Soil pollution constitutes one of the major threats to public health, where spreading to groundwater is one of several critical aspects. In most internationally adopted frameworks for routine risk assessments of contaminated land, generic models and soil guideline values are cornerstones. In order to protect the groundwater at contaminated sites, a common practice worldwide today is to depart from health risk-based limit concentrations for groundwater, and use generic soil-to-groundwater spreading models to back-calculate corresponding equilibrium levels (concentration limits) in soil, which must not be exceeded at the site. This study presents an extensive survey of how actual soil and groundwater concentrations, compiled for all high-priority contaminated sites in Sweden, relate to the national model for risk management of contaminated sites, with focus on As, Cu, Pb and Zn. Results show that soil metal concentrations, as well as total amounts, constitute a poor basis for assessing groundwater contamination status. The evaluated model was essentially incapable of predicting groundwater contamination (i.e. concentrations above limit values) based on soil data, and erred on the "unsafe side" in a significant number of cases, with modelled correlations not being conservative enough. Further, the risk of groundwater contamination was almost entirely independent of industry type. In essence, since neither soil contaminant loads nor industry type is conclusive, there is a need for a supportive framework for assessing metal spreading to groundwater accounting for site-specific, geochemical conditions.

Risks associated with metal contaminated sites are tightly linked to material leachability and contaminant mobility. In this study, metal solubility and transport were characterized within a glass waste landfill through i) lysimeter-collection of pore water and standardized batch leaching tests, ii) soil profiles extending from the landfill surface, through unsaturated soil underneath, and into the groundwater zone, and iii) groundwater samples upstream, at, and downstream of the landfill. The soil analyzes targeted both pseudo-total and geochemically active concentrations of contaminant metals (As, Cd, Pb, Sb) and basic soil geochemistry (pH, org. C, Fe,Mn). Water samples were analyzed for dissolved, colloid-bound and particulate metals, and speciation modelling of the aqueous phase was conducted. The results revealed a highly contaminated system, with mean metal concentrationsin the waste zone between 90 and 250 times the regional background levels. Despite severe contamination of the waste zone and high geochemically active fractions (80–100%) of all contaminant metals as well as elevated concentrations in landfill pore water, the concentrations of Cd and Pb decrease abruptly at the transition between landfill and underlying natural soil and no indication of groundwater contamination was found. The efficient cation retention is likely due to the high pH. However, the sorption of As and Sb is weaker at such high pH,which explains their higher mobility from the pore water zone into groundwater. The field soil:solution for Pb, ranging from 140 to 2,900,000 l kg−1), despite little variability in basic geochemical variables, which we suggest is due to waste material heterogeneity.

Recognising the need for robust models in predicting groundwater contamination risks from metal(loid)s in contaminated topsoil, this study focuses on the geochemical behaviour of As, Cd, Cu, Pb, Sb and Zn in one of Sweden’s most heavily contaminated areas. Samples were collected from the waste zone and underlying subsoil down to 5 meters and batch experiments were carried out to assess pH-dependent solubility. The results indicate that Cd, Cu, Pb and Zn are efficiently immobilized in the waste zone, while As(V) and Sb(V) are more easily leached. With the exception of Pb and Cu at high pH, the mobilized metals appear to be predominantly in a truly dissolved state, as confirmed by ultrafiltration at 10 kDa.

Speciation modelling using Visual MINTEQ did not suggest a significant role of precipitates such as Zn or Pb arsenates and phosphates, although their involvement could not be ruled out. To better understand sorption/desorption patterns, a multi-surface geochemical model was established, drawing on the Stockholm Humic and CD-MUSIC models for organic matter and Fe/Al (hydr)oxide sorption. However, when default parameters were used, the model consistently overestimated the solubility of Cd, Cu, Pb and Zn in both the waste zone and the uncontaminated subsoil. In contrast, As(V) solubility was generally underestimated, also when the reactive surface area of the Fe- and Al (hydr)oxides was decreased in the model. The model's performance was better for Sb(V), though not without imperfections. When the parameters for organic matter were adjusted such that 100% of the solid-phase organic matter was active with respect to ion binding, but only 25% of the dissolved organic matter, the model description improved considerably for Pb and Cu in the upper soil layers. The model revealed distinct differences in the adsorption behaviour of the metal cations, with Pb being sorbed mostly to Fe/Al (hydroxides), whereas a considerable part of Cu was sorbed to organic matter, particularly in the waste zone.

Possibly, the dissolution of easily weatherable metal-containing mineral phases may have contributed to the poor model performance for Cd, Zn and for Cu in the deeper soil layers, although other factors, such as a contribution of hydrous SiO2 or Mn oxides to metal binding, could not be ruled out. Metal sorption to carbonate phases may also have been a contributing factor in the waste zone. Lastly, the reactivity of Fe- and Al (hydr)oxides may have been overestimated by oxalate extraction when default parameters for high-surface-area ferrihydrite were applied. 

These findings provide valuable insights for environmental management and underscore the need for a more detailed characterization of metal(loid) sorption in contaminated soils, as well as the development of improved modelling strategies to enhance solubility predictions.

This study investigates metal contamination patterns and exposure to Sb, As, Ba, Cd and Pb via intake of drinking water in a region in southeastern Sweden where the production of artistic glass has resulted in a large number of contaminated sites. Despite high total concentrations of metals in soil and groundwater at the glassworks sites properties, all drinking water samples from households with private wells, located at a 30-640 m distance from a glassworks site, were below drinking water criteria from the WHO for Sb, As, Ba and Cd. A few drinking water samples showed concentrations of Pb above the WHO guideline, but As was the only element found in concentrations that could result in human exposure near toxicological reference values. An efficient retention of metals in the natural soil close to the source areas, which results in a moderate impact on local drinking water, is implied. Firstly, by the lack of significant difference in metal concentrations when comparing households located upstream and downstream of the main waste deposits, and secondly, by the lack of correlation between the metal concentration in drinking water and distance to the nearest glassworks site. However, elevated Pb and Cd concentrations in drinking water around glassworks sites when compared to regional groundwater indicate that diffuse contamination of the soils found outside the glassworks properties, and not only the glass waste landfills, may have a significant impact on groundwater quality. We further demonstrate that different mobilization patterns apply to different metals. Regarding the need to use reliable data to assess drinking water contamination and human exposure, we finally show that the conservative modelling approaches that are frequently used in routine risk assessments may result in exposure estimates many times higher than those based on measured concentrations in the drinking water that is actually being used for consumption. (C) 2016 Elsevier B.V. All rights reserved.

Risk assessments of contaminated land often involve the use of generic bioconcentration factors (BCFs),which express contaminant concentrations in edible plant parts as a function of the concentration in soil,in order to assess the risks associated with consumption of homegrown vegetables. This study aimed toquantify variability in BCFs and evaluate the implications of this variability for human exposure as-sessments, focusing on cadmium (Cd) and lead (Pb) in lettuce and potatoes sampled around 22 con-taminated glassworks sites. In addition, risks associated with measured Cd and Pb concentrations in soiland vegetable samples were characterized and a probabilistic exposure assessment was conducted toestimate the likelihood of local residents exceeding tolerable daily intakes. The results show that con-centrations in vegetables were only moderately elevated despite high concentrations in soil, and mostsamples complied with applicable foodstuff legislation. Still, the daily intake of Cd (but not Pb) wasassessed to exceed toxicological thresholds for about afifth of the study population. Bioconcentrationfactors were found to vary more than indicated by previous studies, but decreasing BCFs with increasingmetal concentrations in the soil can explain why the calculated exposure is only moderately affected bythe choice of BCF value when generic soil guideline values are exceeded and the risk may be un-acceptable.

The health risk posed by arsenic in vegetables grown in private gardens near 22 contaminated glassworks sites was investigated in this study. Firstly, vegetable (lettuce and potato) and soil samples were collected and arsenic concentrations measured to characterize the arsenic uptake in the selected crops. Secondly, a probabilistic exposure assessment was conducted to estimate the average daily intake (ADI(veg)), which was then evaluated against toxicological reference values by the calculation of hazard quotients (HQs) and cancer risks (CRs). The results show that elevated arsenic concentrations in residential garden soils are mirrored by elevated concentrations in vegetables, and that consumption of these vegetables alone may result in an unacceptable cancer risk; the calculated reasonable maximum exposure, for example, corresponded to a cancer incidence 20 times higher than the stated tolerance limit. However, the characterization of risk depends to a great extent on which toxicological reference value is used for comparison, as well as how the exposure is determined. Based on the assumptions made in the present study, the threshold levels for chronic non-carcinogenic or acute effects were not exceeded, but the cancer risks indicated highlight the need for further exposure studies, as dietary intake involves more than just homegrown vegetables and total exposure is a function of more than just one exposure pathway. In addition, glassworks sites - and contaminated sites in general - contain multiple contaminants, affecting the final and total risk.