CONTENTS

IDENTIFIED REQUIREMENT FOR ATRISKSOIL
IDENTIFIED REQUIREMENT FOR PARKS, PLAYING FIELDS AND OPEN SPACES SSVs
METHODOLOGY FOR DEVELOPMENT OF SOIL SCREENING VALUES
METHODOLOGY FOR DEVELOPMENT OF PARKS, PLAYING FIELDS AND OPEN SPACES SSVS
APPLICATION TO LAND CONTAMINATION PROBLEMS
CURRENTLY AVAILABLE SOIL SCREENING VALUES

IDENTIFIED REQUIREMENT FOR ATRISKSOIL

ATRISKSOIL is an online database providing Soil Screening Values (UK applicable) for common contaminants not currently covered by Soil Guideline Value (SGV) reports issued by DEFRA and the Environment Agency.

The Interdepartmental Committee for the Redevelopment of Contaminated Land (ICRCL) has issued guidance in the past on the safe redevelopment of contaminated land. These guidelines were previously used as a primary tool for assessing the hazards presented by contamination on sites, but were officially withdrawn by the Department of Environment, Food and Rural Affairs (Defra) in December 2002.

In 2002 Defra and the Environment Agency (EA) launched the CLEA (Contaminated Land Exposure Assessment) model software together with associated documentation and a limited suite of Soil Guideline Values (SGVs). The SGVs have been derived using the CLEA model according to three typical land uses and are applicable to long-term human exposure to soil contaminants. The three land uses are:

• Residential (with and without home grown vegetables);
• Allotments; and
• Commercial/industrial.

To date, Defra and the EA have released ten SGVs, seven of which relate to a metal contaminant (arsenic, cadmium, chromium, lead, mercury (inorganic), nickel and selenium) along with toluene and ethylbenzene. There is a requirement for further screening values so that a wider array of commonly encountered contaminants, both organic and inorganic, can be effectively assessed at a generic level.

Where SGVs are not available, Soil Screening Values (SSVs) can be calculated using the CLEA software which follows the methodology laid out in Contaminated Land Reports (CLR) 7-10. However, the CLEA software has limited functionality and contains algorithms which the EA has publicly expressed its intention to update. As a consequence of this, some of the screening values generated by the CLEA software may not adequately reflect site conditions and in some instances are unduly conservative.

Atkins developed a set of screening criteria for the following land uses:

• Residential with plant uptake
• Residential without plant uptake
• Commercial/industrial
• Parks
• Playing Fields and
• Open spaces.

Back To Top

IDENTIFIED REQUIREMENT FOR PARKS, PLAYING FIELDS AND OPEN SPACES SSVs

Defra and the Environment Agency have not developed a conceptual model for recreational open spaces within CLR10 and have therefore not published any Soil Guideline Values (SGVs) for this land-use. The Environment Agency states within the frequently asked questions (FAQ) sheet (available from the EA website, www.environment-agency.gov.uk) that this is for two reasons:

• sufficient information needs to be generated and documented to detail an acceptable conceptual exposure model for this land-use. For example "Who visits parks?", "How long is an average visit to a local park?"
• the diversity in leisure land-uses such city parks, school playing fields and golf courses may not be suited to the derivation of a single standard land-use.

Atkins has found that a number of the projects with which we are involved require screening values for a variety of types of recreational open space. It was therefore decided to explore the potential for developing Soil Screening Values (SSVs) for parks, playing fields and open spaces. Detailed research was carried out by Atkins to collect data to develop conceptual models for parks, playing fields and open spaces. The methodology for the derivation of the values is given below.

Back To Top

METHODOLOGY FOR DEVELOPMENT OF SOIL SCREENING VALUES

Atkins' approach has been to develop a set of Soil Screening Values (SSVs) which can be applied to a broad range of sites. The process followed to derive these values has many similarities with site-specific risk assessment although the conceptual model has been made sufficiently sensitive for the resulting SSVs to be used as relatively conservative screening values.

In November 2005 an update to the CLEA 2002 software was released called CLEA UK. The CLEA UK software was released as a beta version and has currently only been tested by the Environment Agency for a limited range of chemicals and scenarios. Therefore the Environment Agency recommended that at the present time it is used cautiously until a final version is released. Given this statement, Atkins has not derived Soil Screening Values (SSVs) using the CLEA UK model at the time of writing.

RISC Workbench 4.0 (sometimes referred to as BP RISC) was selected by virtue of the algorithms incorporated within it and because it is a scientifically robust and peer reviewed quantitative human health risk assessment tool. Atkins modified the database to ensure the model was consistent with the UK approach. The process included:

• Development of default conceptual site models, including all input parameters;
• Replacing Slope Factors and Reference Doses with Index Doses and Tolerable Daily Intake values in order to be consistent with CLR9 (as per Atkins standard land-uses SSVs). This included incorporation of childhood factors into TDI's to derive TDSI values for the different age ranges. Childhood factors vary with age and therefore this included a substantial proportion of the work for the non-standard land uses;
• Using UK default values, as detailed in CLR10, where possible and appropriate (e.g. soil type, soil porosity, bodyweights, respiration rates); and
• Replacing other parameters within RISC 4.0 with inputs appropriate to the conceptual model using weighted averaging approaches for conversion into a form suitable for input to RISC 4.0.

The direct contact and inhalation exposure routes have been evaluated independently of one another to identify the most sensitive pathway. The SSV adopted for a given land use scenario is the most stringent screening value of all those relevant to the site.

The intention for the non-standard land uses has been to provide a set of SSVs that could be applied to parks, playing fields and open spaces problems for both contaminants for which SGVs are available for standard land uses (except for lead¹ ) and for the current set of SSV contaminants. These contaminants are considered to be applicable to a wide range of sites across the UK and were selected as priorities based on the CLR8 list of potential contaminants for the assessment of contaminated land and client requirements.

The process was subject to rigorous quality control and assurance procedures. The validity of this approach was proven by running the BP RISC 4.0 model in parallel on two computers.

A summary of the critical assumptions incorporated into the Atkins SSVs, which follows a reasonable maximum exposure approach, is included to provide technical backup to the use of the values.

Back To Top

METHODOLOGY FOR DEVELOPMENT OF PARKS, PLAYING FIELDS AND OPEN SPACES SSVS

Atkins approach was to develop a set of Soil Screening Values (SSVs) that can be applied to three different recreational settings:

• parks;
• playing fields; and
• open spaces.

In order to define the conceptual models for these land-uses, data was compiled from the following sources:

• observations made by visiting a wide range of parks, playing fields and open spaces;
• questionnaires/interviewing people in parks, playing fields and open spaces; and
• data provided by Local Councils.

A wide range of information was collected regarding receptors using the sites and the sites themselves, including:

• age of people visiting and assessment of most common age group present;
• number of hours spent during a visit;
• frequency of visits (days/week summer/winter);
• activities undertaken by receptors;
• general observations about the parks, playing fields and open spaces visited, their surroundings and features present.

It was considered appropriate in relation to the conceptual model for everyday activities to use a mean receptor with upper level exposure factors. Descriptive statistics were used to calculate the upper level exposure factors, which were taken as the 90th percentile in most cases. This is consistent with the general level of conservatism adopted for standard land use ATRISKSOIL SSVs.

Back To Top

APPLICATION TO LAND CONTAMINATION PROBLEMS

Available for both commercial/industrial and residential (with and without plant uptake) land use scenarios, ATRISKSOIL enables auditable, consistent evaluation of land contamination problems. SSVs are designed, consistent with SGVs, to provide a preliminary, generic assessment of the risks to human health arising from the presence of soil contamination. The following attributes illustrate the utility of SSVs in practical application.

• Comparison with site data informs decision-making with regard to the need or otherwise for further site evaulation and/or remediation measures.
• Reference to the SSVs enables potential inaccuracies to be identified in screening values applied in submitted reports, assisting the review process.
• SSVs simplify the process for land contamination assessment, saving time, money and effort for contaminated land professionals.

As with SGVs, the Atkins screening criteria apply where actual site conditions are equally or less sensitive than the hypothetical conceptual model and where it is clear that there are no pathways relevant to the site that might have been overlooked or underestimated by the generic assumptions.

The toxicological data used in the assessment are checked against original sources on a regular basis, and where changes are determined, any modifications are made to the SSV values, ATRISKSOIL is updated and subscribers are notified of the adjustment. For example, recent publication of the phenol TOX report by Defra and the EA led to the modification of this SSV; the value for residential with plant uptake changed marginally (increased by approximately 8%).

Back To Top

CONTAMINANTS COVERED BY ATRISKSOIL

Currently available SSVs are listed below. The ATRISKSOIL database is under continual development and further contaminants will be evaluated and provided to subscribers in due course. Additional contaminants can be assessed on request.

SSV's have been generated for 53 contaminants. These include:

• BTEX - benzene, xylene;
• TPH - speciated aliphatic/aromatic and carbon banded to TPHCWG criteria;
• PAHs - acenaphthene, anthracene, benzo(a)anthracene, benzo(a)pyrene, benzo(b)fluoranthene, benzo(ghi)perylene, benzo(k)fluoranthene, chrysene, dibenzo(ah)anthracene, fluorene, fluoranthene, indeno(1,2,3-cd)pyrene, naphthalene, pyrene;
• VOCs - chlorobenzene, trichloroethene, vinyl chloride;
• phthalates;
• phenol;
• free cyanide;
• beryllium;
• vanadium;
• copper;
• zinc;
• 1,1,1- trichloroethane;
• 1,2-dichloroethane;
• Carbon disulphide,
• Carbon tetrachloride;
• Tetrachloroethanes (PCAs) and
• Tetrachloroethene (PCE).

In addition SSVs have been generated for the new Parks, Playing Fields and Open Spaces land uses only for the following contaminants. SGVs are available for the standard land uses for these contaminants:

• Arsenic;
• Cadmium;
• Chromium;
• Ethylbenzene;
• Mercury;
• Nickel;
• Selenium; and
• Toluene.

Back To Top

¹ The approach to assessing lead in the UK is different from other contaminants. This is because there is sufficient information on lead to enable health criteria values to be set on the basis of internal exposure (e.g. uptake) and therefore soil screening values cannot be derived using the RISC model.