青藏高原植被指数最新变化特征及其与气候因子的关系

The Latest Change in the Qinghai-Tibetan Plateau Vegetation Index and Its Relationship with Climate Factors

利用GIMMS/NDVI(全球库存模拟和影像研究/归一化植被指数,Global Inventory Modeling and Mapping Studies,Normalized Difference Vegetation Index)和MODIS/NDVI遥感数据以及青藏高原6个气象代表站的站点数据,结合多种统计和计算方法,分析了青藏高原植被NDVI变化规律及其影响因子。结果表明:1982~2013年青藏高原多年平均植被NDVI的空间分布存在明显的区域差异,总体上呈从东南向西北递减的趋势,而且发现不同地区植被的时间变化规律也不尽相同。根据高原长势最好的6~9月植被NDVI进行经验正交分解,将青藏高原植被分为5个区,并进一步分析了不同分区内植被的变化规律,得出:青藏高原植被NDVI下降最明显的区域在二区的噶尔班公宽谷湖盆地地区和北羌塘高原地区,植被NDVI上升最明显的区域在四区的祁连山东部地区。为了探讨青藏高原不同分区内影响植被NDVI下降的因子,从青藏高原二区、四区、五区各选取NDVI处于下降趋势的两个代表站点。研究分析了各个站点植被NDVI与降水量、平均气温、平均最低气温、平均最高气温、日照百分率5个气象因子的关系,得出:在高原二区日照强度是其它分区的两倍左右,而降水量相对较少导致植被NDVI降低。高原四区由于降水量小、温度高、日照强,导致植被NDVI处于下降趋势;在青藏高原五区虽然降水充足,但日照较弱,限制了植被的正常成长导致NDVI处于下降趋势中;其结果为高原植被退化机制研究及高原植被对大气反馈等奠定了基础。

Based on the remote sensing data of GIMMS（Global Inventory Modeling and Mapping Studies） NDVI （Normalized Difference Vegetation Index） and MODIS （Moderate-resolution Imaging Spectroradiometer） NDVI, as well as data collected at six meteorological stations in the Qinghai-Tibetan Plateau, the authors discussed the spatio-temporal variation of the vegetation in the Qinghai-Tibetan Plateau and its influencing factors using statistical and calculation methods. The results show some obvious differences in the spatial distribution of annual mean Qinghai-Tibetan Plateau vegetation NDVI from 1982 to 2013; the NDVI over all decreased from southeast to northwest. In addition, temporal variations of the vegetation index in different areas were not exactly the same. Using EOF （Empirical Orthogonal Function） analysis, the authors analyzed the regular changing patterns of vegetation coverage over five subregions, which were divided based on NDVI during the growing season from June to September. The results indicate that the largest decrease in NDVI over the Qinghai-Tibetan Plateau occurred in the area of Gar-Lake Bangong Gully and the northern Qiangtang Plateau, while increases in NDVI over the Qinghai-Tibetan Plateau was found in the eastern Qilian Mountain. In order to explore the factors affecting the decline of NDVI in different subregions, two representative stations were selected in subregions II, IV, and V of the Qinghai-Tibet Plateau, where the NDVI showed a declining trend. The authors also discussed the relationship between NDVI and total precipitation, average temperature, mean maximum temperature, and percentage of sunshine duration at each station. Preliminary results show that in subregion II of the Qinghai-Tibet Plateau, the sunshine duration was double that in other subregions while precipitation was relatively low, resulting in vegetation degradation. The vegetation degradation in subregion Ⅳ was attributed to limited precipitation, high temperature, and strong sunshine. In subregion V, precipitation was sufficient but sunshine was weak, which also caused the vegetation degradation. The above results lay a profound foundation for the mechanism study of vegetation degradation in the Qinghai-Tibetan Plateau and its feedback to the atmosphere.