ZURUECK HOCH VOR INHALT SUCHEN

» Back to overview
Proposing Institution

Institut für Informatik der LMU
Project Manager

Dr. Stephan Hachinger
Boltzmannstraße 1
85748 Garching
Abstract
Forecasting severe storms and floods is one of the main challenges of 21th century. Floods are the most dangerous meteorological hazard in the Mediterranean basins due to both the number of people affected and to the relatively high frequency by which human activities and goods suffer damages and losses. The numerical simulations of extreme events which happen over small basins as the Mediterranean ones are need a very fine-resolution in space and time and as a consequence considerable memory and computational power are required.From a mathematical point of view, Numerical Weather Prediction (NWP) is a typical problem determined by its initial and boundary (in case of limited area modeling) conditions. The uncertainty in initial and boundary conditions is the primary source of error in the weather modeling. The meteorological model considered in the project is the Weather Research and Forecasting (WRF) model, a state of the art mesoscale numerical weather prediction system designed to serve both operational forecasting and atmospheric research needs.The proposal describes a research project focused on the data assimilation of observations into the WRF model to reduce uncertainty in the prediction of severe convective events, with special focus on flash-flood producing storms with the use of ensemble assimilations techniques. Furthermore, the project aim to reduce the computational cost of an ensemble approach application through the use of machine learning techniques reducing the number of the ensemble members that diverge from observations.

Impressum, Conny Wendler