U.S.- Italy Research Workshop on the

Hydrometeorology, Impacts, and Management of Extreme Floods

Perugia (Italy), November 1995

ATMOSPHERE MODELING AND HYDROLOGIC PREDICTION UNCERTAINTY

F. Castelli

Institute of Hydraulics

University of Perugia

S.Lucia, 06125 Perugia, Italy

Some of the fundamental principles of Quantitative Precipitation Forecasts (QPF) from operational numerical atmospheric models are briefly addressed in the paper. The attention is then focused on the relations among QPF uncertainty, the subgrid variability of the forecasted precipitation patterns and the uncertainty of the hydrologic prediction. Recognizing the fundamental deterministic nature of numerical model outputs, possible probabilistic parameterizations for the forecasted precipitation are analyzed with the use of a conceptual non-linear hydrologic model. The definition of the scores, which commonly define the quality of the QPF products, is also reviewed in the same probabilistic framework. In addition, some research topics to be developed in the field of QPF related to flood prediction are suggested.