Use of daily precipitation records to assess the response of extreme events to global warming: Methodology and illustrative application to the European region

We introduce a technique for identifying daily precipitation records based on the analysis of maximum daily precipitation events within running 30-day periods. The technique is applied to the E-OBS daily precipitation dataset for the European region spanning the period 1950–2018. We find that, on a Europe wide scale, the number of actual precipitation records as defined above exceeds the theoretical value expected under stationary climate conditions by a factor increasing with time by up to 1.6–1.8. This result, which has substantial subregional variability, is indicative of an increase of daily precipitation extremes consistent with what expected from climate warming conditions. We also apply the technique to the ERA5 reanalysis and four illustrative 21st century projections conducted with two regional climate models under the high end RCP8.5 and low end RCP2.6 greenhouse gas concentration pathways. We find a sustained increase of precipitation records throughout the 21st century with respect to the expected values for the RCP8.5 scenario, and a much reduced signal for the RCP2.6. In both cases the overall increase in actual to expected ratios of precipitation records is consistent with the warming trends found in the two scenarios, although this signal is highly variable in space and depends on the model being analysed. Our technique can be useful in studies of detection and attribution of extreme precipitation events and in the assessment of the impacts of increasing extremes under global warming conditions.