Growing-season vegetation coverage patterns and driving factors in the China-Myanmar Economic Corridor based on Google Earth Engine and geographic detector

Abstract

Although previous studies have shown that the construction of the China-Myanmar Economic Corridor (CMEC) has caused critical deforestation, few studies have monitored the long-term vegetation dynamics in corridor regions. Recent studies have revealed vegetation greening trends at different scales but have failed to further explore whether these trends are due to the restoration of natural vegetation or the large-scale planting of artificial crops. To investigate the vegetation dynamics and influencing factors during the growing season between 2001 and 2020 in the CMEC area, trend analysis, the random forest (RF) classifier at Google Earth Engine platform, and the geographic detector were employed to quantitatively analyze the patterns of distribution and change in vegetation coverage and its driving factors. Note that we distinguished between natural vegetation areas and artificial crop areas using the RF classification algorithm in an attempt to improve the drawback that most previous studies confused the contribution of natural vegetation and artificial vegetation to macro vegetation change. We discovered that (1) the growing-season average fractional vegetation cover (FVC) was 0.7033, which was relatively low in central and southern regions. The FVC increased by 0.21% yr−1 from 2001 to 2020, and areas with an increasing trend in FVC were 16.67% more than those with a decreasing trend. (2) Although forests and grasslands were reduced by 0.68%, the cropland that prominently clustered in the central and southern regions contributed 50.37% to an increase in FVC, and the decreasing effects on natural vegetation were offset by the increase in artificial crops. (4) The factors that influence FVC distribution were ranked in descending order: land use/land cover (LULC) > climate > topography > anthropogenic activity; factors that influence FVC change were ranked as: anthropogenic activity > climate > LUCC. Note that the grain yield influence on the FVC distribution increased 0.31% yr -1, which was also the most important factor driving FVC change, indicating that agricultural activities increasingly facilitated vegetation greening.