基于地理探测器的城市热岛驱动因素分析——以武汉市为例

城市热岛是由地表特征、社会经济和气象等因素综合引起的环境问题，具有显著的季节差异。大量的研究分析了土地覆盖类型、地表覆盖指数、景观指数等和城市热岛之间的相互关系，忽略了气象因素以及气象因素、地表特征与社会经济之间的交互作用对城市热岛的影响。在分析武汉市不同季节热岛强度及空间自相关分析的基础上，采用地理探测器研究了不同季节城市热岛强度的驱动因素及其交互作用和热岛风险区探测。结果表明：武汉市各季节热岛强度在空间上呈现强烈的集聚特征。按春夏秋冬顺序，武汉市4个季节热岛强度的主要驱动因素分别为地表特征、社会经济、社会经济和气象因素，四季对应的主要驱动因子分别是土地覆盖类型、夜间灯光、夜间灯光和气压。在11个驱动因子的交互影响中，四季热岛强度交互影响作用最大的分别为土地覆盖类型与夜间灯光、土地覆盖类型与NDVI、土地覆盖类型与夜间灯光、气压与土地覆盖类型或与NDVI。风险探测结果显示春夏秋冬平均热岛强度最高的分别为城市建成区、城市建成区、城市建成区、裸地和低植被覆盖区，最低的分别是水体、水体、水体、植被；四季的平均城市热岛强度随着夜间灯光的增加而增加；NDVI子区域平均热岛强度呈现先上升后下降的趋势；气压子区域平均热岛强度总体呈现逐步上升的趋势。这些结果揭示了热岛强度的季节变异特征，可为缓解城市热岛效应措施的制订提供依据。

Analysis of Driving Factors of Urban Heat Island Based on Geographical Detector:Taking Wuhan City as an Example

Urban heat islands are environmental problems caused by the combination of surface characteristics, socio-economic and meteorological factors, with significant seasonal differences. A large number of studies have analyzed the relationship between land cover types, land cover indices, landscape indices, and urban heat islands, ignoring the impact of meteorological factors and the interaction between meteorological factors, surface features and socio-economics on urban heat islands. Based on the analysis of the intensity and spatial autocorrelation analysis of heat islands in different seasons in Wuhan, this paper uses geo-detectors to study the driving factors and interactions of urban heat islands in different seasons and the detection of heat island risk areas. The results show that the intensity of heat islands in various seasons in Wuhan shows a strong agglomeration characteristics in space. According to the sequence of spring, summer, autumn and winter, the main driving factors of the four seasons heat island intensity in Wuhan are surface characteristics, socio-economic, socio-economic and meteorological factors. The main driving factors for the four seasons are land cover type, night light, night light and air pressure. . Among the interaction effects of the 11 driving factors, the most important interactions of the four-season heat island intensity are land cover type and night light, land cover type and NDVI, land cover type and night light, air pressure and land cover type or NDVI. The results of risk detection show that the highest average heat island intensity in spring, summer, autumn and winter is urban built-up area, urban built-up area, urban built-up area, bare land and low vegetation coverage area. The lowest are water body, water body, water body and vegetation; the average city of four seasons The intensity of the heat island increased with the increase of nighttime light; the average heat island intensity of the NDVI sub-region showed a trend of increasing first and then decreasing; the average heat island intensity of the gas pressure sub-region showed a gradual upward trend. These results reveal the seasonal variation characteristics of heat island intensity, which can provide a basis for the formulation of measures to alleviate urban heat island effect.