Sensitivity analysis of physical parameterizations in WRF for urban climate simulations and heat island mitigation in Montreal

Abstract

In mesoscale meteorological modeling, physical models (radiation, planetary boundary layer, microphysics, cumulus, and land-surface) are employed to calculate terms of governing equations. We evaluate the sensitivity of near surface air temperature, wind speed, relative humidity and precipitation to different physical models within Weather Research and Forecasting (WRFV3.6.1) model for urban climate simulations and heat island mitigation for Montreal, Canada for period 09–11 Aug-2009. A multi-layer urban canopy model is used to consider the turbulence between buildings in urban areas. The model ensemble with the least error is proposed as an appropriate platform for urban climate simulations to study Urban Heat Island (UHI) mitigation strategies. Increasing surface reflectivity was applied to mitigate the UHI intensity over the domain. The albedo of roofs, walls, and roads increased from 0.2 to 0.65, 0.6, and 0.45, respectively. The results of surface modification are presented. The averaged 2-m air temperature indicates a decrease by 0.2°C. The 10-m wind speed is slightly increased over the domain. The relative humidity decreases as an average of 2.8% and the average precipitation reduced by 0.2mm. An increase of albedo leads to a net decrease of radiative flux into the ground and therefore a decrease of convective cloud formation and precipitation.