Multi-temporal landsat thermal infrared mapping for geothermal assessment in Taiwan

Abstract

Taiwan hosts both active volcanic systems and abundant geothermal resources due to its complex tectonic setting, making it a promising region for geothermal exploration. Land surface temperature (LST) is a key parameter for identifying geothermal prospects and assessing surface heat distribution. In this study, multi-temporal, island-wide LST composites were generated for the first time from cloud-free Landsat thermal infrared data acquired in March 2001, November 2015, and March 2025. Using a seamless mosaicking approach, the thermal images were processed to produce full-coverage LST maps of Taiwan, with non-terrestrial (oceanic) areas masked. Temporal differences among the three datasets were then analyzed to identify surface thermal anomalies and variations. The results reveal distinct patterns of elevated surface temperatures across the island, reflecting both geothermal manifestations and urban heat island (UHI) effects. While geothermal areas exhibit naturally high temperatures due to subsurface heat flow, UHI zones display enhanced warming associated with anthropogenic activities. Differentiating these overlapping influences remains a key challenge in regional thermal analysis. Despite no significant island-wide LST variation over the two-decade period, the detailed LST maps effectively delineate localized thermal anomalies corresponding to known geothermal fields and potential new sites yet to be discovered. This study demonstrates the utility of long-term satellite thermal observations for geothermal resource exploration and monitoring in Taiwan. The mosaicked LST dataset provides valuable spatially continuous thermal information that can be integrated with geophysical and geological data to enhance geothermal assessment, guide future field investigations, and support sustainable geothermal energy development.