Bias Correction of 10m Wind Forecasts near a Coastal Estuary in East Central Florida

The Indian River Lagoon (IRL) is a long, narrow estuary situated alongside central Florida's Atlantic coast. During high-wind events, e.g. extra-tropical and tropical cyclones, set-up and wind-driven waves in the IRL can cause significant economic and ecological damage to bordering property. As part of a NOAA funded CSTAR project, the work presented here is designed to support efforts to extend the Nearshore Wave Prediction System (NWPS) into the coastal estuary. The current suite of operational models does not have the spatial resolution that is needed to resolve the fine scale processes associated with the IRL's wind-driven circulation nor its detailed land-water mask. This ultimately requires the use of high-resolution hydrodynamic models that are, in turn, forced by surface wind output from these operational models. As a result, bias errors present in these wind forcings can be problematic as these errors can propagate into the forecasts of water elevation, wave height, etc. Here we investigate the biases associated with 10 m wind forecasts obtained from the National Centers for Environmental Prediction's (NCEP) Environmental Modeling Center's Short Range Ensemble Forecast System (SREF). The ensemble (10 m) winds from the SREF are compared to three Automated Surface Observing System (ASOS) sites within a few km of the IRL. Biases are identified using both raw ASOS wind observations as well as surface roughness-adjusted (i.e., potential) winds at the three ASOS sites. The biases are removed to produce an improved wind forcing over the IRL and the impact is assessed via hydrodynamic simulations.