| Coalfield fire is one of the grievous disaster record in China’s Agenda of21stcentury, it notonly seriously burnt out a lot of precious coal resource, but also formed a rigorous challenge to oursustainable development of social economy and environment, this issue is one of the populartopics which attracts various attentions from researchers and the whole society, which is a grandtrouble needs to be thrashed out immediately. Remote sensing technology has become one of theimportant methods in the research field of underground coal fire detection and dynamicmonitoring by its unique features of wide-detection range, relatively low cost and time efficiencyetc.This paper takes the coal fire of ShuiXigou mine area in Xinjiang Uygur Autonomous Regionas research area. Coalfield fire information is extracted by use of thermal infrared andmulti-spectrum image data of Landsat TM/ETM+in1990,2001and2011.By means of integratedmethods such as normalized multi-temporal remote sensing image, compared, analyzed andvalidated the surface temperature inversed algorithm, single-band optimal density segmentationand3-D visualization to detect and dynamic monitoring underground coal fires. The keyconclusions are as follow:(1) Land surface temperature inversion for the coalfield fire: Research is based on theobservation data of infrared radiometer and Landsat5image both type of data were collected onon same day which is June31,2011. The author calculated surface temperature at pixel scaleusing several schemes, compared, analyzed and validated the surface temperature of the coal fireareas inversed by mono-window algorithm, generalized single-channel algorithm and Wengalgorithm with TM data. The results show that all the three algorithms have consistent distributionof surface temperature of Shui Xigou underground coal fire areas, and the mono-windowalgorithm and generalized single-channel algorithm have the smallest differences in averagesurface temperature of the whole study area, which is about1.60℃. By comparing the groundmeasurements, a lower difference value is obtained by all the three algorithms, the retrieved databy generalized single-channel algorithm are highly similar to the data which retrieved bymono-window algorithm, regression coefficient is0.886，RMSE is1.48℃. The retrieved results ofgeneralized single-channel algorithm in line with the spatial distribution pattern of underground coal fire temperature, which can be seen obviously from the high-temperature anomalous region.The result of the retrieved data of surface temperature is acceptable and the generalizedsingle-channel algorithm has certain applicability on obtaining the LST of underground coal fireareas. The result provides a reference for the dynamic monitoring and evaluation of undergroundcoal fire areas.(2)Coalfield fire area dynamic monitoring: Attempt to normalize multi-temporal thermalinfrared images and developed a set of methods to detect and extract out thermal anomaliesinduced by underground coal fires using remotely sensed data, and accomplished3-Dvisualization, combined the optimal single-band optimal density split Sec segmentation. Shuixigoumining area is divided into three categories-background area, coal fire than the serious area andthe serious coal fire area. Use high-resolution Worldview-II images and high resolution DEM datato generate the mining area of three-dimensional topographic map, superimposed on the fire zonethermal anomaly information are used to image intuitively the development trend of the coal firearea for qualitative description. Via dynamic monitoring coal fires using remotely sensed data andfield campaign from1990to2011, dynamic variation rules and tendencies of sub coal fires can besummarized. The area of coalfield fire area increased form61200m2in1990to124200m2in2011,2.03times bigger than it used to have, which for first time before thermal anomaly area averagedegree of dynamic change was in1.28. after that average dynamic changes rose to4.57for tenyears. From2001to2011Shuixigou coal fire area, the strong expansion of the momentum of thedisaster in the rate of deterioration has accelerated noticeably. The more serious district coal firecenter of gravity moved to northwest in1990to2001,and from2001to2011moved to thenortheast. The overall trend is the northward expansion of coal fires serious district slightlyrepeated, however the overall trend is still extended northward. Shuixigou Mine area has threeindependent sub-fire zone A, B, C in1990, in2001A and B sub-fire zone merged, and in2011Cfire zone is expand to the A and B fire area. Ultimately, deep understanding of the mechanism andcharacteristics of surface spatial and temporal changes thermal anomalies coal fire in the studyarea, provides an objective basis for the implementation of the project coalfield fire extinguishing,and gives a scientific guidance to fire fighting project. |
PDF Full Text Request