346
Gingele et al.
ity to minimize the preservation effect and to make
more reliable estimations of past productivity.
Barium occurs in the form of discrete barite
microcrystals in the water column, some of which
are transported and preserved in the sediment.
Goldberg and Arrhenius (1958) interpreted the
increase in barium concentrations below the equatorial divergence of the Pacific and Indian Oceans
as a result of intensified biogenic production in the
surface waters. Schmitz (1987) traced the northward shift ofthe Indian Plate during the Cenozoic
on the basis of increased barium concentrations in
the sediment, indicating the position of the plate
below the equatorial high-production belt. Estimations of relative variations in paleoproductivity
based on barium accumulation were carried out in
the Atlantic (Dehairs et al. 1980; Thomson and
Schmitz 1997), Pacific (\/ on Breymann et al. 1992),
Mediterranean (Van Santvoort et al. 1996) and
Southern Ocean (Shimmield et al. 1994) as well
as the Arabian Sea (Shimmield and Mowbray
1991). Quantitative calculations of new productivity and primary production rates from barium accumulation rates were applied to sediments ofthe
Atlantic (Gingele and Dahmke 1994), Mediterranean (Van Os et al. 1994) and Southern Ocean
(Niirnberg 1995; Bonn 1995; Bonn et al. 1998).
They are based on an algorithm developed
by Dymond et al. (1992) which is based on sediment trap data, and correlates the barium flux and
new production. The application of barium for
productivity estimates is not without pitfalls. This
paper will address the problems associated with
and the advantages for the use of barium for
paleoproductivity calculations.
Barium in the Marine System - Generation
of the Productivity Signal
Discrete micron-size barite crystals are the main
carriers of particulate barium in the water column
(Fig. 1). The mechanism of barite formation in
seawater, which is undersaturated with respect to
barite is still poorly understood. Profiles of dissolved
barium in the water column are characterized by
a minimum in the nutrient-depleted euphotic zone
and a continually increasing concentration in
greater depths (Chan et al. 1977), (Fig. 2). Their
similarity to profiles of dissolved silica and alkalinity is taken as evidence for the biogenic association of barium carriers. Maximum concentrations
of particulate barium can be observed just below
the euphotic zone, decreasing slightly with water
depth (Fig. 2). The size distribution of barite particles in the upper water column also suggests that
much of the barite is produced in large decaying
particles in shallow waters and are partly released
as fine particles in the oxygen minimum zone
(Bishop 1988; Dehairs et al. 1990). The nature of
these depth-profiles has led to the conclusion that
biogenic Ba is actively or passively precipitated in
or just below the euphotic zone and is gradually
dissolved with increasing water depth (Chan et al.
1977; Dehairsetal. 1980; Bishop 1988). Since barite
crystals have not yet been found in any major planktonic group, a precipitation in sulfate-barium-rich
microenvironments within decaying organic and
siliceous matter has been suggested (Dehairs et al.
1980; Bishop 1988). The sources of barium and
sulfate have not been determined yet. It has been
suggested that sulfate from decaying labile organic
matter supersaturates microenvironments with respect to barite (Chow and Goldberg 1960; Dehairs
etal. 1980; Bishop 1988) and dissolved barium diffuses from the surrounding seawater to precipitate
in the particles. Bernstein et al. (1992) proposed the
dissolution of acantharian-derived celestite (SrSq)
as the major source of barium and sulfate.
The Relation between Biogenic Ba and
Organic Carbon
Although the mechanism of barite formation in the
water column is not yet completely understood several approaches were made to relate the flux of
particulate barium to that of organic carbon. Dehairs
et al. (1990) relied on the assumption that the
amount of barite precipitated within decaying organic matter is related to the consumption of a certain amount of oxygen to estimate the mass of organic carbon involved in this process. An increasing set of sediment trap data has enabled the establishment of a quantitative relation between the
flux of organic carbon to the seafloor and the flux
of particulate biogenic barium (Dymond et al. 1992;
Francois et al. 1995). The relation between export
Gingele et al.
ity to minimize the preservation effect and to make
more reliable estimations of past productivity.
Barium occurs in the form of discrete barite
microcrystals in the water column, some of which
are transported and preserved in the sediment.
Goldberg and Arrhenius (1958) interpreted the
increase in barium concentrations below the equatorial divergence of the Pacific and Indian Oceans
as a result of intensified biogenic production in the
surface waters. Schmitz (1987) traced the northward shift ofthe Indian Plate during the Cenozoic
on the basis of increased barium concentrations in
the sediment, indicating the position of the plate
below the equatorial high-production belt. Estimations of relative variations in paleoproductivity
based on barium accumulation were carried out in
the Atlantic (Dehairs et al. 1980; Thomson and
Schmitz 1997), Pacific (\/ on Breymann et al. 1992),
Mediterranean (Van Santvoort et al. 1996) and
Southern Ocean (Shimmield et al. 1994) as well
as the Arabian Sea (Shimmield and Mowbray
1991). Quantitative calculations of new productivity and primary production rates from barium accumulation rates were applied to sediments ofthe
Atlantic (Gingele and Dahmke 1994), Mediterranean (Van Os et al. 1994) and Southern Ocean
(Niirnberg 1995; Bonn 1995; Bonn et al. 1998).
They are based on an algorithm developed
by Dymond et al. (1992) which is based on sediment trap data, and correlates the barium flux and
new production. The application of barium for
productivity estimates is not without pitfalls. This
paper will address the problems associated with
and the advantages for the use of barium for
paleoproductivity calculations.
Barium in the Marine System - Generation
of the Productivity Signal
Discrete micron-size barite crystals are the main
carriers of particulate barium in the water column
(Fig. 1). The mechanism of barite formation in
seawater, which is undersaturated with respect to
barite is still poorly understood. Profiles of dissolved
barium in the water column are characterized by
a minimum in the nutrient-depleted euphotic zone
and a continually increasing concentration in
greater depths (Chan et al. 1977), (Fig. 2). Their
similarity to profiles of dissolved silica and alkalinity is taken as evidence for the biogenic association of barium carriers. Maximum concentrations
of particulate barium can be observed just below
the euphotic zone, decreasing slightly with water
depth (Fig. 2). The size distribution of barite particles in the upper water column also suggests that
much of the barite is produced in large decaying
particles in shallow waters and are partly released
as fine particles in the oxygen minimum zone
(Bishop 1988; Dehairs et al. 1990). The nature of
these depth-profiles has led to the conclusion that
biogenic Ba is actively or passively precipitated in
or just below the euphotic zone and is gradually
dissolved with increasing water depth (Chan et al.
1977; Dehairsetal. 1980; Bishop 1988). Since barite
crystals have not yet been found in any major planktonic group, a precipitation in sulfate-barium-rich
microenvironments within decaying organic and
siliceous matter has been suggested (Dehairs et al.
1980; Bishop 1988). The sources of barium and
sulfate have not been determined yet. It has been
suggested that sulfate from decaying labile organic
matter supersaturates microenvironments with respect to barite (Chow and Goldberg 1960; Dehairs
etal. 1980; Bishop 1988) and dissolved barium diffuses from the surrounding seawater to precipitate
in the particles. Bernstein et al. (1992) proposed the
dissolution of acantharian-derived celestite (SrSq)
as the major source of barium and sulfate.
The Relation between Biogenic Ba and
Organic Carbon
Although the mechanism of barite formation in the
water column is not yet completely understood several approaches were made to relate the flux of
particulate barium to that of organic carbon. Dehairs
et al. (1990) relied on the assumption that the
amount of barite precipitated within decaying organic matter is related to the consumption of a certain amount of oxygen to estimate the mass of organic carbon involved in this process. An increasing set of sediment trap data has enabled the establishment of a quantitative relation between the
flux of organic carbon to the seafloor and the flux
of particulate biogenic barium (Dymond et al. 1992;
Francois et al. 1995). The relation between export