THE USE OF δ13C VALUES IN FOSSIL RESINS AS A PROXY FOR CHANGES IN ATMOSPHERIC PO2

Estimating the partial pressure of atmospheric oxygen (pO₂) in the geological past has been challenging due to a lack of reliable proxies. Here we present a technique to estimate paleo-pO₂ using the stable carbon isotope composition (δ¹³C) of modern and fossil plant resins. Resins exhibit a natural variability in δ¹³C of around 8‰, due to local environmental and ecological factors (e.g., water availability, water composition, light exposure, temperature, nutrient availability, etc.). Despite this variability, fossil resins are on average enriched in ¹³C by up to 6‰, compared to modern resins. Shifts in fossil resin isotopic composition follow distinctive trends through time, with the greatest enrichment taking place during the Triassic, mid-Cretaceous, and early Eocene. Experimental evidence and theoretical considerations suggest that neither changes in pCO₂ nor in the δ¹³C of atmospheric CO₂ can account for the observed shifts in resin δ¹³C. The fractionation of ¹³C in resin-producing plants, instead, is primarily influenced by atmospheric pO₂, with more fractionation occurring at higher pO₂. Enriched δ¹³C values suggest that atmospheric pO₂ during most of the Mesozoic and Cenozoic was considerably lower (pO₂ ≥10%) than today. Furthermore, a correlation between resin δ¹³C and the marine δ¹⁸O record implies that pO₂, pCO₂, and global temperatures were inversely linked (i.e., intervals of low pO₂ were generally accompanied by high pCO₂ and elevated global temperatures). Interestingly, intervals with the lowest inferred pO₂ were preceded by large-scale volcanism. This suggests a relationship in which mantle-derived volcanic CO₂ may have triggered an initial warming, leading to an increase in oxidative weathering, and ultimately, a decline of pO₂ during greenhouse periods. Conversely, after volcanic output ceased, pO₂ levels would have increased in response to reduced weathering rates, whereas CO₂ levels and temperatures declined.