1982—2015年中国植被NDVI时空变化特征及其驱动分析

在全球气候变化背景下，多数研究关注植被年际变化趋势及其对降水和气温的响应而忽略了辐射因子的影响。探究持续性和波动变化过程的空间分异特征和定量揭示不同植被类型对季节性多气候要素（降水、气温和辐射）响应规律需进一步深入，并识别植被退化人类活动影响区对中国生态环境保护具有重要意义。该研究基于AVHRR NDVI3g遥感数据、CCI全球土地覆被数据和ERA5-Land数据，采用多元线性回归模型和残差分析等方法，从区域和全国尺度上分析1982-2015年中国植被NDVI（Normalized Difference Vegetation Index）时空变化特征，研究其对季节性气候变化和人类活动的响应。研究结果表明：1）中国的植被变化在空间上具有明显的区域特征，呈现出自东南向西北递减的趋势，变化率范围为-0.016～0.029/a。2）根据Hurst指数分析，80.62%的植被处于持续稳定变化状态。草地和林地的变化趋势趋于不稳定，易受气候变化和人类活动的影响。3）准噶尔盆地、青藏高原以及内蒙古锡林郭勒盟等区域的植被NDVI与气温和太阳辐射呈负相关性，而在云贵高原、黄土高原南部、四川盆地等地区表现为正相关性。4）残差分析结果表明，干旱导致新疆准噶尔盆地和内蒙古中部等干旱区植被退化，而温度和辐射增加是促进东部沿海平原、黄土高原南部、四川盆地和云贵高原西南部等地区植被改善的主要原因。21世纪以来人类活动逐渐加强，包括黄土高原水土保持和退耕还林工程、阿拉善荒漠治理、沿海城市群和东北工业基地的建设等。具体表现为内蒙古阿拉善高原、黄土高原中西部和北部以及华南大部分地区植被改善，东北大、小兴安岭、长白山东部和长江三角洲地区植被退化。该研究对植被变化及其驱动因素的研究，可识别植被退化区，为将来生态修复工程的实施和生态文明建设提供理论依据，助力区域绿色可持续发展。

Spatiotemporal changes of vegetation NDVI and its driving forces in China during 1982-2015

Abstract: Both human activities and climate change have posed great impacts on the vegetation changes in the context of global climate change. It is necessary to identify the areas of man-made vegetation degradation for better environmental protection. However, only limited effort has been focused on distinguishing between the influences of yearly climatic factors and human activities on vegetation growth. Particularly, it is still lacking in the vegetation dynamics under the seasonal climate changes and anthropogenic activities in China. In this study, the spatiotemporal variations in the Normalized Difference Vegetation Index (NDVI) and climate change were investigated using the AVHRR NDVI3g data, land use data and gridded climate data, A multiple linear regression model was then used to determine the driving factors of vegetation change using residual analysis at the regional and national scales from 1982 to 2015. The results show that there was outstanding regional spatial characteristic of vegetation changes with a decreasing trend from the southeast to northwest China, where the annual rate of change ranged from -0.016/a to 0.029/a. An increasing trend was found in the temperature and solar radiation, whereas, the precipitation showed a downward trend in most regions. According to the Hurst index, 80.62% of the vegetation presented a continuous and stable change, while the grasslands and woodland tended to be unstable and vulnerable to climate change and human activities. The NDVIs of Junggar Basin, Qinghai-Tibet Plateau, and Xilin Gol League of Inner Mongolia were negatively correlated with the air temperature and solar radiation, whereas, the NDVIs of Yunnan-Guizhou Plateau, southern Loess Plateau, and Sichuan Basin were positively correlated. The residual analysis showed that the drought resulted in the vegetation degradation in the arid areas, such as Xinjiang Junggar Basin and central Inner Mongolia, while the increase of temperature and radiation was the main driver for the vegetation improvement in the eastern coastal plain, the southern Loess Plateau, Sichuan Basin, and the southwestern Yunnan-Guizhou Plateau. Since the beginning of the 21st century, human activities were gradually strengthened, including soil and water conservation, the conversion of farmland to forest on the Loess Plateau, the desert control in Alxa, as well as the construction of coastal city clusters and northeast industrial bases. Thus, the vegetation improvement was identified in the Alxa Plateau of Inner Mongolia, the central and northern Loess Plateau, and most parts of South China, whereas, the vegetation degradation was detected in the Greater and Lesser Khingan Mountains, the eastern regions of the Changbai Mountains and the Yangtze River Delta. Among them, the national projects of ecological protection were attributed to the significant improvement of vegetation conditions. Consequently, the monitoring of vegetation degradation can provide a theoretical basis for the implementation of ecological restoration and the construction of ecological civilization in the future. The finding can greatly contribute to regional green and sustainable development.