Bistability of atmospheric oxygen, the Great Oxidation and climate
The Neoarchean and Palaeoproterozoic Eras saw fundamental changes to the Earth System. Oxygenic photosynthesis evolved by ~2.7 Ga, the transition from a reducing to an oxidising atmosphere, the "Great Oxidation", was at ~2.4 Ga and the first low-latitude, "Snowball Earth", glaciations were between ~2.4 and ~2.2 Ga. A box model of the global redox system is developed, leading to the identification of a bistability in atmospheric oxygen caused by positive feedback on methane oxidation rates with the formation of the ozone layer. The existence of a low oxygen stable steady state naturally explains the time lag between the origin of oxygenic photosynthesis and the Great Oxidation. The Great Oxidation can be understood as a switch between low and high oxygen states. Applied to the remote sensing of inhabited extrasolar planets, given proposed instrument sen- sitivities, the existence of the low oxygen state presents a possible false negative for extant oxygenic photosynthesis. A new radiative-convective climate model is developed and applied to climate in this period. The paradigm that a strong methane greenhouse was responsible for a warm Neoarchean despite the "Faint Young Sun" is rejected, as is the suggestion that the Great Oxidation caused a collapse of this methane greenhouse, triggering the Palaeoproterozoic glaciations. The advent of photochemical shielding by ozone at the Great Oxidation increases the lifetime of methane and other labile greenhouse gases, causing warming, and is more likely to have assisted recovery from glaciation. Two new hypotheses are suggested: first, that if the atmospheric nitrogen inventory was larger in the Archean, this may have been sufficient to maintain warm conditions by pressure broadening of absorption lines and an increased lapse rate; second, that the Palaeoprotero- zoic and Neoproterozoic Snowball Earth glaciations were caused by hiatuses in volcanism which accompanied low continent formation rates. In developing the radiative-convective model, it was found that the Hadley Centre radiative trans- fer codes perform poorly. This was investigated in detail through comparisons between narrow and broadband versions of the Hadley Centre code and with a line-by-line code. The broadband codes used in the Hadley Centre Unified Model (GCM) underestimate the radiative forcing from carbon dioxide doubling by ~15% at the top of the atmosphere and 23 - 41% at the surface. The radiative forcing of increased water vapour is overestimated. This is likely to have a detrimental effect on results of the Unified Model.
PhD Thesis, University of East Anglia
http://researchpages.net/media...es/2008/04/22/thesis_final.pdf