552
Wagner and Dupont
towards heavier values by I3C-enriched carbon
from terrestrial C4-herbs, which include many
grasses (range -8 to -19 %0, average -12 %0,
Sackett 1989). Terrestrial C4-plants exhibit optimum photosynthesis at temperatures which are
about 10°C higher as compared to C3-plants. They
are thus common in warm tropical and dry to arid
climates. Hence, at appropriate latitudes, such as
the Sahel and Sahara in Africa, masking ofthe bulk
terrestrial signal by C4-plant carbon may be more
important than in mid- and high latitudes (MUller et
a!. 1983; Mariotti eta!. 1991; Wagner 1998). This
effect, for example, is documented in African
lacustrine sediments showing low ol3C mg values
around -28 %0 during forested periods of the
Holocene, whereas Ol3C values were much higher
reaching up to -10 %0 (Talbot and Johannessen
1992) during the Last Glacial Maximum, when open
grass-rich savannah was the dominant vegetation
type. Marine OM of low and middle latitudes, in
contrast, is enriched in J3C carbon and reveals typical isotopic values between -18.5 to -21.5 %0 (e.g.
Sackett 1989). This accounts for a total average
variation of 5-8 %0 between marine and typical terrestrial (C3) OM, at least in low and mid-latitude
oceans (Tyson 1995).
This isotopic offset is traditionally used to determine the relative proportions of marine and
terrigenous organic carbon by applying binary mixing models (e.g. Jasper and Gagosian 1990;
Prahl and Mlihlhausen 1989; Schneider et al. 1994;
RUhlemann et a!. 1996). Definition of marine and
terrestrial endmembers, however, is complicated
due to a number of reasons. First, the variability of
values determined among individual plankton samples broadly scatters between -13 and -31 %0
(Tyson 1995). This signal obviously depends on temperature and PC0 2 (Rau et al. 1991; Fischer et at.
1998). Westerhausen et at. (1993) therefore proposed a model which delineate isotopic
endmembers for quantitative estimates which were
standardized to temperature and water depth. Second, a shift of the terrigenous isotopic signal towards heavier values has to be considered if C4carbon supply is indicated. Third, difficulties are
addressed to the alteration of the isotopic signal
during early diagenesis, although there is no common agreement concerning the importance of such
effects (for further discussion see Tyson 1995).
Fourth, new evidence from the eastern Equatorial
Atlantic indicates, that glacial/interglacial fluctuations in ol3C mg may also be related to variable CO 2
concentrations in past surface waters (MUller et al.
1994). A final potential mechanism is proposed by
Prahl and MUhlhausen (1989) suggesting that
ol3C o ,g of marine deposits may be masked by elevated proportions of proteinaceous components as
amino acids are generally enriched in I3C. The latter factor, however, is not yet well constrained.
In conclusion, it appears questionable whether
the bulk OJ3C mg signal of marine sediments deserves the popularity in the scientific community as
a quantitative proxy for marine vs. terrigenous OM,
since there are various independent mechanisms
which interact and may lead to a modification of
the bulk Ol3C record.
org
Pyrolytic Approaches
A commonly used method to assess terrigenous
OM in marine sediments is Rock-Eval Pyrolysis
(e.g. as part of the standard analytical procedure
of the Ocean Drilling Program). This analytical
device was originally developed to screen the quality and thermal maturity of hydrocarbon source
rocks (Espitalie et at. 1977), but has become a
standard method for the fast characterization of bulk
organic carbon in young, immature sediments which
in general address actual paleoceanographic topics. Three parameters "Hydrogen Index" (HI),
"Oxygen Index" (01), and "T m,," are of special interest for the differentiation of marine- or terrestrial-dominated OM and the identification of reworked, fossil OM. HI and OI are used similar to
Hie and OIC atomic ratios to construct modified
"van Krevelen" diagrams (e.g. Peters 1986;
Delveaux et al. 1990) which enable the determination of kerogen types I to IV (representing a
continuous mixure of biodegraded, marine, terrestrial, and oxidized OM).
A first rough estimate of the relative proportions
of marine and terrigenous OM can principally be
deduced from HI, which displays the amount of
pyrolysable hydrocarbons from kerogen (S2) normalized to TOC (HI=(S2* 100)/TOC). This application uses the fact that OM derived from marine
Wagner and Dupont
towards heavier values by I3C-enriched carbon
from terrestrial C4-herbs, which include many
grasses (range -8 to -19 %0, average -12 %0,
Sackett 1989). Terrestrial C4-plants exhibit optimum photosynthesis at temperatures which are
about 10°C higher as compared to C3-plants. They
are thus common in warm tropical and dry to arid
climates. Hence, at appropriate latitudes, such as
the Sahel and Sahara in Africa, masking ofthe bulk
terrestrial signal by C4-plant carbon may be more
important than in mid- and high latitudes (MUller et
a!. 1983; Mariotti eta!. 1991; Wagner 1998). This
effect, for example, is documented in African
lacustrine sediments showing low ol3C mg values
around -28 %0 during forested periods of the
Holocene, whereas Ol3C values were much higher
reaching up to -10 %0 (Talbot and Johannessen
1992) during the Last Glacial Maximum, when open
grass-rich savannah was the dominant vegetation
type. Marine OM of low and middle latitudes, in
contrast, is enriched in J3C carbon and reveals typical isotopic values between -18.5 to -21.5 %0 (e.g.
Sackett 1989). This accounts for a total average
variation of 5-8 %0 between marine and typical terrestrial (C3) OM, at least in low and mid-latitude
oceans (Tyson 1995).
This isotopic offset is traditionally used to determine the relative proportions of marine and
terrigenous organic carbon by applying binary mixing models (e.g. Jasper and Gagosian 1990;
Prahl and Mlihlhausen 1989; Schneider et al. 1994;
RUhlemann et a!. 1996). Definition of marine and
terrestrial endmembers, however, is complicated
due to a number of reasons. First, the variability of
values determined among individual plankton samples broadly scatters between -13 and -31 %0
(Tyson 1995). This signal obviously depends on temperature and PC0 2 (Rau et al. 1991; Fischer et at.
1998). Westerhausen et at. (1993) therefore proposed a model which delineate isotopic
endmembers for quantitative estimates which were
standardized to temperature and water depth. Second, a shift of the terrigenous isotopic signal towards heavier values has to be considered if C4carbon supply is indicated. Third, difficulties are
addressed to the alteration of the isotopic signal
during early diagenesis, although there is no common agreement concerning the importance of such
effects (for further discussion see Tyson 1995).
Fourth, new evidence from the eastern Equatorial
Atlantic indicates, that glacial/interglacial fluctuations in ol3C mg may also be related to variable CO 2
concentrations in past surface waters (MUller et al.
1994). A final potential mechanism is proposed by
Prahl and MUhlhausen (1989) suggesting that
ol3C o ,g of marine deposits may be masked by elevated proportions of proteinaceous components as
amino acids are generally enriched in I3C. The latter factor, however, is not yet well constrained.
In conclusion, it appears questionable whether
the bulk OJ3C mg signal of marine sediments deserves the popularity in the scientific community as
a quantitative proxy for marine vs. terrigenous OM,
since there are various independent mechanisms
which interact and may lead to a modification of
the bulk Ol3C record.
org
Pyrolytic Approaches
A commonly used method to assess terrigenous
OM in marine sediments is Rock-Eval Pyrolysis
(e.g. as part of the standard analytical procedure
of the Ocean Drilling Program). This analytical
device was originally developed to screen the quality and thermal maturity of hydrocarbon source
rocks (Espitalie et at. 1977), but has become a
standard method for the fast characterization of bulk
organic carbon in young, immature sediments which
in general address actual paleoceanographic topics. Three parameters "Hydrogen Index" (HI),
"Oxygen Index" (01), and "T m,," are of special interest for the differentiation of marine- or terrestrial-dominated OM and the identification of reworked, fossil OM. HI and OI are used similar to
Hie and OIC atomic ratios to construct modified
"van Krevelen" diagrams (e.g. Peters 1986;
Delveaux et al. 1990) which enable the determination of kerogen types I to IV (representing a
continuous mixure of biodegraded, marine, terrestrial, and oxidized OM).
A first rough estimate of the relative proportions
of marine and terrigenous OM can principally be
deduced from HI, which displays the amount of
pyrolysable hydrocarbons from kerogen (S2) normalized to TOC (HI=(S2* 100)/TOC). This application uses the fact that OM derived from marine