植被覆盖度是土地生态的重要指示因子，黄河流域横跨中国地形三大阶梯，是国家重要的生态屏障，同时也是重要的经济地带和“能源流域”。为揭示长时序黄河流域及其煤炭富集地区土地生态变化状况，基于Google Earth Engine(GEE)平台，对1987—2020年黄河流域共40 525景Landsat TM/ETM+/OLI遥感影像进行批量去云、融合和NDVI云计算等处理，获取34 a的植被覆盖度数据。综合利用最大值合成法、像元二分模型、一元线性回归趋势性分析和F检验等方法对黄河流域及流域内煤炭国家规划矿区植被覆盖度的时空变化特征进行定量分析;在此基础上逐一识别地形因子和气候因子对黄河流域及其规划矿区植被覆盖度的影响。结果表明:① 34 a间黄河流域的平均植被覆盖度由1987年的0.457 4上升至2020年的0.581 7，同期流域内煤炭国家规划矿区则由0.355 6增至0.536 1，2者呈现一致的波动上升的趋势;② 时序趋势变化类型构成中，黄河流域植被覆盖度改善的面积(33.19%)远大于植被覆盖度退化的面积(3.55%)。规划矿区内植被覆盖度改善面积占比高于黄河流域，但其植被覆盖度仍明显低于黄河流域平均水平;③ 黄河流域多年平均植被覆盖度空间差异明显，除宁夏平原、河套平原等地外，主要呈现南高北低、由东南向西北递减的趋势;④ 黄河流域及规划矿区地形因子对植被覆盖度的影响表现为高程>坡度>坡向，气候因子的影响表现为气温>降水量。降水量和气温对于规划矿区的植被覆盖度的影响均小于对整个黄河流域的影响。土地利用类型变化促进了黄河流域植被覆盖变化趋势类型的多向性;⑤ 44个煤炭规划矿区中，彬长矿区和乡宁矿区植被覆盖度改善相对更为显著，石炭井矿区和包头矿区退化面积占比相对高。研究结论:应用长时序、多源数据能够客观揭示黄河流域及规划矿区植被覆盖度时序和空间变化异同特征，为科学认识与评价整个流域及规划矿区生态状况、制定生态保护修复政策等提供数据支撑，今后针对植被覆盖度时空差异的形成机理仍需进一步深入分析。

Vegetation coverage is an important indicator for land ecological status.The Yellow River Basin,rich in coal resources and as the important economic development zone,spans China’s topography from the west to the east and across nine provinces.In order to reveal the land ecological status of both Yellow River Basin and its State Planned Coal-mining Areas (SPCA),the authors obtained the vegetation cover data of 34 years based on the platform of Google Earth Engine (GEE) by processing 40525 Landsat TM/ETM+/OLI remote sensing images of the Yellow River Basin from 1987 to 2020 with procedures including batch cloud removal,fusion and NDVI cloud computing.The spatial and temporal variation characteristics of vegetation coverage in both Yellow River Basin and its SPCA are quantitatively analyzed using the methods of maximum synthesis,pixel dichotomy model,trend analysis of linear regression and F-test,etc.Then the topographic factors and climatic factors are identified one by one in terms of the impact on vegetation cover in the Yellow River basin and its SPCA.The study results show as follows:firstly,during the 34 years period,the average vegetation coverage of the Yellow River Basin presents a fluctuating upward trend from 0.457 4 in 1986 to 0.581 7 in 2020,and the average vegetation coverage within its SPCA increased from 0.355 6 to 0.536 1,where they show a consistent increase fluctuation trend.Secondly,the improved area of vegetation coverage,about 33.19%,is much larger than the degraded area,about 3.55% of Yellow River Basin,by the transformation analysis of vegetation coverage change tendency categories in the study area.Though the proportion of improved area of vegetation coverage in the SPCA is higher than that in the Yellow River Basin,but its vegetation coverage in the SPCA is still significantly lower than the average level of the Yellow River Basin.Thirdly,there are significant spatial differences in the mean FVC over the years.Except for Ningxia Plain,Hetao Plain,et al.,the FVC is comparatively higher in the south and lower in the north,and is decreased from southeast to northwest.Fourthly,those terrain factors’influence on vegetation coverage shows in rank as "elevation > slope > aspect" in both Yellow River Basin and its SPCA.The effect of temperature on vegetation coverage is bigger than that of precipitation.The change of land use type promotes the multidirectional trend of vegetation cover change in the study area.Fifthly,among those 44 SPCA,the improvement of vegetation coverage in Binchang mining area and Xiangning mining area is comparatively more significant,and the degradation of vegetation coverage is comparatively evident in Shitanjing mining area and Baotou mining area.At last,the conclusion is drawn that the application of long sequence,multi-source data can reveal the differences and similarities of temporal and spatial change characteristics on vegetation coverage between the Yellow River basin and its SPCA,which is the basis for the scientific understanding,evaluation of their ecological conditions and providing the data support for policy-making on ecological protection and restoration,while the formation mechanism of the temporal and spatial differences of vegetation coverage remains for further in-depth analysis in the future research.

1 研究区概况 

2 数据来源与预处理

3 研究方法

   3.1 植被指数空间数据提取

   3.2 植被覆盖度计算

   3.3 植被覆盖度时序变化特征分析

   3.4 植被覆盖度变化趋势分析

   3.5 植被覆盖度的空间差异

4 结果与分析

   4.1 植被覆盖度时序和空间差异总体特征

   4.2 植被覆盖度随地形因子的变化特征

   4.3 植被覆盖度随气候因子的变化特征

   4.4 植被覆盖度与土地利用变化分析

5 结论与讨论