基于遥感计算云平台高原山区植被覆盖时空演变研究——以贵州省为例

为揭示喀斯特山区植被时空变化规律，选取2000-2018年间1 748景30 m分辨率Landsat-NDVI影像，结合35个气象站点数据，辅以像元二分模型、线性趋势分析及地理探测器等方法，对贵州省19年间年植被覆盖度进行定量估算，分析其植被覆盖度时空变化特征及驱动因素。结果表明:（1）贵州省中、高植被覆盖度以上的区域面积占比约63%，其中高植被覆盖度区域面积占21.16%，主要集中分布于碎屑岩地区。（2）近19年来，贵州省植被覆盖度总体缓慢趋好，年均增长速率为0.4%，严重石漠化样区多年最大植被覆盖度均值始终低于整体植被覆盖度均值。（3）研究期间贵州省植被覆盖度以轻微改善、基本不变两个等级为主，两者面积比重之和约为95.4%，退化区域主要分布在城镇周边，面积比重约为3.8%。（4）气象因素、地理区位各因子间交互作用对植被覆盖度空间格局影响大于单因子作用。综上所述，城镇面积扩展、石漠化治理工程、地理区位及气象因素等是影响植被恢复与生态环境重建的关键要素，研究植被覆盖度多年动态特征力求为相关部门的水土保持、生态环境保护及石漠化治理提供重要的基础数据及科学参考。 

Spatiotemporal variation of vegetation coverage in plateau mountainous areas based on remote sensing cloud computing platform: A case study of Guizhou Province

The purpose of this work is to reveal the spatio-temporal variation of vegetation in karst mountainous areas. We choose the Landsat-NDVI images of 1748 scenes with 30 m resolution during 2000-2018, combined with data from 35 meteorological stations and the pixel dichotomy model, linear trend analysis and geographic detectors to quantitatively estimate the annual vegetation coverage in Guizhou Province in the past 19 years. The spatial and temporal variation and driving forces of vegetation coverage were analyzed. The results show that，(1) the areas of medium-high vegetation coverage account for 63% and the area of high vegetation coverage accounts for 21.16%, which is mainly concentrated in the clastic areas. On the whole, the amount of vegetation coverage is of such an order: clastic rock > dolomite > limestone. (2) In the past 19 years, the vegetation coverage in Guizhou Province has shown slight improvement, with an average annual growth rate of 0.4%. The multi-year average value of maximum vegetation coverage in severe rocky desertification areas is constantly lower than the general vegetation coverage. (3) During the study period, the overall change of vegetation coverage in Guizhou Province was stable, dominated by slight improvement and basically no-change, the sum of which accounts for about 95.4% of the total area. The degraded areas are mainly distributed in periphery of cities and towns, accounting for about 3.8% of the total area. (4) The interaction between meteorological and geographical factors has greater impact on the spatial pattern of vegetation coverage than that of single factor. In summary, the key factors affecting vegetation restoration and ecological environment reconstruction include urban area expansion, rocky desertification control project, geographical location and meteorological factors. The study of multi-year dynamic characteristics of vegetation coverage aims to provide important basic data and scientific reference for water and soil conservation, ecological environment protection decision-making, ecological restoration (rocky desertification control) and sustainable management of relevant departments.