132
Baumann et al.
upwelling intensity, with a deep thermocline and
nutricline and causes the relative abundance of F.
profunda to increase. The opposite scenario consisting in the reduced heat exposure of the African land mass due to an insolation minimum, will
cause an increase of the southeasterly trade wind
component, which is the driving factor for intense
upwelling. Intensified trade winds cause a shoaling
of both nutricline and thermocline. They also increase the upwelling area. The shoaling of the
nutricline provides the upper euphotic zone, where
most of the coccolithophores dwell, with sufficient
nutrients leading to an increased productivity. This
can be seen from both the increased coccolith accumulation rates and minimum values of relative
abundance ofF. profunda (Fig. 10). Most significant maxima in coccolith accumulation occur in late
stage 8, early stage 7 and at the stage 7/6 boundary. Throughout stages 6 and 5, the accumulation
of coccolith remains relatively low demonstrating
only smaller fluctuations. At the end of stage 4, the
coccolith accumulation rates rise again reaching
highest values in stage 3, where fluctuations are
extreme. Since the last glacial, the accumulation
rates have been dropping to very low values in the
Holocene that are comparable to those of stage 5.
In general, abundance fluctuations of F. profunda
show an opposite trend. The majority of species
that cause the maximal coccolith accumulation
rates (E. huxleyi, Gephyrocapsa spp. and C.
leptoporus) have all been reported from areas of
upwelling or higher productivity (Kleijne et al. 1989;
Giraudeau 1992; Knappertsbusch 1993; Flores et
al. 1997). In general, the coccolith accumulation
rates show similar trends as paleoproductivity estimates based on carbonate or organic carbon accumulation rates (Bickert 1992; Riihlemann 1996;
Schneider et al. 1996; Wefer et al. 1996).
The presented results in general confirm those
of Mol fino and McIntyre (1990). However, it can
be concluded that the mechanism that causes a
"F. profunda signal" differs slightly from their conceptual model. Changes in upper euphotic zone productivity control the occurrence of the surfacewater species, whereas the production of F.
profunda should remain stable, as there are always
sufficient nutrients available in its depth habitat.
Moreover, alteration of the communities which are
characteristic for the oligotrophic surface-water
assemblages (such as the decrease in abundance
of U. irregularis and U. tenuis) amplifies the
abundance of F. profunda with respect to the total coccolithophore flora in the sediments below.
The response of F. profunda to a changing
nutricline depth has been reported from the Northwestern Pacific (Ahagon et al. 1993), the Indian
Ocean (Okada and Matsuoka 1996; Beaufort et al.
1997) and the western tropical Atlantic (Bassinot
et al. 1997). In all records, F. profunda showed
cyclicities centered on different frequency bands
which control the nutricline depth in those areas.
This gives further evidence that nutricline dynamics rather than other mechanisms that are influenced by precessional forcing cause variations of
F. profunda.
Sea-Surface Temperatures Reconstructions
In the following we will discuss two different methods for the use of coccolithophores as proxies in
sea-surface temperature reconstructions. These
reconstructions are mainly based on the species E.
huxleyi and those of the genus Gephyrocapsa.
The first method deals with the micropaleontological approach considering the evolutionary development and the downcore fluctuations of
abundances in correlation with the isotope stages
of these taxa. The second method presents a geochemical approach. Here we present and discuss
the results ofMiiller et al. (1997) on the alkenone
temperature method as applied to the abundances
of coccolithophores.
Micropaleontological Approach
Downcore variations in coccolithophore assemblages can be used to determine relative temperature changes, although the paleoclimatic meaning
of some of the coccolithophore taxa, as stated
above, remains controversial.
The highest numbers of coccoliths in core GeoB
1028 (Fig. 9) off Namibia are observed in the relatively warm interglacial stages, except in isotope
stage 6 where high numbers are recorded.
Both concentrations and relative abundances of
coccoliths vary between glacials and interglacials
in concert with sea-surface temperature variations
Baumann et al.
upwelling intensity, with a deep thermocline and
nutricline and causes the relative abundance of F.
profunda to increase. The opposite scenario consisting in the reduced heat exposure of the African land mass due to an insolation minimum, will
cause an increase of the southeasterly trade wind
component, which is the driving factor for intense
upwelling. Intensified trade winds cause a shoaling
of both nutricline and thermocline. They also increase the upwelling area. The shoaling of the
nutricline provides the upper euphotic zone, where
most of the coccolithophores dwell, with sufficient
nutrients leading to an increased productivity. This
can be seen from both the increased coccolith accumulation rates and minimum values of relative
abundance ofF. profunda (Fig. 10). Most significant maxima in coccolith accumulation occur in late
stage 8, early stage 7 and at the stage 7/6 boundary. Throughout stages 6 and 5, the accumulation
of coccolith remains relatively low demonstrating
only smaller fluctuations. At the end of stage 4, the
coccolith accumulation rates rise again reaching
highest values in stage 3, where fluctuations are
extreme. Since the last glacial, the accumulation
rates have been dropping to very low values in the
Holocene that are comparable to those of stage 5.
In general, abundance fluctuations of F. profunda
show an opposite trend. The majority of species
that cause the maximal coccolith accumulation
rates (E. huxleyi, Gephyrocapsa spp. and C.
leptoporus) have all been reported from areas of
upwelling or higher productivity (Kleijne et al. 1989;
Giraudeau 1992; Knappertsbusch 1993; Flores et
al. 1997). In general, the coccolith accumulation
rates show similar trends as paleoproductivity estimates based on carbonate or organic carbon accumulation rates (Bickert 1992; Riihlemann 1996;
Schneider et al. 1996; Wefer et al. 1996).
The presented results in general confirm those
of Mol fino and McIntyre (1990). However, it can
be concluded that the mechanism that causes a
"F. profunda signal" differs slightly from their conceptual model. Changes in upper euphotic zone productivity control the occurrence of the surfacewater species, whereas the production of F.
profunda should remain stable, as there are always
sufficient nutrients available in its depth habitat.
Moreover, alteration of the communities which are
characteristic for the oligotrophic surface-water
assemblages (such as the decrease in abundance
of U. irregularis and U. tenuis) amplifies the
abundance of F. profunda with respect to the total coccolithophore flora in the sediments below.
The response of F. profunda to a changing
nutricline depth has been reported from the Northwestern Pacific (Ahagon et al. 1993), the Indian
Ocean (Okada and Matsuoka 1996; Beaufort et al.
1997) and the western tropical Atlantic (Bassinot
et al. 1997). In all records, F. profunda showed
cyclicities centered on different frequency bands
which control the nutricline depth in those areas.
This gives further evidence that nutricline dynamics rather than other mechanisms that are influenced by precessional forcing cause variations of
F. profunda.
Sea-Surface Temperatures Reconstructions
In the following we will discuss two different methods for the use of coccolithophores as proxies in
sea-surface temperature reconstructions. These
reconstructions are mainly based on the species E.
huxleyi and those of the genus Gephyrocapsa.
The first method deals with the micropaleontological approach considering the evolutionary development and the downcore fluctuations of
abundances in correlation with the isotope stages
of these taxa. The second method presents a geochemical approach. Here we present and discuss
the results ofMiiller et al. (1997) on the alkenone
temperature method as applied to the abundances
of coccolithophores.
Micropaleontological Approach
Downcore variations in coccolithophore assemblages can be used to determine relative temperature changes, although the paleoclimatic meaning
of some of the coccolithophore taxa, as stated
above, remains controversial.
The highest numbers of coccoliths in core GeoB
1028 (Fig. 9) off Namibia are observed in the relatively warm interglacial stages, except in isotope
stage 6 where high numbers are recorded.
Both concentrations and relative abundances of
coccoliths vary between glacials and interglacials
in concert with sea-surface temperature variations