Eco-environment Dynamic Monitoring And Prewarning Study In Panxie Coal Mining Area, Huainan

Coal resources play a significant role in economic and social development in China. However, extensive exploitation of underground coal resources also brought enormous damage to local ecological environment and caused a series of environment problems. In Huainan coal mining area, East China, long-term, large-scale underground coal mining activity has caused severe ground subsidence, which destroyed cultivated lands and formed a large area of waterlogged area. Moreover, influenced by the surrounding industrial, agriculture wastewater and residents living sewage free discharge, the water quality in these waterlogged areas is poor, mainly presented as water eutrophication. Currently, the environment monitoring methods used in coal mining area are so simple and inefficient that they cannot satisfy the demand of dynamic environment monitoring in coal mining area. Thus, this study adopted remote sensing geochemistry methods to analyze the spatial-temporal land use change and water quality of waterlogged area. Then, based on the local ecological environment and previous research, we comprehensively evaluated the ecological environment in Panxie coal mining area and made a prewaming of the ecological environment.Traditional pixel-based and object-based classification method were used for Pleiades and Landsat images to accurate map the land use in Panxie coal mining area. The results showed that object-based image analysis (OBIA) obtained the higher accuracy and the classification map was closer to human interpretation. Compared to pixel-based classification method, OBIA can reduce spectral variability and the heterogeneity of different land use types to avoid the salt-pepper effect during the process of the land use information extraction. The overall accuracy of the land use classification map was more than 90% in Pleiades image. Moreover, Mine productivity volume data were used to find the causes of the land use transformation in Panxie coal mine area, and found significant relationship between land use classes and mine productive growth. The cultivated lands decreased with the mine productive growth, while urban areas, coal and coal gangue, and water growth rapidly. These results indicated that coal mining activity was the major driving factor of the land use change in Panxie coal mining area.Significant temporal and spatial variabilities in constituent concentrations and spectral absorption coefficients of three optically active substances (colored dissolved organic matter, phytoplankton and non-pigment suspended matter) for water in closed and semi-closed waterlogged areas were reported. The descriptive statistics of water- quality parameters measured in Pj and Gq indicated optical conditions typical for turbid productive inland waters. These two waterlogged areas were significantly influenced by non-pigmented components (ad(λ) and aCDOM(λ). The contributions from non-pigmented components were higher in Pj than Gq because of the connection to the Ni River. Based on the optical characters analysis in Pj and Gq, the NIR-red, three-band, and NDCI models were established using in situ hyperspectral data. The three-band model is more accurate for semi-closed waterlogged areas, while the NDCI model is more accurate for closed waterlogged areas. On the other hand, the optimal spectral bands of TSM retrieval models were different in waters with different particle composition and size structure in Pj and Gq. In Pj, the optimal spectral bands were located in red and NIR bands, while the optimal spectral bands in Gq were located in blue and green bands.Models of water quality parameters were established using the HJ-1 data. Comprehensive Trophic State Index method was adopted in this research to evaluate the water quality of the waterlogged areas. Multi-spectral data can not only show the trophic status of the waters and reflect the spatial and temporal distribution of the water quality in waterlogged areas, but also help to found the pollution sources and their transportation characteristics, which cannot be revealed by traditional methods. The results showed that Pj and Gq were eutrophic and revealed different spatial and temporal distribution characteristics. Generally, the trophic status in May was more serious than in November, and the trophic status in Gq was more serious than in Pj.A typical, comprehensive, and operable evaluation index system was built to comprehensively assess the ecological environment of Panxie coal mining area using fuzzy synthetical marking method. The result showed that the eco-environment quality in Panxie coal mining area was weak under the long-term underground coal mining activity, and the comprehensive assessment results from 2008 to 2013 were in grade Ⅳ with a trend of deterioration. According to the prediction results in three years (2016-2018), the eco-environment quality in Panxie coal mining area was at a moderate warning level.