Bruno Guillaume, Arup Geotechnics
In October 2001, the Environment Agency launched proposals to extend its Monitoring Certification Scheme, MCERTS to the chemical testing of soils. The aim of the scheme is to deliver quality environmental measurements with product certification of instruments, the competency certification of personnel and the accreditation of laboratories based on an international standard. In its Land Quality Policy Statement (EAS/2703/1/6/Final 3, available on http://www.environment-agency.gov.uk/commondata/105385/lqpol6re.pdf ), the Agency notes that it will only accept chemical testing data on contaminants in soils that has been produced by laboratories that have been accredited to the BS EN ISO/IEC 17025:2000 quality standard for the testing methods used.
The Agency has provided separate MCERTS performance standards in guidance available from www.environment-agency.gov.uk/business/mcerts. To allow reasonable time for laboratories to complete validation of their testing methods to the Agency’s specification, the Agency will implement this policy over a period of twelve months from 1st April 2002. From the 31st March 2003*, the Agency will require that all chemical testing data on contaminants in soils, which is presented to it in support of regulatory compliance, must have an accompanying estimate of bias and precision and a description of the testing method used, with the laboratory being accredited to the BS EN ISO/IEC 17025:2000 standard for the test method.
The MCERTS proposals were discussed at the annual Contest meeting in June at which I was asked to speak on “the client’s view”. There is no doubt that the risk-based approach to contaminated sites and especially quantitative assessment requires greater confidence in data from site investigations. All laboratories should operate quality control and quality assurance schemes with calibrations, blanks, sensitivity checks and duplicate testing. UKAS accreditation is often quoted as evidence of quality assurance, but it gives no indication of the suitability of test for the intended purpose of the end user. Proficiency testing and participation in schemes such as Contest, Aquacheck or LEAP is a far better indication of a laboratory’s ability to undertake tests reliably. However, the results of proficiency testing are seldom available to third parties, such as those organisations commissioning tests from the laboratories. MCERTS has the potential to provide an indication of data reliability with performance standards set by an authoritative body.
Laboratories have expressed concerns over certain aspects of the MCERTS proposals as applied to the chemical testing of soils. Sampling is potentially a far greater source of data errors than is laboratory analysis, and there is as yet no comparable scheme addressing sampling, in-situ and field tests. The performance standards requested by the Environment Agency are too demanding and prescriptive for certain parameters (e.g. limit of detection of 20mg/kg for sulphates) and ill defined for certain parameters (e.g. are “polyaromatic hydrocarbons” represented by the sum of USEPA priority 16 or defined by other means?).
The Environment Agency is considering certification schemes to address field aspects including sampling. It is worth noting, at this point, that the Agency has produced guidance, (including Technical aspects of site investigation, P5-065/TR and Secondary model procedures for development of appropriate soil sampling strategies, P5-066/TR), though this has been poorly publicised.
Accreditation of laboratories to BS EN ISO/IEC 17025:2000 (General requirements for the competence of testing and calibration laboratories) is significant: clause 4.1.2 states that it is the responsibility of the laboratory to carry out its testing and calibration activities in such a way as to satisfy the needs of the client. The requirement for the laboratory to understand the client’s needs is explicit. Currently, too many clients will commission testing without discussing objectives with the laboratory. Unfortunately, the market is driven by price rather than quality and there are still laboratories that offer methods that are inappropriate. An uninformed client cannot distinguish between good and bad service providers. The introduction of MCERTS and accreditation to BS EN ISO/IEC 17025:2000 could act as a catalyst to encourage the engagement of laboratories earlier in the investigation process, thus ensuring that the laboratory methods are fit for purpose.
MCERTS does not specify the methods of analysis, and proficiency testing shows an astonishingly wide variation in results for certain parameters. This is due to the variety of methods in use and economic pressures in a market where there is over capacity, as well as poor parameter definition and lack of performance standards. Method specification through international standards is an extremely slow process and not favoured in the UK. Research has however been undertaken, funded by the Government and the Environment Agency, on appropriate methods of analysis, and the results of this research are still awaited.
Finally, the risk assessment approach to contaminated land requires further guidance which has yet to be provided by the regulators, including suitability of leaching tests, bio-availability and toxicity assessments. The process of investigation and assessment is complex and potential for errors considerable, but MCERTS should at least address one part of the process and raise the importance of appropriate data.
References: Environment Agency, Performance Standard for Laboratories Undertaking Chemical Testing of Soil, May 2002, Version 1 <http://www.environment-agency.gov.uk/commondata/105385/soiltest.pdf> UKAP, Good Regulation and Competitiveness Network – Environment Sector Study, July 2001, <http://www.chemsoc.org/pdf/ukap/enviro.pdf> Hazel Davidson, VAM Bulletin No. 21, 1999, 4 , <http://www.vam.org.uk/news/news_bulletin.asp>
*Note: The intention to implement to this timetable has been withdrawn to allow a more realistic timescale for the accreditation of laboratories. A statement on the EA website reads:
‘Having recently met with UKAS and discussed the concerns of a number of laboratories with respect to the timescales for compliance, the Agency has decided to revise the phased approach to implementation of its requirements to allow a longer lead in time. In addition, the Agency has decided to take this opportunity to review the detail set out in the performance standard with a view to streamlining the additional requirements over EN ISO/IEC 17025:2000 which are already assessed. We will issue a revised implementation policy and performance standard shortly. In the interim, laboratories are urged to seek the accreditation to EN ISO/IEC 17025:2000 for the chemical testing of soils which is already available.