Impacts of local climate zone mapping quality on urban near-surface air temperature simulation in WRF-UCM

Abstract

Accurate simulation of urban near-surface air temperature (SAT) is essential for understanding climate dynamics. The reliability of the simulated SAT depends largely on the representation of urban morphology. Local Climate Zones (LCZ), as a sort of urban morphology representation, have been widely applied for SAT simulation based on the Weather Research and Forecasting Model coupled with urban canopy model (WRF-UCM). However, how the LCZ mapping quality affects the accuracy of SAT simulation remains underexplored. Here we investigate the relationship between the LCZ mapping quality and the accuracy of SAT simulation in WRF-UCM. The city of Guangzhou, China, is selected as the case study area. Six LCZ classifications are used to simulate SAT at the 1 km resolution, and the results are validated using the meteorological observations. The simulations suggest that the average values of root mean square error decrease if the more accurate LCZ classifications are used. The simulated SAT is also sensitive to the local mapping quality of LCZ classifications within 3 to 7 km windows. Our findings can provide insights for modelers to develop more accurate simulations of SAT, which are fundamental to the understanding of climate change impacts and promote relevant policy making toward sustainable cities.