[Spatiotemporal Dynamics and Driving Factors of Watershed Carbon Storage Based on Remote Sensing Data and Plot Survey...

To investigate the spatiotemporal variations of carbon storage and their driving mechanisms at the watershed scale， this study employs the Yanggong River Basin as a case study. Combined with the results of plot surveys and remote sensing data， we comprehensively applied the PLUS-InVEST-Geodetector model framework to analyze the spatiotemporal evolution characteristics of land use and carbon storage from 1980 to 2020， simulate carbon storage changes under multiple scenarios in 2030 and 2060， and identify driving factors of spatial heterogeneity in carbon storage. The key findings are as follows： ① Forestland remained the dominant land use type throughout 1980-2020. However， significant land use transitions occurred， characterized by the continuous contraction of cropland area， near-doubling of construction land expansion， and gradual stabilization of grassland and shrubland areas. ② Carbon storage exhibited an "initial increase followed by decrease" trend， peaking at 35.53×10⁶ t in 2010 before slightly declining to 35.26×10⁶ t in 2020 due to construction land expansion. Spatially， carbon storage demonstrated relative stability， maintaining a distinct pattern of higher values in peripheral forest/shrubland regions with dense vegetation coverage and lower values in central urbanized areas. ③ Scenario simulations revealed that under natural development， carbon storage would decrease to 35.02×10⁶ t by 2030 before marginally recovering to 35.16×10⁶ t by 2060. Ecological protection scenarios implementing farmland-to-forest conversion and ecological restoration measures could elevate carbon storage to 35.28×10⁶ t （2030） and 35.57×10⁶ t （2060）. Although cropland protection scenarios maintained both cropland area and carbon storage， overall levels remained comparable to those in natural development scenarios. ④ Geodetector analysis identified soil type （q=0.464）， population density （q=0.374）， and NDVI （q=0.361） as dominant factors governing carbon storage spatial differentiation. Factor interactions demonstrated both two-factor and nonlinear enhancement effects， with particularly strong synergistic driving forces between soil type and GDP/NDVI/population density. These findings provide critical data support for optimizing land use management and enhancing terrestrial ecosystem carbon sequestration in the Yanggong River Basin.