|Reeves iii, James|
Submitted to: Journal of Environmental Quality
Publication Type: Peer reviewed journal
Publication Acceptance Date: 9/15/2004
Publication Date: 11/15/2004
Citation: Siebielec, G., McCarty, G.W., Stuczynski, T., Reeves III, J.B. 2004. Use of near- and mid-infrared diffuse reflectance spectroscopy for measuring metal content in soil. Journal of Environmental Quality. 33:2056-2069. Interpretive Summary: The increasing interest in the quality of crops has increased the need to assess and monitor soils within croplands. Detailed monitoring of trace metal contents is needed due to increased risk of elevated metal in feed and in food. This information is especially necessary in regions under present or former influence of industrial activities. Rapid and nondestructive methods such as infrared spectroscopy provide potentially useful alternatives to time-consuming chemical methods. To assess utility of infrared spectroscopy, we compared results obtained using two different regions of the infrared spectrum (near infrared and mid infrared). The soils used were from a diverse collection obtained throughout the metal mining region of Tarnowskie Gory (Upper Silesia/Poland). This region contains soils with elevated contents of zinc, lead, and cadmium due to both natural soil forming processes and to industrial activities. Results obtained using the mid infrared were much better than those using near infrared; this suggests that mid infrared spectroscopy may be useful for accurate measurement of metals in soil. Careful development of calibrations will be required for these measurements to ensure accurate results.
Technical Abstract: Regional assessment of soil quality requires analysis of metal contents in a great number of samples. Rapid and nondestructive methods such as diffuse reflectance infrared spectroscopy provide potentially useful alternatives to time-consuming chemical methods. To assess the utility of near-infrared reflectance spectroscopy (NIRS) and diffuse mid-infrared reflectance spectroscopy (DRIFTS) for soil metal determination, 70 soil samples were collected throughout the metal mining region of Tarnowskie Gory (Upper Silesia/Poland). Soils represented a wide range of pH (4.0-7.6), of total carbon (6.2-62.2 g kg-1), and of textural class (from sand to silty clay loam). Soils had various concentrations of metals - Zn, Pb, and Cd ranged from background concentration to high contamination, while Ni and Cu had close to background concentrations. Soil Zn, Pb, and Cd originated both from industrial contamination and pedogenic sources. Mid-infrared spectroscopy markedly outperformed NIRS. Predictions for Fe, Cd, Cu, Ni and Zn using DRIFTS data were precise [the coefficient of determination (R2) between actual and predicted concentrations were 0.97, 0.94, 0.80, 0.99, 0.96 for those metals, respectively]. Only Pb concentration was predicted poorly (R2=0.66). Calibrations using NIRS were less satisfactorily (R2 = 0.87, 0.54, 0.61, 0.45, 0.84 and 0.67 for Fe, Cd, Cu, Pb, Ni and Zn, respectively). Results indicate that DRIFTS may be useful for accurate predictions of several metals if samples originate from one region. Full and balanced coverage of soil diversity within the area of interest by the collected set of samples will aid in production of useful calibrations.