DERIVATION OF A NEW SOIL RESPIRED CO2 PROXY AND APPLICATION TO PALEO-CO2 RECONSTRUCTION OF THE LATE MIOCENE

Anthropogenically driven climate change is one of the most pressing concerns facing humanity. One way to predict future climate change is to study climate in the geologic past. The main control of the Earth’s temperature over geologic time is the concentration of atmospheric CO₂; however, the precise amount of CO₂ in the past is not fully understood (IPCC, 2007). Therefore, it is essential to quantify the atmospheric CO₂ concentration in the geologic past, especially at times of rapid climate change analogous to current changes.

The widely applied paleobarometer introduced by Cerling (1991) calculates atmospheric CO₂ based on the isotopic ratio of pedogenic carbonates precipitating in equilibrium with soil CO₂. Its primary source of uncertainty is the amount of soil respired CO₂ in the paleosol at the time of pedogenic carbonate formation. To remove the uncertainty from the Cerling paleobarometer, a new soil respired CO₂ proxy was calibrated for soils and paleosols. The relationship between modern soil CO₂ and mean annual precipitation was derived from an extensive literature review of modern carbonate bearing soil CO₂ data. Based measurements of minimum summer soil CO₂ (time and conditions of carbonate formation) from 21 pedogenic carbonate-bearing soils, we find a linear relationship between summer minimum levels of soil respired CO₂ and MAP (R² = 0.57; SE = 722 ppm). For paleosols, past soil respired CO₂ can be predicted using MAP estimates calculated from either the depth to Bk horizon (Retallack, 2005) or the chemical index of alteration without potassium (Sheldon et al., 2002).

Here we use this new soil respired CO₂ proxy to reconstruct atmospheric CO₂ during the late Miocene from paleosols in Montana and compare these results to previously published estimates using other proxies (Demicco et al., 2003; Ekart et al., 1999; Kürschner et al, 2001; Pagani et al., 1999; Yapp and Poths, 1996) from this time period. During the latest Miocene (5.5 Ma), we see atmospheric CO₂ levels approaching preindustrial values of 280 ppm, well in line with the estimates produced by many other proxies. This demonstrates the validity and potential value of this new proxy.