A Statistical Approach to Multisite Downscaling of Daily Precipitation Processes in the Context of Climate Change

The present study proposes a statistical downscaling (SD) approach to represent the linkages between global scale climate predictors and daily precipitation processes at many local sites concurrently. This SD method is based on a combination of two multiple regression models for describing precipitation occurrences and amounts and on the use of the Singular Value Decomposition technique for modelling the stochastic components of these two models. The feasibility of the proposed procedure was assessed using the NCEP/NCAR re-analysis data and the daily precipitation data from a network of 10 raingages located in the Quebec-Ontario region (Canada). Results of this application have indicated the ability of this SD approach to reproduce accurately the observed statistical properties of daily precipitations, and the temporal-spatial dependence and intermittency of the underlying precipitation processes. The proposed SD tool can be used for assessing the climate change impacts on daily precipitations at many different sites over an area of interest.