Effect of local climate zone (LCZ) and building category (BC) classification on the simulation of urban climate and air-conditioning load in Hong Kong

Reliable energy models are required to predict urban-scale air-conditioning (AC) load under extreme outdoor temperatures which is crucial to power security. In this study, we proposed a new landuse/landcover (LULC) scheme for these models by combining the local climate zones (LCZs) from World Urban Database Access Portal Tools (WUDAPT) and building category (BC) map. Apart from evaluating against the conventional LULC of Moderate-resolution Imaging Spectrometer (MODIS) and WUDAPT, the impact of urban canopy parameters and BC on the simulated meteorological conditions and AC load during the 2016 heatwave in Hong Kong was examined by the Weather Research and Forecasting (WRF) model. Incorporating LCZs (BC) reduced the bias of 2-m temperature (T2), 2-m relative humidity (RH2), and AC load intensity by 0.4 degrees C (2.0 degrees C), 1.98% (0.68%) and 6.78 W-2 (13.09 W m-2), respectively. It captured the heterogeneous, intra-urban T2 and turbulence kinetic energy (TKE). Further sub-classifying the BC in LCZs led to WUDAPT/BC which performed even better and significantly improved WRF output. It simulated well the daytime wind speed, mitigated the overprediction (underprediction) of T2 (RH2), and rectified the underpredicted AC load in WUDAPT. Its benefit was prominent in daytime as well as in compact high-rise/mid-rise and open high-rise areas. (200 words).

Automated summary

This study proposed a new land use/land cover (LULC) scheme for predicting urban-scale air-conditioning load. By combining local climate zones (LCZs) and building categories (BC), the model was able to reduce the bias in temperature and humidity, as well as improve the simulation of air-conditioning load. The inclusion of LCZs and BC in the model significantly enhanced the accuracy of the predictions.