Abstract: Fractional vegetation cover,which refers to the fraction of green vegetation seen from the nadir of the total statistical area,is an important parameter in surface process models for numerical weather prediction,regional and global climate modeling,and global change studies. Remote sensing provides the only feasible way to generate fractional vegetation cover products at the regional and global scales because of its advantages to quickly provide broad,impartial and easily available data on the land surface. Especially,the fractional vegetation cover of Shanxi's coalfields is sentive to local eco-environmetal chenges. Therfore,assessing the fractional vegetation cover and its changes over Shanxi's coalfields is an impotant project. In this study,the spatial-temporal pattern and change of fractional vegetation cover of six major coalfields in Shanxi Province of China were analyzed since 2000. The dimidiate pixel model based on the Normalized Difference Vegetation Index(NDVI)calculated by the Moderate Resolution Imaging Spectroradiometer(MODIS)spectral reflectance data was used to estimate fractional vegetation cover of of six major coalfields in Shanxi Province of China from 2001 to 2013. The trend analysis based on linear regression method was also used to analyze the trend of fractional vegetation cover. Results showed that the overall fractional vegetation cover variation presented an ascending trend during the 13-year study period. The fractional vegetation cover of Hedong and Datong coalfields improved dramatically among the six major coalfields. The fractional vegetation cover of the southern coalfields was better than that of the northern coalfields,and better of eastern coalfields than that of western coalfields. The fractional vegetation cover showed degradation trend in spring,obvious improving trend in summer,and slow increase trend in fall. The degradation of the ecological environment caused by the climate change and human activity is the main driving factor for the spatio-temporal changes of fractional vegetation cover in coalfield area. The proposed method is reliable for fractional vegetation cover estimation and can be implemented to detect a long time series changes of regional fractional vegetation cover from MODIS reflectance data. A lot of analysis indicated that the spatial and temporal continuity of the estimates by the proposed method was superior to that by using individual MODIS index. The findings of the paper highlighted that the seasonal and spatial variations in fractional vegetation cover are significantly increased in most coalfields and multiple factors contributed to this changes. Further work should focus on an extensive assessment of the performance of the proposed method using more observed data and analyze the potential contributors to variation in fractional vegetation cover of coalfields.

Key words: MODIS, NDVI, fractional vegetation cover, spatio-temporal variability, coalfields