Heat Risk Assessment in Arid Zones Based on Local Climate Zones: A Case of Urumqi, China

Abstract

Based on the rapid development of urbanization and the increasing severity of extreme heat disasters caused by global warming, it has become increasingly important to enhance the assessment of heat risk. In this study, in response to the urgent need for fine-grained assessment of urban heat risk in arid zones in the context of climate change, an analytical method of dividing Local Climate Zones (LCZs) into street blocks combined with the Hazard–Exposure–Vulnerability–Adaptability (HEVA) heat risk assessment framework is used in Urumqi, a representative city of China’s arid zones. In addition, Shapley Additive Explanations (SHAP) was introduced to quantitatively resolve the driving mechanisms of heat risk in different types of LCZs. The results show that the study area has the largest proportion of bare soil (LCZ F) (37.6%), which is distributed around the built-up types of LCZs, while water (LCZ G) has a very small proportion (0.39%) and only exists in the outskirts of the city. Heat risk was significantly higher in the urban core than in the peri-urban areas, but LCZ F had a very high hazard due to the unique surface characteristics of arid zones, which elevated the heat risk in the peri-urban desertification fringe; SHAP analyses demonstrated that in arid zones, land surface temperature (LST) became a determinant of heat risk for all low-density built-up types of LCZs. This study proposes targeted mitigation strategies for heat risk in arid zones based on the LCZ framework.

Keywords

• local climate zone
• HEVA
• heat risk
• arid zone
• SHAP
• machine learning