Assessing rainfall availability over the Sydney region in a future climate using stochastic downscaling

Coarse resolution of General Circulation Model (GCM) necessitates use of downscaling approaches for transfer of GCM output to finer spatial resolutions for climate change impact assessment studies. This paper presents an application of a stochastic downscaling framework for simulation of multi-site daily rainfall occurrences and amounts for the year 2070. The downscaling framework is developed using current climate reanalysis data and rainfall records at a network of 45 raingauge stations around Sydney in Australia, while information about the atmospheric variables of the CSIRO Mk3.0 GCM (corresponding to the SRES B1, A1B and A2 emission scenarios) is used for downscaling of rainfall for current and future climates. The model calibration and verification results for the baseline period (1960- 2002) indicate that the proposed downscaling model simulates fairly accurately not only the standard rainfall attributes, such as the average number of wet days and rainfall amounts, but also rainfall extremes and the longer time scale variations and, therefore, can be used with confidence for assessing rainfall in a future climate. Downscaled rainfall results for 2070 show large variations across the seasons and emission scenarios ( Table 1). In general, wetter autumn and summer and drier winter and spring with regions along the coast getting wetter and inland regions becoming drier are projected. An increase of 2% by 2070 in annual rainfall is estimated. Amount per wet day also shows a minor increase of 4%. Increased instances of longer dry and wet spells with lighter rains are expected in future. [GRAPHICS] .