关键词: 热环境, 核密度分析, 空间质心模型, PLUS模型

Abstract: Based on Landsat 8 OLI-TIRS remote sensing imagery, the surface temperature data of Nanjing City, China, was inverted using the radiation equation algorithm, the spatial and temporal variability of the thermal environmental landscape in Nanjing during 2013‒2021 is revealed in terms of quantity and rate, and the spatial evolution of the thermal environmental landscape is further revealed using kernel density analysis and spatial centroid migration trajectories. In addtion the driving mechanisms of the thermal environmental landscape from 2013 to 2021 was explored with the help of the PLUS model. The results indicate that: (1) The mesothermal landscape dominated the study area, from 2013 to 2021, and the strongest change in the thermal environmental landscape from 2017 to 2021, with a comprehensive thermal environmental dynamic degree of 23.94%. (2) During 2013‒2021, the density core of the low temperature zone was mainly located in water and woodland, with more pronounced clustering characteristics in the north. The sub low temperature zone was characterized by a denser and more fragmented distribution in northern and southern regions.The medium temperature zone had a "less in the middle and more around" distribution pattern.The sub high temperature zone landscapes showed a trend of multi-core growth, and the high temperature zone had the most significant landscape clustering characteristics, mainly distruibuted on both sides of the Yangtze River and in the centre of urban areas. (3) Generally, the low temperature zone has the fastest rate of centroid migration with an overall movement of 21.30 km to wards the northeast.The rate of centroid migration of the sub low temperature zone changed from slow to fast, during the period, first moving 9.52 km to the northwest and then 17.881 km to the southeast. The centroids of the high temperature zone, sub high temperature zone and medium temperature zone were shifted to the southwest as a whole, with distance of 19.99, 4.77 and 4.04 km respectively. (4) In terms of the driving factors for the expansion of the different thermal environmental landscapes, the expansion of the low temperature zone, sub low temperature zone and medium temperature zone was mainly driven by the elevation, while the main drivers for the expansion of the sub high temperature zone and the high temperature zone were distance from residential buildings and distance from factories.This study provides an important reference for strengthening the monitoring of Nanjing's surface thermal environment and promoting the sustainable development of the urban ecology.

Key words: thermal environment, kernel density analysis, spatial centroid model, patch-generating land use simulation model