Carbonate Dissolution in the Deep-Sea
259
The Use of Bulk Sediment Parameters as
Dissolution Proxies
Quantitative reconstruction of carbonate dissolution to times of the past is not a simple matter. It
requires the determination of the fraction of calcite rained to the sea-floor which survived dissolution. Unfortunately, among the several criteria used
to judge the state of preservation of the calcite most
remain qualitative. One indicator is the weight percentage ofthe coarse fraction (>63 11m). The sand
content of deep-sea carbonates decreases as dissolution progresses (Johnson et al. 1977; Berger et
al. 1982; Wu et al. 1990). The reason for this is that
foraminiferal shells are weakenend by dissolution
and tend to break down into small fragments. Thus,
material moves from the coarse fraction into finer
fractions. Inspection of other dissolution indices,
such as calcareous micro- and nannoplankton dissolution proxies (see chapter below), investigated
in deep-sea sediments by several authors (e.g ..
Hebbeln et al. 1990; Yasuda et al. 1993), shows a
good agreement with the sand content records of
each study. However, other sediment related studies on the deep-sea rise reveal that foraminiferal
fragmentation and hence the percentage of the fine
fraction increases before the significant overall loss
of carbonate begins, and thus may be more sensitive to changes in bottom water or pore water corrosiveness than bulk carbonate. For instance,
Peterson and Prell (1985) showed that about 60 %
of the non-fragmented sand-sized planktic
foraminifera were already broken at the lysocline
level, whereas no more than 20 % to 30 % of carbonate has been lost (cf. Fig. 12). This mismatch
in sensitivities may be due to the transfer of carbonate during the fragmentation from larger to
smaller size fractions. Furthermore, changes in the
rain ratio between nannofossil placoliths and
foraminiferal shells could bias the relative portion
of the coarse fraction without any changes in dissolution (e.g. Bickert and Wefer 1996). Therefore,
the sand content is not an unambiguous proxy for
dissolution and requires calibration with other dissolution indices prior to the interpretation of its
variation with time.
A potentially more quantitative index of calcite
dissolution is the percentage ofCaC0 3 in the sediment. Of course, variations in the percentage of
CaC0 3 in a single core cannot be simply interpreted as an index of preservation because the
relative abundance of carbonate is controlled by the
balance of productivity over dissolution and by dilution due to the influx of non-carbonate sedimentary components. Only in the ideal situation, where
the rain rate of calcitic and of noncalcitic material
are constant in space and time, the amount of calcite lost to dissolution could be calculated from the
percentage CaC0 3 in the sediment. Otherwise, for
each time interval of interest, the calcite content
of the sediment from above the lysocline has to be
used as the reference for the amount of dissolution which has occurred in cores from the transition zone (the "depth-transect approach"; e.g.
Farrell and Prell 1989; Curry and Lohmann 1990;
Bickert et al. 1997). However, the fact that
CaC0 3 -contents of supralysoclinal sediment average to about 90 % in the world ocean (Archer
1996) raises the problem that quite large amounts
of dissolution create only very small changes in the
carbonate content. For example, if a sediment
which, in the absence of dissolution, would have a
calcite content of 90 % were to lose half of its
calcite to dissolution, its CaCq -content would drop
to only 82 %. Because of this, even small variations in the ratio ofthe rain rate of calcite to the
rain rate of non-calcite would lead to substantial
errors in the extent of dissolution.
One way out ofthis dilemma could be the conversion of CaC0 3 % (w/w) to CaC0 3 mass accumulation rate (MAR), which corrects for the
effect of dilution in the sediment. According to van
Andel et al. (1975) the CaC0 3 -MAR is calculated
as:
CaC0 3 -MAR (g/cm2/ky) = CaC0 3 % (w/w) . DBD
(glcm 3 ). SR(cmlky)
(1)
This calculation requires estimates of dry bulk
densities (DBD) for the sediments and sedimentation rates (SR). DBD, if not measured directly,
might be calculated using the empiric equation of
Ruddiman and Janecek (1989). The major problem
comes from the SR, which is commonly derived by
linear interpolation between stratigraphic datums
based on oxygen isotope, paleomagnetic and
259
The Use of Bulk Sediment Parameters as
Dissolution Proxies
Quantitative reconstruction of carbonate dissolution to times of the past is not a simple matter. It
requires the determination of the fraction of calcite rained to the sea-floor which survived dissolution. Unfortunately, among the several criteria used
to judge the state of preservation of the calcite most
remain qualitative. One indicator is the weight percentage ofthe coarse fraction (>63 11m). The sand
content of deep-sea carbonates decreases as dissolution progresses (Johnson et al. 1977; Berger et
al. 1982; Wu et al. 1990). The reason for this is that
foraminiferal shells are weakenend by dissolution
and tend to break down into small fragments. Thus,
material moves from the coarse fraction into finer
fractions. Inspection of other dissolution indices,
such as calcareous micro- and nannoplankton dissolution proxies (see chapter below), investigated
in deep-sea sediments by several authors (e.g ..
Hebbeln et al. 1990; Yasuda et al. 1993), shows a
good agreement with the sand content records of
each study. However, other sediment related studies on the deep-sea rise reveal that foraminiferal
fragmentation and hence the percentage of the fine
fraction increases before the significant overall loss
of carbonate begins, and thus may be more sensitive to changes in bottom water or pore water corrosiveness than bulk carbonate. For instance,
Peterson and Prell (1985) showed that about 60 %
of the non-fragmented sand-sized planktic
foraminifera were already broken at the lysocline
level, whereas no more than 20 % to 30 % of carbonate has been lost (cf. Fig. 12). This mismatch
in sensitivities may be due to the transfer of carbonate during the fragmentation from larger to
smaller size fractions. Furthermore, changes in the
rain ratio between nannofossil placoliths and
foraminiferal shells could bias the relative portion
of the coarse fraction without any changes in dissolution (e.g. Bickert and Wefer 1996). Therefore,
the sand content is not an unambiguous proxy for
dissolution and requires calibration with other dissolution indices prior to the interpretation of its
variation with time.
A potentially more quantitative index of calcite
dissolution is the percentage ofCaC0 3 in the sediment. Of course, variations in the percentage of
CaC0 3 in a single core cannot be simply interpreted as an index of preservation because the
relative abundance of carbonate is controlled by the
balance of productivity over dissolution and by dilution due to the influx of non-carbonate sedimentary components. Only in the ideal situation, where
the rain rate of calcitic and of noncalcitic material
are constant in space and time, the amount of calcite lost to dissolution could be calculated from the
percentage CaC0 3 in the sediment. Otherwise, for
each time interval of interest, the calcite content
of the sediment from above the lysocline has to be
used as the reference for the amount of dissolution which has occurred in cores from the transition zone (the "depth-transect approach"; e.g.
Farrell and Prell 1989; Curry and Lohmann 1990;
Bickert et al. 1997). However, the fact that
CaC0 3 -contents of supralysoclinal sediment average to about 90 % in the world ocean (Archer
1996) raises the problem that quite large amounts
of dissolution create only very small changes in the
carbonate content. For example, if a sediment
which, in the absence of dissolution, would have a
calcite content of 90 % were to lose half of its
calcite to dissolution, its CaCq -content would drop
to only 82 %. Because of this, even small variations in the ratio ofthe rain rate of calcite to the
rain rate of non-calcite would lead to substantial
errors in the extent of dissolution.
One way out ofthis dilemma could be the conversion of CaC0 3 % (w/w) to CaC0 3 mass accumulation rate (MAR), which corrects for the
effect of dilution in the sediment. According to van
Andel et al. (1975) the CaC0 3 -MAR is calculated
as:
CaC0 3 -MAR (g/cm2/ky) = CaC0 3 % (w/w) . DBD
(glcm 3 ). SR(cmlky)
(1)
This calculation requires estimates of dry bulk
densities (DBD) for the sediments and sedimentation rates (SR). DBD, if not measured directly,
might be calculated using the empiric equation of
Ruddiman and Janecek (1989). The major problem
comes from the SR, which is commonly derived by
linear interpolation between stratigraphic datums
based on oxygen isotope, paleomagnetic and