DATE2018-05-31 11:26:14
IDABSTRACT20180531112614-0252
CONTACTdouglas.maraun@uni-graz.at
PRESENTATIONORAL
INVITED0
IDSESSION3
TITLEINFERRING INFORMATION ON FUTURE CHANGES IN REGIONAL PRECIPITATION EXTREMES BY MEANS OF DYNAMICAL AND STATISTICAL MODELING
AUTHORSD Maraun (1)
AFFILIATIONS
  1. University of Graz Wegener Center for Climate and Global Change Graz Austri
ABSTRACTClimate projections of with a very high spatial and temporal resolution are often demanded for assessing the impacts of extreme precipitation in a potential future climate. Yet classical scenario-based approaches of providing such information are still suffering from severe limitations: standard global general circulation models (GCMs) do not realistically simulate the dynamical large-scale fields controlling local-scale precipitation extremes, and operational dynamical regional climate models (RCMs) do neither resolve small-scale convective events nor local complex topography. Often bias correction is applied as a statistical post-processing step to close the gap between model output and user needs. Here we show that bias correction is not capable of mitigating the fundamental GCM and RCM errors mentioned above. Recently, storyline approaches of single events have been suggested as an alternative to provide user-relevant information. The idea is to simulate individual events that occurred in reality under potential future thermodynamic conditions. The focus on single real world events helps to avoid circulation errors, and enables one to conduct very high resolution simulations. Bias correction can be used as add-on to remove residual intensity biases. Dedicated sensitivity studies can also be used to isolate driving processes and to assess added value. Overall, such simulations have the potential to substantially increase the credibility of regional climate change information about extreme precipitation events.
PAGE75
STATE1