Extreme precipitation events: their origins, predictability and societal impacts

Co-operative project funded by NATO Science for Peace Programme

This project is a co-operative action of the Meteorological Institute, University of Bonn (MIUB, Prof. Dr. Clemens Simmer), the P.P.Shirshov Institute of Oceanology, Russian Academy of Science (IORAS, Prof. Dr. Sergey Gulev), the University California at San Diego (UCSD, Dr. Alexander Gershunov), and the Odessa State Ecological University (OSEU, Dr. Sergey Ivanov). The project is focused on the quantitative description and predictability of extreme precipitation conditions over the European and American continents. This problem is of high economic and social importance for all countries, especially the ones of the proposers. The most dramatic recent Western European flood occurred in August 2002, leading to more than 100 fatalities. Economic loss amounted to 14.5 billion ˆ for Germany, Austria and the Czech Republic. High water events at major North American rivers lead to weather-associated economic looses from 2 to 6.5 billion dollars per year during the last two decades, varying tremendously year-to-year, but growing annually at approximately 200 million dollars per year. In contrast to many previous studies, which considered extreme precipitation as one of many manifestations of extreme weather, this project is directly targeted on daily extreme precipitation events in a climatic context and their seasonal prediction. In order to quantify and improve predictability of extreme precipitation events the project participants use for the first time all types of precipitation data, such as station measurements, reanalyses and model products from the leading meteorological centers and microwave radar data. The project methodology includes sophisticated statistical procedures aimed to define local precipitation extremes over Europe and North America, mesoscale modelling to investigate the physical mechanisms of atmospheric moisture transport from the oceans to the continents, and a hybrid dynamical-statistical forecasting methodology in order to predict precipitation extremes.