Quantification and minimization of uncertainty by geostatistical simulations during the characterization of contaminated sites: 3-D approach to a multi-element contamination

Comparison of different interpolation methods and sequential Gaussian simulation to estimate volumes of soil contaminated by As, Cr, Cu, PCP and dioxins/furans

“…During the past years, significant effort have been invested to improve the characterization of contaminated soils and to reduce the costs related to the rehabilitation of these sites by applying geostatistical techniques within the characterization phase. Indeed, rehabilitation of contaminated sites should be based on a precise and accurate characterization of soil contamination to avoid errors in the quantification of pollution, which can have serious consequences for both health and rehabilitation costs (Boudreault et al, 2016). Before starting any sampling campaign, it is important to use the information related to the history of the site, when available, because it provides historical information related to the nature and sometimes the potential location of the pollution (inorganic and/or organic) at a lower cost compared to a systematic characterization campaign on the entire site.…”

“…In this case, non-linear methods such as the indicator kriging or geostatistical simulations are recommended to be used (Juang et al, 2004;Lin et al, 2016). In recent years, many studies have focused on SGS to generate probability maps for assessment of soil pollution and to optimize of sampling plans during the characterization of contaminated sites (Boudreault et al, 2016, Demougeot-Renard et al, 2004.…”

“…In contaminated site characterization, geostatistical methods have been used to estimate the volumes of soils whose concentration exceeds a regulatory criterion, to calculate the probabilities of exceeding regulatory criteria and to evaluate the uncertainty of these estimations (Boudreault et al, 2016).…”

“…This smoothing effect has been clearly demonstrated in the study of Zawadzki et al (2008), which was conducted for reassessment of soil contamination with lead. According to their results, the lead content maps showed that kriged values were smoothed from the range of 1-286 mg.kg -1 to the range of unbiased estimator of contaminant levels above the defined threshold (Boudreault et al, 2016).…”

A machine learning approach for characterizing soil contamination in the presence of physical site discontinuities and aggregated samples

“…In the field of geostatistics, several researchers such as Boudreault et al [1] and Goovaerts [2,3] have employed the Kriging theory to characterize the spatial distribution of contaminant concentration. Historically, Kriging was proposed by Krige and later formalized by Matheron [4].…”