A probabilistic assessment of climate change and related impacts should consider a large range of potential future climate scenarios. State-of-the-art climate models, especially coupled atmosphere-ocean general circulation models and Regional Climate Models (RCMs) can however not be used to simulate such a large number of scenarios. This paper presents a methodology to obtain future climate scenarios through a simple scaling methodology: The projections of several key meteorological variables obtained from a few regional climate model runs are scaled based on different global-mean warming projections drawn in a probability distribution of future global-mean warming. The resulting climate change scenarios are used to drive a hydrological and a water management model to analyse the potential climate change impacts on a water resources system. This methodology enables a joint quantification of the climate change impact uncertainty induced by the global-mean warming scenarios and the regional climate response. It is applied to a case study in Switzerland, a water resources system formed by three interconnected lakes located in the Jura Mountains. The system behaviour is simulated for a control period (1961 - 1990) and a future period (2070-2099). The potential climate change impacts are assessed through a set of impact indices related to different fields of interest (hydrology, agriculture and ecology). The obtained results show that future climate conditions have a significant influence on the system performance and that the uncertainty induced by the inter-RCM variability contributes to a large part to the total impact prediction uncertainty.