R G Clark, CL Associates

Most AGS members are aware or should be aware that ICRCL Guidance Note 59/83, 2nd edition was withdrawn by DEFRA in a letter from Steven Griffiths of the Contaminated Land Branch on 20 December 2002. The letter referred to the fact that the CLEA package (published by DEFRA and the Environment Agency), consisting of Contaminated Land Reports (CLRs) 7 to 10, the CLEA 2002 software, certain toxicological reports (TOX) and certain Soil Guideline Values (SGVs) is considered by DEFRA to represent the key instrument for generic assessment of the human health risks from contaminated land.

The reasoning put forward for the withdrawal of ICRCL 59/83 was that the guideline values are out of date and that they are not in line with the current statutory regime (Part IIA of the Environmental Protection Act 1990) and associated policy.

Up until that time many practitioners had relied heavily on the use of the ICRCL 59/83 guideline values for assessing human health risks associated with contaminated land. Others had already started to use alternative risk based methodologies such as RBCA, R&D P20, SNIFFER and of course CLEA. In many instances ICRCL or Dutch guideline values were used as a first screening before progressing to a quantitative risk assessment (QRA) for those contaminants of particular concern that were above the guideline values for a particular site.

The letter dated 20 December 2002 was immediately followed by a Briefing Note, also dated December 2002, from the Contaminated Land Branch of DEFRA. This briefing note emphasised that ICRCL 59/83, and especially Tables 3 and 4, should no longer be used. It can be noted that other ICRCL guidance has not been withdrawn. The DEFRA Briefing Note recognises the continued use of these other ICRCL guidance documents provided that they are not used as the sole source of information on which decisions are based.

A number of Local Authorities have now adopted the position that for the assessment of contaminated land, only UK based guidance is applicable. They will, therefore, no longer accept both ICRCL 59/83 and the Dutch Guidelines. In effect this means that where no published SGV exists then it is necessary to carry out a site specific QRA for each contaminant even before an initial screening can be undertaken. This constitutes a potential problem where there is also no TOX Report available, in that each organisation that carries out such risk assessments has to source appropriate and verifiable toxicological data.

An update of the existing CLEA Model was published in March 2003. A new version of the CLEA Model (open architecture version) is anticipated which is a spreadsheet version that allows more user-functionality.

SGVs have been published for Arsenic, Cadmium, Chromium, Lead, Mercury, Nickel and Selenium. TOX Reports are available for Benzo (a) pyrene, Benzene, Inorganic Cyanide, Dioxins, Furans and Dioxin like PCBs but no SGVs. The Environment Agency have stated that they will be issuing additional SGVs in batches of 5 or 6 at quarterly intervals over the next 2 years.

But what is to happen in the interim ? – chaos according to some. At various conference and technical committee venues both consultants and Local Authorities have expressed their concerns over the rapid withdrawal of ICRCL 59/83 without a sufficiently comprehensive UK alternative being available.

In conclusion, AGS members should be aware of their responsibilities to their clients. If a geoenvironmental specialist were to continue to use ICRCL 59/83 in reports that they prepare for a client and those reports are subsequently submitted for approval to a regulator (such as part of a Planning Application) or to some other body such as a funding organisation and are rejected due to the use of ICRCL, it may very well be considered that the specialist has failed in its duties to the client because it has used reference material that has been officially withdrawn.

Equally there is a risk that the use of Dutch guidelines could be rejected. This will depend on the approach of the particular regulator. The alternative, apart from using the CLEA Model, is to use SNIFFER etc (see above). However, for all of these methods it is necessary to research toxicological data and to be able to verify these data within a UK context. Site specific criteria are therefore being based on a variety of sources of toxicological data of varying quality. How are those, such as regulators, who have the task of approving these derived values going to make judgements on the reliability and applicability of the toxicological data ?

There is no easy answer at present.