『Abstract
　Marine sediments are the main sink in the oceanic phosphorus (P) cycle. The activity of benthic microorganisms is decisive for regeneration, reflux, or burial of inorganic phosphate (Pi), which has a strong impact on marine productivity. Recent formation of phosphorites on the continental shelf and a succession of different sedimentary environments make the Benguela upwelling system a prime region for studying the role of microbes in P biogeochemistry. The oxygen isotope signature of pore water phosphate (δ¹⁸OP) carries characteristic information of microbial P cycling: Intracellular turnover of phosphorylated biomolecules results in isotopic equilibrium with ambient water, while enzymatic regeneration of Pi from organic matter produces distinct offsets from equilibrium. The balance of these two processes is the major control for δ¹⁸OP.
　Our study assesses the importance of microbial P cycling relative to regeneration of Pi from organic matter from a transect across the Namibian continental shelf and slope by combining pore water chemistry (sulfate, sulfide, ferrous iron, Pi), steady-state turnover rate modeling, and oxygen isotope geochemistry of Pi.
　We found δ¹⁸OP values in a range from 12.8‰ to 26.6‰, both in equilibrium as well as pronounced disequilibrium with water. Our data show a trend towards regeneration signatures (disequilibrium) under low mineralization activity and high Pi concentrations. These findings are opposite to observations from water column studies where regeneration signatures were found to coincide with high mineralization activity and high Pi concentrations. It appears that preferential Pi regeneration in marine sediments does not necessarily coincide with a disequilibrium δ¹⁸OP signature. We propose that microbial Pi uptake strategies, which are controlled by Pi availability, are decisive for the alteration of the isotope signature. This hypothesis is supported by the observation of efficient microbial Pi turnover (equilibrium signatures) in the phosphogenic sediments of the Benguela upwelling system.』

1. Introduction
2. materials and methods
　2.1. Region of the study
　2.2. Retrieval of sediment cores and pore water sampling
　2.3. Quantification of dissolved pore water compounds
　2.4. Isotopic analysis of phosphate
　　2.4.1. Micro extraction of pore water phosphate and isotope ratio mass spectrometry
　　2.4.2. Determination of water oxygen isotopes, calculation of isotopic equilibrium and correction of δ¹⁸OP
　2.5. Modeling of steady state production of dissolved inorganic carbon and phosphate
　2.6. The phosphate oxygen isotope balance of regeneration and microbial turnover
　　2.6.1. Endmember 1: isotope equilibrium in microbial phosphate metabolism
　　2.6.2. Endmember 2: kinetic fractionation and incorporation of water-oxygen during extracellular phosphate regeneration from organic matter
　　　2.6.2.1. Consideration of substrate and enzyme systems for calculation of the regeneration endmember
　　　2.6.2.2. Composition of phosphorus bound to organic matter
　　2.6.3. Construction of the isotope mass balance model
　　　2.6.3.1. The steady state mass balance of Pi in sediment porewater
　　　2.6.3.2. Isotope mass balance model
3. Results
　3.1. Pore water geochemistry
　3.2. Dissolved inorganic carbon and phosphate turnover
　3.3. Water and phosphate oxygen isotopes
4. Discussion
　4.1. Key sources for pore water phosphate in the Benguela upwelling system
　　4.1.1. Ferric oxyhydroxides as Pi source
　　4.1.2. Mineralization of organic matter and preferential regeneration of Pi: relation to carbon-to-phosphorus ratios of organic matter
　　4.1.3. Role of advection and diffusion of Pi from above and below
　4.2. Classification of stations with similar geochemical setting
　4.3. Comparison to results of the mass balance model: efficiency of microbial phosphate cycling
　　4.3.1. Preservation of regeneration signature (disequilibrium) at low activity, low Pi sites
　　4.3.2. Equilibrium signature at high activity, high Pi mudbelt sites
　　4.3.3. Unequal display of slope sites with intermediate DIC production
　4.4. Apparent inconsistencies between geochemical data and oxygen isotope mass balance regarding enhanced Pi regeneration
　　4.4.1. Microbial Pi transport systems: consequences for phosphate isotope biosignatures
　　4.4.2. The mudbelt as a unique environment for marine P cycling
5. Conclusion
Acknowledgments
Appendix A. Supplementary data
References