Response of vegetation cover to CO2 and climate changes between Last Glacial Maximum and pre-industrial period in a dynamic global vegetation model

Climate and atmospheric CO2 strongly influence the vegetation distribution and the terrestrial carbon storage. Process-based dynamic global vegetation models (DGVM) are important tools for simulating past vegetation dynamics and carbon cycle; yet the link between spatial gradients of climate and vegetation cover in geological past has received less attention. In this study, we simulate the distribution of vegetation under three CO2 levels for two climate states, the Last Glacial Maximum (LGM) and Pre-industrial (PI) climate with fire activated or deactivated using the ORCHIDEE-MICT DGVM. Results show that elevated CO2 and warmer climate promote global total tree cover but the impacts are different between forest biomes. Regional tree cover is highly regulated by mean annual precipitation (MAP) especially in the tropics, and by temperature for the boreal-arctic tree line. Based on quantile nonlinear regressions, we analyze the MAP threshold at which maximum tree cover is reached. This threshold is significantly reduced with elevated CO2 for tropical and temperate trees. With higher CO2, increased tree cover leads to reduced fire ignition and burned area, and provides a positive feedback to tree cover, especially in Africa. Besides, in our model, increasing CO2-induced enhancement of gross primary productivity (GPP) is more prominent for tropical trees than for temperate and boreal trees, and for dry regions than wet regions. This difference explains why CO2 is the major factor influencing forest cover in the tropics. It also highlights that special attention should be paid to collect paleo-vegetation data across savannas-forest transition in dry regions. (C) 2019 Elsevier Ltd. All rights reserved.