Estimating long-term soil respiration rates from carbon isotopes occluded in gibbsite

Carbon occluded in the soil gibbsite crystal structure at the Panola Mountain Research Watershed, Georgia, U.S. is presumed to be in isotopic equilibrium with the CO2 respired from soil organics by microbes and plant roots. Fitting of the stable carbon isotopic data to a Fickian diffusion-based depth function results in an estimate of 47 gC m−2 y−1 for the long-term soil respiration rate. A numerical model that includes depth-dependent production and diffusion terms results in estimates of 28–12 gC m−2 y−1. These values range from 15 to 50 times less than the average of modern values for mixed deciduous forests in wet temperate climates. This disparity has several implications for our understanding of the geologic record of climate change, which include: (1) evidence for a cooler and seasonally drier climate during the mid-Holocene in the southeastern U.S., or (2) fluxes of carbon from the soil pool as recorded by soil mineral proxies (i.e., long-term soil respiration rates) under estimate atmosphere annual carbon flux measurements (i.e., short-term measures), and (3) the need to refine soil respiration models used to relate paleosol stable carbon isotopic measurements to paleo-atmospheric PCO2 estimates.