A number of models have been developed to simulate hypoxia in the Chesapeake Bay, but these models do not agree on what processes must be included. In this study we implemented a previously published biogeochemical (BGC) code developed for open-ocean waters that includes “cryptic” microbial sulfur cycling, a process that can increase denitrification and anammox rates in anoxic waters. We ran this BGC code within the ChesROMS physical model of the Chesapeake Bay, then compared the results to those of a ChesROMS simulation with an estuarine BGC code previously implemented and calibrated in the Bay. The estuarine BGC code neglects sulfur cycling but includes burial of particulate organic matter (POM) and cycling of dissolved organic matter (DOM) and uses different values for many parameters governing phytoplankton growth and particle dynamics. At a key test site (the Bay Bridge Station), the model with sulfur cycling gives better results for oxygen and nitrate. However, it also gives a worse overprediction of ammonium-suggesting that its greater accuracy in predicting these two variables may result from cancellation of errors. By making comparisons among these two models and derivatives of them, we show that the differences in modeled oxygen and ammonium are largely due to whether or not the BGC codes include cycling of DOM and sedimentary burial of POM, while the differences in modeled nitrate are due to the other differences in the modeled biogeochemical processes (sulfur cycling/anammox/optics). Changes in parameters used in both BGC codes (in particular particle sinking velocities) tended to compensate the other differences. Predictions of hydrogen sulfide (H 2 S) within the Bay are very sensitive to the details of the simulation, suggesting that it could be a useful diagnostic.