Key points are not available for this paper at this time.
Warm-season quantitative precipitation forecasts (QPFs) are the poorest performance area of forecast systems worldwide. They stubbornly fall further behind while other aspects of weather prediction steadily improve. Unless a major effort is mounted to overcome the impediments to improved prediction, it is certain to remain the Achilles' heel of weather prediction, at a progressively greater cost to society. For these reasons and others, the Office of the Lead Scientist, U.S. Weather Research Program (USWRP), commissioned a workshop to examine future courses of action to improve understanding and prediction of heavy warm-season rainfall and associated flood forecasts. The workshop was held in Boulder, Colorado, in March 2002. It was attended by 75 people and produced numerous “white papers” and panel reports, all of which are readily available to the reader. Herein the major findings of the workshop are summarized, including an overarching strategy to achieve improved predictive skill and recommendations for future research and development. Improving warm-season QPFs requires a substantial and sustained commitment of resources focusing on a complex suite of issues. The basic strategy is to take those steps that will facilitate forecasting deep, moist convection in a fully probabilistic manner wherein the statistical properties of the forecast convection are similar to those observed in nature. A warm-season QPF program should be inclusive of a testbed framework, wherein development and testing of each and all components of the forecast system can be conducted; impediments to operations can be identified and corrected; and socioeconomic value, at the margin, can be researched and identified.
Fritsch et al. (Thu,) studied this question.