Abstract
Advanced diagnostic methods for the evaluation of quantitative precipitation forecasts are developed and tested using an object-oriented approach. Both the forecast and observed fields are resolved into objects and forecast and observed objects are then compared. Objects can be detected in data fields by a simple convolution filter based approach. Simple geometric shapes can be fitted to objects to represent them in a simplified or schematic fashion. Objects can be matched or merged according to closeness of attributes.
| Original language | English |
|---|---|
| Pages (from-to) | 3223-3228 |
| Number of pages | 6 |
| Journal | Bulletin of the American Meteorological Society |
| State | Published - 2004 |
| Event | Combined Preprints: 84th American Meteorological Society (AMS) Annual Meeting - Seattle, WA., United States Duration: Jan 11 2004 → Jan 15 2004 |