masses that sink and spread at depth. These young,
relatively well oxygenated and [CO
2À
3 ]-enriched
waters tend to depress the depth of the saturation
horizon and allow carbonate to accumulate over much
of the Atlantic basin, as manifested by a deep lysocline
and CCD. In contrast, neither the Indian nor Pacific
Oceans today experience surface conditions that allow
deep or bottom waters to form; water masses at depth
in these basins largely originate in the Atlantic sector as
part of what is sometimes described as the ocean’s
conveyor belt circulation, with a general upwelling of
waters from depth balancing the formation and sinking
of deep waters in the Atlantic source areas. Since deep
and bottom waters in the Indian and Pacific Oceans are
further removed from their modern source areas in the
Atlantic, they tend to be CO 2 -enriched and [CO 3
2À
]depleted because of their greater age and the cumulative effects of organic matter remineralization along
their flow path. In particular, the in situ decrease in
[CO 3
2À ] concentration leads to an increase in undersaturation of the water masses and a progressive
shoaling of the saturation horizon (Figure 3). Thus,
Indian and Pacific deep waters are generally more
corrosive to the biogenic carbonate phases than Atlantic waters at comparable depth, the lysocline and
CCD are shallower, and a smaller area of the seafloor
experiences conditions suitable for carbonate preservation and accumulation. This pronounced modern
pattern of basin-to-basin fractionation is illustrated by
the fact that roughly 65% of the present Atlantic seafloor is covered by carbonate ooze, while only 54% of
the Indian Ocean floor and 36% of the Pacific Ocean
floor share that distinction. Naturally, if thermohaline
circulation patterns have changed in the past, then
carbonate preservation and accumulation patterns will
change accordingly. The mapping and reconstruction
of such trends has emerged as a powerful paleoceanographic tool.
Temporal Changes in Carbonate
Accumulation and Preservation
The patterns of carbonate accumulation and preservation in the deep sea contain important information about the chemistry and fertility of ancient
oceans. Numerous studies have now shown that
variations in the carbonate system have occurred on
a variety of timescales, both within and between
ocean basins. On a local or even regional scale,
such variations can often be used as a correlation
tool. This has come to be known as ‘preservation
stratigraphy’.
A number of criteria have commonly been used as
indicators of the intensity of carbonate dissolution in
deep-sea sediments. Variations in the measured carbonate content of sediments are commonly used to
correlate between cores in a region, but are difficult
to interpret strictly in terms of dissolution and
changing deep-water chemistry. This is because the
weight percent carbonate content of a sample can
also be affected by changing carbonate input (i.e.
surface production) and by dilution from noncarbonate sources. More useful are indices based on
some direct measure of preservation state, such as
the percentage of foraminiferal fragments in a sample relative to whole shells (Figure 5). However,
while clearly recording dissolution, preservationbased indices can also be affected by other factors,
including ecologic changes that may introduce variable proportions of solution-susceptible species into
a region over time.
Because carbonate dissolution is a depth-dependent
process, it is best studied where existing seafloor
topography allows for sampling of sediments over a
broad depth range. Given this sampling strategy, one
way to circumvent the problems of using measured
carbonate content and other relative dissolution indices (e.g. fragmentation) is to calculate carbonate
accumulation histories for the individual sampling
locations and examine depth-dependent differences
in accumulation rates and patterns. To do so requires
an accurate knowledge of sedimentation rates (e.g.
cm per thousand years) and measurements of sediment bulk density (in g cm
À3 ), in addition to the data
on carbonate content. The product of these three
measures yields a mass accumulation rate for the
carbonate component expressed in g per cm
2 per
thousand years. Differences in accumulation between
depth-distributed sites can provide insights into dissolution gradients and carbonate loss.
As the relative importance of calcium supply from
weathering and carbonate production vary through
time, the depth of the CCD must adjust to control
dissolution and to keep calcium levels in balance.
Studies of CCD behavior during the Cenozoic
(Figure 6) have generally shown that CCD fluctuations were similar in the various ocean basins and
were likely to have been driven by a global mechanism, such as a change in sea level and/or hypsometry of the ocean basins or a change in supply of
calcium to the oceans. There are, however, clear
ocean-to-ocean differences in this general pattern
that are likely to have been the result of changes in
regional productivity and the interbasinal exchange
of deep and surface waters. By examining such differences, estimates of past circulation and of the
relative differences in carbonate productivity in different regions can be determined from regional offsets in the depth of the CCD.
342 CALCIUM CARBONATES
relatively well oxygenated and [CO
2À
3 ]-enriched
waters tend to depress the depth of the saturation
horizon and allow carbonate to accumulate over much
of the Atlantic basin, as manifested by a deep lysocline
and CCD. In contrast, neither the Indian nor Pacific
Oceans today experience surface conditions that allow
deep or bottom waters to form; water masses at depth
in these basins largely originate in the Atlantic sector as
part of what is sometimes described as the ocean’s
conveyor belt circulation, with a general upwelling of
waters from depth balancing the formation and sinking
of deep waters in the Atlantic source areas. Since deep
and bottom waters in the Indian and Pacific Oceans are
further removed from their modern source areas in the
Atlantic, they tend to be CO 2 -enriched and [CO 3
2À
]depleted because of their greater age and the cumulative effects of organic matter remineralization along
their flow path. In particular, the in situ decrease in
[CO 3
2À ] concentration leads to an increase in undersaturation of the water masses and a progressive
shoaling of the saturation horizon (Figure 3). Thus,
Indian and Pacific deep waters are generally more
corrosive to the biogenic carbonate phases than Atlantic waters at comparable depth, the lysocline and
CCD are shallower, and a smaller area of the seafloor
experiences conditions suitable for carbonate preservation and accumulation. This pronounced modern
pattern of basin-to-basin fractionation is illustrated by
the fact that roughly 65% of the present Atlantic seafloor is covered by carbonate ooze, while only 54% of
the Indian Ocean floor and 36% of the Pacific Ocean
floor share that distinction. Naturally, if thermohaline
circulation patterns have changed in the past, then
carbonate preservation and accumulation patterns will
change accordingly. The mapping and reconstruction
of such trends has emerged as a powerful paleoceanographic tool.
Temporal Changes in Carbonate
Accumulation and Preservation
The patterns of carbonate accumulation and preservation in the deep sea contain important information about the chemistry and fertility of ancient
oceans. Numerous studies have now shown that
variations in the carbonate system have occurred on
a variety of timescales, both within and between
ocean basins. On a local or even regional scale,
such variations can often be used as a correlation
tool. This has come to be known as ‘preservation
stratigraphy’.
A number of criteria have commonly been used as
indicators of the intensity of carbonate dissolution in
deep-sea sediments. Variations in the measured carbonate content of sediments are commonly used to
correlate between cores in a region, but are difficult
to interpret strictly in terms of dissolution and
changing deep-water chemistry. This is because the
weight percent carbonate content of a sample can
also be affected by changing carbonate input (i.e.
surface production) and by dilution from noncarbonate sources. More useful are indices based on
some direct measure of preservation state, such as
the percentage of foraminiferal fragments in a sample relative to whole shells (Figure 5). However,
while clearly recording dissolution, preservationbased indices can also be affected by other factors,
including ecologic changes that may introduce variable proportions of solution-susceptible species into
a region over time.
Because carbonate dissolution is a depth-dependent
process, it is best studied where existing seafloor
topography allows for sampling of sediments over a
broad depth range. Given this sampling strategy, one
way to circumvent the problems of using measured
carbonate content and other relative dissolution indices (e.g. fragmentation) is to calculate carbonate
accumulation histories for the individual sampling
locations and examine depth-dependent differences
in accumulation rates and patterns. To do so requires
an accurate knowledge of sedimentation rates (e.g.
cm per thousand years) and measurements of sediment bulk density (in g cm
À3 ), in addition to the data
on carbonate content. The product of these three
measures yields a mass accumulation rate for the
carbonate component expressed in g per cm
2 per
thousand years. Differences in accumulation between
depth-distributed sites can provide insights into dissolution gradients and carbonate loss.
As the relative importance of calcium supply from
weathering and carbonate production vary through
time, the depth of the CCD must adjust to control
dissolution and to keep calcium levels in balance.
Studies of CCD behavior during the Cenozoic
(Figure 6) have generally shown that CCD fluctuations were similar in the various ocean basins and
were likely to have been driven by a global mechanism, such as a change in sea level and/or hypsometry of the ocean basins or a change in supply of
calcium to the oceans. There are, however, clear
ocean-to-ocean differences in this general pattern
that are likely to have been the result of changes in
regional productivity and the interbasinal exchange
of deep and surface waters. By examining such differences, estimates of past circulation and of the
relative differences in carbonate productivity in different regions can be determined from regional offsets in the depth of the CCD.
342 CALCIUM CARBONATES