Forecasting hypoxia in Corpus Christi Bay, Texas by model fusion

Hypoxic or dead zones, which occur when dissolved oxygen levels in water drop below 2 mg/L, are prevalent worldwide. An example of such an hypoxic zone forms intermittently in Corpus Christi Bay (CCBay), Texas, a USEPA-recognized estuary of national significance. Hypoxia in CCBay occurs primarily due to inflow of hypersaline waters that enter from adjacent bays and estuaries, fluctuations in oxygen levels due to the oxygen production-consumption cycle of the aquatic flora and fauna, seasonal fluctuations, and lastly, discharges from several wastewater treatment plants. The aim of this paper is to create more accurate and efficient near-real-time forecasts of hypoxia that will give researchers advance notice for manual sampling during hypoxic events. The approach involves developing and testing a suite of data-driven model fusion approaches with the help of cyberinfrastructure. The models that will be used as a part of this study are: a data-driven k-nearest neighbor model (Coopersmith, 2008), and a physics-based valve model (Sin Chit To, 2009). The k-nearest neighbor predicts the probability of occurrence of hypoxia 24 hours later, and the levels of dissolved oxygen. On the other hand, the valve model determines the distance traveled by a gravity-current upon entering Corpus Christi Bay from adjacent estuaries and bays and the period of time it persists. A comparison will be made between the results of the fused model and those of the individual models to test the effectiveness of model fusion in predicting the estuarine condition in the model.