The Wgne Assessment of Short-term Quantitative Precipitation Forecasts

APRIL 2003 AMERICAN METEOROLOGICAL SOCIETY | O f all the weather elements for which forecasts are provided to the public, rainfall is perhaps of the greatest interest. While most people simply want to know whether they will need an umbrella that day, there is growing demand from industry, agriculture, government, and many other sectors for more detailed rainfall predictions. Unfortunately, rainfall is certainly among the most difficult weather elements to predict correctly. Rainfall has greater spatial and temporal variability than most other meteorological quantities of interest. Many processes can lead to rain, including large-scale ascent of moist air, convection caused by heating of moist air near the surface, convergence of moist air in a baroclinic zone, and orographic lifting. These processes must all be represented in numerical weather prediction (NWP) models, whose output forms the basis for most rainfall forecasts. It is of great interest to assess how well we can meet the need for timely and accurate rainfall forecasts using operational NWP models. In the mid-1990s, the Working Group on Numerical Experimentation (WGNE), established under the World Meteorological Organisation’s World Climate Research Programme (WCRP) and Commission for Atmospheric Sciences (CAS), turned its attention to quantitative precipitation forecasts (QPFs). Since accurate prediction of rainfall depends critically on the accurate prediction of atmospheric motion and moisture content, it is reasonable to expect that a good forecast of rainfall over a large domain indicates a good forecast overall. Indeed, many operational centers use QPF skill as a critical measure of model health. Accumulated precipitation can be verified (albeit imperfectly, given its highly variable nature) using rain gauge networks. Knowledge of a model’s QPF behavior not only helps model developers but also users of the QPFs to understand the reliability of the model output. At the 10th annual WGNE meeting it was recommended that QPFs from several operational NWP models be evaluated in different areas of the globe THE WGNE ASSESSMENT OF SHORT-TERM QUANTITATIVE PRECIPITATION FORECASTS