Submitted to: Soil Science Society of America Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: September 4, 2001
Publication Date: April 30, 2002
Interpretive Summary: Increasing the carbon dioxide (CO2) content of the atmosphere from human activities is causing the global temperature to increase. This has stimulated research in assessing the role of land-based ecosystems that can remove CO2 from the atmosphere. Considerable interest exists in using production agriculture as a carbon sink to help remove excess atmospheric CO2. In this way agriculture can benefit by playing an important role in a carbon credit system designed to regulate CO2 emissions. For agriculture to take advantage of this opportunity, however, new technologies are needed to measure carbon stocks in agricultural landscapes. A new technology for direct, nondestructive measurement of carbon in soil was investigated. This technology is based on measurement of light reflected from soil that has been illuminated by infrared light. In preliminary studies, this method provided rapid, accurate measurements of carbon in soil samples. Further development of this technology will enable agricultural to play an inportant role in the emerging carbon economy.
Ability to inventory soil carbon and estimate carbon sequestration on landscapes is limited by ability to rapidly measure the carbon content in soil. Easier and more rapid methods for soil carbon measurement are needed to better ascertain spatial structure and measure changes in carbon storage. Diffuse reflectance spectroscopic analysis in the near-infrared (400 to 2500 nm) and mid-infrared (2500 to 25000 nm) regions provides a means for rapid measurement of soil carbon. These spectral regions were compared for their ability to provide useful information related to the quantity of total, organic, and inorganic carbon in 273 soils from a large geographic region in the central United States. The soils tested ranged considerably in properties including organic (0.23 to 98 mg C g-1 soil) and inorganic carbon content (0 to 65.4 mg CO3-C g-1 soil). These soil samples were analyzed with and without an acid treatment for removal of carbonate. Both spectral regions contained information on organic and inorganic carbo content of the soils studied. The mid-infrared analysis substantially out-performed analysis in the near infrared region. The superior performance of the mid-infrared regions likely reflects a higher quality of information for soil carbon in this spectral region. The ability of mid-infrared spectroscopy to quantify carbon in diverse soils collected over a large geographic region suggested that only regional calibrations may be required.