Detection of Geothermal Anomaly Areas With Spatio-Temporal Analysis Using Multitemporal Remote Sensing Data

Abstract

Thermal infrared (TIR) remote sensing is an important technology for detecting geothermal anomalies. However, detection results have been found to have a certain dependency on the distribution of ground objects and imaging conditions at different times, and pseudo-anomalous areas are easily extracted. To solve this problem, a new geothermal anomaly detection method is proposed in this article and implemented in the Jiaonan uplift in the Yishu fault zone. A temperature inversion experiment is carried out on TIR remote sensing images based on the radiation transfer equation. Then, a gradient operator is used to extract high-temperature regions in various periods, and geothermal anomaly areas are selected through the spatio-temporal analysis method proposed in this article after excluding the influence of impervious surfaces, water bodies, and vegetation. The temperature anomaly points, which are all high-temperature points in each inversion result and geothermal anomaly areas extracted by the proposed method, are compared with the five geothermal anomaly points, which are determined based on 150 geothermal wells and the geological structure in the study area. The spatial locations of the temperature anomaly points and geothermal anomaly areas are close to those of the geothermal anomaly points. Compared with the mean grad method, the proposed method is found to effectively delete some pseudo-anomaly areas under the premise of ensuring the extraction accuracy of geothermal anomaly areas.