The History of Atmospheric Oxygen: Regulation by Plants and Fire

Abstract

Since the emergence of land plants ~400 million years ago, atmospheric oxygen has varied remarkably little, remaining dynamically stable between bounds of approximately 17-35% vol. O2. Understanding how and why it has done so has led to a problem termed the ‘oxygen puzzle’. It has long been suggested that fire provides a negative feedback to the regulation of atmospheric O2, through altering the abundance of terrestrial vegetation and therefore organic carbon burial, which is the main oxygen source over long geological timescales. However, varying the strength of this fire feedback in biogeochemical models that predict the abundance of oxygen in the atmosphere, leads to varying predictions that pose questions as to how strong this fire feedback might be. Therefore, despite being a long-standing problem, the oxygen puzzle persists. This thesis updates the LPJ-LMfire Dynamic Global Vegetation Model used to simulate process-based, large-scale representations of terrestrial vegetation dynamics, allowing it to account for the effects of varying atmospheric O2 levels on fire and photorespiration feedbacks on vegetation biomass. This is then used to test the proposed strength of regulatory mechanisms on the abundance of oxygen in the atmosphere over the Phanerozoic. Results from the LPJLM simulations found that forests were able to persist at 35% vol. O2 due to high fuel moistures in low latitudes, contradicting long-standing assumptions, indicating that fire feedbacks on oxygen are weaker than previously considered and that additional mechanisms are required to provide strong regulation. Photorespiration in high oxygen could also influence total terrestrial biomass so by subsequently accounting for increased rates of photorespiration under rising atmospheric O2 levels, it was found that this had a similar magnitude effect to fire. However, by combining fire-photorespiration effects a greater than additive effect on removing global vegetation due to interactions between the two processes was found. This suggests that a combined fire-photorespiration negative feedback would be stronger than considering the effects separately and should be considered in order to provide a more comprehensive understanding of the global impacts of increased atmospheric O2 on vegetation. By including this coupled feedback in a series of biogeochemical models it was found that this reduced extreme predictions in atmospheric O2 variation through geological time, keeping O2 levels below 25% vol. O2. This new lower upper limit must be key to maintaining successful planetary functioning and has likely determined evolutionary trajectories of Earth ecosystems over time.