PII S0016-7037(01)00668-8 Biological productivity during sapropel S5 formation in the Eastern Mediterranean Sea: Evidence from stable isotopes of nitrogen and carbon ULRICH STRUCK, 1, *KAY-CHRISTIAN EMEIS, 1 MAREN VOß, 1 MICHAEL D. KROM, 2 and GREG H. RAU 3 1 Baltic Sea Res. Inst. Warnemuende, Seestrasse 15, D18119 Rostock, Germany 2 School of Earth Sciences, Leeds University, Leeds LS2 9JT, UK 3 Institute of Marine Sciences, University of California, Santa Cruz, CA, 95064, U.S.A. (Received April 26, 2000; accepted in revised form April 24, 2001) ²Present address: Dep. of Paleontology and Historical Geology, Munich University, Richard Wagner Stra e 10, D-80333 Munich, Germany Abstract—We determined 15 N/ 14 N ratios in modern surface and sapropel S5 sediments of the Mediterranean Sea to clarify differences in the nutrient regime associated with sapropel formation. In the modern situation, high  15 N of unused nitrate (15–20 ‰) remaining in the surface waters during the winter phytoplankton bloom evidences P-limitation of biological production in winter.  15 N of surface sediments decrease towards the east of the basin (5 to 2.5‰). This is a consequence of either eastward increasing nitrogen fixation during the summer months, or of particulate matter being supplied predominantly by the P-limited winter bloom. Very low (-1–1‰)  15 N values in sapropel S-5 from four locations require a very light source of nutrient-N assimilated at a minimum of ten times the modern export flux. Because the isochronous records show no spatial gradient in  15 N, we exclude both Ekman-type upwelling and direct riverine discharge as likely sources of nutrients. Our data are consistent with an anti-estuarine thermohaline circulation in the upper 500m during S5 time, allowing for the trapping of nutrients in the eastern basin. The most likely scenario for S5 is that phosphorus release from a relatively shallow redox boundary resulted in an imbalanced supply of N:P (16:1) to the photic zone. The result was a slow assimilation of carbon during summer stratification and extensive N 2 -fixation providing the majority of the export flux from a N-limited system. Copyright © 2001 Elsevier Science Ltd 1. INTRODUCTION The modern Mediterranean Sea is a highly oligotrophic (nutrient poor) marine environment and the flux of organic carbon from the sea surface is very low and is controlled by the addition of new nutrients from the continent (Be ´thoux, 1989). Gross primary productivity in the Eastern Mediterranean is limited by phosphorus (Krom et al., 1991), and is based mainly on regenerated nutrients (Krom et al., 1992; Yacobi et al., 1995; Zohary and Robarts, 1998). The unusual P limitation of the Eastern Mediterranean may be caused by unequal supply of N and P from external sources (Herut et al., 1999), by adsorp- tion of phosphate by inorganic particles (Krom et al., 1991), or by N 2 -fixation similar to that observed in the Central Pacific (Karl et al., 1997; see also Wu et al., 2000). The sediments in the Eastern Mediterranean Sea, however, intermittently recorded dramatic increases in the burial of or- ganic carbon since 5.3 Million years ago (and possibly earlier). Sapropels (Kidd et al., 1978; Kullenberg, 1952; Olausson, 1961) containing organic carbon concentrations of up to 30% were deposited during these periods, and each sapropel event lasted several thousand years (Vergnaud-Grazzini et al., 1977). The accumulation rate of organic matter increased abruptly, which has been attributed to higher biologic production (Calvert et al., 1992), better preservation of organic carbon (Cita and Grignani, 1982), or both (Emeis et al., 2000a). Sedimentological (Emeis et al., 1996; Kidd et al., 1978), geochemical (Nijenhuis et al., 1998) and micropaleontological (Schmiedl et al., 1998) facies are consis- tent with anoxic conditions in deep water bodies and H 2 S may have even been present in the euphotic zone during some episodes of sapropel deposition (Passier et al., 1999). The reasons behind increased organic carbon burial are at the heart of the dispute over the significance of sapropels and their environment of formation. If the productive surface layer was replete with nutrients, then where did they come from? If it was not, then how could organic matter sedimentation and burial increase so enormously? Our aim was to reconstruct the nutri- ent regime during formation of one sapropel (S5, deposited during marine isotope stage 5e in the interval from 128 –123 kyr) and our tools are isotope ratios of 13 C/ 12 C and 15 N/ 14 N. Nitrogen sources associated with the increased export flux during sapropel formation might be identified based on distinc- tive isotopic ratios. In accordance with the Raleigh model of isotope fractionation (e.g., Mariotti et al., 1981), biotic utiliza- tion of a nutrient pool such as nitrate in which the utilization is selective for the lighter isotope of nitrogen, 14 N, will result in N products (biomass) which are depleted in 15 N relative to the source nitrate. In a closed system, as such N utilization pre- cedes the remaining nitrate will become progressively enriched in 15 N, in turn enriching the biomass subsequently produced. Nearing complete N utilization, the 15 N/ 14 N of the accumulated products must, via mass balance considerations, approach that of the original source nitrate. Under circumstances where biotic N utilization is relatively non-isotopically selective and the N substrate is abundant (e.g., N fixation) the 15 N/ 14 N of the biomass produced will consistently differ little from that of the N substrate (e.g., N fixer  15 N -2– 0‰). *Author to whom correspondence should be addressed (u.struck@ lrz.uni-muenchen). Pergamon Geochimica et Cosmochimica Acta, Vol. 65, No. 19, pp. 3249 –3266, 2001 Copyright © 2001 Elsevier Science Ltd Printed in the USA. All rights reserved 0016-7037/01 $20.00 + .00 3249