Comparing urban canopy models for microclimate simulations in Weather Research and Forecasting Models

Abstract

The modelling of the urban heat island (UHI) phenomenon is becoming critical to allow better prediction and assessment of mitigation strategies. This paper compares different urban canopy models to characterize the UHI in Toronto. In particular, the effects of urban parameterization modeling are investigated using the state-of-the-art Weather Research and Forecasting Model (WRF). The WRF is coupled separately with three Urban Canopy Models (UCMs) to predict the heat and moisture fluxes from the canopies to the atmosphere. The three UCMs are a slab (SB), single-layer (SL), and multi-layer (ML) representation of the canopy. The SB, SL, and ML respectively consider buildings as increased roughness elements, a single orientation of the two-dimensional approximation, and the three-dimensional urban surfaces that also accounts for the vertical exchange of heat, moisture, and momentum. The WRF-UCMs are used to investigate a specific heat wave period. The simulations are evaluated comparing the results of the air temperature, wind speed, relative humidity, and dew point temperature with the measurements from several weather stations. The correlation (r-squared) of the air temperature and dew point temperature was close to 0.9 and 0.85, whereas the correlation of the wind speed and relative humidity was 0.6 and 0.7 respectively. The UHI magnitude is also estimated by the air temperature changes between urban and rural areas. The results illustrate that the daily heat island magnitude is around 1.2–1.5 °C, while the daytime is near 0.7 °C. The UHI is magnified during the nighttime and is up to 2 °C. The analysis indicated that the SL is reliable for climate simulations, but for the evaluation of the UHI magnitude and to analyze more sophisticated structures, the ML urban canopy model has to be applied as it is critical to account for the air turbulences and multi-reflections in the urban canopy.