Assessing the Risk of Soil Pollution around an Industrialized Mining Region Using a Geostatistical Approach

Abstract

Knowledge of mobility of some heavy metals in coal mining areas is fundamental in order to understand their toxicity and geochemical behavior. This paper aims to map pollution and assess the risk to agricultural soils in a wider lignite opencast mining and industrial area. Geochemical data related to environmental studies show that the waste characteristics favor solubilization and mobilization of inorganic contaminants. The geochemical distribution of soil pollution is studied by the application of the Bayesian Maximum Entropy (BME) procedure, a versatile extension of geostatistics which allows merging spatial and temporal estimations in a single model. Results reveal a correlation range of contaminants concentrations up to 5000 m and indicate a potential forecasting range of up to five years. Inspection of the produced spatiotemporal maps indicates that the whole study area is contaminated by As and various heavy metals, a situation which seems to be more or less stable over time.

Spatial Patterns and Risk Assessment of Heavy Metals in Soils in a Resource-Exhausted City, Northeast China

“…Unlike classic modeling approaches, spatial interpolation methods incorporate information about the geographic positions of the sample points [ 20 ]. The rationale behind spatial interpolation is that parameters for points that are close to each other are more likely than parameters for points further away to correlate and show similarities[ 21 ]. In the study presented here, the ordinary kriging method was used to determine the spatial distribution patterns for individual heavy metals and radial basis functions to express the Nemerow synthetic pollution index.…”

Geostatistical estimation of risk for soil and water in the vicinity of olive mill wastewater disposal sites

“…In international literature, there are many examples considering the use of geostatistics to assess risk for soils and waters in waste disposal sites or in the vicinity of industrial sites [26][27][28][29][30][31]. To our knowledge, this is the first attempt to use geostatistics and estimate risk for both soil and water in the vicinity of an OMW disposal site.…”

Bayesian inference-based environmental decision support systems for oil spill response strategy selection

“…Jakeman et al (2006) identify eight distinct families of model methodologies but one proven successful modelling approach that accommodates Romero and Young's (1997) environmental systems' defining characteristics, and also readily satisfies all but one of the desirable features they propose for an EDSS, is Bayesian inference based modelling. (The Bayesian approach discussed in this paper does not intrinsically represent spatial data, however work by Aps et al (2009a) has successfully integrated it with Geographical Information Systems and Modis and Vatalis (2014) use the technique of Bayesian Maximum Entropy that enables the integration of spatial and temporal data into a single model). Frequently used in decision analysis, Bayesian inference is the basis for Bayesian Belief Networks (BBNs) and Inference Diagrams (IDs) (DSL, 2010) collectively referred to as Bayesian networks.…”

“…Modelling habitat and population variability of at-risk species (Marcot et al, 2001). Understanding soil pollution and toxicity due to mining using a geostatistical approach (Modis and Vatalis, 2014). Using expert knowledge to estimate rivers' carrying capacity for Atlantic salmon smolt (Uusitalo, 2007).…”

“…Bayesian probability techniques have been found to be very effective in oil spill contingency planning and environmental risk assessment (i.e. before a pollution incident occurs), post-spill event analysis (after a pollution event) and in environmental management and ecology (Aps et al, 2009a,b;Carriger and Barron, 2011;Chen and Pollino, 2012;Colombo et al, 2012;Ellison, 2004;Marcot et al, 2001Marcot et al, , 2006McCann et al, 2006;Modis and Vatalis, 2014;Montewka et al, 2013;Uusitalo, 2007). However, very little evidence has been found in the published literature for its use during the emergency response phase of a pollution incident, when the ability to handle uncertainty and integrate multiple inputs could bring major benefits to the effort to reduce environmental damage.…”