Abstract:

As a critical frontline in China's battle against desertification， the Horqin Sandy Land's vegetation dynamics serve as a vital indicator of the efficacy of regional ecological governance. Leveraging the Google Earth Engine （GEE） platform， this study retrieved a long-term time-series dataset of fractional vegetation cover （FVC） during the growing seasons from 2000 to 2023. By integrating Sen-MK trend analysis， the coefficient of variation， and the Hurst exponent， we systematically investigated the spatiotemporal evolution， stability， and future trajectory of FVC. Furthermore， partial correlation analysis and a land use transition matrix， complemented by redundancy analysis （RDA）， were employed to quantitatively disentangle the driving forces of climate and human activities. Our key findings reveal： （1） There was a significant upward trend in growing-season FVC， with an average annual growth rate of 0.43%. The FVC structure improved substantially， as the proportion of extremely high coverage soared from 17.45% to 49.78%. Spatially， FVC presents a mosaic pattern of widespread high-coverage areas interspersed with patches of low coverage， showing no clear directional differentiation. （2） Vegetation change was stable in 47.89% of the area， while unstable changes， concentrated in patchy zones within the western region， characterized 44.13%. The Hurst exponent suggests strong persistence in future trends， with 90.13% of the region expected to maintain its current trajectory 63.28%，exhibiting extremely strong persistence （H>0.7）， indicating that the ongoing improvement is likely sustainable. （3） Precipitation emerged as the dominant climatic factor， positively correlating with FVC in 74.05% of the study area （12.46% showing significant positive correlation， P<0.05）. Human activities exerted bidirectional impacts： land conversions from bare land to grassland/forest land yielded substantial ecological benefits， while transitions from farmland to construction land and grassland to bare land negatively affected local vegetation. Notably，RDA quantitatively identified human activities as the core driver shaping the spatial pattern of vegetation， accounting for 67.44% of the observed changes. This study provides crucial scientific evidence and a methodological framework for guiding ecological governance in the Horqin Sandy Land and for monitoring vegetation dynamics in arid and semi-arid regions globally.

Key words: Horqin Sandy Land, vegetation coverage, temporal and spatial variation, drivers