What is the way forward for contaminated land?
Developments in contaminated land have tried to provide a more consistent approach to risk assessment, but these changes may not be the nirvana that was hoped for. Clive Griffiths explains whyOver the past five years there have been major changes in the field of contaminated land, not least with move towards the contaminated land exposure assessment (CLEA) probabilistic model. The model is used to generate soil guideline values (SGVs) against which the assessment of data produced from site investigations can be made.
The introduction of CLEA however has not met with widespread praise in the industry - not least from the aspect that the computer modelling system is difficult to understand and the site-specific assessment criteria obtained are not particularly reproducible.
In parallel with the introduction of CLEA, the MCERTS scheme for quality control within chemical analysis of soils has also been progressing. The scheme can be summarised as a system with controls to produce results of soils analysis which are precise, accurate and reproducible.
Too much room for error
However, the errors on sampling contaminated soils, not least from the frequency of locations chosen, are immense. Using the analysis of metals as an example, toxicology changes are dependent on what compound the metal is present as - for instance, nickel oxide will have a different toxicology to nickel sulphate.
Since the routine total analysis of metals is, in reality, that of acid extractable metal, how can the resultant data be compared against the limits generated under CLEA? I believe there is a fundamental problem here and there are strong arguments against the need to spend more time and effort on greater analytical precision in the laboratory.
Defra has now released a consultation document The Way Forward for contaminated land which promotes, in part, a move to a more deterministic model for the generation of SGVs which may change the whole assessment approach.
Defra asked for comments on the document but by early February this year, only two responses had been received. The old ICRCL limits that became defunct in 2002 were a deterministic approach to assessment. They were removed primarily because they were regarded as unscientific, and replaced by the probabilistic approach being the CLEA route.
While this was a positive move to a more scientific approach, it had difficulties in that legislation under Part IIA of the Environmental Protection Act had generated a formal definition of contaminated land that included the word 'significant'. The CLEA approach is based purely on harm to human health and does not include the term 'significant'. It should not be used to determine whether land is contaminated, but unfortunately, has been by some local authorities which has resulted in some sites being wrongly diagnosed.
So what is the way forward? One route is to provide an additional set of guideline values for those determinants where there is a significant risk. But what risk levels do we currently use? There is already a dual standard in the CLEA approach between radioactivity and other contamination in that the acceptable risk levels for the former are set at three orders of magnitude higher than the latter.
A call for consistency
This will change but it is not planned to provide full consistency between the two. But consistency is what is needed. The industry has always bemoaned the inconsistencies in approach between different regulators around the country but can you blame the regulators when the risk ratings are different.
If we move towards two guideline values for different circumstances are these to be used in the same way as the old ICRCL limits were designed to be? In other words, the threshold limit - below which the material is white and represents tolerable risk, and the action limit - above which the material is black and represents a significant risk. The shades of grey allowed consultants to use their professional judgement widely.
The ICRCL limits were first released 24 years ago and one wonders why we appear to be going full circle and moving away from the more risk-based assessments of the past five years. Could this be the death knell of quantitative risk assessment? If so, that would be a shame.
Moving to a simple comparison system may be easier to understand, but will perhaps bring us back to viewing the assessment of contaminated land as a 'black art'. This will not enhance the reputation of an industry that has historically suffered from numerous individuals and firms offering services in areas where they do not have the expertise or experience to provide sound advice.