China's large-scale tree planting programs are crucial for achieving carbon neutrality by 2060. This study develops a machine learning framework integrating environmental variables to quantify tree growth suitability and its relationship with tree numbers and biomass carbon stocks. Mapping carbon sink potentials under different scenarios, including one aligned with China's ecosystem management policy, reveals that 44.7 billion trees could be planted, sequestering 5.9 ± 0.5 PgC. Tree densification within existing forests is identified as a particularly effective and economically viable strategy.

Ling Yao, Tang Liu, Jun Qin, Hou Jiang, Lin Yang, Pete Smith, Xi Chen, Chenghu Zhou, Shilong Piao