RESEARCH ON THE VARIATION CHARACTERISTICS OF VEGETATION CHANGE AND ITS TERRAIN INFLUENCE FACTORS IN ALPINE GRASSLAND IN THE SOURCE REGION OF THE YELLOW RIVER

The source region of the Yellow River is an important ecological barrier in China. Since the 1990s, under the influence of various factors, its runoff has decreased and the ecological environment has deteriorated, which has attracted great attention of governments. Vegetation is the general term of plant communities covering the surface, which is the natural link between the atmosphere, water and soil. It plays an important role in air regulation, soil conservation and maintaining the stability of the whole ecosystem. Therefore, the study on the differential characteristics and topographic influencing factors of alpine grassland change in the source region of the Yellow River can provide reference for the ecological environment protection in -this region in the future. Based on the remote sensing data of MODIS NDVI, we studied the spatial distribution pattern and temporal variation characteristics of vegetation coverage in the source region of the Yellow River from 2000 to 2009 by means of maximum synthesis method, pixel dichotomy model method and trend analysis method. Furthermore, combined with DEM topographic elevation data, we discussed the influence of topographic factors such as elevation, slope and aspect on fractional vegetation cover (FVC) change. The results showed that: (1) From 2000 to 2009, the FVC in the source region of the Yellow River fluctuated upward, showing the spatial distribution characteristics of the southeast region > the central region > the northwest region; (2) During the ten years from 2000 to 2009, the almost constant of FVC accounted for the largest area, and the areas with increasing FVC mainly concentrated in the northeast and northwest of the studied region, the areas with reduced vegetation coverage were scattered in various places in the studied region, mainly in the form of minimal patches or spots; (3) Different terrain factors had different effects on FVC and the effect of altitude on FVC changes was significantly greater than that of slope and aspect. With the increase of altitude, the proportion of significant reduction in FVC increased firstly and then decreased, while the proportion of significant increase in FVC was decreasing and the proportion of almost unchanged in FVC was increasing; With the increase of slope, the proportion in the above three cases (significant reduction, significant increase and almost unchanged in FVC) was increased, decreased gradually and fluctuated respectively. Under different aspect, the proportion in the above three cases varied little, fluctuating only within 1%. * Corresponding author

sensing data of MODIS NDVI, we studied the spatial distribution pattern and temporal variation characteristics of vegetation coverage in the source region of the Yellow River from 2000 to 2009 by means of maximum synthesis method, pixel dichotomy model method and trend analysis method. Furthermore, combined with DEM topographic elevation data, we discussed the influence of topographic factors such as elevation, slope and aspect on fractional vegetation cover (FVC) change. The results showed that: (1) From 2000 to 2009, the FVC in the source region of the Yellow River fluctuated upward, showing the spatial distribution characteristics of the southeast region > the central region > the northwest region; (2) During the ten years from 2000 to 2009, the almost constant of FVC accounted for the largest area, and the areas with increasing FVC mainly concentrated in the northeast and northwest of the studied region, the areas with reduced vegetation coverage were scattered in various places in the studied region, mainly in the form of minimal patches or spots; (3) Different terrain factors had different effects on FVC and the effect of altitude on FVC changes was significantly greater than that of slope and aspect. With the increase of altitude, the proportion of significant reduction in FVC increased firstly and then decreased, while the proportion of significant increase in FVC was decreasing and the proportion of almost unchanged in FVC was increasing; With the increase of slope, the proportion in the above three cases (significant reduction, significant increase and almost unchanged in FVC) was increased, decreased gradually and fluctuated respectively. Under different aspect, the proportion in the above three cases varied little, fluctuating only within 1%. ecological environment and has an important impact on the global energy cycle and the biochemical cycle of matter (Jiang et al, 2015). In recent years, the environmental conditions of alpine grassland vegetation have been seriously destroyed by the influence of climate and human activities. The grassland in the source region of the Yellow River has been seriously degraded, the water conservation capacity has been reduced, and many times the flow has been cut off at the source, which seriously threatens the sustainable development of economy and society in the source region of the Yellow River and its surrounding regions, and accelerates the decline process of the alpine grassland.
In recent years, many scholars have done a lot of research on the relationship between vegetation coverage and terrain factors, but the results will vary with different research areas, different data, different research methods and different time scales (Cai et al, 2014;Xiong et al, 2018;Chen et al, 2006

Data preprocessing:
(1) Due to the large amount of data in this study, the data is processed in batch processing. First, the MRT toolbox is used to perform format conversion, projection conversion and resample, which makes the data standard 250m, the format was converted to Geotiff, and the original projection was converted to Albers equal area (Chen et al, 2015). In ENVI 5.3, the administrative vector boundaries of the source region of the Yellow River are used to tailor the data required for the study area, and the maximum synthesis method was used to process the image to reduce the impact of clouds and atmosphere.
(2) The ArcGIS is used to perform mosaic and reprojection, the slope, aspect are extracted by the surface analysis function of Spatial Analysis Tools, and then reclassify the slope and aspect, the DEM is classified according to elevation, obtain elevation, slope and aspect grade distribution map of the study area.         thus affect the growth and development of vegetation (Li et al, 2016). In order to studied the relationship between FVC and altitude in the source region of the Yellow River, the altitude gradient was divided into five grades: 2564~3600m, 3600~4000m, 4000~4300m, 4300~4500m and 4500~6295m, and the change of FVC under each altitude gradient was obtained.

Referring to the《Classification Standard of Soil Erosion》(SL
From Table 2, it can be seen that the proportion of high altitude areas larger than 3600m in the source region of the Yellow River reaches 84.51%, accounting for the vast majority of the whole basin. The change of vegetation coverage (P < 0.01) will be different with the change of altitude. When the altitude is less than 4300m, the proportion of significant reduction and almost unchanged in FVC gradually increased with the increase of altitude. On the contrary, the proportion of significant increase gradually decreased with the increase of altitude.
When the altitude is more than 4300m, the proportion of From Table 3, it can be seen that the slope of the studied region is the largest in the range of 5°~20°, which accounting for 58.23% of the total area of the studied region. The central part of the studied region is mainly steep slope and dangerous slope.
As shown in Table 3, FVC shows different trends with the increase of slope (P < 0.01), the proportion in the three cases (significant reduction, significant increase and almost unchanged in FVC)are increased, decreased gradually and fluctuated around 80% respectively.

Effect of aspect on vegetation coverage change：Aspect
indicates the change direction of each grid elevation change, the solar radiation and soil moisture received from different aspect are different, thus affects the distribution characteristics of vegetation (Wang et al, 2018). In this paper, the aspect is As shown in Table 4, under different aspect, the proportion in the above three cases varied little, fluctuating only within 1% (only the other 8 levels except the flat were analyzed). The percentage of significant increase in FVC in the southwest was the highest, followed by the south and the west, and the smallest was the north; the highest percentage of significant reduction in FVC was the north, followed by the northeast and the northwest, and the smallest was the south; the highest percentage of almost unchanged in FVC was the north, and the smallest was the southwest. Overall, the proportion of vegetation coverage increase on sunny slope is larger than that on shady slope, and the proportion of vegetation coverage decrease is smaller than that on shady slope, because the solar radiation energy received on sunny slope and shady slope is different.