Development of a predictability index for severe weather events over Europe (T01)

• Subject Area: Atmospheric Science
• Term: from 2017 to 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 257899354

Accurate and timely prediction of high impact weather, and in particular intenseprecipitation events, is essential to limit losses of life and property. Although predictabilityis fundamentally limited by the chaotic nature of the atmosphere, recent research inatmospheric dynamics has provided new insights into the processes that limit the accuracyof forecasts, which may be of significant use in operational practice. The aim of this projectis to evaluate some recently developed measures of how different atmospheric processesinfluence predictability, and to combine them to formulate a new forecasting methodologyto provide guidance in the interpretation of forecasts of heavy precipitation.Rossby wave packet diagnostic tools, developed in the framework of Waves to Weather,will be applied to ECMWF analysis and forecast products in order to obtain a robustquantification of the presence and impact of upper level precursors to the heavyprecipitation events. The statistics of past basin-area heavy precipitation events will beinvestigated using the newly released high resolution precipitation dataset ArCIS(Climatological Archive for Central Northern Italy). A convective adjustment time scale willbe computed to provide a measure of how closely the precipitation is coupled to the largescaleweather pattern. Combining these statistics, we aim to define a predictability indexthat reflects the degree to which predictability is imposed by the large scale, butdiminished by the unpredictability of the small scale. This index, designed for operationalusage, will provide forecasters and other end users with a measure of which scales ofmotion can be accurately predicted for a particular weather forecast.

• DFG Programme: CRC/Transregios (Transfer Project)
• Subproject of: TRR 165: Waves to Weather
• Applicant Institution: Ludwig-Maximilians-Universität München
• Project Heads: Professor Dr. George Craig; Professor Dr. Volkmar Wirth