Analysis Of Vegetation Coverage Changes On Alpine Grassland Along The Qinghai-Tibet Railway Based On Remote Sensing Images

The Qinghai-Tibet railway is the key project connecting Qinghai-Tibet area and the mainland, and its construction has a certain influence on the local vegetation. Using remote sensing technique can monitor local vegetation conditions easily and timely. Moreover, it can provide reliable decision in determining vegetation restoration along the Qinghai-Tibet railway, and it would be more helpful to the sustainable development of the western region.In this thesis, Wudaoliang area along the Qinghai-Tibet railway was selected as the research area and two Landsat satellite images acquired in the year of 2000 and 2013 were selected as the remote sensing data sources. We extracted the vegetation coverage information along the Qinghai-Tibet railway and analyzed changes of vegetation coverage. The main work and conclusions in this thesis can be summarized as,(1) The Landsat 8 remote sensing image in 2013 was executed decision tree classification using the ENVI software to generate the vegetation classification image. This image was comprised of Stipa purpurea and Carex moorcroftii steppe, Poa rossbergiana and Littledalea racemosa meadow, Kobresia pygmaea meadow, Kobresia schoenoides meadow, Arenaria musciformis and Androsace tapete cushion vegetation. The classification precision evaluation results showed that the overall accuracy was 63.82%, the Kappa coefficient was 0.5575;(2) The vegetation coverage in study area was extracted by multiple endmember spectral mixture analysis(MESMA) using VIPER Tools. Then we made the correlation analysis between the extracted vegetation coverage and the normalized difference vegetation index(NDVI), and the correlation coefficient was 0.9957. It showed that the extracted vegetation coverage using MESMA had higher reliability to some extent;(3) We made 0.5km, 1km, 2km, 3km, 5km, 10 km, 15 km, 30 km and > 30 km buffer zones from the track on the east and west separately, and the average vegetation coverage in each buffer zone in 2000 and 2013 was calculated. The results showed that the Qinghai-Tibet railway had caused the vegetation coverage decrease on both sides. It was 1.7%-3.6% on the west from 0 to 10 km off the track, and it was 4.7%-5.6% on the east from 0 to 3 km off the track.(4) We counted the vegetation coverage changes of different types in each buffer, and got the following conclusions: the Qinghai-Tibet railway had caused that the vegetation coverage of alpine steppe decreased by 20.8% on the west from 0 to 2km off the track, and 8.2% on the east from 0 to 3km off the track. Furthermore, it led to vegetation coverage of Kobresia pygmaea meadow dropping 9.0% on the west from 0 to 15 km off the railway, and 20.0% on the east 0 to 5km off the railway; In terms of Poa rossbergiana and Littledalea racemosa meadow, the vegetation coverage reduced by about 3.8% on the east from 0 to 2km off the railway, but no changes in other area; The Qinghai-Tibet railway caused that the vegetation coverage of alpine marsh meadow decreased by 2.4% on the east from 0 to 5km off the track, and no significant changes on the west. The influence of the railway on alpine cushion vegetation mainly was that it lead to the vegetation coverage reducing 2.0% on the west 0 to 1km off the railway, and 4.9% on the east 0 to 0.5km. The vegetation coverage had no changes in other regions.