Long-term controls on ocean phosphorus and oxygen in a global biogeochemical model

In this study, we use a biogeochemical ocean general circulation model (HAMOCC), originally developed for the carbon and silicon cycles, and expand it with a description of the sedimentary phosphorus (P) cycle. The model simulates the release of reactive P by aerobic and anaerobic degradation of organic matter in the sediment, as well as formation and burial of Fe-oxide bound P and authigenic Ca-P minerals. We also include pre-anthropogenic inputs of P from atmospheric dust, which is mostly in the form of detrital apatite. Model predicted total P concentrations and rates of reactive P burial for the deep sea agree reasonably well with observations in open ocean and near continental margin sediments. As part of a sensitivity analysis, we assess the long-term response of ocean productivity and deep water oxygenation to increases in the riverine input of P and preferential release of P from sediments. The simulations show that the feedback from preferential P regeneration accelerates the expansion of suboxia (O-2 < 25 mu M) along continental margins and in the naturally suboxic areas in tropical-subtropical regions on timescales of 10-100 ka. For a case in which maximum P regeneration from sediments is enabled, a large-scale pattern of bottom water suboxia (30% of the total ocean area) develops over the southeastern, tropical and northern Pacific Ocean sectors.