Detecting the dynamics of vegetation disturbance and recovery in surface mining area via Landsat imagery and LandTrendr algorithm

作者：Yang, Yongjun; Erskine, Peter D.; Lechner, Alex M.; Mulligan, David; Zhang, Shaoliang^*; Wang, Zhenyu

来源：Journal of Cleaner Production, 2018, 178: 353-362.

DOI：10.1016/j.jclepro.2018.01.050

• Environmental assessment
• Land cover change
• Remote sensing
• Mining
• Rehabilitation
• Cumulative D/R ratio

The vegetation cover dynamics such as the spatial extent and pattern of disturbance and recovery are necessary information in mining regulation and environmental sustainability assessment. The acquisition and further analysis of these dynamics are hampered due to that the vegetation cover in mining area changes both spatially and temporally. This paper aims to assess the capability of LandTrendr algorithm and Landsat imagery in surface mining area to detect the vegetation change and characterize the historical dynamics. Curragh coal mine site in Australia was taken as an illustrative application. The spatial extent, change pattern and attributes of vegetation disturbance and recovery during 1989-2014 were detected and mapped. The overall accuracy for disturbance and recovery classification are 85.21% and 86.59% respectively. The results show that more than 2982 ha out of 4573 ha of disturbed land in entire Curragh had been covered by vegetation, while 95% of the oldest part of the operation (Central Curragh) has been seeded and rehabilitated. The average value of cumulative Disturbance to Recovery ratio for entire Curragh mine site was about 59%. This suggests that the Curragh mine has been complying with mining legislation by undertaking progressive rehabilitation and vegetation establishment efforts. The advantage of LandTrendr is that it relies on free data from Landsat archive and has straightforward operating procedures. It can provide environmental assessment and rehabilitation with serviceable data including raster maps, yearly data, cumulative Disturbance to Recovery ratio and change attributes. The LandTrendr algorithm and Landsat imagery could be supplemented with field validation and study such as recovery success analysis to facilitate better environmental assessment and mining sustainability.

• 出版日期2018-3-20
• 单位中国矿业大学（北京）