1 Generating projected rainfall time series at sub-hourly time scales using statistical and stochastic downscaling methodologies S. Molavi 1 , H. D. Tran 1,2 , N. Muttil 1 1 School of Engineering and Science, Victoria University, PO Box 14428, Melbourne, VIC. 8001, Australia; 2 Institute of Sustainability and Innovation, Victoria University, Melbourne, Australia Email: shahram.molavi@live.vu.edu.au ; dung.tran@vu.edu.au ; nitin.muttil@vu.edu.au Abstract Many climate studies in the recent past have revealed an obvious variation in climate compared to the past. Recent extreme events such as flash flooding, bushfires and drought provide ample evidence of these variations. In addition to the natural cycle of the climate, the anthropogenic effects of human development are no longer negligible. Emission of greenhouse gases and other aerosols into the atmosphere have led to warmer temperatures and consequently to more extreme events. Urban stormwater systems will particularly be influenced by climate change. Conventionally, to design and develop the stormwater collection systems, it has been assumed that events that occurred in the past would happen in the future. The change in climatic patterns has led to a new approach of considering the future variation in climate for the assessment of stormwater systems. In this study statistical and stochastic approaches to downscaling climate variables from Global Climate Models (GCMs) are discussed and a convenient approach to downscale daily rainfall data into sub-hourly timescales is presented. The Statistical Downscaling Model (SDSM) has been selected to downscale GCMs spatially at the site location. SDSM provides results in a daily time base. A disaggregation approach has been modified and simplified to generate sub hourly time scale rainfalls from daily rainfalls. Availability of these data is essential for the assessment of stormwater system functionality against future variability. This research is an ongoing attempt to develop a new statistic stochastic approach to increase the accuracy of the model in re-sampling of the observed data especially at the sub-hourly time scale. Key Works: Climate Change, Global Climate Models, Statistical Downscaling Model SDSM, Stochastic Disaggregation. Introduction Urban drainage systems have been designed and developed to handle large amounts of rainfall that is converted to stormwater runoff after a catchment reaches its infiltration capacity. The anthropogenic effects of imperviousness of catchments, mostly due to urbanization, have led to stormwater being concentrated and collected more rapidly. Conventionally, the recorded historic data of rainfall and flows from the catchment have been treated as a reliable basis for the design of stormwater collection system infrastructure. The design rainfalls applied were obtained through frequency analysis of the stationary