Dynamics of aboveground biomass over the past four decades in China’s forests

As one of the most important carbon sinks and carbon pools globally, forest ecosystems play a critical role in absorbing and storing carbon dioxide from the atmosphere. The implementation of eco-engineering has transformed China into one of the hotspots in global greening. However, great differences in natural conditions between subregions, function transformation of forest ecosystem between carbon gain and carbon loss due to rapid land use changes, and uncertainties in the stability and sustainability of forest ecosystem functions resulting from climate change, lead to the large-scale forest carbon sink capacity and future carbon sink potential remaining largely unclear. This lack of clarity is not conducive to the formulation of climate change mitigation strategies in China. Therefore, it’s urgent to undertake the quantification and assessment of forest aboveground biomass carbon with high spatiotemporal resolution. Here, using a machine learning model, six bands of Landsat images, along with 3 indicators derived from the bands by raster calculators at a resolution of 30m × 30m, were used to train the estimate model of aboveground biomass carbon density in combination with the adjusted aboveground biomass / carbon products from 2019. Subsequently, the carbon density from 1985 to 2023 at a 30m × 30m resolution were predicted. The model’s RMSE was 9.03 MgC ha^-1 and the R² of test datasets stabilized around 0.77. We found that forest aboveground biomass carbon stock decreased first and then increased during the period. Despite a decreasing trend in the area of stable forests, the carbon stock increased from 7.50 PgC to 8.05 PgC, at a rate of 0.015 PgC yr^-1. The area of secondary forests, however, showed the most rapid regrowth in carbon density during the period, with a rate of 0.46 MgC ha^-1 yr^-1. Over the past about four decades, carbon loss due to deforestation amounted to approximately 1.49 PgC, while carbon gain from plantation sumed to 4.55 PgC. Spatial and temporal high-resolution data of forest aboveground biomass carbon serve as an invaluable resource for identifying areas with significant carbon stocks and high carbon sink potential, and allows an in-depth understanding of the differences in dynamic patterns over time in China’s forest and provides a scientific reference for optimizing land management.