Changes in both vegetation type and productivity can affect the structure, functioning, and services of terrestrial ecosystems. Understanding vegetation dynamics and their drivers is critical for biodiversity conservation and ecosystem management. Although land cover datasets, climate-vegetation models, and remote sensing vegetation indices have been frequently used to reflect vegetation dynamics, they generally lack biological information about vegetation, such as species compositions, community structures, and succession status. Yunnan, recognized as the most biodiverse province in China, has undergone considerable vegetation changes over recent decades. However, the roles of climate change and human activities remain unclear. This study integrated detailed vegetation maps, climate factors, and vegetation indices obtained in the 1980s (1986–1995) and 2010s (2006–2015) to comprehensively evaluate the coverage transformations and productivity changes of vegetation types in Yunnan and unravel the drivers of decadal vegetation changes. The results indicated that: 1) A greening trend was observed across all vegetation types, particularly in coniferous and temperate forests. 2) The decadal vegetation changes were dominated by: the restoration of savanna and shrubland, cropland expansion, and artificial afforestation, accounting for 23.7 %, 22.9 %, and 19.1 % of all observed changes, respectively. 3) Conservation and restoration efforts dominated vegetation greening in Yunnan (55 %), followed by artificial afforestation (23 %), and agricultural expansion (16 %). By comparing vegetation maps of separated stages, the inclusion of vegetation class information provided critical knowledge for a better understanding of the intrinsic mechanisms behind vegetation greening and range shifting. Our study highlighted the significant role of ecological conservation and restoration policies and practices in influencing the spatiotemporal dynamics of biodiversity and ecosystem functioning at a regional scale, within merely a few decades.
Read full abstract