149-2 Towards a Rapid Method of Measuring Labile Soil C Using Infrared Spectroscopy.
Poster Number 948
See more from this Division: SSSA Division: Soil Biology & BiochemistrySee more from this Session: Soil Biology & Biochemistry: II
Monday, November 3, 2014
Long Beach Convention Center, Exhibit Hall ABC
Permanganate oxidizable carbon (POXC) gives a measure of a relatively processed but labile fraction of soil C. POXC has been related to soil microbial biomass and particulate organic C, which are fundamental for soil nutrient cycling and mineralizable C. We analyzed POXC, soil organic C (SOC), and total N (TN) on a large sample set that encompassed many soil types and managed plant communities (36 total sites from 7 different research projects, n = 519). The soils were scanned in Fourier transformed Mid-Infrared spectroscopy (DRIFTS, 4000-400 cm-1) and Near-Infrared (NIR, 10000-4000 cm-1), and the spectral data was used to develop calibrations for POXC, SOC, and TN using Partial Least Squares regression (PLS). Our objectives were to 1) determine if DRIFTS and NIR could be used to determine POXC, SOC, and TN on a wide range of soils, 2) compare the calibration quality of the DRIFTS and NIR, 3) test if specific spectral regions can be used to develop calibrations instead of the full DRIFTS spectrum. Our results show that DRIFTS spectra produced calibrations for SOC, TN, and POXC, with R2 of 0.77, 0.73, and 0.85 respectively. Using the NIR spectra for the calibrations resulted in lower R2 scores of 0.52, 0.50, and 0.70 for SOC, TN, and POXC. Calibrations were also calculated using selected spectral regions. The region between 3425-2800 cm-1 encompasses absorbance for aliphatic CH and OH/NH. This region is mainly organic, and has little mineral interference in soils that are low in carbonates. The region between 1750-1325 cm-1 includes what is termed the organic fingerprint region. The PLS regression with the selected regions had R2 results equivalent to those using the full DRIFTS spectrum. Using the selected regions has the advantage that the PLS runs are quicker, given the smaller number of spectral points used for the analyses.
See more from this Division: SSSA Division: Soil Biology & Biochemistry
See more from this Session: Soil Biology & Biochemistry: II