Future Changes in Day-to-Day Precipitation Variability in Europe

Abstract

Abstract We analyzed regional patterns of day-to-day precipitation variability across Europe and assessed their future changes using Coordinated Regional Climate Downscaling Experiment (CORDEX) regional climate models. A discrete Markov chain process was employed to calculate transition probabilities from wet and dry states, and the precipitation variability was quantified using the proposed variability index (IVAR, the sum of wet-to-dry and dry-to-wet transitions divided by the total number of transitions). The IVAR is, in general, lowest in southern Europe and gradually increases northward in the observed data. Performance of the regional climate models is season dependent: They capture IVAR reasonably well in summer, but higher simulated variability was found for the winter season. The IVAR trends computed for the 2006–95 period suggest decreasing day-to-day precipitation variability over southern Europe, especially in summer under the high-concentration RCP8.5 pathway. By contrast, increased variability is projected in northern Europe. Between these two regions, future IVAR trends are less clear because they strongly depend on the selection of driving global model, hinting of an uncertain future hydroclimate in the central European region. Significance Statement In a warming world, water availability will play a key role in ecosystem productivity. Although future changes in rainfall amounts have been studied extensively, much less attention has been given to changes in their temporal distribution and variability. Because grouping wet or dry days into sequences vitally contributes to characterizing floods or droughts, we aimed to study future changes in these tendencies. We found that although future changes in wet or dry days grouping tendencies are mostly driven solely by change in their frequency, climate models do not agree on the change in the frequency of wet days over large parts of continental Europe. This leaves major uncertainties in a future European hydroclimate and implications for impact modeling.