BASIN-SCALE EVALUATION OF RCM BIAS USING RAINFALL OBSERVATION NETWORKS

The local downscaling of rainfall predictions provided by climate models is crucial for the assessment of climate change impacts on most ecological processes related to the land/water cycle, such as vegetation dynamics, soil-bacteria activity, and ecological response of water-bodies. In this study, we present a methodology to analyze the predictive performance of a Regional Climate Model (RCM) with regard to daily rainfall fields. A comparison between statistical properties of rainfall observations and model control simulations was performed through a robust and meaningful representation of the precipitation process. Our objectives were, first, to evaluate RCM bias data at basin-scale against daily rainfall records coming from a rain gauge network, and then to propose a simple framework to investigate possible alterations of the daily rainfall occurrence and intensity under climate change by way of a stochastic model suitable to investigate both ordinary regimes and extreme climate events.