『Abstract
　Benthic nitrogen (N) cycling was investigated at six stations along a transect traversing the Peruvian oxygen minimum zone (OMZ) at 11゜S. An extensive dataset including porewater concentration profiles and in situ benthic fluxes of nitrate (NO3^-), nitrite (NO2^-) and ammonium (NH4⁺) was used to constrain a 1-D reaction-transport model designed to simulate and interpret the measured data at each station. Simulated rates of nitrification, denitrification, anammox and dissililatory nitrate reduction to ammonium (DNRA) by filamentous large sulfur bacteria (e.g. Beggiatoa and Thioploca) were highly variable throughout the OMZ yet clear trends were discernible. On the shelf and upper slope (80-260 m water depth) where extensive areas of bacterial mats were present, DNRA dominated total N turnover (≦2.9 mmol N m^-2 d^-1) and accounted for ≧65% of NO3^- + NO2^- uptake by the sediments from the bottom water. Nonetheless, these sediments did not represent a major sink for dissolved inorganic nitrogen (DIN = NO3^- + NO2^- + NH4⁺) since DNRA reduces NO3^- and, potentially NO2^-, to NH4⁺. Consequently, the shelf and upper slope sediments were recycling sites for DIN due to relatively low rates of denitrification and high rates of ammonium release from DNRA and ammonification of organic matter. This finding contrasts with the current opinion that sediments underlying OMZs are a strong sink for DIN. Only at greater water depths (300-1000 m) did the sediments become a net sink for DIN. Here, denitrification was the major process (≦ 2 mmol N m^-2 d^-1) are removed 55-73% of NO3^-and NO2^- taken up by the sediments, with DNRA and anammox accounting for the remaining fraction. Anammox was of minor importance on the shelf and upper slope yet contributed up to 62% to total N2 production at the 1000 m station. The results indicate that the partitioning of oxidized N (NO3^-, NO2^-) into DNRA or denitrification is a key factor determining the role of marine sediments as DIN sinks or recycling sites. Consequently, high measured benthic uptake rates of oxidized N within OMZs do not necessarily indicate a loss of fixed N from the marine environment. 』

1. Introduction
2. Study area
3. materials and methods
　3.1. Sampling and geochemical analysis
　3.2. Numerical modeling
　　3.2.1. Coupling reaction and transport
　　3.2.2. Reaction network
　　3.2.3. Constraints on the rates of N cycling
　　3.2.4. Boundary conditions and model solution
4. Results and discussion
　4.1. Sediment geochemistry and POM degradation
　4.2. Nitrogen turnover processes along the transect and their regulation
　4.3. Relative importance of denitrification, DNRA, and anammox in N cycling
　4.4. Potential importance of anammox on the shelf
5. Conclusions
Acknowledgments
Appendix A. Supplementary data
References