Abstract: Urban heat island effect, a typical feature of urban climate, plays an important role in urban planning and people’s production and life. Based on the Landsat 8 TIRS data, we adopted the split window algorithm to estimate the surface temperature of the central urban area of Chengdu City at four different time points, and then analyzed the spatial and temporal dynamics of heat island effect and its influence factors. The algorithm achieved an accuracy of 1.6 K without synchronous meteorological observation data to adjust the atmospheric transmissivity. There was an obvious heat island effect in the central area of Chengdu, with the area of such lands accounting for about 30% of the total area. The spatial variation of surface temperature in spring and summer was obviously larger than that in winter. The heat island effect out of the third-ring was more obvious than that of the area within it. There was a high temperature area between Wuhou District and Shuangliu District. The negative correlation between normalized difference vegetation index(NDVI) and land surface temperature (LST) was the most obvious in summer, suggesting that urban heat island effect could be alleviated by planting green plants. The correlation between LST and modified normalized difference water index (MNDWI) in summer was lower than that in spring and winter, indicating that water may have limited effect in alleviating urban heat island effect in summer. The positive correlation between normalized difference build-up index (NDBI) and LST at four time points showed that human activities may be an important cause of urban heat island effect, and thus NDBI is an effective index for urban heat island effect.

Key words: potassium level, absorption., ¹³C, ¹⁵N, apple