Clues to Ocean History: a Brief Overview of Proxies
33
effect apparently is not large enough to explain the
discrepancy between Cd/Ca and /iI3C values. As
already mentioned, C. wuellerstorfi tends to show
light /iI3C values below regions with high primary
productivity (Vincent et al. 1981; Sarnthein et al.
1988). From investigations of protoplasm-filled
tests, Mackensen et al. (1993) were able to demonstrate that C. wuellerstorfi below high-productivity areas in the Southern Ocean have significantly
lower /iI3C values than LC0 2 from the water column. Changes in the carbonate chemistry could
also be responsible (Bijma et al. this volume; Lea
et al. this volume).
Additional clues on the question ofNADW influence in the Southern Ocean are provided by
231Pa/2 3 0Th measurements (e.g. Kumar et al.
1995). Analysis of radionuclides in the water column and in surface sediment transects in the South
Atlantic have shown that the meridional transport
ofNADW and Circumpolar Deep Water (CDW)
influences the distribution of dissolved and particlereactive radionuclides (Rutgers van der Loeff and
Berger 1993). The measurements indicate a depletion of 231Pa in the Atlantic and an excess in the
Southern Ocean, which is tied to particle transport
within the NADW (Yu et al. 1996). Because patterns of depletion and enrichment are the same in
the Holocene and the LGM, the influx ofNADW
into the Southern Ocean is taken to persist during
the LGM. Apparently, the 231 paJ230Th ratios do not
record paleoproductivity in the Southern Ocean
(Walter et al. this volume).
In addition to C. wuellerstorfi, other benthic
foraminifersl species have been considered as
proxies for past nutrient conditions. Mackensen et
al. (1993) reported isotope measurements on
Nutallides umbonifera which does not record the
low /iI3C values in high productivity areas as C.
wuellerstorfi does. This species may be better
suited to register the /iI3C values of the bottom
water. The disadvantage is, however, that this species is not as widely distributed asc. wuellerstorfi.
Other species or genera may also become useful.
Benthic foraminifers apparently exhibit no variation in /iI3C in the course of their growth (ontogeny) as has been reported for planktic foraminifers.
Dunbar and Wefer (1984) analyzed nine species
and six size classes from the upwelling area off
Peru and found virtually no influence by size on the
composition ofthese species, with the possible but
doubtful exeption of Uvigerina species. The Cd/
Ca ratio recorded in benthic foraminifers also appears to be independent of size (Boyle 1995).
The application of Boyle's method to planktic
foraminifers has proved difficult because of local
differences in phosphate-cadmium relationships in
the surface water (Boyle 1995). An additional factor is that planktic foraminifers are not anchored
to a single position as are the benthics, and can
therefore incorporate varying Cd concentrations
during fonnation of their tests, especially during the
crystallization of crusts. Boyle and Rosenthal
(1996) suggest that the use of this method be limited to bottom-water investigations.
Nitrogen Isotopes
A rather new development in the use of stable isotopes to obtain clues on ocean productivity and the
nutrient utilization is the analysis ofthe l SN/14N ratio
in organic matter (see Holmes et al. this volume).
The fractionation dynamics are analogous to those
of carbon isotopes. Assimilation of nitrate by
phytoplankton is accompanied by nitrogen isotope
fractionation and produces a strong gradient in/iISN
as the source nitrate is consumed and organic
matter is exported from the euphotic zone (e.g.
Owens 1987; Wada 1980). As a result, temporal
and spatial changes in the balance between nitrate
advection and consumption in the upper ocean will
lead to corresponding changes in the 8 15 N of
particulate nitrogen settling out of the productive
zone towards the sediment. Such changes in nutrient utilization typically occur in plankton ecosystems, whether rapid and intennittent (e.g. stonn induced; Montoya et al. 1991), slow and periodic (e.g.
by monsoonal upwelling; Schafer and Ittekkot
1995), or during the spring bloom (Altabet et al.
1991; Altabet 1996). Altabet et al. (1991), for instance, observed seasonal variations in/i 15 N in the
order of 7 0 / 00 in various types of particles associated with seasonal changes in the surface water
nitrate concentration and particle flux during the
JGOFS North Atlantic Spring Bloom Experiment.
33
effect apparently is not large enough to explain the
discrepancy between Cd/Ca and /iI3C values. As
already mentioned, C. wuellerstorfi tends to show
light /iI3C values below regions with high primary
productivity (Vincent et al. 1981; Sarnthein et al.
1988). From investigations of protoplasm-filled
tests, Mackensen et al. (1993) were able to demonstrate that C. wuellerstorfi below high-productivity areas in the Southern Ocean have significantly
lower /iI3C values than LC0 2 from the water column. Changes in the carbonate chemistry could
also be responsible (Bijma et al. this volume; Lea
et al. this volume).
Additional clues on the question ofNADW influence in the Southern Ocean are provided by
231Pa/2 3 0Th measurements (e.g. Kumar et al.
1995). Analysis of radionuclides in the water column and in surface sediment transects in the South
Atlantic have shown that the meridional transport
ofNADW and Circumpolar Deep Water (CDW)
influences the distribution of dissolved and particlereactive radionuclides (Rutgers van der Loeff and
Berger 1993). The measurements indicate a depletion of 231Pa in the Atlantic and an excess in the
Southern Ocean, which is tied to particle transport
within the NADW (Yu et al. 1996). Because patterns of depletion and enrichment are the same in
the Holocene and the LGM, the influx ofNADW
into the Southern Ocean is taken to persist during
the LGM. Apparently, the 231 paJ230Th ratios do not
record paleoproductivity in the Southern Ocean
(Walter et al. this volume).
In addition to C. wuellerstorfi, other benthic
foraminifersl species have been considered as
proxies for past nutrient conditions. Mackensen et
al. (1993) reported isotope measurements on
Nutallides umbonifera which does not record the
low /iI3C values in high productivity areas as C.
wuellerstorfi does. This species may be better
suited to register the /iI3C values of the bottom
water. The disadvantage is, however, that this species is not as widely distributed asc. wuellerstorfi.
Other species or genera may also become useful.
Benthic foraminifers apparently exhibit no variation in /iI3C in the course of their growth (ontogeny) as has been reported for planktic foraminifers.
Dunbar and Wefer (1984) analyzed nine species
and six size classes from the upwelling area off
Peru and found virtually no influence by size on the
composition ofthese species, with the possible but
doubtful exeption of Uvigerina species. The Cd/
Ca ratio recorded in benthic foraminifers also appears to be independent of size (Boyle 1995).
The application of Boyle's method to planktic
foraminifers has proved difficult because of local
differences in phosphate-cadmium relationships in
the surface water (Boyle 1995). An additional factor is that planktic foraminifers are not anchored
to a single position as are the benthics, and can
therefore incorporate varying Cd concentrations
during fonnation of their tests, especially during the
crystallization of crusts. Boyle and Rosenthal
(1996) suggest that the use of this method be limited to bottom-water investigations.
Nitrogen Isotopes
A rather new development in the use of stable isotopes to obtain clues on ocean productivity and the
nutrient utilization is the analysis ofthe l SN/14N ratio
in organic matter (see Holmes et al. this volume).
The fractionation dynamics are analogous to those
of carbon isotopes. Assimilation of nitrate by
phytoplankton is accompanied by nitrogen isotope
fractionation and produces a strong gradient in/iISN
as the source nitrate is consumed and organic
matter is exported from the euphotic zone (e.g.
Owens 1987; Wada 1980). As a result, temporal
and spatial changes in the balance between nitrate
advection and consumption in the upper ocean will
lead to corresponding changes in the 8 15 N of
particulate nitrogen settling out of the productive
zone towards the sediment. Such changes in nutrient utilization typically occur in plankton ecosystems, whether rapid and intennittent (e.g. stonn induced; Montoya et al. 1991), slow and periodic (e.g.
by monsoonal upwelling; Schafer and Ittekkot
1995), or during the spring bloom (Altabet et al.
1991; Altabet 1996). Altabet et al. (1991), for instance, observed seasonal variations in/i 15 N in the
order of 7 0 / 00 in various types of particles associated with seasonal changes in the surface water
nitrate concentration and particle flux during the
JGOFS North Atlantic Spring Bloom Experiment.