Pilot Region: the upper Durance catchment. Understanding trends in hydrologic regimes

The work carried out by AdaptAlp partners allowed gathering a hydrologic dataset aimed at detecting climate-related trends in the hydrologic regime of Alpine catchments [see AdaptAlp report, Bard et al., 2011, for more details]. The analysis of the series from this dataset revealed significant trends affecting certain aspects of Alpine hydrologic regimes [e.g. decreasing severity of winter droughts, earliness of snowmelt-related flows, increasing runoff for glacier regimes, see Bard et al., 2011]. The aim of this study was to evaluate whether these trends correspond to the expected response of the catchments to the evolution of climate forcings for a few catchments located in the Upper Durance region. The strategy used to achieve this objective was to simulate the streamflow response to observed climate forcings using the hydrologic model SIM, and to compare the trends detected in observed and simulated streamflow. Results from this analysis revealed a lack of consistency between trends affecting observed and simulated streamflow, suggesting that non-climatic factors play a role in the evolution of observed streamflow. However, this conclusion should be qualified for at least two reasons: (i) Non-homogeneities in the forcing data may dampen or exacerbate genuine trends caused by the evolution of forcings; (ii) Although the standard Nash-Sutcliffe efficiency measure suggests an overall good performance of the hydrologic model, further analyses revealed stronger deficiencies for specific aspects of the hydrologic regimes (e.g. high/low flow). Although the results of this study are somehow inconclusive, they illustrate the difficulty to disentangle the roles of various possible causes of change in hydrologic series.