Detecting Geothermal Resources in a Plateau Area: Constraints From Land Surface Temperature Characteristics Using Landsat 8 Data

Abstract

Geothermal energy is a kind of clean energy, which attracts more attention. The detection of geothermal resources is inseparable from regional geothermal prospects. Land surface temperature (LST) is an indispensable parameter for geothermal exploration, but the retrieval accuracy of LST for complex and remote areas is currently a major challenge. In this article, based on Landsat 8 remote sensing data, the characteristics of surface temperature retrieval methods are systematically reviewed, and the differences among these three algorithms are researched by using them to detect the surface temperature in the study area, which is Kangding County, Sichuan Province, China. Then the experimental results of the three algorithms are verified by using long-time (more than 1 year) measured data from the two monitoring sites, and the monitoring sites are situated in Zhonggu and Lao Yulin area of Kangding County. The results show that the radiative transfer equation (RTE) has the highest accuracy, and the mean error is 0.372°C; mono-window algorithm (MW) has a mean error of −0.606°C; and the split-window (SW) algorithm has the lowest accuracy, with a mean error of −2.07°C. The experimental results were used to select an algorithm with relatively high accuracy and low sensitivity. At the same time, a time series was used to perform temperature retrieval for this study area from November 2016 to December 2017 to evaluate the applicability of the method. The result shows that the RTE has the highest accuracy in mid-winter and a relatively low accuracy in summer in Kangding County. The purpose of this article is to establish a suitable method for high-precision surface temperature retrieval in plateau areas and to provide technical support for exploring geothermal resources or evaluating geothermal potential in these areas.