Photochemical modelling of atmospheric oxygen levels confirms two stable states

<div> <div> <div> <p>Various proxies and numerical models have been used to constrain O<sub>2</sub>&nbsp;levels over&nbsp;geological time, but considerable uncertainty remains. Previous investigations using 1-D photochemical models have predicted how O<...

Full description

Bibliographic Details
Main Authors: Gregory, BS, Claire, MW, Rugheimer, S
Format: Journal article
Language:English
Published: Elsevier 2021
Description
Summary:<div> <div> <div> <p>Various proxies and numerical models have been used to constrain O<sub>2</sub>&nbsp;levels over&nbsp;geological time, but considerable uncertainty remains. Previous investigations using 1-D photochemical models have predicted how O<sub>3</sub>&nbsp;concentrations vary with assumed ground-level O<sub>2</sub>&nbsp;concentrations, and indicate how the ozone layer might have developed over Earth history. These classic models have utilised the numerical simplification of fixed mixing ratio boundary conditions. Critically, this modelling assumption requires verification that predicted fluxes of biogenic and volcanic gases are realistic, but also that the resulting steady states are in fact stable equilibrium solutions against trivial changes in flux.</p> <p>Here, we use a 1-D photochemical model with fixed flux boundary conditions to simulate the effects on O<sub>3</sub>&nbsp;and O<sub>2</sub>&nbsp;concentrations as O<sub>2</sub>&nbsp;(and CH<sub>4</sub>) fluxes are systematically varied. Our results suggest that stable equilibrium solutions exist for trace- and high-O<sub>2</sub>/O<sub>3</sub>&nbsp;cases, separated by a region of instability. In particular, the model produces few stable solutions with ground O<sub>2</sub>&nbsp;mixing ratios between&nbsp;<span tabindex="0" data-mathml="&lt;math xmlns=&quot;http://www.w3.org/1998/Math/MathML&quot;&gt;&lt;mn is=&quot;true&quot;&gt;6&lt;/mn&gt;&lt;mo is=&quot;true&quot;&gt;&amp;#xD7;&lt;/mo&gt;&lt;msup is=&quot;true&quot;&gt;&lt;mrow is=&quot;true&quot;&gt;&lt;mn is=&quot;true&quot;&gt;10&lt;/mn&gt;&lt;/mrow&gt;&lt;mrow is=&quot;true&quot;&gt;&lt;mo linebreak=&quot;badbreak&quot; linebreakstyle=&quot;after&quot; is=&quot;true&quot;&gt;&amp;#x2212;&lt;/mo&gt;&lt;mn is=&quot;true&quot;&gt;7&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;">6&times;10&minus;7</span>&nbsp;and&nbsp;<span tabindex="0" data-mathml="&lt;math xmlns=&quot;http://www.w3.org/1998/Math/MathML&quot;&gt;&lt;mn is=&quot;true&quot;&gt;2&lt;/mn&gt;&lt;mo is=&quot;true&quot;&gt;&amp;#xD7;&lt;/mo&gt;&lt;msup is=&quot;true&quot;&gt;&lt;mrow is=&quot;true&quot;&gt;&lt;mn is=&quot;true&quot;&gt;10&lt;/mn&gt;&lt;/mrow&gt;&lt;mrow is=&quot;true&quot;&gt;&lt;mo linebreak=&quot;badbreak&quot; linebreakstyle=&quot;after&quot; is=&quot;true&quot;&gt;&amp;#x2212;&lt;/mo&gt;&lt;mn is=&quot;true&quot;&gt;3&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;">2&times;10&minus;3</span>&nbsp;(<span tabindex="0" data-mathml="&lt;math xmlns=&quot;http://www.w3.org/1998/Math/MathML&quot;&gt;&lt;mn is=&quot;true&quot;&gt;3&lt;/mn&gt;&lt;mo is=&quot;true&quot;&gt;&amp;#xD7;&lt;/mo&gt;&lt;msup is=&quot;true&quot;&gt;&lt;mrow is=&quot;true&quot;&gt;&lt;mn is=&quot;true&quot;&gt;10&lt;/mn&gt;&lt;/mrow&gt;&lt;mrow is=&quot;true&quot;&gt;&lt;mo linebreak=&quot;badbreak&quot; linebreakstyle=&quot;after&quot; is=&quot;true&quot;&gt;&amp;#x2212;&lt;/mo&gt;&lt;mn is=&quot;true&quot;&gt;6&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;">3&times;10&minus;6</span>&nbsp;and 1% of present atmospheric levels). A fully UV-shielding ozone layer only exists in the high-O<sub>2</sub>&nbsp;states. Our atmospheric modelling supports prior work suggesting a rapid bimodal transition between reducing and oxidising conditions and proposes&nbsp;Proterozoic&nbsp;oxygen levels higher than some recent proxies suggest. We show that the boundary conditions of photochemical models matter, and should be chosen and explained with care.</p> </div> </div> </div>