Analysis of Grassland Desertification in Open-Cast Coal Mines Based on Remote Sensing and Its Countermeasures——An Example from Huolinhe Open-Cast Coal Mine

The desertification grassland in Huolinhe open-cast coalmine was extracted from remote sensing imageries using a decision tree and back-propagation(BP) neural network,and the desertification during the past 16 years in the research field was analyzed.The results show that 1) there is a strong to weak,then weak to strong evolution process for the total grassland desertification,which can be attributed to climate changes and human activities,especially to the influences of climate changes;2) there is the similar strong to weak, then weak to strong evolution process for the area of middle and slight desertification,which can be mainly ascribed to climate changes,then to the over-grazing,uprooting of the medicinal materials and over-reclamation;3) due to the coal exploitation,the severe desertification is always distributing around the mining land,and developing to the east quickly with the help of monsoon;4) the area of severe grassland desertification is on the increase during the past 16 years,and there is much correlation between the mining land and severe grassland desertification area.According to the characteristics of grassland desertification in the opencast coalmioe,the strategies of controlling desertification such as planting shifting-sand control and wind break trees,reinforcing land reclamation,changing unadvanced ability of the production and management and dividing the whole region into sub-regions to maximize the grassland resources are put forward.