Sensitivity of the Weather Research and Forecasting (WRF) model to downscaling ratios and storm types in rainfall simulation

摘要: Accurate information of rainfall is needed for sustainable water management and more reliable flood forecasting. The advances in mesoscale numerical weather modelling modern computing technologies make it possible to provide simulations forecasts at increasingly higher resolutions space time. However, being one the most difficult variables be modelled, quality products from model remains unsatisfactory hydrological applications. In this study, sensitivity Weather Research Forecasting (WRF) investigated using different domain settings various storm types improve performance simulation. Eight 24-h events are selected Brue catchment, southwest England, with spatial temporal distributions intensity. Five configuration scenarios designed gradually changing downscaling ratios used run WRF ECMWF 40-year reanalysis data periods eight events. A two-dimensional verification scheme proposed evaluate amounts simulated both dimensions. consists categorical continuous indices a first-level assessment quantitative evaluation rainfall. results reveal general improvement as we downscale outermost innermost domain. Moderate 1:7, 1:5 1:3 found perform better giving reasonable than smaller ratios. For study on types, shows best reproducing evenness observed rainfall, whereas type highly concentrated time trickiest case handle. Finally, efficiencies several variability verified categorising basis evenness, which could way event classification that facilitates further studies. It important similar studies carried out other catchments geographic climatic conditions, so error patterns can improvements made models. Copyright © 2011 John Wiley & Sons, Ltd.