Estimation of future precipitation Return-Levels in Europe for the adaptation of industrial facilities

Industrial facilities, like any building or installation, are designed to withstand defined levels of natural hazards during their lifetime. These levels are requested by the regulations, and are commonly estimated by use of the Statistical Extreme Value theory ahead of the building in order to support the design of the planned asset. However, for long lasting installations, climate change may change the frequency of the level defined at the time of the building, so that the protection is not as high as initially expected. This work presented here aims at describing and testing a way to estimate Return Levels for precipitation in different locations in Europe at the 2050 time horizon. The methodology is based on the definition of a variable whose extremes can be considered as stationary, so that future Return Levels are obtained from those of this variable and the climate model changes in mean, standard deviation and rainy day frequency at the desired future horizon (Acero et al. 2017). The methodology is first tested in a cross-validation setting over the historical period using 15 rainfall observation time series in Europe provided by the ECA&D dataset and CMIP5 climate model simulations. Then, estimates of the 50-year Return Levels in 2050 are computed. The methodology is then applied to the gridded E-OBS dataset with the objective of producing risk maps at the European scale. The first step is then to compare the estimations previously obtained for the station time series to those obtained for the nearest E-OBS grid points, in order to assess the ability of gridded data to faithfully represent the behavior of the extremes. Depending on the results, advices can be given about the most suited way to map future rainfall extremes in Europe in relation to the adaptation of industrial facilities.

 

Reference:

Acero F.J., Parey S., Hoang T.T.H., Dacunha-Castelle D., Garcia J.A. and Gallego M.C.: Non-stationary future Return Levels for extreme rainfall over Extremadura (SW Iberian Peninsula). Hydrological Sciences Journal, 2017, DOI: 10.1080/02626667.2017.1328559