This dataset covers above-ground and below-ground vegetation carbon stocks across China from 2002 to 2021. By integrating high-resolution active microwave, passive microwave, and optical remote sensing data with ground plot measurements, regression and machine learning algorithms were employed to generate spatiotemporally continuous data with a spatial resolution of 1/120 degree (approximately 1 km) and annual temporal resolution. Data are stored in TIF format. Above-ground carbon stock files are named “AGBCY + year,” while below-ground carbon stock files are named “BGBCY + year.” Compared with existing datasets, this dataset more accurately depicts the spatial distribution and interannual variation of China's forest carbon stocks, supporting quantitative research on the impacts of climate change and ecological restoration on carbon storage.