Integrating satellite remote sensing techniques for detection and analysis of uncontrolled coal seam fires in North China

摘要

China is the biggest producer of coal in the world and mines about 1000 Mt of raw coal per year. Approximately 70% of China's energy consumption is covered by coal. At the same time, it is estimated that about 20 Mt of coal are being burnt in uncontrolled coal fires in China each year. Because these coal fires are spread out over the whole northern part of the country, stretching from Xinjiang province in the West to the Pacific coast in the East, it is extremely difficult to keep an overview of the development of known fires as well as of newly developing ones. Satellite remote sensing offers a powerful tool to observe and monitor such large regions; however, special methods and techniques have to be derived to accurately detect and monitor near surface coal seam fires. In this paper, an integrated satellite remote sensing approach is described allowing detection and monitoring of near surface coal seam fires by observing subtle land surface changes induced by the fires. These changes include thermal surface anomalies, changes in spectral surface characteristics as well as land subsidence caused by the fires. The methods comprise the radar interferometric generation of digital elevation models (DEMs) for geometric referencing and orthorectification of satellite imagery, dedicated analysis of thermal satellite data (daytime and nighttime), and multispectral analysis of land surface properties as well as the mapping of small-scale land subsidence by means of differential radar interferometry. It is shown how different satellite remote sensing methods can be synergistically combined to detect, analyze, and monitor near surface coal seam fires in arid or semiarid areas of North China. First results show the successful application of the methods and, furthermore, a comparison with ground measurements is given. While thermal and optical analysis of the fires can be considered robust methods, the assessment of coal seam fires using differential radar interferometry still has to be further developed in order to serve as a reliable analysis and monitoring tool. Within this work, a special focus is given to aspects of automation of the coal fire detection and analysis by means of satellite remote sensing in order to allow fire mapping in large areas with only minimal operator interaction.