Development of a verification methods testbed at the WRF DTC

Mike Baldwin
Purdue University


Traditional methods of measuring the performance of forecasts often fail to provide meaningful information when applied to forecasts containing realistic, small-scale features. There is a great need within both the research and operational numerical weather prediction communities for new verification methods to verify high-resolution forecasts that are currently being produced routinely by operational centers as well as many research groups around the world. Several researchers have proposed improved verification techniques, including object-oriented, fuzzy neighborhood, scale decomposition, and field comparison approaches. A major issue facing developers of new verification methods is the time that it takes to establish credibility for new techniques, which delays the adoption of these methods throughout the wider community of users. In this work, an accelerated path to establishing credibility for new verification techniques is proposed. Specifically, this involves applying new techniques to a database of operational and experimental forecasts that cover a period of several years. This archive will reside at the WRF Developmental Testbed Center (DTC) and will result in the establishment of a testbed for verification methods. Issues related to determining whether new verification methods fit into the general framework for verification that was originally defined by Murphy and Winkler (1987) will also be discussed in this presentation.