Current numerical weather prediction (NWP) ensemble systems are biased and under dispersive. While statistical methods have been developed to correct for these errors, the resulting post processed forecasts are typically available at observing stations only. Yet post processed forecasts are needed everywhere. Additionally, the horizontal grid spacing of current ensemble systems is too coarse to resolve the Indian River Lagoon (IRL), a coastal estuary in central Florida that impacts the local wind climatology. Thus, computationally reasonable approaches are needed that simultaneously downscale and post process ensemble wind forecasts. Currently no approach exists to do this, particularly in coastal areas such as the IRL. This dissertation presents two techniques for this purpose.

The first technique bias corrects and downscales wind speed over water bodies that are unresolved by NWP models and analyses. The dependency of wind speeds over water bodies to fetch length is investigated as a predictor of model wind speed error. Because model bias is found to be related to the forecast wind direction, a statistical method that uses the forecast fetch to remove wind speed bias is developed and tested. The method estimates wind speed bias using recent forecast errors from similar stations, i.e., those with comparable fetch lengths. As a result, the bias correction is not tied to local observations but instead to locations with similar land-water characteristics. Thus, it can also be used to downscale wind fields over inland and coastal water bodies. Applied to three NWP analyses in Florida, the fetch method yields a bias near zero and results in a reduction of the mean absolute error that is comparable to other methods.

Second, a method is developed that performs gridded post processing of ensemble wind vector forecasts. Antecedent to operational calibration, the Weather Research and Forecasting (WRF) model is utilized to downscale an expansive set of idealized wind profiles to a 333 m2 grid over a coastal region marked by an intricate land / water mask. In real time, ensemble model output statistics (EMOS) is used to calibrate ensemble wind vector forecasts at observing stations, after which EMOS predictive parameters (i.e., mean and variance) are spread from locations with observations to those without utilizing flowdependent statistical relationships extracted from the WRF downscaled wind fields, resulting in calibrated forecasts across the 333 m grid. Compared to an operational ensemble, the approach improves both the deterministic and probabilistic forecast skill. Analysis of multivariate rank histograms indicate the post processed forecasts are calibrated.