~5. -co,
-Ts
'
-30 -4S -~0 -TS
-oo,
-oo ,
,
~
,
,
y
,
=
,
,
o,~0)
,
,
,
oqoo
,
,
,
oW
=o,
gO
eo.
~O 4~ ~0 t~ 0
o'
--30 ,.j
.%/,>),
-eo. -7~.
I,
:',
I
o
I
I ;.J__4.' I? \ k ~ ' z ,
tl
J.'_;__(
--eO
-QO 1
,
oo~o
,
~
t
I
o
poo
I
I
~
o
i
,
o~
z)o, 7s' oo. 4~' ao,
90 7G ~D 4~ 30
o,
D
-ao -4S
' '
-co,
-Ts
•
-ao,
~oo
,,,o
,7o i)o.
,.o
~o
~o
, MID-MONTH
,~o
,,,o
~o
~o
,6o
4°
V
40
I°
INSOLATION
,k . . . .
~o
V
'
-O0
'
--90
,~o
7
o
JANUARY
~O 7S; 80 4~ 30
eoaoIsO-30
-
-,Is
-
-eo
-
-?i; -oo
0 --IG
If(All1 I ' ' ' I
LU
eo
I::)
4~ :)o
'
I
°°
I
, °
I
, , .~r-~ I 0
O0
I
~,, ...... '
--00
I° ° ° ° ° i
I
I
I gO 7~; O0 4~ 30 IG
o
0
'" ":° -7g
--30
-
I °
oo ,
I
.
i
e o
~ o
I
,
~ i D
o
o I
--7~; --r~O
o ,
I
I
I
I
60
7~ 60 4~ 30 o.
0
I I / "~ / -~s
t I,.)o'~ ,,.,j i
,"
~"
~._..,,o d
"'
"
•
b I oo
eo
eo
70
YEARS
SO
SO
BEFORE
40
30
aO
I0
-
--00
-
-90
0
PRESENT
Figures 2.a-b. Long-term variations of the deviations (from their present value~) of mid-month daily insolations for January. These values are given here in Wm ~ and for periods extending from i00 to 200 kyr BP (part a) and from present to 100 kyr BP (part b). For each time period, the top panel is for the insolation at the top of the atmosphere, the middle one for the incident insolation at the surface and the bottom one for the absorbed insolation at the surface. The solid lines are for the positive deviations (insolation higher than today) and the dashed lines characteristize the insolation below their present values.
144
"7°
'~, '.° <=<',7,'~ .,,.
=o-
';'°,.'r . . . . . . .
'r.
'~<' <
'~"
'~<~ ':<' <,~,,,.,.%,~,,
'7 <' .
~
eo-
4s-
Qo
w eo 4s ~o
Iso-is -4s
-
-0o
-
ts a
%
1 MID-MONTH i
~ o
INSOLATION
v
I
-
,
o
JULY
I
-Qo
I
I
o
I 0
,
-
~
oo
I
I o
¢BO
imo.
7seo.4g:~,
60 4G ~0
ig-
0 -30
I
o'
I
°'
I
~
1
l,
l
, o
l,,,~.
1
l ~0 7~
7S+ eo"
~0 4~
30" ISO"
30 I15 0
-3o
-
~;30
-eo -TS
" "
--GO
a zoo
~7o
Jeo
,oo,
leo
~o , ~o o'o~°
teo
, .,~°
~so
oooo
oooo
-N
~4o
~o o
~o
~o o
~o
~o o, o
~ so
~o
too
o
,,o
-
oooo gO 7K 60 30 iS 0
o. -,~s, -eo.
-=o.
MID-MONTH I
I
INSOLATION ,~
I
I
~gO
JULY t
I
t
I
I
v
eO 4=
4~ ~0
~0 IS 0
~'~ .j
-~o
-00
-
I
t
f
I
I
t
I
t
1
,
l
o'
I
,,
o gO aO ,4S 30 II; 0 • -30
b I oo
~o
eo
~o
YEARS
~o
so
BEFORE
40
~o
~o
1o
o
PRESENT
Figures 3.a-b. Long-term variations of the deviations (from their present values) of mid-month daily insolations for July. These values are given here in Wm - and for periods extending from i00 to 200 kyr BP (part a) and from present to 100 kyr BP (part b). For each time period, the top panel is for the insolation at the top of the atmosphere, the middle one for the incident insolation at the surface and the bottom one for the absorbed insolation at the surface. The solid lines are for the positive deviations (insolation higher than today) and the dashed lines characteristize the insolation below their present values.
145
To
link insolation
insolation
to climate,
is as important
the variations
(if not more)
as the
of the
latitudinal
local variations
gradient
of
of insolation.
Instead of using a local gradient, as done by Ohmura et al° (1984), a large-scale gradient between
tropical
and high latitudes
has been defined by calculating
the
difference between insolation at 30 ° and 70 ° in each hemisphere. Figure 4a shows the long-term variations of January mid-month insolation over the last 200 kyrs for 30°N and 30°S. Three curves as in figures 2 and 3 (respectively extraterrestrial tion, incident at and absorbed by the surface) are again drawn for each
insola-
latitude.
Figure 4.b gives similar curves for 70°N and 70°S and Figure 4.c gives the variations of the large-scale gradient of insolation, from Fig. 4.a.
obtained by substracting Fig. 4.b
Finally, Figures 5.a, 5.b and 5.c are similar figures but for July.
For 30°N and 30°S, Fig. 4.a and 5.a again illustrate the similar decrease of the variations of the absolute
insolations
incident and absorbed at the surface, when
the atmospheric attenuation is taken into account. These curves are almost identical because of the small surface albedo in the tropical regions. On the contrary, Fig. 4.b and 5.b show the importance of the larger surface albedo at 70°N (around 25 per cent in July) and 70°S (around 55 per cent in January). As mentioned above, the difference between
incident and absorbed radiation is especially visible for 70°S
January because of the relatively larger surface albedo. the variations of insolation in high latitudes
in
In the winter hemisphere,
(70°N in Fig. 4.b and 70°S in Fig.
5.b) are almost continuously equal to zero because the insolation is equal or near zero over the whole period at these latitudes. This peculiarity helps to interpret partially the variations of the large-scale gradient of insolation given by Fig. 4.c and 5.c.
For
the winter hemisphere
5.c) the variation of the gradient
(northern for Fig.
4.c and southern
for Fig.
is given by the variation of the insolation at
30 ° alone (compare the three upper curves of Fig. 4.a with the three upper curves of Fig. 4.c and the three lower curves of Fig. 5.a with the three lower curves of Fig. 5.c). For
the s u ~ e r
hemisphere,
interpret. Interestingly,
the variation
of the
for the north-south gradients
gradient
is not
so
easy
to
in summer hemispheres (Fig.
4.c SH and 5.c NH), the characteristic frequencies of the absorbed insolation at the surface are no more those of the extra-terrestrial sis, Berger
and Pestiaux
(1984) have indeed
shown
insolation. From spectral analythat extraterrestrial mid-month
insolation at 30°N and 30~S presents mainly peaks around 19 and 23 kyrs, while midmonth insolation at high latitudes shows in sun~ner an additional peak around 41 kyrs (such features can be deduced from analysis of Fig. 4.a, 4.b, 5.a and 5.b). As far as the variation
of the latitudinal
gradient is concerned,
Fig. 4.c and 5.c show
that extraterrestrial insolation is characterized by a periodicity of about 40 kyrs, whereas the absorbed insolation exhibits a higher frequency which corresponds
to a
period in the range of 23 kyrs. The incident insolation at the surface follows the
146
MID-MONTH
INSOLATION
JANUARY
o ~ f - ~ 7 1 ~ - - . . . ~ . - ~ f ' ~ ~ _ ~ " - ~ ~
-so ~
~
~
....... "-~'~_.j
~"~/-"~.~j" - - ~ - ~ f - " ~
~ W
_oo
a 200
100 BEFORE
YEARS
MID-MONTH
0 PRESENT
INSOLATION
JANUARY
50
O-
50.
50
O
50
=
~----~,----- ~ , " - ~ w ~ - ~ / ' - ~ , ~
b
,
,
I
J
,
~ f
=
100
200 YEARS
BEFORE
PRESENT
Figures 4.a-b. Lorkg-term variations of January mid-month insolation over the last 200 kyr, in Wm ~, for 30°N and 30°S (part a) and for 70°N and 70°S (part b). The left scale gives the deviation from the present values. For each latitude in both hemispheres, the upper curve gives the variations for the solar radiation at the top of the atmosphere, the middle curve the variations for the incident radiation at the Earth's surface and the lower curve the variations for the absorbed radiation at the Earth's surface.
147
MID-MONTH
INSOLATION
JANUARY
I
t
~0
~,O
50-
O
50 ._/
W
.
.
.
.
.
.
,
.
.
.
.
.
.
.
.
I
200
1OO
YEARS
O
BEFORE
PRESENT
Figure 4.c. Long-term variations of January mid-month large-scale gradient of insolation over the last 200 kyr, in Wm -, for the northern hemisphere (the three upper curves), and the southern hemisphere (the three lower curves). The left scale gives the deviation from the present values of the gradient defined as the difference between insolation at 30 ° and 70 °. For each hemisphere, the upper curve gives the variations for the solar radiation at the top of the atmosphere, the middle curve the variations for the incident radiation at the Earth's surface and the lower curve the variations for the absorbed radiation at the Earth's surface. same
frequency
behaviour
hemisphere
su~er
hemisphere
sunm~er (Fig.
Fig. 4.a means
(30°S) and 4.b
January
northern
southern
hemisphere
difference plained
(Fig.
(given
by noting
in time
about 40 kyrs
5.c)
for
4.c~.
If
- upper
panel
similarity
-
is
at
least
for
the
less
clear
for
the
summer
5.a (30°N) and 5.b
lower
of Fig.
that the successive
panel 5.a
and
maximal
of the location
of these
4.a
5.b),
in the variations
large-scale
latitude.
the available
tudes than in the tropics.
the
4.b
- and
the periodicity
(in
maxima
latitudinal
southern curves
absolute
of (it
July
of
their
can be exvalue)
from
it is the opposite. The induces
gradient.
On
a periodicity the contrary,
of for
the successive maxima are, most of the time, located
gradient
These
are due to the
of
and
panel, respectively)
deviations
relative
insolation
northern
(70°N) are compared
of Fig.
larger at 30 ° than at 70 ° , and sometimes
at 30 ° and the
reduces more
this
radiation,
the extraterrestrial
(70°S) or Fig.
at the surface,
insolations
absorbed
hemisphere
the insolations
tion at that
the
in 4.c lower panel and 5.c upper
today are sometimes change
than
exhibits
larger
larger
atmospheric
insolation
the same periodicity
deviations
from today attenuation
(and therefore
than the insola-
values in high
its variations)
in the
tropical
latitudes
which
at these lati-
148
o
o
.so ] " ~ ' / - ~ " - v "/r'~..__//~" J/-'~'-.~1 ' I a
= ,
MID-MONTH ,
,
,
i
,
i
,
200 YEARS
o
~
INSOLATION i
i
JULY
,
100 BEFORE
v ~
,
i
,
,
i
i
i
,
PRESENT
~
~o" ~ , / - ~ . _ - - ' ~ v ~ , / - ~ ~ ~ ~ 50.
50. MID-MONTH
b
INSOLATION
JULY
I
200 YEARS
100 BEFORE
PRESENT
Figures 5a-b. ~ong-term variations of July mid-month insolation over the last 200 kyr, in Wm -~, for 30°N and 30°S (part a) and for 70°N and 70°S (part b). The left scale gives the deviation from the present values. For each latitude in both hemispheres, the upper curve gives the variations for the solar radiation at the top of the atmosphere, the middle curve the variations for the incident radiation at the Earth's surface and the lower curve the variations for the absorbed radiation at the Earth's surface.
149
o-
so
""~.v,"-"-v
-J'-~...~f-~'- f ' - ~ - ~ j ~ ~ _ j ' - " ~ ' -
~v
"~-'~'-.._~,/'-'~'v' / ' ~ " ~ ' ~ . _ . / ~
~'v
-~o r ~ , , . . . . , r " - ~
v
"~-'~'-_.~'f~'--~-.j/'~" MID-MONTH
~
INSOLATION
|
..........
~
''~'-~
JULY
100
200
O
YEARS BEFORE PRESENT
Figure 5.c. Long-term variations of J u~y mid-month large-scale gradient of insolation over the last 200 kyr, in Wm--, for the northern hemisphere (the three upper curves) and the southern hemisphere (the three lower curves). The left scale gives the deviation from the present values of the gradient defined as the difference between insolation at 30 o and 70 ° . For each hemisphere, the upper curve gives the variations for the solar radiation at the top of the atmosphere, the middle curve the variations for the incident radiation at the Earth's surface and the lower curve the variations for the absorbed radiation at the Earth's surface.
5. Conclusions
According
to the
in the geometry Pleistocene
astronomical
ice ages. Accurate
values
and related monthly insolations Even
if recent
theory of paleoclimates,
of the Earth's orbit
climatic
models,
orbital parameters
have modulated to do
which link insolation
variations
and
the climate during no human
of these
orbital
quantitative,
allow,
of climatic changes
the whole Quaternary
interferences),
to climate variations
in particular,
developed
further.
through time,
elements
show that
(and will
at the astronomical
and more
the
the exact mechanisms
cies are not yet totally known. Both simple models, which would reproduce mic behaviour
of
for the last 2 to 3 million years.
both qualitative
so assuming
the long-term variation
cause of the succession
for the variations
are now available
probably
continue
is the fundamental
sophisticated
frequenthe dyna-
models which
to test the validity of the first for selected dates,
must be
150
In this paper, an analysis of the impacts of the insolation forcing on the insolation available at the Earth's surface has been made by comparing, in the time and frequency domains, variations of the extraterrestrial radiation to variations of the incident and absorbed radiations at the Earth's surface.
Considering the potential
importance of the insolation during summer (which could prevent or allow snow melting), results for July for northern hemisphere and January for southern hemisphere have been
stressed.
The atmospheric attenuation essentially reduces the absolute
variations of the incident solar radiation at the Earth's surface as compared to the variations of the extraterrestrial radiation. Over the last 200 kyr, these two kinds of insolation generally present maximal variations
in high
latitudes
in relation
with the variation of the obliquity. On the contrary, the absorbed radiation at the Earth's surface has always maximal variations in tropical and middle latitudes related to the increase of the surface albedo with latitude.
Finally, in the summer hemisphere, the large-scale gradient of insolation between the tropics and the polar regions shows deviations from its present-day value which characteristic frequencies depend upon the type of insolation considered : (i) for the extraterrestrial
insolation~
the main
frequency of the variations
of the
large-scale latitudinal gradient is about 40 kyrs, whereas (ii) for the incident and mainly
the absorbed
addition,
insolation at the surface the large-scale gradient
quasi-periodicity of
about
23 kyrs;
this
shows,
in
difference is related to the
atmospheric attenuation which reduces more strongly the variations of insolation at the surface in high latitudes than in tropics.
References
Berger, A. 1977 : Support for the astronomical theory of climatic change. Nature, 269, 44-45. Berger, A. 1978 : Long-term variations of daily insolation and Quaternary climatic changes. Journal of Atmospheric Science, 35(12), 23622367. Berger, A. 1979a : Insolation signatures of Quaternary climatic changes. II Nuovo Cimento ~, series i, 2, 63-87. Berger, A. 1979b : Spectrum of climatic variations and their causal mechanisms. Geophysical Surveys, 3(4), 351-402. Berger, A. 1984 : Accuracy and frequency stability of the Earth's orbital elements during the Quaternary. In : A. Berger, J. Imbrie, J. Hays, G. Kukla, B. Saltzman (Eds) : Milankovitch and Climate, pp. 3-40, D. Reidel Publ. Company, Dordrecht, Holland. Berger, A., Imbrie, J., Hays, J., Kukla, G., Saltzman, B. (Eds) 1984 : Milankovitch and climate, D. Reidel Publ. Company, Dordrecht, Holland. Berger, A. and Pestiaux, P. 1984 : Accuracy and stability of the Quaternary terrestrial insolation. In : A. Berger et al. (Eds) : Milankovitch and Climate, pp. 83-112, D. Reidel Publ. Company, Dordrecht, Holland. Berger, A., and Tricot, Ch. 1986 : Global climatic changes and astronomical theory of paleoclimates. In : A. Cazenave (Ed.) : Earth Rotation : Solved and Unsolved Problems, pp. 111-129, D. Reidel Publ. Company, Dordrecht, Holland. Berlyand, T.G., and Strokina, L.A. 1980 : Zonal cloud distribution on the Earth. Meteor. Gidrol., 3, 15-23.
151
Braslau, N., and Dave, J.V. 1973 : Effect of aerosols on the transfer of solar energy through realistic model atmospheres. Part I : Nonabsorbing aerosols, J. Appl. Meteor., 12, 601-615. Chou, S.H., Curran, R.J., and Ohring, G. 1981 : The effects of surface evaporation parameterization on climate sensitivity to solar constant variations. J. Atmos. Sci., 38, 931-938. Fouquart, Y. 1986 : Radiative transfer in climate modeling. In : M. Schlesinger (Ed.) : Proceedings of the NATO-ASI on Physically-Based modeling and simulation of climate and climatic change, Erice, 11-23 May 1986, to be published by D. Reidel Publ. Company, Dordrecht, Holland. Fouquart, Y., and Bonnel, B. 1980 : Computations of solar heating of the Earth's Atmosphere : a new parameterization. Beitr. Phys. Atmos., 53, 35-62. Hays, J.D., Imbrie, J., and Shackleton, N.J. 1976 : Variations in the Earth's orbit : pacemaker of the ice ages. Science, 194, 1121-1132. Hartmann, D.L., Ramanathan, V., Berroir, A., and Hunt, G.E. 1986 : Earth Radiation Budget-Data and climate research. Rev. Geophys., 24, 439-468. Imbrie, J. and Imbrie, J.Z. 1980. Modeling the climatic response to orbital variations. Science, 207, 943-953. Joseph, J.H., Wiscombe, W.J., and Weinman, J.A. 1976 : The delta-Eddington approximation for radiative flux transfer. J. Atmos. Sci., 33, 2452-2459. Jouzel, J., Merlivat, L., and Lorius, C. 1982 : Deuterium excess in an East Antarctic ice core suggests higher relative humidity at the oceanic surface during the last glacial maximum. Nature, 299, 688-691. Kutzbach, J.E., and Street-Perrott, F.A. 1985 : Milankovitch forcing of fluctuations in the level of tropical lakes from 18 to 0 kyr BP. Nature, 317, 130-134. Lenoble, J. 1977 : Standard procedures to compute atmospheric radiative transfer in scattering atmospheres. Proc. IAMAP Radiation Commission, NCAR, Boulder, 125pp. Lorius, C., Raynaud, D., Petit, J.R., Jouzel, J., and Merlivat, L. 1984: Late glacial maximum-Holocene atmospheric and ice thickness changes from Antarctic ice core studies. Annals of Glaciol., 5, 88-94. Milankovitch, M.M. 1941 : Kanon der Erdbestrahlung. Beograd. KSninglich Serbische Akademie. 484pp. (English translation by isra~l program for Scientific Translation and published for the U.S. Department of Commerce and the National Science Foundation). Ohmura, A., Blatter, H., and Funk, M. 1984 : Latitudinal variation of seasonal solar radiation for the period 200,000 years BP to 20,000 AP. In : G. Fiocco (Ed.) : IRS 84 : Current problems in Atmospheric Radiation, Proceedings of the International Radiation Symposium, Perugia, Italy, 21-28 August 1984, A. Deepak Publ., 338-341. Oort, A.H. 1983 : Global atmospheric circulation statistics, 1958-1973. NOAA Professional Paper 14, U.S. Gov. Printing office, Washington, 180pp. Ou, S.S., and Liou, K.N. 1984 : A two-dimensional radiation turbulence climate model. I. : Sensitivity to Cirrus radiative properties. J. Atmos. Sci., 41, 2289-2309. Peng, L., Chou, M.D. and Arking, A. 1982 : Climate studies with a multilayer energy balance model. Part I : Model description and sensitivity to the solar constant. J. Atmos. Sci.., 39, 2639-2656. Pestiaux, P., Duplessy, J.CI., van der Mersch, I., and Berger, A. 1987 : Paleoclimatic variability at frequencies ranging from i cycle per i0,000 years to i cycle per 1,000 years : evidence for nonlinear behavior of the climate system. To be published in Climatic Change. Rennick, M.A. 1977 : The parameterization of tropospheric lapse rates in terms of surface temperature. J. Atmos. Sci., 34, 854-862. Robock, A. 1980 : The seasonal cycle of snow cover, sea ice and surface albedo. Mon. Wea. Rev., 108, 267-285. Smith, W.L. 1966 : Note on the relationship between total precipitable water and surface dew point. J. Appl. Meteor., 5, 726-727. Stephens, G.L., Campbell, G.G. and Vonder Haar, T.H. 1981 : Earth Radiation Budgets. J. Geophys. Res., 86, 9739-9760. Van Heuklon T.K. 1979 : Estimating atmospheric ozone for solar radiation models. Solar Energy, 22, 63-68. Willson, R.C., S., Hanssen, M., Hadson, H.S., and Chapman, G.A. 1981 : Observations of solar irradiance variability. Science, 211, 700-702.
152
WMO 1981 : Aerosols and Climate. report of the Meeting of JSC experts, Geneva, 27-31 October 1980. WCRP report, WCP-12, 60p. WMO 1986 : A preliminary cloudless standard atmosphere for radiation computation. WCRP report, WCP-II2, WMO/TD-n°24.
CAUSES AND EFFECTS OF NATURAL CO 2
VARIATIONS
DURING THE GLACIAL-INTERGLACIAL CYCLES
Ulrich Siegenthaier Physics Institute, University of Bern Bern, Switzerland
i. Observations
Although volume
carbon
concentration
dioxide
is
being only
a minor
atmospheric
0.03 percent,
constituent,
it plays
its
a very funda-
mental role in nature - for the earth's climate via the greenhouse effect,
for vegetation
as
a basic
substance
for photosynthesis.
Varia-
tions of its concentration may therefore have great impact on the environmental
conditions
on e~rth,
several years ago analyses during
glacial
time,
70 % of its recent raised cial
the
the
observational
tions,
age climate. then
the
shift,
In the
evidence
CO 2
level.
(I) what was
CO 2 concentration ice
atmospheric
pre-industrial
two questions:
the
and it was
an exciting
of samples of old polar
and
for
(2)
the
ideas
This
the cause
following,
present-day
level
what
finding when
ice indicated
was
reduced
observation
that about
immediately
of the glacial-postgla-
are
the
I shall
implications
first discuss
glacial-interglacial
about
to
possible
CO 2
causes
and
for
shortly variafinally
the estimated effects on global climate. The principle formation
of
of the ice core measurements
ice by sintering
of cold
firn,
is simple.
air bubbles
During the
are enclosed
which have the composition of atmospheric air. Extraction of the trapped air and analysis in the laboratory chromatography
or of infrared
the concentration if
several
(usually by means either of gas
spectroscopy)
permit
to determine
of CO 2 - and of other trace gases - in these samples
of the ancient atmosphere. only
laser
conditions
However, are
meaningful
fulfilled.
results can be obtained
First,
melting
must
never
154
have occurred
in the ice,
that melting For
this
reason,
below about
because
CO 2 is highly
of ice and refreezing only
-20°C,
ice
i.e.
from
leads
sites
from polar
soluble
to layers
with
a mean
regions,
in water,
enriched
annual
temperature
can be used for such ana-
lyses; at warmer sites, melting events do occur in summer. determination skill.
of the concentration
laboratory
For avoiding CO 2 dissolution effects,
boratory
is done by
ice typically
CO 2
sorption
and
water
crushing
contains
sample of typically addition,
of
in the ice at
100 ml of air or
to
behave
-20°C.
30 ~i o f
desorption and
from walls
CO 2 is
other air components,
indecently
in
occur,
the
special
in the la-
One kg
of polar
CO2,
so that an ice
rather
little CO 2. In
laboratory
especially
preferentially
by water vapour
Second,
requires
air extraction
e.g.
10 g to 500 g mass contains
tends
vapour,
the
so
in CO 2.
systems.
in the
transported,
in extraction
Ad-
presence
compared
systems
to
[Zumbrunn
et al., 1982]. Figure
1 shows
CO 2 concentrations
from Dye 3, Greenland between
ice
measured
[Stauffer and Oeschger,
from postglacial
and
glacial
on the
time,
at ca.
the values decrease from about 300 ppm to 200 ppm. have been observed in only one ice core, caused
by
due
alkaline
to
some
concentration far,
two
interaction
ice from glacial
change
from
between
has been
Greenland
and
time.
four
from
and ice properties
et al.,
et al.,
1982, Stauffer
observed
concentrations
most
1985, likely
do
part
roughly
of
the
14,000
W~rm
The concentration partly because tervals, values
300
indicate
tion.
not
yr
B.P.,
Thus,
it
were
near
200
ppm
at Dye
during
have
3,
so
very
1980, Neftel Thus,
atmospheric
the
values.
during
whole
Holocene
and
again
low values
seems
that
atmospheric
in
large be
[1987] W~rm
the
during
rose
of ca.
corresponding
I, may
et al.
and
between 280 ppm.
is not well established,
in the
Figure the
of
1987].
value
undergone
of Barnola the
pattern
representing
ppm
to the Holocene
to
e.g.
CO 2 concentration during the la-
ice quality
seem
low values as
represent
ice,
ice cores
[Delmas et al.,
history during the Holocene
The measurements
concentrations 120,000
B.P.
of brittle
does
above
events.) core
but
glaciation
and i0,000
m depth,
that it is
a similar
six different
Barnola et al.,
The records show that the atmospheric ter
suspect
Antarctica,
different temperatures
1810
surrounding
However, in
ice core
If this shift would
one might
CO 2 and the
observed
deep
1985]. At the transition
due
to
summer
melt ice
similar
CO 2
interglacial,
ca.
the penultimate
CO 2 was
in(The
on the Vostok
glacial,
last
depth
variations.
generally
glacial periods and relatively high during interglacials.
low
glaciaduring
155 (%0)
-25
180/160 ratio
-30
T
I
~IT
T
-35
(pprn) 340
,/
CO2 concentration (pprn)
320 '
/
/
300,
280' 260 240, 220 200 180
*
1~oo
1~5o
tg~o
1~5o 2o%o Depth below surface (m)
Figure l: CO 2 concentrations measured on the deep ice core from Dye 3, Greenland [Stauffer and Oeschger, 1985]; top: 180/160 ratios, indicating climate change (measured by Dansgaard and colleagues). Time scale: the ±°O/±°O shift at ca. 1782 m depth occurred about 10,000 years B.P., that at ca. 1806 m depth about 13,000 years B.P., the interval 1860-1885 m depth (arrows) to 30,000-40,000 years B.P.
A problem of special interest is the rate of concentration change with time.
According to the available data the full shift from 200 to
280 ppm at the end of the
last glaciation
years.
intervals
ppm,
In the Dye have
been
3 core,
found
during
30,000 to 40,000 yr B.P. surements
indicated
occurred within
of relatively
glacial
(open circles
time,
in
high
the
about
CO2,
period
4000
near of
250
about
in Figure i), and detailed mea-
that the CO 2 concentration
shifted,
together
with
156
8180
and
other
[Stauffer assumed, within
et
ice
properties,
al.,
1984].
abruptly
If
an
then the changes of CO 2 and, a
few
considerable
centuries
or
implications
less.
at
these
undisturbed
depths
ice
simultaneously,
Such
rapid
2) is
of 6180 occurred
variations
for their possible causes.
(Figure
stratigraphy would
However,
have
thorough
and dense measurements
on the ice core from Byrd Station,
Antarctica,
do
rapid
1987].
not
show
situation considered are
due
similar
is,
therefore,
that to
the
rapid
disturbed
variations
not
clear;
[Neftel and
transitions
stratification
et
the
al.,
possibility
observed
in the Dye
(e.g.
"sandwiching"
The
must
be
3 ice core of
older
between younger ice) of the glacial ice in Greenland.
[ppm]
a) O
CO 2 concentration
o
O
0
2,0
o~°
oA O
~o° ~00
oI
j~°N°
one8°
]
o l-I
1
l°° °
oOo
°o~ I
oVo o
I
°AI-
O,o
o
O
o 8 O i
1860
U l '
1870
l
1880
1890 [m]
[%o]
-32
-34
-36 !
1880
!
1870
1880
1890 [m] depth below surface
Figure 2: CO 2 concentration and 180/160 ratio on a section of the ice core from Dye 3, Greenland, of ca. 30,000 - 40,000 years B.P. Note the synchronicity of the rapid changes in the two parameters [Stauffer et al., 1984].
t57
2. P o s s i b l e c a u s e s of the CO 2 v a r i a t i o n s
2.1
The o c e a n i c c a r b o n cycle
The CO 2 v a r i a t i o n s must be due to a r e d i s t r i b u t i o n of carbon between
its
major
global
and o c e a n sediments. sidered
here,
since
they
scales > 106 years. ning about
- atmosphere,
ocean,
are
slow
and
only
become
The most important reservoir
60 times
CO 2 c o n c e n t r a t i o n
reservoirs
land
biosphere
V o l c a n i s m and tectonic p r o c e s s e s need not be c o n important
on
is the ocean,
as much carbon as the atmosphere.
The
time
contai-
atmospheric
is d i r e c t l y r e g u l a t e d by the c h e m i c a l c o m p o s i t i o n of
sea surface water.
It is therefore n e c e s s a r y to c o n s i d e r
the c a r b o n a t e
c h e m i s t r y of sea water.
ATMOSPHERE CO2
;s
solubility = f(T)
dissolved
CO2 "-"
HCO3
( 1% )
~
CO~
(90%)
( 9% )
OCEAN Figure 3: S c h e m a t i c plot of the CO 2 e x c h a n g e b e t w e e n air and sea. Total CO 2 (ZC02) includes d i s s o l v e d CO 2 gas, HCO 3- and CO3 = ions.
Dissolved 90 %),
CO3 =
equilibrium
inorganic (= 9 %)
between
Ca, and in water, Cw = s
c a r b o n or "total CO2"
and d i s s o l v e d air
and
CO 2 gas
water,
the
(ECO2)
(= 1 % ;
includes HCO 3cf.
concentrations
Figure of
3).
(= At
CO 2 in air,
Cw, are p r o p o r t i o n a l to each other:
Ca
(i)
s is the solubility.
It d e c r e a s e s
wer
temperatures
Cw,
to lower a t m o s p h e r i c
during
glacial
with times
C02 levels.
increasing should
have
temperature, led,
for
so lo-
constant
During the glacial m a x i m u m
18,000
158
years
ago,
the
average
sea
surface
than now, which would have caused
temperature
was
about
some CO 2 decrease.
1.5 K
lower
The combined ef-
fect of lower temperature and enhanced ocean salinity can be estimated to have led to an atmospheric less than the observed changes
in
Cw,
i.e.
CO 2 reduction of only about i0 ppm, much
80 ppm change.
in the
oceanic
Therefore,
carbonate
one has to look for
chemistry,
in order
to
understand that change.
According
to
proportionally
the
relation
to a change
(1),
ved CO 2 gas in surface water. ter;
in addition,
partitioned
can be determined
relations
are
considered.
The
of bicarbonate
affected
by
the
we
concentrate A
more
HCO 3-
governing
alkalinity
the charges
fects.
CO2,
if another quantity,
equilibrium
will
on how the dissolved
dissolved
the
formation on
or
constant spheric
in
is
aqueous
approximately ions
and dissolution changes
discussion
Siegenthaler
alkalinity
and
CO 2 concentration
level
changes
C w of dissol-
[1986].
of of If
temperature), increases
ZCO 2 may therefore have considerable
inorganic
and CO3 =.
This
of
given
Cw
by
carbonate
marine
chemistry the
sum
and
Here,
other
ef-
chemistry
in Broecker
increases
of
it is
shells.
dominate
carbonate
by roughly
is
is known and if
carbonate
ZCO 2 that
ZCO 2
carbon
partitioning
(HCO 3- + 2 C03=);
connection with atmospheric CO 2 can be found e.g. [1982]
CO 2
the alkalinity, the
and carbonate
the
detailed
atmospheric
C w mainly depends on ZCO 2 in surface wa-
it depends
between
the
in the average concentration
by
and Peng
1
consequently i0 %. Minor
in
%
(with
the
atmo-
changes
of
effect for the atmospheric compo-
sition.
0
pmol/kg 2000 --~...~
%
2200 ,
,
2400 ,
I
I
I
0
,umo[/kg 1 ,
o
-1
0
I
2
I
I
I /
I
0 '"'"~.
I
I
I
6
2
3
I
I
I
I
~2 £ v 'I" l-r,
~4
81
~. CO 2 6
!
I
6
I
PO 4 ........
I
Figure 4: Depth distribution of ~CO2, 613C of ~CO 2 and phosphate in the mid-latitude North Pacific, after Broecker and Peng (1982). The gradient between surface and deep water is caused by biological activity.
159
Now
~CO 2
(Figure formed whre
in
4),
surface
because
is
are
oxidized
lower
than
atmospheric
In average
CO 2 concentration,
are
mainly
and
and
transport
biological nitrate,
the
also
productivity
which
to deeper
mechanism
is
values,
water is
are depleted
termed
therefore
by
limited
layers,
often
the
by
to
%
the
interplay
circulation.
there
ocean shells
ZCO 2 is 10-20
and
regulated
by
average
carbonate
surface water,
surface
areas,
phosphate
This
The
activity
for
and
sink from the surface
sea water.
biological
oceanic
than
particles
or dissolved.
carbon pump.
in deep
smaller
organic
by by small organisms
they
the biological
of
water
dead
In
the
large
nutrients
nearly
zero
(cf.
Figure 4). In these areas the biological pump strength is closely linked to water
circulation:
and therefore a
proportional
concentrations
2.2
a change
in the
in the supply of nutrients change
in
biological
vertical
activity,
so
layers,
that
the
[1982] was the first who presented a coherent hypothesis at the end of the
last glaciation.
that
times,
oceanic
glacial
the
trients was considerably higher have
led,
chemical
CO 2 changes
for the CO 2 increase during
rate,
involves
at the surface remain unaffected.
Possible causes of glacial-postglacial Broecker
circulation
from deeper
for unchanged
average
than during
water
He suggested
concentration
the Holocene,
circulation,
to higher
of
nu-
which would
biological
pro-
ductivity and therefore to lower ZCO 2 in surface water and a lower atmospheric
C02
occurred
when,
level.
(because
of
formation
rich in organic nental
Broecker
at the
shelves
of
matter were
suggested that
beginning huge
of
continental
and nutrients
transported
such a nutrient
glaciation, ice
from the
global
sea
masses)
and
conti-
This hypothe-
change in CO 2 and would also
be compatible with changes of 613C observed in deep-sea sediments below).
However,
process
and depends
within relatively trast with
deposition
and erosion
on major
is
based
evidence on
the
concentration
of cadmium,
in
ocean
the
times
world [Boyle,
of
shelf
sediments
in sea level,
while
short time. The shelf-sediment hypothesis
recent
that the
ice-age ocean was presumably finding
changes
was
overall
of
sediment
which behaves about
phosphate
not much different
result
the
same
sank
sediments
freshly exposed
to sea and redissolved.
sis could explain the glacial-postglacial
increase
level
is a slow CO 2 changed is at con-
content
of the
than at present.
studies
that
(see
the
This
average
very similarly
as phosphorus,
in
in
glacial
1987]. At the same time, however,
as
these
postglacial
(and other)
data
160
clearly
show
glaciation North
that
than
the
now,
Atlantic
ocean
circulation
especially
Ocean,
which
that is
pattern
formation
very
was of
different
deep
important
during
water
at
in
the
present,
was
considerably reduced. Another the
hypothesis
high-latitude
is that
oceans.
atmospheric
There
the
CO 2 reacted
vertical
water
to changes
exchange
since it is not impeded by a thermal density stratification, rich in nutrients latively enough
high to
so
consume
that
all
slown down
then
the
available
productivity
nutrients
(and
at a re-
is
the
not
high
corresponding
5). If the vertical water circulation would
an unchanged
depletion
biological
fast
and water
and CO 2 wells up from depth to the surface
rate,
amounts of C02; cf. Figure
stronger
is
in
of ~CO 2
biological
productivity
(and nutrients)
sequently to an atmospheric
CO 2 decrease.
a reduction
circulation
of the vertical
would
lead to a
in surface water, Model calculations
in high latitudes
and conshow that
by a factor
two could have caused a CO 2 decrease by some 50 ppm [Knox and McElroy, 1984; Sarmiento and Toggweiler,
0
}J m0! I kg 2000
1984; Siegenthaler and Wenk,
1984].
}J tool / kg 2200
2400
0
1 !
"~
2
3 !
--4
v 1-
IM O
Antarctic.
4
:N Pac p
ECO 2 6
'"
I
I
1
I
i
I
Figure 5: Typical vertical profiles of ZCO 2 and phosphate in the midlatitude North Pacific and in the Antarctic Ocean. In the Antarctic, phosphate concentrations and ~CO 2 are relatively high at the surface.
ment
As mentioned
above,
studies
the
that
deed significantly have
affected
hypothesis, North
in
Atlantic
there
formation CO2,
direction
Ocean
seems
evidence
of North
reduced during
atmospheric the
is clear
according
to be
Atlantic
the glacial
observed; too
from deep-sea
time.
Deep Water This
small
the
involved
to explain
was
in-
indeed could
to the high-latitude but
sedi-
area
- ocean in
the
the whole
el-
161 fect.
The
Southern
Ocean
area and the required evidence
around
Antarctica
oceanographic
has
properties,
for reduced vertical circulation
the
necessary
large
but there is no clear
in the Southern Ocean during
glacial time.
I
I
I
I
I
I
]
I
!
I
I
I
I
I
I
I
I
I
I
I
I
A S 13C
8 180 I
I
I
I
20
I
I
40
I
60
I
I
80
100
120
140
160
Age (1000 y e a r s )
Figure 6: 813C difference between surface-dwelling and bottom-dwelling foraminifera, and 8180 in bottom-dwelling foraminifera from a deep-sea sediment core from the eastern equatorial Pacific [Shackleton et al., 1983]. 8180 essentially gives a record of global ice volume (high 8180: large ice volume), because iow-6180 water was transferred from ocean to land ice during glaciation. The hypotheses ple be
checked
the influence
about the cause of the CO 2 changes
by means
ratio of dissolved
inorganic
viation of the 13C/12C seen
of data
of the biological
in Figure
4.
ratio
Plants
12C during photosynthesis, compared oxidation
on
carbon
organic
isotope
ratio
(expressed
preferentially
take
13C/12C,
since
in the 13C/12C
by 813C = relative
from that of a standard, up
the
in permil), lighter
deas
isotope
and marine organic matter has 613C = -20 %,
to 613C = 0 %o of total of
the
pump is also reflected
can in princi-
particles
CO 2 in sea water.
leads
to
lower
613C
Consequently, at depth
than
the in
162
surface
water.
is roughly
The
proportional
for e s t i m a t i n g profiles found sea
(Figure
(tiny
6).
of
pump m e c h a n i s m
from
160,000
is exactly
deep-sea
in
the
magnitude
two
ice
mentioned
hypotheses
hypothesis,
ocean
water
and
foraminifera
813C in
seem
CO 2 trapped
in
et al.
Vostok,
al.,
1987],
with
that
CO 2.
The
confirm
if indeed
ice
should
0.5 %o
the biological CO 2 variations.
model
the
calculations
to
the
higher
the
to the CO 2
covering
past agree
for
the
high-latitude
in
ice-age
data
expectation.
give
the
difference
~613C variations
available
this
on
and the last interglacial
According
atmospheric
and
the
have
and ben-
larger by about
the
the been
ice-age
water
Antarctica,
have
(1983)
nicely parallel
and
what
predict.
to
was
varies
should
not
surface
is expected
from
et
approximately
ocean
et al.
from 813C
of planktonic
sediments
water
fact may be used
in the g l a c i a l - i n t e r g l a c i a l
core
[Barnola
in
the Holocene
what
is involved
years
Shackleton
living
and deep
in the past
shells
of foraminifera
time than during
This
This
- depth
carbonate
The A813C curve of Shackleton record
surface
animals
the two kinds
surface
~CO 2 difference.
on the
respectively)
during glacial
between
in deep sea sediments.
foraminifera
floor,
to the
analyses
A813C b e t w e e n
in 813C
ZCO 2 differences
measured
by 813C
thonic
difference
on planktonic
813C
required
-
surface
analyses
on
information
to
check this reliably. Thus,
both
existing
cial-interglacial with
the
CO 2
existing
variations
are
data.
to
have
it
the
about
their
seems
the cause
problems
rather
biological
likely
carbon
of the gla-
when
compared
that
pump
the
in the
CO 2
ocean,
is not yet fully understood.
Climatic effects of the CO 2 v a r i a t i o n s It is
now g e n e r a l l y
terglacials
accepted
is
connected
orbital
parameters
of climate
records
earth's lysis
hypotheses
While
connected
the precise m e c h a n i s m
3.
model
variations
periodicities orbital
- ca.
elements
21,
tronomical
variations linking
insolation
the
pattern
1976].
The
- is,
While
and
external by
- are
climate
cause
of
itself,
of
earth
- changes enough
are
in the
the ana-
in the
between
well established,
on
in-
the same
as present
the connection
strong
and
By spectral
and elsewhere,
appears
forcing
ages
variations
found
changes
not
ice
hypothesis).
sediments
i00 yr
and global
cycle
quasi-periodic
in deep-sea 41 and
external
understood.
the
(Milankovitch
[Hays et al.,
mechanisms from well
to
that
still
asthe far
seasonal
to produce
the
163
ice
ages;
internal
strongly amplified
processes them,
within
the
and atmospheric
global
climate
must
have
carbon dioxide variations
are
such an internal process.
ENERGY
BALANCE OF THE EARTH
SOLAR RADIATION
INFRARED
Figure 7: Energy balance of the earth. Incoming solar radiation = I00 units. Left side: short-wave length radiation; right side: infrared.
CO 2 influences
the global
climate via the greenhouse
gure 7 shows the radiation balance of the earth, ming solar radiation, earth's
surface.
tion;
the
from earth
rate of
infrared
to the fourth power perature
to space
emission
i.e.
considerably
rature
of
15°C.
the
atmosphere,
transparent radiation.
The
because
for
sunlight most
within
the
but of
flux of
surface
temperature.
radia-
is proportional
The equilibrium
to the absorbed sunlight
tem-
is 255 K
lower than the actual mean surface tempe-
difference
Thus,
a corresponding
in the form of infrared
from the
of its absolute
within
absorbed
equilibrium,
of the earth corresponding
or -18°C,
Fi-
30 are reflected back to space by clouds and the
In a radiative
energy is emitted
effect.
of I00 units of inco-
is
water
not
the
atmosphere
for
due the
infrared by
to
vapour,
these
energy CO 2 and
longer-wave emitted
at
transfer other length the
infrared-active
effects
gases
are
infrared
surface gases,
is
which
164
re-emit space
upwards at
the
sunlight
the
cf.
In
consequence,
atmosphere
Figure
7),
is
but
concentration
of
CO 2
the
equal the
show that
the
flux
changes,
to
net
absorbed
surface
receives
from the atmosphere. then
and consequently the surface temperature,
Model calculations
infrared
to
earth's
95 units of infrared back-radiation
atmospheric
effect,
downwards.
of
(70 units,
additional the
and
top
the
If
greenhouse
is modified.
if the CO 2 level would double,
then
the mean surface temperature would rise by 1.2°C as a direct response. However, tive
several
feedback
stant relative (which
is
Another when leads
feedback
is due
humidity,
also
a
earth
to
The
tendency
so that the absolute
effect
gas)
comes
becomes
a decrease
surface.
amplify the warming.
atmosphere's
greenhouse
feedback
the
effects
to the
of
the
the
the aibedo
overall mean
content
increases
from
warmer;
as
of
of
(refiectivity increase
snow
snow for
a con-
in water vapour
temperature
melting
reduction
temperature
A strong posi-
to maintain
goes
up.
sea
ice
and
and
ice
sunlight)
cover of the
for a CO 2 doubling,
as
calculated by models of various complexity and for different model assumptions,
is 2 to 4°C, with larger changes in high latitudes.
The question as to the contribution of the reduced CO 2 concentration to the global cooling during the ice age, as simple as it sounds, cannot be answered cessary are
to
simultaneously
conditions the
in this
specify are
CLIMAP
necessary.
peratures, estimated
mean
information
sea
surface
which
B.P.
of continental
the global cooling,
it is incomplete. climatic
additional
project,
18,000 years
extent
which,
and
Such
1981)
because
and
changing,
(1976,
glacial maximum,
form,
whether,
Manabe,
and
1987];
has
become
be sea
only
are now
from
the
last
sea surface
tem-
temperature
was
1.5°C
colder
than
The
today;
discussed
[Hansen
et al.,
are especially
1984;
Broccoli
and
both used general circulation models of the atmosphere,
simulation
of
layer of the ocean the
seasonal
(which is important
temperature
as prescribed boundary conditions,
prescribed
surface
for
and sea ice, and surface albedo.
al. run their model with the CLIMAP-reconstructed conditions
ice-age
available
reconstructed,
(before present),
on
including continents, may have been about 4°C.
coupled to a mixed surface a realistic
mechanisms
information
Two model studies of the climate 18,000 years B.P. interesting
It is ne-
feedback
continental
temperature
could be compared with
which
ice
and
was
thus
land
cycle).
Hansen
for et
land and sea surface
while Broccoli and Manasurface
simulated
the CLIMAP reconstruction.
but
not
model
and
In both papers,
the
by
albedo, the
165
sensitivity of the simulated
climate
to variations
in individual
fac-
tors, amongst other atmospheric CO2, is studied.
Hansen (without
et
al.
other
listed
the
feedbacks)
to
contributions
the
overall
of
individual
difference
in
changes
mean
global
temperature between 18,000 years B.P. and today as follows: Changed water vapour and clouds
1.4-2.2°C
Extent of continental ice
0.9°C
Extent of sea ice
0.7oC
CO 2 concentration
0.5oC
(280 ÷ 200 ppm)
Vegetation
0.3oC
Orbital variations
0±0.2°C
The total calculated cooling is 3.8 to 4.6°C. This value, as
the
size
of
the
individual
contributions,
clearly
as well
depends
on
the
specific model structure and assumptions.
The
largest
concentration
single
contribution
and cloud cover.
it is not a primary
or
This
external
stems
from changed
is particularly
effect,
but
water
vapour
interesting
a purely
internal
since feed-
back! The relatively large range has to do with uncertainty in the reaction of clouds,
which is still an unsolved problem in climate model-
ling.
The
next
important
change
which
led
to a considerably
is in the extent
higher
continental
age. The change in type of vegetation
affected
areas not covered by ice, but with minor of the CO 2 concentration
albedo
during
ice,
the ice
the albedo of the land
influence.
change by a factor
!0 percent of the overall cooling;
of continental
The direct effect
1.4 amounts
to only about
but a CO 2 decrease is automatically
amplified by changed water vapour and cloud cover which are fast feedbacks. Thus,
a sudden change in CO 2 from 280 to 200 ppm, with present-
day continental cooling increase
by and
and sea
about
ice distribution,
l°C.
As
therefore
a
global
albedo,
again
the long run, also polar and continental the different
climate
factors
would
consequence,
induce
however,
a global mean sea
a positive
ice might grow.
are interdependent
ice
would
feedback.
On
In this way,
and it is not always
possible to distinguish between cause and effect. Broccoli and Manabe wing
changes,
separately
(1987) considered the influency of the folloand
in combination:
CO 2, and albedo of non ice-covered sea
ice
were
allowed
to
adjust
land.
continental
Atmospheric
themselves,
i.e.
ice
sheets,
water vapour and
acted
as
positive
166
feedbacks, mean
but
cooling
by Hansen
et al.
their mean nabe and
cloud
cover
for the
was
combined
(1984),
perhaps
only for regions
obtained present
the
prescribed changes
because
Broccoli
temperature
(influence
of
way.
smaller
free of c o n t i n e n t a l
following
conditions
in a fixed
is 2.8°C,
and Manabe
ice.
albedo
global
computed
Broccoli
and Ma-
between
ice-age
differences
land
The
than obtained
is
omitted
here,
since rather small):
Southern Northern iHemisphere H e m i s p h e r e
Global
Extent of continental ice CO 2 c o n c e n t r a t i o n (300 ÷ 2 0 0 Combined changes
Thus,
according
and changed to the tant;
results the
Broccoli esting ice
results
is,
et
the
the
difference nearly
(1984).
ice
not
the effect
there
cooling
Manabe,
land
course,
the
the
the
imporbecause
Most
the
sheets
result
hand,
more
inter-
hemispheres. to
ice
this
in both hemispheres.
that CO 2 was responsible
two
sheets
according
compared,
restricted large
other
be
ice
while
somewhat
directly
Hemisphere; On
continental effect,
seems
between no
0.3 ° 1.3 ° 1.9 °
of fast feedbacks.
fully
were
Southern
al.
same
sults indicate
same global
of
is
because in
Hansen
and
the
et al.,
the
effect
Hemisphere,
Broccoli
include
however,
sheet
about
can,
2.4 ° 1.1 ° 3.9 °
about
of Hansen
and Manabe
Antarctica by
to
CO 2 produce
1.3 ° 1.2 ° 2.8 °
ppm)
was
outside also
of
obtained
CO 2 decrease
Broccoli
The
Northern
causes
and Manabe's
for most of the cooling
re-
in the
Southern Hemisphere. This finding to the
available
is most important evidence,
nous in the two hemispheres. the M i l a n k o v i t c h insolation, is not ween
theory
which
synchronous
the
discussed
two
is
the
hemispheres
climate
model
CO2, the c o n c e n t r a t i o n
This
of the
supposed
in
in view of the fact that according
the glaciations
two is
could
not be e x p e c t e d
ice ages, to be
since
the
results,
the
ultimated
hemispheres. therefore
seem to have
it might
and
be found
of which is h o m o g e n e o u s
a priori
seasonality driving
A coupling
needed,
been synchrofrom
of the
mechanism,
mechanism
according
to
betthe
in the atmospheric
over both hemispheres.
167
4.
A coupled global climate-CO 2 system
The climate significant
model
role
hand, as discussed rying
conditions
events, tion. did
led
the
to the
CO 2 change
the
lankovitch-type
ocean
observed thus
surface,
atmospheric cycle.
of
or was
which
CO 2 played On
the
modified
variations
linked
in atmospheric
was
the CO 2 change
insolation
probably
variations
arises
because
factor,
indicate that
glacial-interglacial
the
a
other
in section 2, the current understanding is that vaat
The question
feedback
studies
for
the
hen
climate
to
climatic
CO 2 concentra-
and which
and
initiated
then
egg:
acted
directly
and then caused
the
as
a
by the Mi-
a shift
in glo-
bal climate? At present,
this question
The ice core work, on
the
Vostok
1987],
core
indicates
cannot yet be resolved with certainty.
especially covering
that by the French-Soviet the past
160,000
years
[Barnola
et
al.,
that the CO 2 concentration varied remarkably parallel
to climate.
This demonstrates
that both are intimately
it does
itself
to
by
collaboration
not
allow
decide
about
the
connected,
cause-effect
but
rela-
tionship.
One plausible nal cause
hypothesis
explaining
the CO 2 changes by an exter-
is that the seasonally varying
tudes might be involved. that e.g.
during the
insolation
Knox and McElroy
(1984)
last glacial maximum,
in early or late winter,
i.e.
in the high
put forward
high intensity of sunlight
a time when deep water
have resulted in higher biological
lati-
the idea
is formed,
might
activity and thus in lower CO 2 con-
centrations in surface water and in the atmosphere.
It is difficult to
test this hypothesis.
that light-limita-
tion
may
would
steer
not
expect
could be
1987].
which
the
for
fully
If the
the idea is attractive
biological
that
responsible
transition al.,
the
While
the
took
productivity
slow
variation
CO 2 increase place within
rapid CO 2 variations
in Greenland ice [Stauffer et al., the Milankovitch-type cause.
In any case,
insolation
in
of
high
the
at the about
one
parameters
glacial-postglacial
4000
observed
latitudes,
orbital years
during
[Neftel
glacial
et
time
1984] can be shown to be real, then cycles
must
clearly
it seems that oceanographic
be
ruled
out
as
factors must have been
involved, at least as strong feedback effects.
The
current
ideas
are
that
insolation
changes
in high
latitudes
of the Northern Hemisphere were the main primary cause for the ice age cycles.
There
is clear
evidence
that
formation
of deep
water
in the
168
Northern Atlantic
was reduced
time (e.g. Boyle,
1987), which must have been connected to colder tem-
peratures
and different
and considerably larger extent of sea ice. Conditions
Southern
Ocean
around Antarctica,
ged,
there
is no evidence
but
water
formation
spheric Ocean.
than today during glacial
there.
CO 2 varied The
problem
Thus
in
is
sea
ice,
for comparably
it seems
response
here
e.g.
to
that
certainly
drastic
reasonable
the
in
events
the
North
to in
in the
also
changes
assume the
in deep
that atmo-
North
Atlantic,
chan-
Atlantic
the
area
of
surface water in rapid contact with the deep sea seems to be too small to
effect
some
the
indirect
atmospheric effects,
composition
e.g.
a
in
the
teleconnection
observed to
the
way,
so
Southern
that
Ocean,
must also be involved. While
it is not yet possible
to give a consistent
explanation
of
all events at the onset of an ice age, it is tempting to sketch a scenario. Figure 8 represents
a personal view of how the combined climate
and CO 2 changes might have a~isen
and be connected.
According
to this
view, CO 2 changed in response to oceanographic events in the North Atlantic,
induced by the Northern Hemisphere
it then
acted
as a climatic
and CO 2 may have been rather (1984)
the
hundred
oceanic
years
to
feedback. small;
carbonate
changed
as shown by Wenk
system
adjusts
circulation
CO 2 data from the glacial-postglacial
cooling and ice growth,
The time delay between
or
and
climate
and Siegenthaler
itself
productivity.
within The
a
few
available
transition do not allow to deci-
de, within a few centuries, whether CO 2 preceded or followed climate.
Further modelling mate system. and deep-sea pes,
nutrient
tion are questions.
work is necessary
Above all, however, sediments
on the detailed
concentrations
needed
in order
in
to understand
more experimental history
the oceans
to gain more
of CO 2 and its
and of
insight
the CO 2 - cli-
data from ice cores isoto-
the ocean circula-
into these
fascinating
169
POSSIBLE CONNECTION CLIMATE TRANSITION
CO 2
-
INTERGLACIAL --~ GLACIAL
I
Insolation I (Milankovitch)
t Temperature t I Polar water in ........ North Atlantic ®
_ Continental ALBEDO
,t
,c:l
i North Atlantic ........... = Deep Water ®
GREENHOUSE I 002 ....EFFECT .....[ concentration ® I
Figure 8: Scenario of the possible connection between processes in the combined climate-CO 2 system at the onset of an ice age. Insolation changes are the external driving force, all other compartments and processes belong to the interactive internal system.
REFERENCES
Barnola, J.M., Raynaud, D., Korotkevitch, Y.S., Vostok ice core: a 160,000-year record of Nature, in press. Boyle, E.A. 1987: Cadmium: chemical tracer graphy. Paleoceanography, in press.
Lorius, C. atmospheric
of deep water
1987: CO 2 .
paleoceano-
Broccoli, A.J., Manabe, S. 1987: The influence of continental ice, atmospheric CO2, and land albedo on the climate of the last glacial maximum. Climate Dynamics i, 87-99.
170
Broecker, W.S. Cosmochim.
1982: Ocean chemistry Acta 46, 1689-1705.
during
Broecker, W.S., Peng, T.-H. 1982: Tracers Lamont-Doherty Geological Observatory, CLIMAP project members, ce 191, 1131-1137.
1976:
The surface
glacial
time.
Geochim.
in the Sea. Eldigio Press, Palisades, NY 10964. of the ice-age
earth.
Scien-
CLIMAP project members, 1981: Seasonal reconstruction of the earth's surface at the last glacial maximum. Geol. Soc. Am. Map Chart Set. MC-36. Delmas, R.J., Ascencio, J.M., Legrand, M. 1980: Polar ice evidence that atmospheric CO 2 20,000 y B.P. was 50 % of present. Nature 284, 155-157. Hansen, J., Lacis, A., Rind, D., Russel, G., Stone, P., Fung, I., Ruedy, R., Lerner, J. 1984: Climate sensitivity: analysis of feedback effects. In: Climate Processes and Climate Sensitivity, Geophysical Monograph 29, 130-163. American Geophysical Union. Hays,
J.D., Imbrie, J., Shackleton, N.J. 1976: Variations in the earth's orbit: pacemaker of the ice age. Science 194, 1121-1132.
Knox,
F., McElroy, of the marine 4637.
M.B. 1984: Changes in atmospheric CO2: Influences biota at high latitude. J. Geophys. Res. 89, 4629-
Neftel, A., Oeschger, H., Schwander, J., Stauffer, B., Zumbrunn, R. 1982: Ice core sample measurements give atmospheric CO 2 content during the past 40,000 yr. Nature 295, 220-223. Neftel, A., Oeschger, H., Staffeibach, T., Stauffer, cord in the Byrd ice core 50,000-5000 years press.
B. 1987: CO 2 reB.P. Nature, in
Sarmiento, J.L., Toggweiler, J.R. 1984: A new model for the role of the oceans in determining atmospheric pCO 2. Nature 308, 621-624. Shackleton, N.J., Hall, M.A., Line, J., Shuxi, C. 1983: carbon isotope data in core V19-30 confirm reduced carbon dioxide concentrations in the ice age atmosphere. Nature 306, 319-322. siegenthaler, U. 1986: Carbon dioxide: its natural cycle and anthropogenic perturbation. In: Buat-M~nard, P. (ed.), The Role of AirSea Exchange in Geochemical Cycling, 209-247 (Reidel). Siegenthaler, U., Wenk, T. 1984: Rapid atmospheric ocean circulation. Nature 308, 624-625.
CO 2 variations
and
Stauffer, B., Hofer, H., Oeschger, H., Schwander, J., Siegenthaler, U. 1984: Atmospheric CO 2 concentration during the last glaciation. Ann. Glaciol. 5, 160-164. Stauffer, B., Neftel, A., Oeschger, H., Schwander, J. 1985: CO 2 concentration in air extracted from Greenland ice samples. In: Geophysical Monograph 33, 85-94, American Geophysical Union.
171
Stauffer, B., Oeschger, H. 1985: Gaseous components in the atmosphere and the historic record revealed by ice cores. Ann. Glaciol. 7, 54-59. Wenk,
T., Siegenthaler, U. 1984: The high-latitude ocean as a control of atmospheric CO 2. In: The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present, Geophysical Monograph 32, 185-194. American Geophysical Union.
Zumbrunn, R., Neftel, A., Oeschger, H. 1982: CO 2 measurements on l-cm 3 ice samples with an IR !aserspectrometer, Earth Planet. Sci. Let. 60, 318-324.
172
SUBJECT INDEX
A e r o s o l p a r t i c l e s , e f f e c t s of Albedo Anthropogenic aerosol Astronomical theory Atmospheric chemistry and climate Biological Biological
1 59-161
carbon pump processes
119
13C in s e d i m e n t s C a r b o n d i o x i d e (C02): - c l i m a t i c e f f e c t of - d o u b l i n g of c o n t e n t - d u r i n g ice age general aspects ocean cycle CLIMAP Climate and atmospheric chemistry Climate modelling: - climate applications main problems man made changes - meteorological applications natural system - numerical errors ocean models - parametrization errors Climatic feedback effects Cold summers -
-
-
-
-
-
Deep sea sediments Deep water formation Dendroclimatology Dendro-ecological diagram
87-99,160-165 162-169 127 153 4,126-127,162-169 157-159 165 128 119-120 3 125 118-119 123 120 1t9 120 165-167 72 83-97,160,165 83-101,160,168 2,35-56 36 26-30 26-30 23-26
E1Ni~o ENSO phenomenon Eruptions Faculae Flares F o r a m i n i f e r a in s e d i m e n t s Forcing factors Glacial - interglacial Global water cycle Grape harvest dates Greenhouse effect G r o w t h r i n g s of t r e e s
128 134,164-166 129 4,132-152 128
transition
Hemispheric temperatures High altitude aerosol High Middle Ages High energy flares Historical sources Homogeneity problems
10 10 8 3 - t 04 22 95,155,169 126 59,76 163-169 35 18 13 57 10 60-63 18
173 Ice a g e c l i m a t e Ice a g e s , a s t r o n o m i c a l Ice cores Ice v o l u m e , g l o b a l Insolation Isotope stages Isotopic equilibrum Little
Ice
theory
162-169 142-150,162,167 154-156 83,99,159 132,136,167-168 85-104 83-87
of
15
Age
Man's impact on climate Man-made climatic change, detection Maximum density (tree rings) Meteorological observations Milankovitch theory North Atlantic deep Norwegian sea Numerical modelling
water
of
125-130 130 36 19 132,141 ,162,167 83 83 117-131
1 8 0 :
155-156 83-I 04 83,16t
- in i c e c o r e s - in o c e a n w a t e r in s e d i m e n t s Ocean: sediments models Orbital variations -
3
-
119,126 132,147
-
Pointer years (dendroclimatology) Polar ice cores Precipitation: - patterns - reconstruction Proxy data Radiation: - absorbed at e a r t h - in t h e a t m o s p h e r e Ring width
surface
Sea ice Sea level variations Sea surface temperature Solar activity Solar constant Solar irradianoe: measurements - power spectra Soot content Southern oscillation Spectra of motion Sunspots
71-74 39 36
137,141 133,163 36 84,95,100 99,159 19-22,95-103 12 1,6-17 8
-
Temperature: annual estimates anomalies deep water global data - hemispheric data - reconstruction from - variations
42 4
11 129 26-30 121 10
-
-
-
-
conifers
20 45-54,65 83-I 04 18 18 37 2
174
Trace gas content Tree ring data
126 63 23-26
Volcanic
effects
Warm anomalies Warm period Weather p a t t e r n s
68-71 2,57 58
175
ADDRESSES
A. Berger, Institut d'Astronomie Lemaitre, Universit@ Catholique Cyclotron, B - 1 3 4 8 L o u v a i n - l a - N e u v e , P.L. Blanc, France
CEN/FAR,
J.C. Duplessey, mixte CNRS-CEA,
B.P.
Centre F-91190
6-92265
et de G@ophysique Georges de Louvain, 2 Chemin du Belgium Fontenay
aux
Roses
des Faibles Radioactivit@s, Gif sur Yvette, France
Cedex,
Laboratoire
C. Fr5hlich, Physikalisch-Meteorologisches Observatorium, R a d i a t i o n Center, CH-7260 Davos Dorf, S w i t z e r l a n d
World
H. Grassl, Forschungszentrum Geesthacht, Max-Planck-Strasse D -2054 Geesthacht, Federal Republic of G e r m a n y
I,
P.D. Jones, Climatic Research Unit, Sciences, University of East Anglia, Kingdom
School Norwich
of Environmental NR4 7TJ, United
P.M. Kelly, Climatic Research Unit, Sciences, U n i v e r s i t y of East Anglia, Kingdom
School Norwich
of Environmental NR4 7TJ, United
L. Labeyrie, Centre des Faibles Radioactivit@s, mixte CNRS-CEA, F - 9 1 1 9 0 Gif sur Yvette, France
Laboratoire
C. Pfister, H i s t o r i s c h e s strasse 4, CH-3012 Bern,
Engehalden-
Institut, U n i v e r s i t ~ t Switzerland
F. Schweingruber, Swiss Federal CH-8905 Birmensdorf, S w i t z e r l a n d
Institute
of
U. Siegenthaler, Physikalisches Institut, S i d l e r s t r a s s e 5, CH- 3012 Bern, S w i t z e r l a n d C. Tricot, Institut d'Astronomie Lemaitre, Universit@ Catholique Cyclotron, B-1348 Louvain-la-Neuve, H. Wanner, Hallerstrasse
Geographisches 12, CH-3012 Bern,
Bern,
Forestry
Research,
Universit~t
Bern,
et de G@ophysique Georges de Louvain, 2 Chemin du Belgium
Institut, Switzerland
Universit~t
Bern,
