Wednesday, 14 January 2004
Application of Scale-Recursive Estimation to Ensemble Forecasts: A Comparison of Coarse and Fine Resolution Simulations of a Deep Convective Storm
The Advanced Regional Prediction System (ARPS) is used to produce multiple resolution ensemble forecasts for a tornadic thunderstorm case that occurred over Fort Worth, TX on 28-29 March 2000. A five-member scaled lagged average forecasting (SLAF) ensemble is generated for each resolution ranging from coarse (24 km) to fine (storm-scale, 3 km). A single forecast at fine resolution (3 km) is also generated. Scale-recursive estimation (SRE) is then applied to merge the model precipitation output by blending the coarse resolution ensembles with statistical information from the fine resolution forecast and/or high resolution NEXRAD Level-II radar data (remapped to 1.5 km grid-scale ARPS domain) to form a precipitation estimate with statistical information comparable to a more expensive fine resolution ensemble forecast. The potential applications of this merging technique in QPF verification and storm-scale rainfall prediction are examined.