Contaminated land: chemistry and toxicology aspects of chemical risk assessment
Meeting report by James Lymer
ECG Committee Member
ECG Bulletin January 2011
ECG Committee Member
ECG Bulletin January 2011
Exploring the science applied in models used to assess the risk to long-term human health from contaminants in soil”
This one-day meeting, organised jointly by the RSC Environmental Chemistry Group and the RSC Toxicology Group took place in the Chemistry Centre at Burlington House, Piccadilly on 28th September, 2010.
The meeting was aimed at practitioners involved in assessing and remediating contaminated land, and focused on risk assessment models used to estimate chronic exposure from oral, inhalation and dermal pathways. Forty-three delegates attended from a wide range of organisations including consultancy firms, regulatory authorities, and academia.
Dr Sohel Saikat (Health Protection Agency) opened the proceedings with a presentation entitled Oral bioaccessibility data: missing the point in exposure assessment of soil-borne chemicals. He introduced the concepts of contaminant bioavailability and bioaccessibility and described methods of measurement and applications of data used for contaminated land risk assessment models. Using metal exposure as an example, the analysed total metal concentration in soil does not take into consideration metal speciation, which determines bioavailability and bioaccessibility. Hence total metal concentration ≥ acid extractable ≥ bioaccessible ≥ bioavailable metal concentration.
In contaminated land risk assessment, there is a general conservative assumption that a contaminant is 100% bioavailable following exposure from soil. Exposure models such as CLEA (Contaminated Land Exposure Assessment Model), which can take into account the bioaccessibility of a contaminant, provide a better method for assessing the risks to human health from exposure to contaminated land.
Dr Chris Collins’ (University of Reading) presentation, Modelling plant uptake of organic chemicals – current status and future needs, also discussed CLEA and, in particular, the role that the uptake of organic contaminants by plants has in this model. Routes of transport for organic compounds in different types of plants were discussed and the influence that chemical properties (lipophilicity) has on cellular transport mechanisms described.
Plant uptake models used to predict the uptake of non-ionic chemicals in soil were compared. These range from simple regression models (Travis and Arms) to mechanistic (Trapp and Matthies) and fugacity models (Hung and Mackay). The more complex models tend to require a significant amount of data which are unlikely to be measured by those engaged in contaminated land risk assessment. These models are more sensitive to data variation.
Continuing the theme of the uptake of chemicals by plants, Alan Dowding (Food Standards Agency) reviewed The Food Standards Agency’s approach to assessing uptake of chemicals by crops from contaminated soil. He discussed sampling and analysis of chemicals in soil and crops, including allotments, as part of the regulatory role of the FSA.
Dowding also described the PRISM model, which is used by the FSA to predict inter alia uptake by plants of inorganic chemicals in soil. The PRISM model incorporates data for 42 inorganic contaminants and 129 radionuclides, and uptake by plants is modelled for various transport mechanisms including deposition (wet and dry), absorption, translocations (in xylem and phloem) and root uptake.
The FSA are planning to determine levels of total and inorganic arsenic in fruit and vegetables grown in the UK and to investigate the relationship between arsenic levels in soil and crops, and the extent to which arsenic is taken up by different crop types. The focus will be on soil samples from locations with high arsenic levels of geochemical, mining or industrial origin.
Katy Baker (Arcadis) in her talk Vapour Intrusion in the UK – Where do we go next? described how vapour migration from the contamination of soil by volatile organic compounds (VOCs) is measured and then modelled for CLEA and other exposure models. Inhalation of vapours from volatile organic compounds in soil can be a significant exposure pathway. Exposure to VOCs is assessed by analysing soil and groundwater samples.
The Johnson and Ettinger Vapour Intrusion Model has been incorporated into CLEA. Other models used to estimate vapour intrusion, if the site of interest does not match the generic CLEA model, include CSOIL, VOLASOIL and Biovapour.
In exposure models, all VOCs have traditionally been assessed in the same way irrespective of the different physico-chemical properties of VOCs and how these may affect their environmental impact. This was illustrated with two case studies: (i) a petrol filling station (petroleum hydrocarbons); and (ii) an active industrial estate (chlorinated solvents in the subsurface). When the measured soil vapour concentrations were compared with modelled concentrations, the model tended to over predict the petroleum hydrocarbon concentrations and under predict the chlorinated solvent concentration, highlighting the variation and influence of contaminant properties.
Inhalation as an exposure route continued with a talk by Camilla Pease (Environment Agency) entitled Adverse lung effects and the derivation of inhalation Health Criteria Values (HCVs). Lung toxicology was illustrated with examples from chronic tobacco smoking, occupational exposure (coal mining) and asbestos exposure. The UK risk assessment framework for deriving a HCV is detailed in the Environment Agency guidance document SR2. HCVs may be estimated from in vivo inhalation and dermal studies or from epidemiological data.
Simon Firth (Firth Consultants) (Chemical exposure via inhalation of soil derived dust – modelling versus measurement) explained how the inhalation of soil-derived dusts and their contaminants can be measured, what models are adopted by CLEA, and for which contaminants does the inhalation of dust become significant in terms of affecting the Generic Assessment Criteria (GAC) derivation.
The CLEA model algorithm is used to estimate chronic exposure from outdoor and indoor dust. A parameter in CLEA is used to estimate the proportion of indoor soil derived dust i.e. the Transport Factor which is default at 0.5 in CLEA and indicates that 50% of the PM10 particulates in dust are derived from soil. Firth has derived a value of 0.8 for the Transport Factor from his experiments, which is larger than the CLEA default value. Hence the Transport Factor can vary significantly from site to site and from contaminant to contaminant.
The last two presentations concerned the dermal route of exposure. Alison Mckay (Mckay Environmental Ltd) (The dermal pathway - A touchy subject!) reviewed the dermal exposure pathway, the available data, and uncertainties in the contaminated land risk assessment framework. Estimating dermal exposure to soil is complex, and assumptions are usually made about human behaviour to establish what the frequency and duration of exposure is, and the type of human activities that could lead to exposure to soil. Other considerations are the properties of the skin, the soil, and the contaminants. The skin acts as a living barrier, and uncertainties in estimating dermal exposure include variability in the barrier function. The CLEA model has adopted the Dermal Absorption Fraction value, derived from experimental data for each contaminant. Currently, there is a lack of experimental data on skin absorption; as more data become available, and uncertainties diminish for other exposure pathways, the relative contribution of the dermal pathway will become more important.
A presentation from Tayo Adedeji (Atkins) on Dermal Exposures to soil: metabolism, toxicity and acute duration exposures provided further information on the dermal uptake pathway and the toxic effects of dermal absorption. Effects from dermal exposure may be local or systemic, reversible or non-reversible. Local effects include irritation and defatting by solvents and systemic effects include sensitisation (nickel) and carcinogenesis (benzo[a]pyrene).
The meeting concluded with a question and answer session with the eight speakers as the panel and Simon Firth chair. Given the positive response to this event and to September 2009’s Contaminant Transport and Fate meeting, the ECG would welcome ideas for topics for a third contaminated land forum, again with a chemistry focus, for 2011/2012.
Please email James Lymer via the ECG website with your suggestions: www.rsc.org/Membership/Networking/InterestGroups/Environmental/committee.asp
Copies of the speakers’ presentations from this meeting are available on the ECG website www.rsc.org/ecg.
This one-day meeting, organised jointly by the RSC Environmental Chemistry Group and the RSC Toxicology Group took place in the Chemistry Centre at Burlington House, Piccadilly on 28th September, 2010.
The meeting was aimed at practitioners involved in assessing and remediating contaminated land, and focused on risk assessment models used to estimate chronic exposure from oral, inhalation and dermal pathways. Forty-three delegates attended from a wide range of organisations including consultancy firms, regulatory authorities, and academia.
Dr Sohel Saikat (Health Protection Agency) opened the proceedings with a presentation entitled Oral bioaccessibility data: missing the point in exposure assessment of soil-borne chemicals. He introduced the concepts of contaminant bioavailability and bioaccessibility and described methods of measurement and applications of data used for contaminated land risk assessment models. Using metal exposure as an example, the analysed total metal concentration in soil does not take into consideration metal speciation, which determines bioavailability and bioaccessibility. Hence total metal concentration ≥ acid extractable ≥ bioaccessible ≥ bioavailable metal concentration.
In contaminated land risk assessment, there is a general conservative assumption that a contaminant is 100% bioavailable following exposure from soil. Exposure models such as CLEA (Contaminated Land Exposure Assessment Model), which can take into account the bioaccessibility of a contaminant, provide a better method for assessing the risks to human health from exposure to contaminated land.
Dr Chris Collins’ (University of Reading) presentation, Modelling plant uptake of organic chemicals – current status and future needs, also discussed CLEA and, in particular, the role that the uptake of organic contaminants by plants has in this model. Routes of transport for organic compounds in different types of plants were discussed and the influence that chemical properties (lipophilicity) has on cellular transport mechanisms described.
Plant uptake models used to predict the uptake of non-ionic chemicals in soil were compared. These range from simple regression models (Travis and Arms) to mechanistic (Trapp and Matthies) and fugacity models (Hung and Mackay). The more complex models tend to require a significant amount of data which are unlikely to be measured by those engaged in contaminated land risk assessment. These models are more sensitive to data variation.
Continuing the theme of the uptake of chemicals by plants, Alan Dowding (Food Standards Agency) reviewed The Food Standards Agency’s approach to assessing uptake of chemicals by crops from contaminated soil. He discussed sampling and analysis of chemicals in soil and crops, including allotments, as part of the regulatory role of the FSA.
Dowding also described the PRISM model, which is used by the FSA to predict inter alia uptake by plants of inorganic chemicals in soil. The PRISM model incorporates data for 42 inorganic contaminants and 129 radionuclides, and uptake by plants is modelled for various transport mechanisms including deposition (wet and dry), absorption, translocations (in xylem and phloem) and root uptake.
The FSA are planning to determine levels of total and inorganic arsenic in fruit and vegetables grown in the UK and to investigate the relationship between arsenic levels in soil and crops, and the extent to which arsenic is taken up by different crop types. The focus will be on soil samples from locations with high arsenic levels of geochemical, mining or industrial origin.
Katy Baker (Arcadis) in her talk Vapour Intrusion in the UK – Where do we go next? described how vapour migration from the contamination of soil by volatile organic compounds (VOCs) is measured and then modelled for CLEA and other exposure models. Inhalation of vapours from volatile organic compounds in soil can be a significant exposure pathway. Exposure to VOCs is assessed by analysing soil and groundwater samples.
The Johnson and Ettinger Vapour Intrusion Model has been incorporated into CLEA. Other models used to estimate vapour intrusion, if the site of interest does not match the generic CLEA model, include CSOIL, VOLASOIL and Biovapour.
In exposure models, all VOCs have traditionally been assessed in the same way irrespective of the different physico-chemical properties of VOCs and how these may affect their environmental impact. This was illustrated with two case studies: (i) a petrol filling station (petroleum hydrocarbons); and (ii) an active industrial estate (chlorinated solvents in the subsurface). When the measured soil vapour concentrations were compared with modelled concentrations, the model tended to over predict the petroleum hydrocarbon concentrations and under predict the chlorinated solvent concentration, highlighting the variation and influence of contaminant properties.
Inhalation as an exposure route continued with a talk by Camilla Pease (Environment Agency) entitled Adverse lung effects and the derivation of inhalation Health Criteria Values (HCVs). Lung toxicology was illustrated with examples from chronic tobacco smoking, occupational exposure (coal mining) and asbestos exposure. The UK risk assessment framework for deriving a HCV is detailed in the Environment Agency guidance document SR2. HCVs may be estimated from in vivo inhalation and dermal studies or from epidemiological data.
Simon Firth (Firth Consultants) (Chemical exposure via inhalation of soil derived dust – modelling versus measurement) explained how the inhalation of soil-derived dusts and their contaminants can be measured, what models are adopted by CLEA, and for which contaminants does the inhalation of dust become significant in terms of affecting the Generic Assessment Criteria (GAC) derivation.
The CLEA model algorithm is used to estimate chronic exposure from outdoor and indoor dust. A parameter in CLEA is used to estimate the proportion of indoor soil derived dust i.e. the Transport Factor which is default at 0.5 in CLEA and indicates that 50% of the PM10 particulates in dust are derived from soil. Firth has derived a value of 0.8 for the Transport Factor from his experiments, which is larger than the CLEA default value. Hence the Transport Factor can vary significantly from site to site and from contaminant to contaminant.
The last two presentations concerned the dermal route of exposure. Alison Mckay (Mckay Environmental Ltd) (The dermal pathway - A touchy subject!) reviewed the dermal exposure pathway, the available data, and uncertainties in the contaminated land risk assessment framework. Estimating dermal exposure to soil is complex, and assumptions are usually made about human behaviour to establish what the frequency and duration of exposure is, and the type of human activities that could lead to exposure to soil. Other considerations are the properties of the skin, the soil, and the contaminants. The skin acts as a living barrier, and uncertainties in estimating dermal exposure include variability in the barrier function. The CLEA model has adopted the Dermal Absorption Fraction value, derived from experimental data for each contaminant. Currently, there is a lack of experimental data on skin absorption; as more data become available, and uncertainties diminish for other exposure pathways, the relative contribution of the dermal pathway will become more important.
A presentation from Tayo Adedeji (Atkins) on Dermal Exposures to soil: metabolism, toxicity and acute duration exposures provided further information on the dermal uptake pathway and the toxic effects of dermal absorption. Effects from dermal exposure may be local or systemic, reversible or non-reversible. Local effects include irritation and defatting by solvents and systemic effects include sensitisation (nickel) and carcinogenesis (benzo[a]pyrene).
The meeting concluded with a question and answer session with the eight speakers as the panel and Simon Firth chair. Given the positive response to this event and to September 2009’s Contaminant Transport and Fate meeting, the ECG would welcome ideas for topics for a third contaminated land forum, again with a chemistry focus, for 2011/2012.
Please email James Lymer via the ECG website with your suggestions: www.rsc.org/Membership/Networking/InterestGroups/Environmental/committee.asp
Copies of the speakers’ presentations from this meeting are available on the ECG website www.rsc.org/ecg.