Abstract
Abstract. Information on the frequency and intensity of extreme precipitation is required by public authorities, civil security departments, and engineers for the design of buildings and the dimensioning of water management and drainage schemes. Especially for sub-daily resolutions, at which many extreme precipitation events occur, the observational data are sparse in space and time, distributed heterogeneously over Europe, and often not publicly available. We therefore consider it necessary to provide an impact-orientated data set of 10-year rainfall return levels over Europe based on climate model simulations and evaluate its quality. Hence, to standardize procedures and provide comparable results, we apply a high-resolution single-model large ensemble (SMILE) of the Canadian Regional Climate Model version 5 (CRCM5) with 50 members in order to assess the frequency of heavy-precipitation events over Europe between 1980 and 2009. The application of a SMILE enables a robust estimation of extreme-rainfall return levels with the 50 members of 30-year climate simulations providing 1500 years of rainfall data. As the 50 members only differ due to the internal variability in the climate system, the impact of internal variability on the return level values can be quantified. We present 10-year rainfall return levels of hourly to 24 h durations with a spatial resolution of 0.11∘ (12.5 km), which are compared to a large data set of observation-based rainfall return levels of 16 European countries. This observation-based data set was newly compiled and homogenized for this study from 32 different sources. The rainfall return levels of the CRCM5 are able to reproduce the general spatial pattern of extreme precipitation for all sub-daily durations with Spearman's rank correlation coefficients >0.76 for the area covered by observations. Also, the rainfall intensity of the observational data set is in the range of the climate-model-generated intensities in 60 % (77 %, 78 %, 83 %, 78 %) of the area for hourly (3, 6, 12, 24 h) durations. This results in biases between −16.3 % (hourly) to +8.2 % (24 h) averaged over the study area. The range, which is introduced by the application of 50 members, shows a spread of −15 % to +18 % around the median. We conclude that our data set shows good agreement with the observations for 3 to 24 h durations in large parts of the study area. However, for an hourly duration and topographically complex regions such as the Alps and Norway, we argue that higher-resolution climate model simulations are needed to improve the results. The 10-year return level data are publicly available (Poschlod, 2020; https://doi.org/10.5281/zenodo.3878887).
Highlights
Sub-daily precipitation extremes affect our daily lives with a wide range of consequences that can have impacts on infrastructure, economy, and even health
For each duration we store one file with five columns containing the return level based on the median of the 50-member Canadian Regional Climate Model version 5 (CRCM5)-LE and the 5 % quantile and the 95 % quantile of the ensemble at each grid cell as well as the geographical coordinates
Single members of the CRCM5-LE show this smooth regional distribution, but the use of the median of 50 SMILE members enhances this behaviour, as it filters out the internal variability in the climate system within individual 30-year periods
Summary
Sub-daily precipitation extremes affect our daily lives with a wide range of consequences that can have impacts on infrastructure, economy, and even health. Heavyrainfall events of several hours and up to days can lead to river flooding or coastal flooding as a singular trigger or as a contributing process of compound flooding events such as rain-on-snow or coastal compound floods due to joint river runoff and storm surge (Bevacqua et al, 2017, 2019; Cohen et al, 2015; van den Hurk et al, 2015; Poschlod et al, 2020) These hazards have large impacts on the European infrastructure of urban drainage systems, roads and railroads, waterway transport, electricity, and communication networks (Forzieri et al, 2018; Groenemeijer et al, 2015; Nissen and Ulbrich, 2017). Over 4600 people have lost their lives because of these hazards
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
