The Thermal Anomaly Monitoring And Trend Analysis Of Coalfields Of Shanxi Province In 2001-2010 Based On Landsat TM

The geological disasters of Shanxi are Main located in the coalfield field, and it mainly includes ground fissures, ground collapse, landslide and spontaneous combustion of coal seam, etc. Spontaneous combustion of coal seam is one of the main geological disasters in coalfield area and a common thermal anomaly.Coalfields exist all kinds of the thermal anomalies, and coal fire is one of the phenomena which is most common. Coal fire may cause the fire hazard which is one of the serious disasters in mines in our country all the year round. Shanxi Province has an investigation into Spontaneous combustion disaster of coal seam for the six coalfields area in 2012-2014. The results preliminary show that the total area of coal seam spontaneous combustion is about 56.59 square kilometers, the total loss of coal resources is about 250 million tons. The statistic is the census results mainly based on the thermal infrared remote sensing for coal seam spontaneous combustion area, not for all coal thermal anomalies, and lack of perennial dynamic detection research.This paper states the thermal anomaly extraction algorithm in detail, and it uses the thermal anomaly extraction algorithm for detecting the thermal anomaly area of the six coalfields of Shanxi Province by collecting the Landsat-5 TM which covers Shanxi Province from 2001 to 2010 in winter. Distribution characteristics and change tendency of the thermal anomaly area of Shanxi Province’s coalfields can be obtained through analyzing the thermal anomalies testing results. The key research content of the paper includes the following three aspects:( 1) Thermal anomaly region extraction algorithm is based on the differences of the reflection spectrum of various kinds of objects in the multispectral data to extract the coal region and it is based on the relative relationship between the target pixel and its neighboring pixels to obtain the thermal anomaly area by the convolution operation. To analysis the feasibility of thermal anomaly extraction algorithm which is based on multispectral analysis technique and convolution operation analysis technique, this paper compares the results of thermal anomalies area extracted by Landsat-5 TM and MOD14A2. The comparison suggests that the Landsat-5 TM and MOD14A2 to extract the thermal anomaly area are basically consistent on the spatial position and distributional shape. Its false alarm rate and missing rate are within an acceptable range. It shows the feasibility of using thermal anomaly region extraction algorithm to detect thermal anomaly of coalfield area in Shanxi Province.(2)The thermal anomaly extraction algorithm is used respectively for detecting the thermal anomaly area about the preprocessing Landsat-5 TM, by collecting the data covering the Shanxi Province from 2001 to 2010 in winter. The resulting data about the thermal anomaly of the same year is handled with such as removing the jag, merging the data, and using the vector boundary of mines to shear the whole data. Ultimately, it obtains the thermal anomaly distribution of the six coalfields of Shanxi Province in 2001-2010 and those white dots are the thermal anomaly area, the gray dots are the secondary thermal anomaly area.(3)This paper analyzes the distribution characteristics and the change trend of the thermal anomaly of every coalfield and mine in Shanxi Province from 2001 to 2010 respectively. The results show that the coalfield thermal anomaly area overall in Shanxi has a trend of weakening. The thermal anomaly area of Ningwu coalfield,Hedong coalfield and Qinshui coalfield are larger than others, the thermal anomaly area of Datong coalfield is smaller. The trend of Datong coalfield, Ningwu coalfield and Xishan coalfield is declining,whereas the trend other coalfield is opposite. The thermal anomaly area of Xiangning mine, Xuangang mine, Huozhou mine, Jincheng mine and Yangquan mine are larger than others, the thermal anomaly area of Shuonan mine is smaller. The trend of Datong mine, Pingshuo mine, Xuangang mine, Lanxian mine, Xishan mine, Liliu mine, Wuxia mine,Yangquan mine and Dongshan mine is declining, whereas, the trend other mine is opposite.