ESTIMATING TYPICAL FOUR-SEASON THERMAL DISTRIBUTION AND SELF-ADAPTIVE GRADIENT BASED THRESHOLDING PARAMETERS OVER COAL FIRE AREAS IN CHINA USING ASTER AND LDCM TIRS-A CASE STUDY IN THE WUDA COAL FIELD, CHINA

摘要

Coal fire results in significant environmental impacts and coal losses in many countries, including the United States, India, and China. Multiple fire detection methods have been proposed. Many of these rely on thermal infrared (TIR) imagery. This study results from previous research on TIR, including our development of a self-adaptive gradient based thresholding method for coal fire delineation. We used field measurements and images acquired by the ASTER sensor onboard NASA's Terra satellite and the TIRS onboard Landsat 8 to derive calibration parameters for a threshold estimation algorithm considering different solar radiation intensities, which impact radiance estimations from coal fire. We designed a simultaneous ASTER-field measurement plan in the Wuda coal field (China) and scheduled image collection for four periods, including the winter and the summer solstices (least and most intense solar radiation periods). Collection also included the vernal/autumnal equinoxes. Land surface temperature (LST) was collected before and after each satellite overpass in planned intense sampling block areas. LST field samples were integrated into 90-100 m TIR pixels. Data were combined with coal fire boundaries collected in the field and were used to validate the coal fires retrieved from four calibrated image by our temperature retrieving method, and the gradient-based threshold method. Results are a series of adjustment parameters for the fire detection method for four typical seasons. Correction parameters estimated by our method at the Wuda coal field can be extended to other fire areas lacking detailed studies, thus supporting surface temperature retrieval and underground coal fire delineation.