Author(s): P. Rahman; A. Sharma; G. Smith

Linked Author(s):

Keywords: Rainfall; Climate Change; Downscaling; Disaggregation; Method of Fragments

Abstract: Assessing catchment flood response to extreme future storms remains a challenge. Traditional “event based” flood prediction methods are pursued assuming climate stationarity, utilising statistical information which is invalid with a changing climate. In the absence of consensus regarding predicting rainfall at the fine time steps typically required to make catchment scale hydrological predictions, planning in NSW has centred on applying sensitivity testing by scaling design rainfall by 10-30%. The analysis presented here will apply stochastic downscaling to an urban catchment to derive instances of future rainfall for use with an existing flood model. Methodology will involve reinterpreting GCM simulations (for the SRES A2 emissions scenario) using a scaling method reflecting regionally observed climatology at a daily time step and applying a non-parametric disaggregation, which maintains within-day temporal dependence, to derive rainfall at sub-daily time steps. The study is expected to provide a useful comparison between the variability in flood estimates based on the aforementioned approach and IFD scaling.

DOI:

Year: 2011