Contaminated sites are often characterized by the presence of a multitude of chemicals. Although it is possible that each chemical on its own does not give rise to human health risks, the question arises whether the assessment of combined (or cumulative) exposure to certain chemicals would not change the outcome of the risk assessment. Review of approaches in cumulative risk assessment reveals a wide range of methodologies, varying from consideration of the organ where the effect occurs till detailed evaluation of mode of action. Cumulative risk assessment should consider both aspects of exposure and risk characterization. In contaminated sites assessment in Flanders, the exposure assessment typically is of a cumulative type, so the focus of the developed methodology is on the risk characterization part. The assumption of additivity (no interaction) was chosen as this is the most appropriate option in case of low dose exposures. For each regulated chemical, critical and subcritical endpoints of effects were listed by exposure route. Distinction was made between local and systemic effects. Endpoints were grouped into effect classes, which were the basis to combine individual risks. For non-threshold carcinogens, no distinction is made with regard to affected organ. The key table with effect classes and allocation of each chemical to its relevant classes per route of exposure forms the basis of the further methodology, that describes approaches to look at cumulative risk assessment in a) site screening based on remediation values, b) site risk assessment, and c) urgency for remediation. As introduction of cumulative risk assessment in contaminated sites assessment could have a vast impact on dossier outcomes, an approach for use in current practice focusing on remediation urgency was developed. In parallel, a preferred approach was developed as well for use throughout all steps of the site assessment (screening, risk assessment, urgency). The latter approach, which extends on the first approach, can be adopted gradually as part of a review of chemical information and remediation values.
The approach will be demonstrated for a number of cases.
Background
It is well established that total concentrations of soil contaminants are useful to indicate pollution; however they do not necessarily indicate risk. Alternative measures can be used to denote the bioavailable fraction, the so-called bioavailability of soil contaminants. If bioavailability is accounted for in risk assessment models, the accuracy will be improved and this would lead to more reliable decisions on the need to remediate and on how much soil needs to be remediated. Risk-based management approaches based on bioavailability principles have a potential to be more cost effective than conventional approaches based on total concentrations. Furthermore, it will open up for alternative site specific management methods based on immobilization of contaminants (reducing bioavailability). To date, many soil testing methods have been developed to predict uptake, toxicity and degradation potential of soil contaminants, but no generally accepted methodology to incorporate contaminant bioavailability in risk assessment models exist. Within the SNOWMAN funded project IBRACS we have evaluated the option to use a passive sampler method, in combination with the equilibrium partitioning theory, as a basis for a risk assessment framework. The proposed framework, with examples, will be presented at the conference.
Methods
An equilibrium passive sampler polyoxymethylene (POM) was used to assess the bioavailability of native polycyclic aromatic hydrocarbons (PAHs) in 22 diverse historically contaminated soils (coke work, gas work and wood tar sites), alongside the lipid concentrations in exposed worms (Enchytraeus crypticus). For details about methods and results, see Arp, et al. (2014).
Results and discussion
The soils studied covered a wide range in soils properties, including texture, pH and organic carbon content. The amount of total organic carbon in the soils (TOC) varied from 2 – 49%. Some samples were low in black carbon (4% of TOC), whereas others, particularly those from coking sites, were rich in black carbon (26 – 95% of TOC). Total concentrations of PAHs in soils varied considerably (0.27 - 2651 µg/g); so did the corresponding POM derived pore water concentrations (0.02 - 460 µg/). One major finding was that the TOC normalized partition coefficients for PAHs was about one order of magnitude higher than those recommended by national agencies, like the United States Environmental Protection Agency (USEPA) for sediments and the Netherlands' National Institute for Public Health and the Environment (RIVM) for soils and sediments, i.e. the sorption of PAHs was significantly stronger in the historically contaminated soils than in “spiked soils” normally used in toxicity experiments. This illustrates the need to actually measure pore water concentrations in historically contaminated soils as a first step in a site specific risk assessment that accounts for bioavailability.
Soil quality standards and critical limit values for non-polar organic compounds, like PAHs, are in most countries based on the assumption of equilibrium partitioning. According to this theory, freely dissolved PAHs in the pore water are in equilibrium with both the soil organic matter component and the lipid phase of soil organisms. Our results support that the assumption of equilibrium partitioning also holds for diverse historically contaminated soils; we found strong correlations between pore water concentrations and lipid concentrations for the investigated PAHs.
A key issue in a risk assessment framework that uses chemical methods for assessing a “bioavailable” concentration or fraction is to develop a reference system to which this concentration or fraction can be related. In this respect we draw on a recent RIVM compilation (Verbruggen, 2012). Here, “critical lipid concentrations” for a wide range of organisms (soils, sediments and waters) were presented. The critical lipid concept is based on the assumption that toxicity of individual PAHs is similar after entering the cell membrane. The RIVM compilation resulted in two proposed “critical lipid concentration”, corresponding to two sets of critical pore water concentrations for individual PAHs, indicating “no risk” (Maximum Permissible Concentration, SRC) or “serious risk” (Serious Risk Concentration, SRC).
We propose the following scheme to include equilibrium-based chemical bioavailability tests in site specific ecological risk assessments of PAHs contaminated soils: 1) Determine pore water concentration of freely dissolved PAHs, 2) Relate individual concentrations to risk limits (e.g. RIVM’s MPC or SRC values), using the toxic unit approach, 3) Assume additive effect and calculate the toxic unit value (if > 1, risk). This procedure is in line with the one proposed by Brand et al. (2013). The procedure was applied on two Swedish PAH contaminated sites and the outcome was compared with an assessment based on the Swedish generic guideline values. The comparison showed that the number of samples indicating risk to soil organisms decreased from 80% to 20% at one site, and from 90% to 55% at the other, when applying the proposed procedure. Accordingly, the time and money invested in extra POM analyses, which are similar to or cheaper than soil analysis, are likely to be paid off during the remediation phase.
References
Arp, H. P. H., S. Lundstedt, S. Josefsson, G. Cornelissen, A. Enell, A.-S. Allard and D. B. Kleja (2014). Environmental Science & Technology, 48, 11187−11195.
Brand, E., Lijzen, J., Peijnenburg, W., Swartjes, F. 2013. J. Hazard. Mater. 261, 833−839.
Verbruggen, E. M. J. (2012). RIVM report 607711007.
In the Netherlands, in case of expected ecological risks caused by soil pollution, the Triad approach can be used to assess site specific ecological risks. This approach combines and integrates three lines of evidence from a site to give a complete and realistic overview of the site specific ecological risks. The three fields of expertise that are combined are environmental chemistry, toxicology (bioassays) and ecological field observations. Although this method is fully validated and has been used in the Netherlands ever since it’s development (Chapman et al. 1987, Nobis 98-1-28, Jensen and Mesman 2006) a standard and legal protocol for this type of research was missing.
From an evaluation of all Triad projects in the Netherlands (project management by Bioclear) it was concluded that authorities find it hard to judge on specialized work which is not formalised in legal protocols (Wagelmans et al., 2009) which hampers decision making. In 2009 the Dutch Council of State (Raad van State) has rejected the results of a specific Triad research. Because no protocol or guidelines for sampling and sampling strategy for Triad research were present, the research protocol followed was compared to the protocol for standard chemical soil research. However, the goals of Triad research significantly differ from the goals of a standard soil research. Therefore, the protocol for standard soil research is not applicable to Triad research and alignment of protocols for standard chemical soil research and Triad research was needed.
In order to prevent future rejections of Triad research a technical guideline for sampling and sampling strategy was developed by Bioclear commissioned by SIKB (SIKB BRL protocol 2301). This protocol describes how many samples need to be taken in a given situation, how samples for different parts of the Triad need to be taken, and how to choose the reference sample. It also describes which choices need to be made and how these choices need to be documented in the sampling strategy.
At the same time a Dutch National Standard (NEN) was developed by RIVM, Alterra and NEN - in collaboration with Grontmij, Tauw, Dienst Vastgoed Defensie, Province Zuid Holland and Bioclear - named “Soil- Process of site specific ecological risk assessment of soil pollution”. This standard describes the process of ecological risk assessment (from the beginning of the project until the final decision). Based on soil use, the ecological constraints are determined by the project group (consisting of stakeholders, authorities and scientists) as well as critical ecological aspects of the polluted site. Subsequently scientists compose a research plan (based on above mentioned protocol SIKB BRL 2301). Together with stakeholders and authorities test criteria per test are determined. Agreements are made about the use of site specific references, weighing of results, method of risk assessment, site specific risk boundaries and handling uncertainties. After that, the site specific risk assessment is carried out by the scientists. The standard is meant to bring scientists, stakeholders and decision makers together. Because decision makers are involved in the project from the beginning, it is easier to make well funded decisions based on site specific ecological risk assessment studies. At this moment ISO is using both the standard and the technical guideline to make an international standard for Triad research.
During development of both the technical protocol and the Dutch National Standard we’ve tested the draft versions on practical applicability on the topsoil of a former landfill polluted with heavy metals. The site is being used for agriculture, grazing of sheep and cows. Workshops have been organized with all stakeholders (farmer, owner of the site, water regulatory authority, two different local authorities, the provincial authority and scientists). After this workshop the ecological risk assessment has been carried out.
The Dutch National Standard on ecological risk assessment and the technical guideline on sampling and sampling strategy are valuable tools in risk communication. It increases acceptance for the research plan, the results and the final decision because stakeholders have influence on the process. In the process their questions and objections will be answered and solved in an early stage. Because of this acceptance by stakeholders, it is easier for decision makers to make a final decision based on site specific research, especially now it has been formalized in standard protocols. By using the standard and technical guideline risk assessment is not a project for only scientists anymore but it becomes a process for both decision makers and scientists together. Also legal authorities can determine whether the research was performed correctly following the technical guidelines which is positive for future acceptance of Triad projects.
Risk assessment of Urban Gardening in Copenhagen
Marlies Warming a,b, Mette G. Hansen a, Peter E. Holm a, Jakob Magid a, Thomas H. Hansen a, Stefan Trapp b,*
a Department of Plant and Environmental Sciences, Faculty of Science, University of Copenhagen, Thorvaldsensvej 40, 1871 Frederiksberg C, Denmark.
b Department of Environmental Engineering, Technical University of Denmark, Miljoevej 113, 2800 Kongens Lyngby, Denmark.
* sttr@env.dtu.dk
Urban gardening is hip, and people in the metropoles all over the world start to grow their own food. At the same time, most if not all urban soils are polluted with heavy metals and other contaminants. In continued projects, we have for several years measured produce from the City of Copenhagen, such as potatoes, carrots, kale, radish and apples. The concentrations of arsenic (As), cadmium (Cd), copper (Cu), chromium (Cr), nickel (Ni), lead (Pb) and zinc (Zn) were measured by ICP-OES or ICP-MS. Concentrations of the platinum group elements platinum (Pt), palladium (Pd), iridium (Ir), rhodium (Rh) and ruthenium (Ru) were also measured, but with lower frequency.
Concentrations in soil were partly below the soil quality standards set by the Danish EPA. The most polluted soils were, however, clearly above, with Pb up to 600, As 22, Cd 2.6 and Zn 1200 mg/kg dw. Concentrations of the platinum group elements in soil were low (Pt, Ir, Rh, Ru < 0.05 mg/kg) except Pd (max 1.4 mg/kg).
The European Union released legal standards for Cd (0.05 to 0.2 mg/kg dw) and Pb (0.3 mg/kg dw) in vegetables. We found very few concentrations above (Cd in spinach, 1 case). For all other heavy metals, we conducted a risk assessment based on established acceptable daily intake ADI. It was assumed that urban gardeners supply 10% of their vegetables and fruit consumption from urban gardens, and that overall consumption was the same as the Danish average. Summarized, we found no reason of concern. For no heavy metal, the intake with urban vegetables and fruits was above ADI or even close. Intake of Pb and As would increase, compared to Danish average, while intake of Cd and Ni was less than with commercial products. We did not find legal standards or ADI for the platinum group elements. Concentrations in lettuce were higher than in spinach or radish and reached 0.11, 3.81 and 1.37 mg/kg dw for Pt, Pd and Ir, but were ≤ 0.01 mg/kg for Rh and Ru.
The consideration of direct soil ingestion by adults (50 mg/d) and children (4-13 years, 200 mg/d) gives an intake of Pb above ADI at the medium and highest polluted soils. In addition, the contribution of As via soil ingestion is of relevance (40% of ADI). This risk assessment assumes daily ingestion and lifelong exposure, which is probably not the case for urban gardening.
In conclusion, we did recommend the people of Copenhagen to continue with their urban garden activities, but to take measures to reduce the attachment and ingestion of soil.
Pyrite Cinder Waste Deposition Scherpekamp
Introduction: Worldwide millions of tons of industrial solid waste such as blast furnace slag, steel slag, non-ferrous metallic slag, coal ash, coal cinder, mining waste rock, mill tailings, etc. is generated from industrial production activities every year. In the past huge amounts of these wastes were used as backfill in open mine pits often containing hazardous heavy metals with potential environmental risks. In developing countries this probably is still the case.
Problem: Back in the 1970s pyrite cinder waste from the Duisburger Kupferhütte (Germany) were deposited in clay pits at the premises of a brick factory on the riverbank of the Rhine in the Netherlands. This iron oxide pyrite cinder waste contained varying concentrations of heavy metals and metalloids and notable contained lead, zinc, arsenic and copper. Previous soils surveys carried out over the years led to the conclusion that there was a high risk of contaminants spreading in groundwater. These studies were carried out using standard soil survey techniques such as auger drilling.
ARCADIS has studied several remedial options: in situ immobilization, excavation and isolation. In order to select the right solution, evaluation of these techniques was based on a detailed geochemical study. Questions to be answered were:
• What is the real risk of leaching?
• What natural processes take place in the soil?
• How will the geochemical system respond to the remediation technique chosen?
The study was conducted recruiting qualified personnel for every stage of the work. Drilling and sampling was performed under supervision of qualified geochemists. Chemical analyses were carried out using a portable XRF (X-ray fluorescence) to gather a high resolution dataset. XRD (X-ray diffraction) to determine primary and secondary cinder and soil mineralogy in order to estimate the vulnerability of contaminants for leaching. SEM (scanning electron microscope) for measuring the distribution of heavy metals and arsenic in the detected mineralogy. All analytical work was performed or checked by qualified geochemists. Based on the various datasets a geochemical reconstruction of events was made.
Conclusion: The cinder studied was high in lead, zinc, arsenic and copper containing levels usually considered economically suitable for extraction. Most important minerals still present in the slag were: iron oxides (magnetite/maghemite/hematite), anglesite (lead sulphate), scorodite (iron arsenate) and beudantite (lead-iron-sulfate-arsenate). All these minerals are insoluble at given conditions. Scorodite might decompose under reducing circumstances.
In the first years after deposition an acidic liquor of acidic Zinc sulphate started to leach out and reacted with calcite and clay in the subsoil forming voluminous zinc carbonates, zinc silicates and gypsum. As a result an impermeable mineral layer was formed preventing any further contact between groundwater and pyrite cinder.
All together the geochemical situation is judged to be stable and no further action is considered to be necessary. Most heavy metals in the cinder are locked away in hardly soluble or insoluble minerals and the self-sealing layer at the cinder/soil boundary prevents further migration. The observed contamination of groundwater was predominantly caused by drilling undertaken during the initial site assessment phases.
Presentation: We will present the results of the geochemical study and the consequences of it for the conceptual site model as well as the remedial options considered.