Numerical simulation models have a long history as research tools for the study of coastal marine ecosystems, and are increasingly being used to inform management, particularly related to nutrient-fueled eutrophication. Demand for modeling assessments is rapidly increasing, and managers need generally applicable tools that can be rapidly applied with limited resources. Additionally, a variety of calls have been made for the development of reduced complexity models for use in parallel with more complex models. We propose a simplified, empirically constrained modeling approach that simulates the first-order processes involved in estuarine eutrophication, contains a small number of aggregated state variables and a reduced set of parameters, and combines traditional mechanistic formulations with robust, data-driven, empirical functions shown to apply across multiple systems. The model was applied to Greenwich Bay, RI (USA), a subestuary of Narragansett Bay, and reproduced the annual cycles of phytoplankton biomass, dissolved inorganic nutrients, and dissolved oxygen, events including phytoplankton blooms and development of hypoxia, and the rate of annual primary production. While the model was relatively robust to changes in parameter values and initial conditions, sensitivity analysis revealed the need for better constraint of the phytoplankton carbon-to-chlorophyll ratio, temperature dependence of phytoplankton production, and parameters associated with our formulations for water column respiration and the flux of phytoplankton carbon to the sediments. This reduced complexity, hybrid empirical-mechanistic approach provides a rapidly deployable modeling tool applicable to a wide variety of shallow estuarine systems.