Diagnostic Screening of Urban Soil Contaminants Using Diffuse Reflectance Spectroscopy

Abstract

There is increasing demand for cheap and rapid screening tests for soil contaminants in environmental consultancies. Diffuse infrared reflectance spectroscopy in the visible-near-infrared (vis-NIR) and mid-infrared (MIR) has the potential to meet this demand. The aims of this chapter are to develop diagnostic screening tests for heavy metals and polycyclic aromatic hydrocarbons (PAHs) in soil using vis-NIR and MIR diffuse reflectance spectroscopy. Cadmium, copper, lead, and zinc were analysed, as were total PAHs and benzo [a]pyrene. An ordinal logistic regression technique was used for predictions in the screening tests and to determine false-positive and false-negative rates. Zinc had the best prediction accuracy (89%) and copper predictions were consistently above 75%. Cadmium and lead were the least well predicted of the heavy metals (67 and 70%, respectively), and PAH predictions averaged 78.9%. MIR analysis (average prediction accuracy of 79.9%) was only slightly more accurate than vis-NIR analysis (average prediction accuracy of 77.1%), but the latter may currently be used in situ, thereby reducing cost and time of analysis and providing diagnosis in real time. Diffuse reflectance spectroscopy in the vis-NIR can substantially decrease both the time and the cost associated with screening for soil contaminants.

Visible and Near Infrared Spectroscopy in Soil Science

“…Wu et al (2007) produced some further data suggesting that the correlation between heavy metals and spectrally active soil components such as organic matter and clay but particularly Fe is the major predictive mechanism. Bray et al (2009) used a vis-NIR and mid-IR diagnostic screening procedure for detection of contamination by heavy metals (Cd, Cu, Pb, and Zn) and polycyclic aromatic hydrocarbon (PAH) in urban soils. They used ordinal logistic regressions for predictions of either contaminated or uncontaminated soil at different thresholds and found that the mid-IR technique produced only slightly better prediction accuracies.…”

Comparison between inductively coupled plasma and X-ray fluorescence performance for Pb analysis in environmental soil samples

“…(Zia et al 2011). The accumulation of Pb in the kidney, liver, teeth and bones can be associated with negative health effects including inhibited brain development and haematological effects, such as anaemia (Bray et al 2009;WHO 2010). Since 1889 the emission of dust particles from South Australia's smelter led to identify three different area, according to levels of risk: high, medium and low, based on the mean BPb, an accurate bioindicator of environmental lead exposure detected on children under five years (Maynard et al 2005).…”

Modelling Diverse Soil Attributes with Visible to Longwave Infrared Spectroscopy Using PLSR Employed by an Automatic Modelling Engine

“…This can be the reason why the CEC prediction was worse (RPD = 1.48, R 2 test = 0.62). For heavy metals, the results are comparable or even better than those published by [10] when employing SVM modelling together with the 1st derivative as a pretreatment technique or by Bray et al (2009) [62] who used an ordinal logistic regression technique to predict Zn, Cu, Pb and Cd.…”