Key points are not available for this paper at this time.
There are large uncertainties in the future evolution of water resource availability over western and central Europe. This availability, characterized by total runoff, is the result of two driving variables, precipitation and evapotranspiration. The signal-to-noise ratio for changes in precipitation is low. Changes in evapotranspiration are strongly influenced by uncertainties associated with the modelling of land-atmosphere interactions.Here we highlight the diverging hydrological behavior of models from the 6th phase of the Coupled Model Intercomparison Project over western and central Europe. Some models project a marked decrease in runoff in the late 21st century while others simulate no change or a slight increase. The decrease in runoff is driven either by a decrease in precipitation and evapotranspiration or by an increase in precipitation and an even larger increase in evapotranspiration. The mechanisms responsible for the inter-model spread are investigated with experiments from the Model Intercomparison Project on Land Surface, Snow and Soil Moisture (LS3MIP, van den Hurk et al. 2016) and on the coupled Climate-Carbon Cycle (C4MIP, Jones et al. 2016). The change in large-scale atmospheric circulation is shown to be an important driver of the inter-model spread in precipitation changes in winter. In summer, the physiological effect of CO2 has a direct influence on evapotranspiration and precipitation changes. Models with a strong biogeochemical effect of CO2 show a strong decrease in evapotranspiration and precipitation in summer.The potential relationship between the mean present-day state and trends, and the inter-model spread in the future changes in the hydrological cycle projected by the CMIP6 models is also examined. The models that project a large decrease in runoff are often associated with a large climatological value of evapotranspiration. Among them, some models also show strong positive trends in precipitation and evapotranspiration that are consistent with their future response at the end of the 21st century. However, it remains difficult to judge the realism of models on the basis of this assessment alone, in particular due to large observational uncertainties.van den Hurk et al.: LS3MIP (v1.0) contribution to CMIP6: the Land Surface, Snow and Soil moisture Model Intercomparison Project aims, setup and expected outcome, Geosci. Model Dev., 9, 28092832, https://doi.org/10.5194/gmd-9-2809-2016, 2016.Jones, C. D et al.: C4MIP The Coupled ClimateCarbon Cycle Model Intercomparison Project: experimental protocol for CMIP6, Geosci. Model Dev., 9, 28532880, https://doi.org/10.5194/gmd-9-2853-2016, 2016.
Deman et al. (Fri,) studied this question.