Skip to main content
ARS Home » Research » Publications at this Location » Publication #155696

Title: MAPPING SOIL CARBON WITH ON-THE-GO NEAR INFRARED SPECTROSCOPY

Author
item CHRISTY, COLIN
item LUND, ERIC
item Laird, David

Submitted to: Advances in Terrestrial Ecosystem Carbon Inventory Measurements and Monitor
Publication Type: Proceedings
Publication Acceptance Date: 10/17/2003
Publication Date: 10/17/2003
Citation: CHRISTY, C., LUND, E., LAIRD, D.A. MAPPING SOIL CARBON WITH ON-THE-GO NEAR INFRARED SPECTROSCOPY. ADVANCES IN TERRESTRIAL ECOSYSTEM CARBON INVENTORY MEASUREMENTS AND MONITOR. 2003. AVAILABLE FROM: http://www.oznet.ksu.edu/ctec/Fall_Forum.htm.

Interpretive Summary:

Technical Abstract: We tested the ability of a newly developed field-mobile, on-the-go, near infrared spectroscopy (NIRS) system to measure the distribution of soil organic C in two 40-acre agricultural fields. The system, developed by Veris Technologies, Salina, KS, consists of a steel shank that can be pulled through the soil by a tractor at a controlled depth, an optical window on the base of the shank, a built-in light source, a fiber optic cable, an NIR spectrometer, a GPS unit, and a computer. Correlation coefficients between measured and predicted values of organic C concentration were higher for the Baker field (R2=0.785) than the Hogranch field (R2=0.543) because of a much broader range in organic C in the Baker field (0.9 to 6.8% C) than in the Hogranch field (1.1 to 2.3% C). On the other hand, Root Mean Standard Error of Cross Validation (RMSECV) values were larger for the Baker Field (RMSECV=0.579) than the Hogranch field (RMSECV=0.149) due to non-linearity which was caused by very high organic C levels in some samples from the Baker field. The results indicate that the NIRS system is capable of rapidly and accurately quantifying the distribution of organic carbon in surface soils of agricultural fields. Several things are needed before the NIRS system can be used routinely, including: further improvements in the instrumentation, development of national or regional calibration data bases, development of robust calibration models, and development of calibration transfer functions.