Emergent constraints on historical and future global gross primary productivity

Terrestrial gross primary productivity (GPP) is the largest carbon flux in the global carbon cycle and plays a crucial role in terrestrial carbon sequestration. However, historical and future global GPP estimates still vary markedly. In this study, we reduced uncertainties in global GPP estimates by employing an innovative emergent constraint method on remote sensing-based GPP datasets (RS-GPP), using ground-based estimates of GPP from flux towers as the observational constraint. Using this approach, the global GPP in 2001-2014 was estimated to be 126.8 ± 6.4 PgC year^-1, compared to the original RS-GPP ensemble mean of 120.9 ± 10.6 PgC year^-1, which reduced the uncertainty range by 39.6%. Independent space- and time-based (different latitudinal zones, different vegetation types, and individual year) constraints further confirmed the robustness of the global GPP estimate. Building on these insights, we extended our constraints to project global GPP estimates in 2081-2100 under various Shared Socioeconomic Pathway (SSP) scenarios: SSP126 (140.6 ± 9.3 PgC year^-1), SSP245 (153.5 ± 13.4 PgC year^-1), SSP370 (170.7 ± 16.9 PgC year^-1), and SSP585 (194.1 ± 23.2 PgC year^-1). These findings have important implications for understanding and projecting climate change, helping to develop more effective climate policies and carbon reduction strategies.