Underground Coal Fire Detection and Monitoring Based on Landsat-8 and Sentinel-1 Data Sets in Miquan Fire Area, XinJiang

Underground coal fires have become a worldwide disaster, which brings serious environmental pollution and massive energy waste. Xinjiang is one of the regions that seriously affected by underground fires. After years extinguishing, fire areas in not been significantly reduced yet. To extinguish fires, it critical to identify monitor them. Recently, remote sensing technologies showing great potential fires’ identification monitoring. The thermal infrared technology usually used detect anomalies areas, Differential Synthetic Aperture Radar Interferometry (DInSAR) for detection related ground subsidence. However, non-coal caused objects with low specific heat capacity, surface subsidence mining crustal activities accuracy areas. improve using technologies, this study firstly obtains temperature, normalized difference vegetation index (NDVI), information based on Landsat8 Sentinel-1 data, respectively. Then, multi-source strength weakness constraint method (SWCM) proposed analysis. results show SWCM has highest among employed methods. Moreover, can reduce commission omission error fire-related Specifically, 70.4% average, 30.6%. Based results, spatio-temporal change characteristics obtained. In addition, found there significant negative correlation between time-series temperature rate (R2 reaches 0.82), indicates feasibility both