Original title: Bay Assessment Model: A Python-based Salinity Projection Tool for Florida Bay, USA
Erik Stabenau, Joseph Park · September 2026
Too much salt in the water harms Florida Bay. In the late 1980s, dropping freshwater flows made the bay hypersaline, meaning extremely salty. This triggered a large seagrass die-off. To help restoration planners protect the bay, researchers built a new computer tool called the Bay Assessment Model. The model breaks the bay into 54 connected basins. It tracks water movement across shallow shoals by looking at water height differences. It also factors in local rain, evaporation, upstream freshwater levels, and ocean tides. Testing showed the model predicts salinity accurately on average, with an overall mean salinity bias of 0.00 ppt. However, model bias shifts systematically between dry and wet periods. When scientists tested uniform water offsets along the Everglades boundary, salinity changed by 2-4 ppt in the northeastern bay, with weaker effects closer to the sea. The tool is open source and coded completely in Python.
Imagine you are managing a series of connected fish tanks. Each tank has its own saltiness, and water flows between them through small pipes. The challenge is that the flow depends on how full each tank is, and you need to keep the saltiness just right for the fish. Here, the tanks are the basins of Florida Bay, the pipes are the shoals, and the model calculates how water and salt move to predict saltiness under different restoration plans.
Florida Bay experienced a massive seagrass die-off in the 1980s because salinity got too high. Restoration planners now need to predict how changes in freshwater flow will affect salinity. This open-source Python model gives them a transparent tool to test different restoration scenarios and avoid future ecological disasters.
The model's bias varies with hydrologic conditions, increasing from drought to wet periods and reversing sign in one basin, which limits the reliability of anomaly-based metrics. Additionally, the full analysis of restoration and sea level scenarios is not included in this paper but reported separately.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Bay Assessment Model: A Python-based Salinity Projection Tool for Florida Bay, USA — Erik Stabenau, Joseph Park
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.