Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: March 12, 2012
Publication Date: N/A
Interpretive Summary: The capacity of soils in Oklahoma to store carbon (C) has become an important issue. Carbon storage could be used as an aid in reducing greenhouse gasses in the atmosphere, and may provide cash flow for land managers in the State. Describing the amount of C in soils, both background amounts and change in amounts over time, is a key part of monitoring projects. Description of stored C by traditional laboratory-based techniques requires accurate estimates of two soil properties to provide estimates on a metric ton per acre basis: concentration of C, and density of the soil for the chosen depth being examined. Both techniques require large amounts of labor, time, and equipment, which can result in high costs for estimates. We undertook a study to describe the capacity of near infrared reflectance spectroscopy (NIRS) to provide direct estimates of C in a series of soils located in a set of experimental pastures in central Oklahoma. Included were three pastures of native prairie managed under different grazing regimes during 1978 to 2004, and one pasture of winter wheat used for grazing. We collected samples from the A horizon (top 10 inches) of soils along an elevation gradient common to all pastures. The soil samples were processed, then scanned by reflectance spectrophotometer, and analyzed by standard laboratory techniques to develop the data required for analysis. We then tested the capacity of an equation developed by NIRS to describe amounts of C in soil based on reflected wavelengths that were recorded by NIRS. The study showed an equation, based on laboratory measurements and wavelengths, could be developed that defined 93% of the information related to C in soil. An equation that can be applied to a wider section of central Oklahoma will require samples from additional landscapes, soil types, and forms of management.
Technical Abstract: Interest in carbon (C) storage within agricultural soils of Oklahoma as an aid in reducing atmospheric greenhouse gasses, and cash flow land managers might access, has increased recently. Description of C mass requires measurement of both bulk density and C concentration, but the techniques used are labor intensive, and expensive. This study defined the capacity of NIRS to directly predict total C mass of soils in experimental paddocks of native prairie (n=3) and winter wheat (n=1) in central Oklahoma, under different forms of long-term (1978 to 2004) management. Samples were collected along 150 m transects (n=1 per paddock) situated between a ridge and toe slope. The A horizon was divided into sections (0- 5, 5-10, and 10-25 cm), reflectance (R) measurements (log 1/R) collected, and absorption spectra (450-2500 nm) developed for a randomly selected subset (n=209). A calibration equation between wavelengths and laboratory-measured C mass in 5 cm depth increments (n=94) were developed by multivariate partial least squares regression, and tested with a validation dataset (n=101). The relationship between laboratory-measured and NIRS estimated C mass generated by the calibration equation was significant (R^2= 0.96; P<0.01), and approximated unity (slope=1.01). Application of the calibration equation to the validation dataset generated a different slope (0.97) with a lesser but significant relationship (R^2=0.93; P<0.01). Results suggest the developed equation could provide useful predictions of C mass. A more accurate and broadly applicable equation will require samples from a wider range of landscapes, soil types, and management regimes.