Simulating sub-daily Intensity-Frequency-Duration curves in Australia using a dynamical high-resolution regional climate model
Mantegna, GA and White, CJ and Remenyi, TA and Corney, SP and Fox-Hughes, P, Simulating sub-daily Intensity-Frequency-Duration curves in Australia using a dynamical high-resolution regional climate model, Journal of Hydrology, 554 pp. 277-291. ISSN 0022-1694 (2017) [Refereed Article]
Climate change has the potential to significantly alter the characteristics of high-intensity, short-duration rainfall events, potentially leading to more severe and more frequent flash floods. Research has shown that future changes to such events could far exceed expectations based on temperature scaling and basic physical principles alone, but that computationally expensive convection-permitting models are required to accurately simulate sub-daily extreme rainfall events. It is therefore crucial to be able to model future changes to sub-daily duration extreme rainfall events as cost effectively as possible, especially in Australia where such information is scarce. In this study, we seek to determine what the shortest duration of extreme rainfall is that can be simulated by a less computationally expensive convection-parametrizing Regional Climate Model (RCM). We examine the ability of the Conformal Cubic Atmospheric Model (CCAM), a ∼10km high-resolution convection-parametrizing RCM, to reproduce sub-daily Intensity-Frequency-Duration (IFD) curves corresponding to two long-term observational stations in the Australian island state of Tasmania, and examine the future model projections. We find that CCAM simulates observed extreme rainfall statistics well for 3-h durations and longer, challenging the current understanding that convection-permitting models are needed to accurately model sub-daily extreme rainfall events. Further, future projections from CCAM for the end of this Century show that extreme sub-daily rainfall intensities could increase by more than 15% per °C, far exceeding the 7% scaling estimate predicted by the Clausius-Clapeyron vapour pressure relationship and the 5% scaling estimate recommended by the Australian Rainfall and Runoff guide.