Fig. 2: The constrained global gross primary productivity (GPP) and ecosystem respiration (ER).
From: Vegetation biogeography is a main source of uncertainty in modelling the land carbon cycle
a The comparison of global GPP estimates from dynamic global vegetation models (DGVMs) from TRENDY v9 project, the constrained GPP using remote sensing-based PFT maps (DGVMRS), and the GPP estimates from other approaches, including upscaled eddy covariance fluxes using machine learning20,21 (MLEC), solar-induced fluorescence19,80,81 (SIF), soil respiration39 (SR) and plant carbonyl sulfide40 (OCS). The MLEC includes the RS_METEO output of FLUXCOM – the ensemble estimates upscaled using different machine learning methods and different meteorological forcings20, as well as the X-BASE product from the FLUXCOM-X21. The solid lines represent the mean and the error bars represent one standard deviation of the GPP estimates. b The comparison of global ER estimates from DGVMs, DGVMRS, and MLEC. c The relationship between global ER and GPP across DGVMs. d The total GPP changes driven by elevated CO2 (CO2_total), the total GPP changes driven by CO2-driven PFT changes (CO2_PFT), the total GPP changes driven by climate change (CLI_total) and the total GPP changes driven by climate change-driven PFT changes (CLI_PFT).
