Wednesday, November 7, 2007 - 2:00 PM
290-8
A comparison of Soil Carbon Characterization using py-MBMS and py-GC/MS.
Kim Magrini, National Renewable Energy Laboratory, 1617 Cole Blvd, Golden, CO 80401-3393, A. Stuart Grandy, Crop and Soil Sciences, Michigan State University, East Lansing, MI 48824, and Eldor Paul, Colorado State University, 843 Rossum Dr., Loveland, CO 80537.
A critical need exists to better understand both the amount of soil organic carbon (SOC) as a result of land use and management practices and its chemical (structural/molecular) composition. Recent research has suggested that soil biological processes may be a critical driver of C structure and, particularly in disturbed systems, largely regulate SOC chemistry and turnover rates. Two techniques that can provide detailed chemical information that can be linked via biological markers to soil communities use analytical pyrolysis coupled with 1) molecular beam mass spectrometry (py-MBMS) and 2) gas chromatography/mass spectrometry (py-GC/MS). Py-MBMS analysis of well-characterized native prairie soils recently demonstrated that SOC, soil microbial biomass, and mineral associated carbon (Cmin) could be chemically distinguished. Currie-point Py-GC/MS has also been used to distinguish different soils and unique fractions within soils to show how decomposition dynamics vary across ecosystems and intensities of disturbance. This method provides a molecular fingerprint of SOC and in part because it uses a consistent pyrolysis temperature results are highly reproducible. We used both of these methods to characterize the molecular structure of agricultural soils from Hoytville, OH, the W. K. Kellogg Biological Station, MI, and Lamberton, MN. Curie-point Py-GC/MS showed considerable variation among these managed soils. Lignin concentrations varied from 1.48% in the Wooster soils to 4.81% at KBS while lipid concentrations ranged from 2.16% at Hoytville to 11.53% at Lamberton. Polysaccharide contents were high in all of the samples (>50%), which may have been a function of soil management or the preferential recovery of these structures by py-GC/MS. We compare these results to those using py-MBMS and show that while each technique has advantages and drawbacks both can be used to understand management effects of SOC structure and the potential for biological processes to mediate structural transformations.