干旱区生态环境知识资源中心

Given its proximity to an artificial oasis, the Donghu Nature Reserve in the Dunhuang Oasis has faced environmental pressure and vegetation disturbances in recent decades. Satellite vegetation indices (VIs) can be used to detect such changes in vegetation if the satellite images are calibrated to surface reflectance (SR) values. The aim of this study was to select a suitable VI based on the Landsat Climate Data Record (CDR) products and the absolute radiation-corrected results of Landsat L1T images to detect the spatio-temporal changes in vegetation for the Donghu Reserve during 1986-2015. The results showed that the VI difference (VI) images effectively reduced the changes in the source images. Compared with the other VIs, the soil-adjusted vegetation index (SAVI) displayed greater robustness to atmospheric effects in the two types of SR images and was more responsive to vegetation changes caused by human factors. From 1986 to 2015, the positive changes in vegetation dominated the overall change trend, with changes in vegetation in the reserve decreasing during 1990-1995, increasing until 2005-2010, and then decreasing again. The vegetation changes were mainly distributed at the edge of the artificial oasis outside the reserve. The detected changes in vegetation in the reserve highlight the increased human pressure on the reserve.