ANCIENT ATMOSPHERIC CARBON DIOXIDE FROM COOLER LATE OLIGOCENE AND WARMER EARLY MIOCENE TIMES, ESTIMATED FROM FOSSIL LEAVES AND CARBON ISOTOPES

Plants use atmospheric carbon dioxide to produce their own food via photosynthesis. Recent variations in the concentration of atmospheric carbon dioxide (pCO₂) have been shown to affect this process and are correlated with anatomical and physiological changes observed in leaves. These include changes to stomatal frequency and size, as well as in the ratio of the stable carbon isotopes, ¹³C and ¹²C, all of which can be measured on leaf fossils. Thus, fossil leaves can be used as proxies for Earth’s atmospheric carbon dioxide history. This study tests the link between pCO₂ and global temperature for the late Oligocene and early Miocene, two time slices shown by marine isotope records to document warming, and is important for understanding current climate change.

Late Oligocene and early Miocene atmospheric CO₂ concentrations are estimated herein using the Franks et al., (2014) model, a model that relates pCO₂ with rate of CO₂ assimilation by leaves, conductance to CO₂ diffusion into leaves, and ratio of leaf internal CO₂ concentration to atmospheric CO₂ concentration. Well-dated late Oligocene (27.23 ± 0.01 Ma) and early Miocene (21.73 ± 0.01 Ma) strata containing exquisitely preserved compressed leaves are found in the northwest and central Ethiopian Highlands. Stomatal density, carbon isotope composition of leaves, and conductance to CO₂ diffusion into leaves were quantified from the fossil leaves, and late Oligocene and early Miocene pCO₂ were estimated using the Franks et al., (2014) model.

The results of a two-sample T-test indicate a statistically significant difference between the estimated late Oligocene and early Miocene CO₂ concentrations; early Miocene pCO₂ values are significantly higher than late Oligocene pCO₂ values. These values are consistent with a positive correlation between pCO₂, a greenhouse gas, and global average temperature, as independent studies of δ¹⁸O from benthic foraminifera document warming for the same time interval studied herein (Zachos et al., 2001). Thus, the presence of a link between pCO₂ and global temperature is supported for the late Oligocene to early Miocene.