Bulk Sediment 15N as a Tool in Paleoceanography

R. Francois Dept. of Marine Chemistry and Geochemistry, W.H.O.I., Woods Hole, MA 02543, USA

rfrancois@whoi.edu

M. A. Altabet Dept. of Chemistry, University of Massachusetts, Dartmouth, MA 02543,USA

D. Sigman Dept. of Geology and Geophysics,W.H.O.I., Woods Hole, MA 02543, USA

Because of isotopic fractionation during nitrate uptake by phytoplankton, the nitrogen isotopic compositon of oceanic particles sinking from surface waters (sink) reflect nitrate utilization by phytoplankton (u; i.e. the fraction of nitrate reaching the euphotic zone and utilized by phytoplankton before its removal by physical processes such as mixing, advection or convection), the fractionation factor during nitrate uptake (*u), and the initial isotopic composition of the nitrate added to surface waters (NO3in):

sink = NO3in + (*u(1-u)ln(1-u)/u)

A fraction of sinking particulate N is eventually buried in sediments providing a record of past variations in sink, after correction for a diagenetic overprint of 3 to 4 permil. Available estimates of *u indicate that it varies little within broad geographic regions. Likewise, NO3in varies within a narrow range (4-5 permil), unless denitri-fication occours in the water column. The letter process is characterized by a large isotopic fractionation factor, producing an 15N enrichment in the residual nitrate which eventually reaches the euphotic zone.

Downcore variations in bulk sediment 15N can thus be interpreted in two different ways, depending on the location of the study. In regions far remowed fromzones of denitrification (e.g. Southern Ocean), NO3in and *u can be taken as constant. Variations in sediment 15N thus primarily reflect changes in nitrate utilization by phytoplankton (u). In regions where water column denitrification occurs (e.g. Arabian Sea, Eastern Equatorial Pacific), sink can potentially be affected by changes in both u and NO3in, the latter reflecting changes in the intensity of denitrification in
intermediate waters. Deconvolving these two variables is possible by building regional synoptic maps. Denitrification zones are invariably associated with upwelling regions. These regions display characteristic increases in sediment 15N away from thew upwelling center, reflecting the
gradual depletion of surface nitrate. Changes in NO3in due to changes in denitrification would evenly affect sediment 15N over the entire region, while changes in upwelling geometry and nitrate utilization by phytoplankton would produce
distinct geographic patterns in the 15N distribution of sediment.

Results obtained to date by several groups indicate an increase in nitrate utilization during the last glacial maximum in the southern ocean, and a decrease in denitrification in the Arabian Sea and the Eastern Pacific. Both observations have imtportant implications for our understanding of the involvement of the carbon cyle in late Quaternary climatic variations. Lower denitrification suggests a significant increase in the oceanic nitrate during glacial periods and a more efficient biological pump, while increased nutrient utilization in the southern ocean would contribute to further lowering atmospheric CO2.