Development and Application of an Evaluation Method for the WRF Mesoscale Model

摘要

This report establishes automated methods of obtaining and archiving initialization and time-dependent lateral condition data for the Weather Research and Forecast (WRF) mesoscale model. The data includes Eta forecast data, MADIS surface data, and upper-air sounding data. Statistical analysis methods that compare surface and upper-air data between model calculation and observation have also been established and are described in detail. In this study, the WRF was applied to model domains surrounding the National Training Center in southern California. A set of 24-hr forecast data of 2-way, coupled domains (with grid resolutions of 18 and 6 km) were statistically compared with surface and upper-air observation data.