CR 53:141-156 (2012)  -  DOI: https://doi.org/10.3354/cr01091

ABSTRACT: Extreme weather events present a risk to urban communities, which under future climate change may require greater resilience to reduce susceptibility to potential damage and loss of life. Improved resilience requires detailed information on both current and future levels of weather-related hazards; however, climate models do not provide output at a sufficiently high resolution to provide usable information at the city scale. Furthermore, uncertainties in future climate projections need to be communicated in an appropriate manner to facilitate better local decision-making and adaptation. This paper addresses the problem of providing local-scale projections of hazards associated with high wind speeds over south-east London, along with quantified estimates of uncertainty. It describes a 2-step method for downscaling daily mean wind speed from a 25 km regional climate model (RCM) perturbed physics ensemble. This links climatological bias correction and engineering downscaling approaches to derive projections of maximum hourly mean wind speed and maximum 3 second gust speed at a resolution of 1 km. This employs temporal downscaling factors that have previously been used to estimate structural loading on new buildings but are here implemented at the city scale. The ensemble indicates that future changes in mean daily wind speed for the 2050s are not significantly different to those for a baseline period (1961−1990). Further, whilst the area is projected to experience an increase in exposure to events with the potential to cause damage to buildings, changes are small and considerable uncertainties remain in modelling such extremes.

KEY WORDS: Wind · Downscaling · Regional climate models · UKCP09 · Ensemble · Extreme weather · Hazard