Oxygen Isotope Values of Planktic Foraminifera
171
Southern Ocean, because large-sized specimens do
not occur continually during glacial periods. If the
same size fraction is always used for each species
for the temperature calculation, the ontogenetic offset can be neglected.
Usually, a range for the "oxygen isotope disequilibrium" is given in literature for the most species
(Table 2). For G. bulloides (230 to 250 /lID), for
instance, the "oxygen isotope disequilibrium" ranges
between -0.5 %0 (culture experiments: Spero and
Lea 1996) and +0.5 %0 (plankton tows: Ganssen
1983). Depth habitat observations by net-tows for
G. bulloides have shown a maximum occurrence
between 20 and 50 m (Ottens 1992), consequently
an "oxygen isotope disequilibrium" of+O.5 %0 must
be assumed. Therefore, we used this "oxygen isotope disequilibrium" forGo bulloides (Hemleben
et al. 1989), which was subtracted from the measured values prior to temperature calculation.
If the relevant past "ice-effect" is known (the
"ice-effect" is governed by the differential loss of
160 as compared to 180, when isotopically light
water accumulates on land in the form of ice during glacial periods), the sea-surface temperature
(eg. Emiliani 1955; Ganssen 1983; Keigwin 1996),
or the salinity (Rostek et al. 1993; Niebler 1995,
Maslin et al. 1995; Wolffet al. this volume) can be
estimated from the shell 0 18 0 calc;te of surface-dwelling planktic foraminifera.
After the correction of the "ice-effect" and the
"oxygen isotope disequilibrium", the oxygen isotope
temperatures were calculated using the paleotemperature equation ofShakleton (1974), because
this equation (1) is also calibrated for low temperatures:
where T is the temperature during calcification, 0 c
is the isotopic composition of the calcite of the
foraminiferal shell in %0 relative to the Pee-DeeBelemnite (PDB) standard and Ow is the isotopic
composition ofthe ambient sea-water in %0 (PDB).
For the purpose oftemperature reconstruction, the
o value has to be converted from the PDB standw
ard to the Vienna Standard-Mean-Ocean-Water
(V-SMOW) (Coplen et al. 1983, Hut 1987):
(2)
where 0180 w is the isotopic composition of the
ambient sea-water in %0 (V -SMOW). To calculate
the 0180 w value, the Southern Oceans surface water relationship between 0180 w and salinity as reported by Duplessy et al. (1991) was used:
1) 18 0 w =-18.791 +O.546xS
(3)
where S denotes the salinity. Salinity data are mean
summer values for a water depth of30 m and 250
m, respectively, and were taken from Levitus and
Boyer (1994) for the area north of 45°S and from
Olbers et al. (1992) for the area south of 45°S. This
method of 0180 w reconstruction depends on the
assumption that the spatial patterns in ocean salinity do not change significantly with time. A potential input of 160-rich meltwater in the South Atlantic Ocean at the end of the Last Glacial as proposed
by Francois et al. (1997) would affect the temperature reconstruction. But this scenario is controversially disscussed. The method ofo 180 w reconstruction is discussed in detail by Wolff et al. in this volume.
As measured oxygen isotope values of seawater were not available for comparison purposes,
we calculated the Oc value for calcite formed in
equilibrium with sea-water by reconverting equation(1):
I) =1) +21.9- .J310.61 + 10 x T
c
w
Oxygen Isotope Composition of Planktic
Foraminifera across a N-S Profile
(4)
To characterize the 0180 cald ,e fractionation of
planktic foraminifera, the isotope composition in five
size fractions of24 species was determined at five
locations between the Antarctic Polar Front and
the southernmost Subtropical Gyre, and additionally at the W alvis-Ridge. These six locations were
selected, in particular to obtain information of every
characteristic Zone in the ACC and the adjacent
SG. The diversity of species and morphotypes decreases southwards, towards lower water tem-
171
Southern Ocean, because large-sized specimens do
not occur continually during glacial periods. If the
same size fraction is always used for each species
for the temperature calculation, the ontogenetic offset can be neglected.
Usually, a range for the "oxygen isotope disequilibrium" is given in literature for the most species
(Table 2). For G. bulloides (230 to 250 /lID), for
instance, the "oxygen isotope disequilibrium" ranges
between -0.5 %0 (culture experiments: Spero and
Lea 1996) and +0.5 %0 (plankton tows: Ganssen
1983). Depth habitat observations by net-tows for
G. bulloides have shown a maximum occurrence
between 20 and 50 m (Ottens 1992), consequently
an "oxygen isotope disequilibrium" of+O.5 %0 must
be assumed. Therefore, we used this "oxygen isotope disequilibrium" forGo bulloides (Hemleben
et al. 1989), which was subtracted from the measured values prior to temperature calculation.
If the relevant past "ice-effect" is known (the
"ice-effect" is governed by the differential loss of
160 as compared to 180, when isotopically light
water accumulates on land in the form of ice during glacial periods), the sea-surface temperature
(eg. Emiliani 1955; Ganssen 1983; Keigwin 1996),
or the salinity (Rostek et al. 1993; Niebler 1995,
Maslin et al. 1995; Wolffet al. this volume) can be
estimated from the shell 0 18 0 calc;te of surface-dwelling planktic foraminifera.
After the correction of the "ice-effect" and the
"oxygen isotope disequilibrium", the oxygen isotope
temperatures were calculated using the paleotemperature equation ofShakleton (1974), because
this equation (1) is also calibrated for low temperatures:
where T is the temperature during calcification, 0 c
is the isotopic composition of the calcite of the
foraminiferal shell in %0 relative to the Pee-DeeBelemnite (PDB) standard and Ow is the isotopic
composition ofthe ambient sea-water in %0 (PDB).
For the purpose oftemperature reconstruction, the
o value has to be converted from the PDB standw
ard to the Vienna Standard-Mean-Ocean-Water
(V-SMOW) (Coplen et al. 1983, Hut 1987):
(2)
where 0180 w is the isotopic composition of the
ambient sea-water in %0 (V -SMOW). To calculate
the 0180 w value, the Southern Oceans surface water relationship between 0180 w and salinity as reported by Duplessy et al. (1991) was used:
1) 18 0 w =-18.791 +O.546xS
(3)
where S denotes the salinity. Salinity data are mean
summer values for a water depth of30 m and 250
m, respectively, and were taken from Levitus and
Boyer (1994) for the area north of 45°S and from
Olbers et al. (1992) for the area south of 45°S. This
method of 0180 w reconstruction depends on the
assumption that the spatial patterns in ocean salinity do not change significantly with time. A potential input of 160-rich meltwater in the South Atlantic Ocean at the end of the Last Glacial as proposed
by Francois et al. (1997) would affect the temperature reconstruction. But this scenario is controversially disscussed. The method ofo 180 w reconstruction is discussed in detail by Wolff et al. in this volume.
As measured oxygen isotope values of seawater were not available for comparison purposes,
we calculated the Oc value for calcite formed in
equilibrium with sea-water by reconverting equation(1):
I) =1) +21.9- .J310.61 + 10 x T
c
w
Oxygen Isotope Composition of Planktic
Foraminifera across a N-S Profile
(4)
To characterize the 0180 cald ,e fractionation of
planktic foraminifera, the isotope composition in five
size fractions of24 species was determined at five
locations between the Antarctic Polar Front and
the southernmost Subtropical Gyre, and additionally at the W alvis-Ridge. These six locations were
selected, in particular to obtain information of every
characteristic Zone in the ACC and the adjacent
SG. The diversity of species and morphotypes decreases southwards, towards lower water tem-