GEOGRAPHICAL RESEARCH ›› 2021, Vol. 40 ›› Issue (3): 856-868.doi: 10.11821/dlyj020190975

• Articles • Previous Articles     Next Articles

Spatial pattern analysis and quantitative detection of multi-factor influence for urban heat island effect in a mountainous city: A case study of Chongqing metropolitan circle

WANG Yang1,2,3(), YANG Dan1, MIN Jie1,2,3(), ZHAI Feitong1, WANG Yu1, WU Xiaojiao1, ZHANG Hongrui1   

  1. 1. School of Geography and Tourism, Chongqing Normal University, Chongqing 401331, China
    2. Key Laboratory of GIS Application, Chongqing Municipal Education Commission, Chongqing 401331, China
    3. The Three Gorges Reservoir Area Surface Processes and Remote Sensing Municipal Laboratory, Chongqing 401331, China
  • Received:2019-11-11 Accepted:2020-07-01 Online:2021-03-10 Published:2021-05-10
  • Contact: MIN Jie;


Urban heat island effect (UHI) is affected by multiple factors on the urban surface, while the situation of a mountainous city is more complicated. To detect the cause of UHI, this article takes a typical mountainous city of Chongqing as an example. Firstly, we collected multi-source spatial data based on Landsat8 OLI/TIRS images, high-precision vector building, etc. as basic dataset. Then, the model method of urban land surface temperature retrieval (LST), normalized vegetation index (NDVI) and the sky view of factor (SVF) etc. are applied to obtain spatial pattern of each factor. Finally, with focus on the urban built-up area, the binding force of each factor on the UHI is analyzed by the method of geographic detector. Through the above steps, this study found that: (1) Around the whole metropolitan area, the spatial heterogeneity of urban thermal field is significant. From the spatial pattern of each factor, we can see that some factors, like vegetation coverage (NDVI), urban surface elevation (CSE) and the sky view of factor (SVF), have global binding performance for UHI, while others like building density (BD), building volume rate (BVR) and road network distance (RD) have local binding performance for UHI. (2) Within the urban built-up area, the top three factors influencing the spatial pattern of UHI separately are vegetation coverage (q=0.782), urban surface elevation (q=0.499) and building density (q=0.496). Besides, the local constraint factors, like building density (BD), building volume rate (BVR) and road network distance (RD), performance strong binding on UHI, yet among them show little difference. In addition, for there is no significant spatial difference in urban sky horizon within the high-density built-up area, the overall effect of the sky view factor (SVF) on urban heat island is relatively small (q<0.1). (3) Through interactive detection and analysis, the results suggested that each influencing factor shows overlapping constraints on the spatial distribution of UHI within the built-up area of the city. From the criterion q value of the factor binding force, we can see that interaction of two factors will increase the interpretation of the UHI. In other words, the superposition explanatory degree (q(XiXj) is stronger than the independent interpretation degree, whose superposition interpretation degree q value is between 0.50 and 0.82.

Key words: mountainous urban heat island effect, multi constraint factor detection, spatial pattern analysis, geographic detector, Chongqing