Thermal Infrared Remote Sensing Based Coal Spontaneous Combustion Detection Method And Change Detection For Coal Fire Areas

The previous temperature inversions and coal fire detection methods using thermal infrared remote sensing were studied.Then a novel coal fire detection method:self-adaptive gradient based thresholding(SAGBT)method was proposed.For verifying this SAGBT method we designed simultaneous satellite-field observations in four typical seasons.ASTER and Landsat TIR data was obtained,and coal fires delineated by SAGBT were compared with the field measured land surface temperature(LST)samples.Using these field measurements and a solar irradiation simulation,a series of adjustment parameters for the temperature inversion method for four typical seasons were calculated.SAGBT was applied to images spanning over 10 years(2001-2011)and specifically in about one year(2007-2008)to retrieve the changes of coal fires in the Wuda coal field.As a result of the ten year change detection in 2001?2011,the area of coal fires shows a positive correlation with coal productivity in the Wuda coal field.Coal fire centers were observed migrations from deep to shallow or along the strike from the annual change detection in 2007?2008.Combining the coal fire area in different stages by change detections,combustion experiment based CO2 emission rates and gas content by analysis of coal fire vents,we estimated the annual coal loss at 44.3×103 tons and CO2 emission equivilant to be 92.7 x 103 tons induced from coal fires in the Wuda coal field.