Remote Sensing Detection of Vegetation and Landform Damages by Coal Mining on the Tibetan Plateau

Abstract

In order to satisfy the needs of constant economic growth, the pressure to exploit natural resources has been increasing rapidly in China. Particularly with the implementation of the National Western Development Strategies since 1999, more and more mining activities and related infrastructure constructions have been conducted on the Tibetan Plateau (TP). Mining activities are known to have substantial impacts on plant dynamics and hence the water and energy cycles. Identifying mining activities and quantifying their effects on vegetation cover are critical to the monitoring and protection of the pristine TP environment. Thus, this study aims to develop an automated approach that detects the timing of initial mining development and assess the spatial distribution of mining-ruined vegetation. The Breaks for Additive Seasonal and Trend (BFAST) algorithm was used to decompose the signal in the normalized difference vegetation index (NDVI) time series derived from high-frequency MODIS images, and to detect abrupt changes of surface vegetation. Results show that the BFAST algorithm is able to effectively identify abrupt changes in vegetation cover as a result of open-mining development on the studied alpine grassland. The testing study in Muli Town of Qinghai Province shows that the mining development began in 2003 and massive destructions of vegetation cover followed between 2008 and 2012. The integrated use of Landsat imagery and multi-temporal DEMs further reveals detailed areal and volumetric changes in the mining site. This study demonstrates the potential of applying multi-mission satellite datasets to assess large-scale environmental influences from mining development, and will be beneficial to environmental conservation and sustainable use of natural resources in remote regions.