THE POTENTIAL OF NEAR- AND MID-INFRARED SPECTROSCOPY FOR THE DETERMINATION OF SOIL CARBON

Authors: Reeves Iii, James; McCarty, Gregory; FOLLETT, R - SPNR; KIMBLE, J - NRCS; MAC MADARI, B - EMBRAPA, BRAZIL

Submitted to: Meeting Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 12/21/2004
Publication Date: N/A
Citation: N/A

Interpretive Summary: Carbon (C) sequestration is the concept of removing from the atmosphere and storing it in the soil. Due to the interest in sequestering C in soil as a means of reducing or limiting the buildup of atmospheric CO2 from the use of fossil fuels, it has become apparent that a tremendous number of soil samples will need to be analyzed to make accurate estimates. It has also become apparent that the currently used laboratory methods are inadequate for the task in terms of cost, capacity per instrument per day, toxic wastes generated, etc. As a result, new, rapid, low cost per sample, high capacity methods will be needed to meet the sample analysis requirements. Near-infrared reflectance spectroscopy (NIRS) has been used for the last couple of decades to determine the composition of food products, animal feeds and many other organic materials. Using light beyond the range of human sight, NIRS has been coupled to various mathematical in a process now called chemometrics, to find relationships between the spectra of samples and composition. More recently diffuse reflectance Fourier Transform spectroscopy using light in the 2500 to 25,000 nm range has similarly been used. While DRIFTS appears to be the more accurate method, for other than laboratory analysis NIRS has some advantages. These include: 1. The building of field portable instruments is easier and cheaper as fiber-optic cables are more readily available and more rugged than corresponding mid-IR systems. 2. The scanning of non-homogeneous samples, including soil cores, would be easier in the NIR due to the larger spot size and sample holders available compared to the static cups and very small spot size for DRIFTS units. Portable FTIRs are available which might be adaptable for scanning larger or non-homogeneous samples although the spot size is still small. 3. While the same principles appear to apply to developing calibrations using mid-IR spectra as for NIR spectra, the database of knowledge available for NIRS is much larger. In conclusion, both NIRS and DRIFTS have a tremendous potential for increasing our ability to analyze soil samples in a more rapid, cost efficient matter.

Technical Abstract: Due to the interest in sequestering C in soil as a means of reducing or limiting the buildup of atmospheric CO2 from the use of fossil fuels, it has become apparent that a tremendous number of soil samples will need to be analyzed to make accurate estimates. It has also become apparent that the currently used laboratory methods such as dry combustion or chromate oxidation are inadequate for the task in terms of cost, capacity per instrument per day, toxic wastes generated, etc. As a result, new, rapid, low cost per sample, high capacity methods will be needed to meet the sample analysis requirements. Near-infrared (NIR) reflectance spectroscopy (NIRS) has been used for the last couple of decades to determine the composition of food products, animal feeds and many other organic materials. Using light in the range of 400 to 2498 nm, NIRS has been coupled to various regression methods in a process now called chemometrics, the objective of which is to find relationships between the spectra of samples and composition. More recently diffuse reflectance Fourier Transform spectroscopy using light in the 2500 to 25,000 nm range has similarly been used. While DRIFTS appears to be the more accurate method, for other than laboratory analysis NIRS has some advantages. These include: 1. The building of field portable instruments is easier and cheaper as fiber-optic cables are more readily available and more rugged than corresponding mid-IR systems. 2. The scanning of non-homogeneous samples, including soil cores, would be easier in the NIR due to the larger spot size and sample holders available compared to the static cups and very small spot size for DRIFTS units. Portable FTIRs are available which might be adaptable for scanning larger or non-homogeneous samples although the spot size is still small. 3. While the same principles appear to apply to developing calibrations using mid-IR spectra as for NIR spectra, the database of knowledge available for NIRS is much larger. In conclusion, both NIRS and DRIFTS have a tremendous potential for increasing our ability to analyze soil samples in a more rapid, cost efficient matter.