Assessing the Impact of Land Use Patterns, NDVI, and NDBI on Urban Land Surface Temperature in Taipei, Taiwan, Using Landsat Data

Abstract In recent years, cities worldwide have faced extreme heat, making it essential to study the geographical characteristics of urban environments to develop sustainable urban strategies. This research employs Landsat data from 1984 to 2024, a ArcGIS Pro (version 3.4.2) and Python (version 3.12.1) to thoroughly analyze spatial changes in surface urban heat island (SUHIs) in subtropical Taipei City. We examined the relationships among the normalized difference vegetation index (NDVI), normalized difference built-up index (NDBI), and land surface temperature (LST) concerning land use and land cover. The results show that the Kappa coefficient for the years 1988, 2001, 2007, and 2022 was 0.93, 0.93, 0.94, and 0.92, respectively. LST increased significantly during the study period, with a maximum increase of 12.2°C from 1984 to 2022. Moreover, the growth rate of building area from 1984 to 2024 was 209.3%; this substantial increase in vertical building development substantially impacted the urban thermal environment. LST showed a moderately strong negative correlation with NDVI and a moderately strong positive correlation with NDBI, with both correlations reaching statistical significance. These findings emphasize the urgency of sustainable urban planning controls, especially in high-temperature and high-humidity climatic conditions, to mitigate the impacts of SUHIs.