APPLICATION OF STATISTICAL DOWNSCALING MODEL FOR LONG LEAD RAINFALL PREDICTION IN KURAU RIVER CATCHMENT OF MALAYSIA

Abstract

The climate impact studies in the hydrology are often relying on the climate change information at a fine spatial resolution. However, Global Climate Models (GCMs) which is regarded as the most advanced models yet for estimating the future climate change scenarios are operated on the coarse spatial resolution and not suitable for climate impact studies. Therefore, in this study, the Statistical Downscaling Model (SDSM) was applied to downscale rainfall from the GCMs. The data from single rainfall station located in the Kurau River were used as input of the SDSM model. The study included the calibration and validation with large-scale National Centers for Environmental Prediction (NCEP) reanalysis data, and the projection of future rainfall corresponding to the GCMs-variables (HadCM3 A2). The study results shows that during the calibration and the validation stage, the SDSM model can be well acceptable in regards to its performance in the downscaling of the daily and annual rainfall. For the future period (2010- 2099), the SDSM model estimates that there were increases in the total average annual rainfall and generally, the area of rainfall station become wetter.