Content uploaded by Stephan Kempe
Author content
All content in this area was uploaded by Stephan Kempe on Nov 17, 2016
Content may be subject to copyright.
FNotur
w ssenschaflen
Naturwissenschaften 77,426-427 (1990) © Springer-Verlag 1990
Alkalinity: the Link Between Anaerobic Basins and
Shallow Water Carbonates?
S. Kempe
Institut fiir Biogeochemie und Meereschemie der Universit~it, D-2000 Hamburg
Anaerobic ocean basins were common
features in earth history. Today, only
one large anoxic basin exists, the deep
Black Sea (480 000 km3). Massive cal-
careous deposits are often found in
close regional and stratigraphic associa-
tion with deposits of anoxic basins. The
Aptian/Albian and Cenomanian/Tu-
ronian anoxic events were, for ex-
arnaple, accompanied by widespread
epicontinental chalk and reef limestone
deposition.
Therefore, one may ask if there is a
causal relation between the two facies.
In fact, such a connection has been
proposed (e.g., [3, 6]). These models
describe the general implications of
shifting chemoclines and explore con-
sequences of a Ca input, for example,
from hydrothermal sources. Here, I
present an alternative view which sug-
gests that small changes in the concen-
tration of the total dissolved inorganic
carbon
(~]CO2)
could be responsible.
Shifts in ~CO2 affect the supersatura-
tion of CaCO 3 more effectively and
therefore faster than the same change
in Ca because the Ca concentration in
seawater is about ten times that of
~-]CO2.
Anaerobic basins are governed by
sulfate reduction which mineralizes
sinking algal organic matter (i.e., of a
Redfield composition [ 12]):
53 SO42- + C106Hz63OlloN16P ] +
14 H20
53 H2S + 106 HCO~- +
HPO 2- + 16NH 2 + 14 OH- (1)
424 e- are used in the reduction and to
transfer negative charge from sulfate to
bicarbonate. H2S and HCO~- are
426
produced at a molar ratio of 1:2. The
addition of HCO~- increases both
~CO2 and alkalinity (i.e., the sum of
the charges of all anions of weak acids,
HCO~- + CO~-+ H(BO3)- + OH-
etc.).
Turning to the deep Black Sea, we find
that the ratio between H2S and ~CO 2
is higher than 1:2 (Fig. 1). Either HzS
has been removed or HCO~- was gen-
erated in excess to Eq. (1). Both possi-
bilities are realistic. H2S is, in fact, re-
moved by the formation of settling in-
soluble metal (mainly Fe) sulfides, and
CaCO 3 dissolution would increase
~CO 2 and alkalinity.
Probably both processes occur simulta-
neously. Our sediment trap experi-
ments allow one to estimate the FeS2
flux to 690 nag m -2 a ] or ca. 2 % of
--4200
Z
=:L
4000
3800
o
3600
3400
I..--
o:
3200
r I ~ i ~ 2000~
1 oo t T r ~ ! l
0 1 O0 200 300
Sulfide sulfur
[FH]
Fig. 1. Comparison of actual ]~CO2/H2S
molar ratio (points) with the 2:1 ratio of
sulfate reduction (lower line) in the Black
Sea. The hatched area represents excess
~CO 2 gained by either calcite dissolution or
pyrite formation (altered after [12], data by
[11])
Naturwissenschaften 77 (1990)
the total vertical particle flux of 36g
m -2
a-1 (Kempe and Enaeis, unpubl.
EDX data, 1983/84, station 40km
north of Amasra/Turkey). The total
vertical particle flux in the Black Sea is,
however, highly variable seasonally,
interannually, and between different
locations [5]. An average deposition
rate is therefore better estimated from
sediment cores, it amounts to roughly
190g m -2 a- ] [4]. Thus, total FeS 2 flux
may amount to 3.8 g
m -2 a-1
(64mmol
S na -2 a-1). For comparison, Tambiev
[13] estimated the pyrite flux to 2.9g
na- 2 a- 1 (48 mmol m- 2 a- 1). Sulfate
reduction amounts to 2.5 mol m -2 a -1
of which 2.4mol m -2 a-1 are oxidized
again [8]. The remainder, 100mmol
m -2 a-1, minus the pyrite flux created
the present H2S pool (600mol m -2) in
the Black Sea. These figures suggest
that the pyrite flux accounts for the re-
moval of roughly half of the net
production of HzS and must play a key
role in building up excess ~CO2.
The deposition rate of CaCO 3 is ca.
100g m -2 a -1 [4], compared to 7g
m -2 a -1 recorded by sediment trap
(1983/84). This illustrates the size of
medium-term changes in the CaCO3
production in the Black Sea. How
much of the sinking CaCO 3 dissolves is
unknown. At present, possibly none
because the water below a depth of
60m is just saturated with respect to
calcite (unpubl. data), but could have
been corrosive in the recent past. This is
in accordance with the sediment rec-
ord: During the last 1 000 years a coc-
colith carbonate ooze was deposited
(varve counts by Degens et al. [2], and
Hay, pers. comm.). It rests on a marine
sapropel free of carbonate, deposited
5 000- 1000 years •.P., at a time when
the Black Sea was already anoxic but
calcite undersaturated. The observed
~CO2 surplus may also be a relic from
that time.
The ]]CO 2 difference between the oxic
surface layer and the bottom of the
Black Sea is 1 mmol/kg, i.e., five times
larger than in the world oceans
© Springer-Verlag 1990
(0.22 mmol/kg; Fig. 2) [ 1]. Comparing
the ages of anaerobic conditions in the
Black Sea (ca. 5 000 a) with the average
age of deep oceanic waters (ca. 1 000 a)
and taking the different mean depths of
both water bodies into account (factor
of 2) we can tentatively deduce that
~CO 2 accumulation by aerobic and
anaerobic respiration relate like 1 to 2.
Such rates would deplete the sulfate
supply (28mmol/kg) of an oceanic
basin within a few 205 a.
Most striking is, however, that the sur-
face of the Black Sea has a much higher
ECO2 than the world ocean (Fig. 2).
Apparently, ~CO 2 is "leaking" from
below the chemocline to the surface
layer by upwelling and/or eddy diffu-
sion. This shows that the reverse pro-
cess to sulfate reduction, i.e.,
HzS ox-
idation,
HzS + 20 2 --+ SO 2- + 2H + ;
H + +
HCOf --+ H20 +
CO 2
(2)
is not 100 °70 effective to turn all of the
HCOf back to CO2 which would then
degas to the atmosphere. Only if a
sizable part of the sulfide was removed
to sediments or if oxidation produces
elemental S, which sinks back into the
anoxic zone, the alkalinity generated by
sulfate reduction would be charge-bal-
anced by cations and is protected from
degassing.
Whatever the exact reaction pathways
are, the Black Sea managed to increase
its surface ECO2 much above the mix-
ture of the two end members which
deliver water to the Black Sea, i.e.,
Mediterranean water with 2.18 to
2.36mmol/kg [9] and rivers with a dis-
charge weighted average of 1.3
mmol/kg [10].
This high ]~CO2 is, however, paired
with a low Ca 2+ concentration. For Ca,
end-member mixing leads to a much
lower concentration than seawater be-
cause of the low salinity of the Black
Sea. Thus, the ion activity product of
Ocean, surface l
/
0
2.23 &30
Fig. 2. Comparison of ~CO 2 (in mmol/kg)
between the surface and bottom waters of
the world ocean and the Black Sea (data
after [1] and [12] quoted in [12])
Naturwissenschaften 77 (1990)
CO 2
Ca 2÷
k,,,~t k
i
O~ C02
Fig. 3. Scheme of fluxes in the coupled anaerobic basin - epicontinental carbonate platform
model
[Ca2+][CO3 -] in Black Sea surface wa-
ters is comparable to average oceanic
surface water. If, however, the Black
Sea would have an oceanic salinity
(35°/00), then the surface would be
highly supersaturated with respect to
calcite, high enough to possibly cause
spontaneous aragonite formation or to
impede growth of most marine species.
An example of such a high supersatura-
tion environment is the seawater-filled
Satonda Crater lake. There an increase
in carbonate alkalinity to only
3.4mEq/kg proved enough to cause
cyanobacterial stromatolites and ara-
gonite precipitation, and to exclude all
marine biomineralizing species, except
for one gastropod [7]. In Satonda the
logarithmic index of calcite saturation
increased to +0.8 compared to +0.5 in
the Black Sea and to +0.4- +0.6 in
normal sea-surface waters.
The Black Sea shows that stagnation
periods of a few thousand years allow
the generation of high ECO 2 and al-
kalinity not only in anaerobic bottom
waters, but also in aerobic surface wa-
ters. If such waters are then advected to
epicontinental seas, further CO2 is
extracted by phytoplankton and
epiphytes. This causes a CaCO 3 super-
saturation stress, which, in turn, leads
to increased rates of biomineralization
forming either chalks or reef limestones
(Fig. 3). Paleogeographically the Cre-
taceous anaerobic basins were caused
by the opening of the Atlantic and a
high sea level provided for extensive
epicontinental seas. In such a geo-
graphic setting three different modes of
sedimentation can be discerned: 1)
Platform carbonates and basin sedi-
ments can form at the same time but
spatially apart; 2) at the
H2S/O 2
inter-
face alternating columns of anaerobic
sediments and carbonate deposits may
form induced by changes in sea level or
© Springer-Verlag 1990
changes in the depth of the aerobic
layer; and 3) the aeration of the total
basin may cause deep water carbonate
deposits on top of deep-water
anaerobic sediments.
Received March 8 and May 22, 1990
1. Broecker, W., Peng, T.-H.: Tracers in
the Sea. Palisades: Lamont-Doherty
Geol. Observ., Cotumbia Univ. 1982
2. Degens, E. T., Michaelis, W., Garrasi,
C., Mopper, K., Kempe, S., Ittekkot,
V. : N. Jb. Geol. Pal~iont. Mh.
1980
(2),
65
3. Degens, E. T., Stoffers, P.: Nature
263,
22 (1976)
4. Hay, B. J.: Dissertation, Woods Hole
Oceanographic 1987
5. Hay, B. J., Honjo, S., Kempe, S., Ittek-
kot, V., Degens, E. T., Konuk, T.,
Izdar, E.: Deep-Sea Res. (submitted)
6. Kazmierczak, J., 2ttekkot, V., Degens,
E. T.: Pal~ont. Z.
59
(1 -2), 15 (1985)
7. Kempe, S., Kazmierczak, J.: Chem.
Geol. (in press)
8. Lein, A. Yu., Ivanov, M. V., Vainsh-
tein, M. V., in: 2nteractions of Bio-
geochemical Cycles in Aqueous Systems
Pt.7 (Degens, E. T., Kempe, S., Lein,
A., Sorokin, Y., eds.). Mitt. Geol.-Pa-
l~ont. Inst. Univ. Hamburg, SCO-
PE/UNEP Sonderband.
71
(in press)
9. Millero, F., Morse, J., Chen, C.-T.:
Deep-Sea Res.
26A,
1395 (1979)
20. Shimkus, K. M., Trimonis, V. V.: Am.
Ass. Pet. Geol. Mere.
20,
249 (1974)
21. Skopintsev, B. A., et al., in: The Sea,
Vol. 2 (Hill, M.N., ed.). New York:
Wiley Interscience 1966
12. Stumm, W., Morgan, J. J.: Aquatic
Chemistry. New York-London: Wiley
Interscience 1970
23. Tambiev, S. B., in: Particle Flux in the
Ocean (Degens, E. T., Izdar, E.,
Honjo, S., eds.). Mitt. Geol.-Pal~tont.
Inst. Univ. Hamburg, SCOPE/UNEP
Sonderband.
62,
42 (1987)
427