Atmospheric oxygen of the Paleozoic

The evolution of Earth's atmosphere is of extreme importance to the lithosphere, hydrosphere and biosphere. This is particularly true for the role the biosphere played in the inceptual formation of atmospheric oxygen, and subsequently in its potential role in the evolution of higher life forms. Quantifying the change in atmospheric oxygen with geologic time is extremely challenging, with the redox sensitive isotope and element models subject to contradictory outcomes. Here, we present a new approach in determining past atmospheres, by a) directly measuring (DM) the amount of atmospheric oxygen trapped in fluid-gas inclusions of primary and unaltered halite, and by b) using a back-calculation method (BCM) applied to trapped halite fluid inclusion gas contents subjected to post-depositional biogeochemical reactions to calculate the amount of atmospheric oxygen.

On average (±standard error), atmospheric oxygen content during the latest Ediacaran was 17.4 ± 2.1% compiled from DM of 17.4 ± 1.3% and from BCM of 17.3 ± 2.9%. For the earliest Cambrian average oxygen was 19.3 ± 1.4% compiled from DM of 18.8 ± 1.5% and from BCM of 19.8 ± 1.3%. The oxygen content during the mid and late Ordovician was relatively invariant with 15.6 ± 1.8% [15.4 ± 1.4% (DM), 15.8 ± 2.2% (BCM)], and 16.2 ± 1.2% [15.7 ± 1.7% (DM), 16.6 ± 0.7% (BCM)], respectively. The relatively stable atmospheric oxygen levels continued into the Silurian of 15.9 ± 1.1% (12.9 ± 0.4 to 16.5%, DM; 14.3 ± 1.7 to 19.8 ± 2.1% BCM) except for a peak to about present levels during the late Silurian of 23.2 ± 1.9% (21.6% DM, 24.7 ± 3.7% BCM). Early Carboniferous atmospheric oxygen returned to relatively lower and invariant levels at about 15.3 ± 0.7 to 15.7 ± 1.0% (15.0%, 15.5 ± 0.1% DM, 9.4 ± 3.9 to 15.8 ± 1.9% BCM). Similarly, the average atmospheric oxygen content in fluid inclusion of halites from the late Permian was similar with 15.7 ± 1.3% with 16.9% (DM) to 14.5 ± 2.6% (BCM). Indeed, our study suggests that atmospheric oxygen was relatively constant for most of the Paleozoic at about 16.5% (±0.6 SE, ±2.2 SD). This puts our proxy result at odds with the values and trends suggested by the COPSE, GEOCARBSULF AND GEOCARBSULFOR models for the early Paleozoic atmospheric oxygen, and continues the discord with the GEOCARBSULF and GEOCARBULFOR models during the late Paleozoic.