Analyzing Rainfall Uncertainty on Salinity Forecasting Within the Barataria Bay Estuary

This study focuses on using NEXRAD radar-rainfall information to investigate the impact of rainfall spatial variability and limited sampling on salinity prediction in an estuarine system. The site of this study is the Barataria basin, which is a wetland-dominated estuarine ecosystem in southwest Louisiana. Salinity prediction was found to rely heavily upon accurately estimating basin rainfall, due to rainfall being the largest source of freshwater and the most variable component in the net supply of fresh water to the basin. Rain gauge density scenarios of limited rainfall samplings were simulated from the fully-distributed radar data and corresponding salinity predictions were assessed. Results indicated that a high degree of uncertainty existed in salinity prediction associated with the typical average U.S. rain gauge density (1.3 gauges/1000 km²). By slightly increasing rain gauge density beyond the typical density, a significant amount of salinity prediction uncertainty could be reduced.