The use of an index-based regression model for precipitation analysis on the Iberian peninsula

A large number of atmospheric circulation classification techniques have been developed in the investigation of synoptic controls on regional rainfall. Often the rationale is to aid efforts to downscale GCM output for the purpose of producing more confident climate change impact scenarios. Discrete weather typing techniques, although proven to be successful do not capture weather type intensity and within-type variability can often be high. In this study an objective indexing method, developed for Egypt and the British Isles area is applied to the Iberian peninsula. Air flow index values are then used as predictor variables in simple linear regression models to estimate monthly mean grid point rainfall amounts. Separate models are evaluated for the winter and summer halves of the year and also for surface and mid-tropospheric flow (500 hPa). The models are evaluated and compared indicating that the index values provide good estimation of rainfall but variability in performance between season and site is noted.