Combined Effects of Oil Concentration, Clay and Moisture Contents on Diffuse Reflectance Spectra of Diesel-Contaminated Soils

Abstract

Removal of petroleum hydrocarbon (PHC) contamination that is hazardous and often prevalent in soils would benefit from a rapid detection technique. Visible and near-infrared spectroscopy (VIS-NIRS) has a large potential as a rapid detection technique for PHC in soils. Nevertheless, the combined influence of oil concentration, moisture content and clay content on soil reflectance spectra and the accuracy of the technique have yet received little attention. The objective of this study was to investigate the combined influence of oil concentration and moisture and clay contents on the spectral characteristics of diesel-contaminated soils and the quality of calibration models developed for polycyclic aromatic hydrocarbons (PAH) in soils using VIS-NIRS. With partial least-squares regression data from a systematic experimental design using 150 artificially contaminated soil samples, results showed that soil diffuse reflectance decreased with increasing oil concentration, clay and moisture contents. The trend was less defined in relation to moisture and clay due mainly to the interaction effects of the soil matrices as mediated by the oil. The PAH partial least squares cross-validation showed best performance with the lowest oil concentration and clay content at 20 % moisture with r 2 of 0.89, root mean square error of prediction of 0.201 mg/kg and ratio of the standard error of prediction to the standard deviation of the reference data in the validation set of 2.75. Analysis of variance showed that the interaction effects of oil concentration, moisture and/or clay content significantly (p < 0.05) affected the quality of the PAH models.