A new methodology for the assessment of climate change impacts on a watershed scale

Many climate change impact studies have been conducted using a top-down approach. First, outputs from global circulation models (GCMs) are considered which are downscaled in a second step to the river basin scale using either a statistical/empirical or a dynamic approach. The local climatic signal that is obtained is then used as input into a hydrological model to assess the direct consequences in the basin. Problems related to this approach include: a high degree of uncertainty associated with GCM outputs and an increase in uncertainty due to the downscaling approach. An inverse approach is proposed in this article to improve the understanding of the processes leading to hydrological hazards including both flood and drought events. The approach analyses of existing guidelines and management practices in a river basin with respect to critical hydrological exposures that may lead to failure of the water resources system or parts thereof. Critical hydrologic exposures (flood levels for example) are then transformed into corresponding critical meteorological conditions (extreme precipitation events for example). These local weather scenarios are then statistically linked to possible large-scale climate conditions that are available from the GCMs. The proposed procedure allows for the assessment of the vulnerability of river basins with respect to climate forcing. It also provides a tool for identifying the spatial distribution of the vulnerability and risk.