| The research analyze the MODIS NDVI/EVI vegetation index during2000to2011to represent the recent changes of vegetation resources in the Three-RiversHeadwaters (TRH) region, based on which to discuss how the vegetation resourcevariations of local grassland response to hydrothermal conditions and humanactivities, and also, to find the potential driving factors those influencing the spatialdistribution of the grassland resources in the TRH region.Our results showed that the grassland vegetation index in the TRH regiongenerally increased from2000to2011. In every county (town), the NDVI and EVIaslo significantly increased. Hence, it is clear that the TRH vegetation is on the wayto be gradually recovered. There is a strong spatial heterogeneity of vegetation index,with high values in the eastern and central south regions, but low values in thenorthern and western range. Furthermore, spatial heterogeneity also influences thevegetation index dynamics in the counties.We also analyzed the response mechanisms of temporal and spatial changes ofvegetation index to hydrothermal conditions in the TRH region. There wassignificantly positive correlation between NDVI/EVI and precipitation in all11yearsexcept for2002. Hence, the prominent increase of precipitation was highly beneficialto the growth of vegetation. The temperature in6stations and the precipitation in2stations had significantly correlation with vegetation index, showing that changes ofhydrothermal conditions had some spatial differences on the influence of vegetationrestoration in the region. It should be mentioned that, at the spatial level, thecorrelation between vegetation index and precipitation was closer than those betweenvegetation index and temperature. Contrarily, on the temporal level, the correlationbetween vegetation index and temperature was closer than those between vegetationindex and precipitation.The buffer analysis found that both NDVI and EVI mean decreased from theinner layer to the outer layer. The NDVI of0～5km,5～10km buffer layerssurrounding the residential places significantly increased, so did that of0～1km,1～ 2km,2～3km,3～4km buffer layers along the roads. The nearer to the residentialplaces or the more distant from the roads, the higher increased ratio of vegetationindex. This showed that the herdsmen like to inhabit and make producing activities inthe places where have better vegetation conditions, and also indicated that reasonablegrazing intensity and management measures has played an important role in therestoration of vegetation after the project in the Three-Rivers Headwaters region.The results of the spatial heterogeneity of vegetation index responsing to otherfactors indicated that: vegetation index in the two kinds of grasslands hadsignificantly negative correlations with altitude, wind speed and ultraviolet radiation,while had significantly positive correlations with moisture index and Nitrogen. Therewere no correlations between aspect, solar radiation and vegetation index in the twokinds of grasslands. The slope had positive correlation with vegetation index inutilizable grassland, but not with that in unavailable grassland. The population density,GDP density and road density had significantly positive correlations with vegetationindex in utilizable grassland, but not with that in unavailable grassland. The results ofPrincipal Component Analysis indicated that, the factors which had supremecontribution to grassland vegetation index in the TRH region were altitude, nitrogenand ultraviolet radiation.In a word, there were considerable temporal and spatial heterogeneity ofvegetation index as well as their dynamics in the grassland in the TRH region, andalso, this vegetation index as well as their dynamics could response to variousenvironmental and human activity factors. Finally we conclude that, theenvironmental factors were the main driving factors of temporal and spatial change ofvegetation index in the TRH region, which can provide scientific basis for grasslandresource managing in the future. |
PDF Full Text Request