SOIL CO2 MEASUREMENTS IN MODERN VERTISOLS AND IMPLICATIONS FOR ESTIMATIONS OF PALEOATMOSPHERIC CO2 FROM PEDOGENIC CARBONATES

Accurate estimates of Earth’s ancient atmospheric CO₂ concentrations ([CO₂]_atm) are essential for understanding the geologic carbon cycle, indentifying relationships between geologic events and the evolution and extinction of organisms and predicting the magnitude of future climate change. The majority of previously reported, proxy-based pre-Cenozoic values of [CO₂]_atm come from the application of an isotope mass balance equation called the paleosol carbonate CO₂ barometer. We are currently calibrating this equation for Vertisols (smectitic, clay-rich soils that swell during seasonal soil saturation), the ancient equivalents of which are abundant in the rock record. The soil respired CO₂ concentration during carbonate formation (S(z)), an input variable required in the paleosol barometer, is poorly constrained and is the target of our study. We hypothesize that closing of soil macropores during wet conditions allows high S(z) values to develop in the subsoil, whereas cracking upon drying opens these conduits, resulting in the release of respired CO₂ and large decreases in S(z) values which drive calcite precipitation. We infer that decreased soil respiration rates, increased pore space volume, increased diffusivity and barometric pumping through cracked soil all act to decrease S(z) values during Vertisol drying. We began monthly monitoring of S(z) values in modern Vertisols (in the Bkss horizons, ~100 cm depth) from central Texas in January 2010 and currently have a times series extending through July 2010. The S(z) values measured for this study range from 1-9% by volume. Thus far, S(z) values are higher in Vertisol microhighs than in microlows, which is the opposite of what we expected and may be the result of anoxia limiting soil respiration rates in microlows. For all locations, the measured S(z) values are at or substantially above the maximum S(z) values typically used in the paleosol barometer. We have not yet been able to fully test the hypothesis about the importance of cracking, but we observed large (up to 3% by volume) decreases in S(z) values as the soil surface dried and began to crack between the end of April and the middle of May. Understanding the dynamics of soil CO₂ in modern Vertisols may help us understand soil carbonate formation and develop proxies or accurate estimated values for S(z).