Work is ongoing at the former Vospers shipyard (Woolston Riverside, Southampton) to develop a solution for contamination issues that include asbestos in soils. CampbellReith , on behalf of the South East England Development Agency (SEEDA), has implemented a series of trials and research experiments to help clarify the level of risk presented by varying levels and types of asbestos in soils.
This will aid regulator approval on the appropriate level and type of remediation in the absence of standard assessment criteria for this commonly occurring contaminant. This article describes the current state of progress and considers forthcoming developments in the methods of testing for, and assessing the significance of, asbestos in soils by the Environment Agency (EA) and Health and Safety Laboratory (HSL).
The remediation of the Woolston site is at the tender stage and a further technical report will be produced following remediation in 2008-09.
The former shipyard site occupies a central location on the River Itchen in Southampton. Primarily a commercial shipyard, the materials used in the fabric of the site buildings and products associated with ship construction were used to raise levels on the south part of the site.
Normal investigations using trial pits and boreholes, identified up to 8m of made ground containing small quantities of products that also incorporated varying types of asbestos. Due to proposed basement excavations in this area, large amounts of made ground are arising that could be reused elsewhere on site where levels are being raised. Examining methods for describing asbestos levels in soils will help appraise site-based sustainable ex-situ remediation technologies and ways of validating the remediation at the site.
This work has highlighted the technical challenges of relating asbestos materials in soils – and the hazard these present – to a risk-based remedial threshold. These challenges have the potential to limit the remediation options available for ex-situ treatment and redeposit of soils with a content of asbestos containing materials (ACMs).
Scope of works
An attempt has been made to quantify asbestos in soils and judge the risk of different asbestos types and products, evaluate the level of acceptable risk of asbestos in soils to benchmark treatment, obtain appropriate regulatory approvals and trial remediation processes for soils with ACMs.
The delineation operation comprised extensive trial pitting, supervised by a specialist asbestos consultant under controlled conditions, combined with both commercial and specialist soils analysis (the latter by the HSL).
A remedial process, including mechanised screening and soil washing, was trialled to separate the bulk of asbestos from soils by Heijmans Blackwell Remediation. Soil washing is routinely employed in the Netherlands by Hiejmans for contaminants including asbestos.
Technical overview and context
Asbestos is a routine analyte in contaminated land investigations and is a potential contaminant for a range of historic land uses due to its widespread use as described in CLR 8: Potential Contaminants for the Assessment of Land and historic Department of Environment Industry Profiles.
For all contaminated land investigations it is important to express the amount of a contaminant in a quantity of soil and to calculate associated uncertainties within the investigation itself (Ramsey et al 2007). These appraisals are combined with conceptual site models to determine an appropriate numeric risk threshold for remediation. This approach is not available for asbestos, making it difficult to quantify and risk assess.
Quantifying asbestos in soils
Normal commercial analysis for asbestos, for example, HSG 248, involves identification of the asbestos type under a microscope to provide a semi-quantitative judgment of the amount of asbestos in a given ACM product, for example trace or 5%. However, these tests do not relate the product and asbestos type to the amount of soil in the sample.
Several techniques have been developed to quantify the level of asbestos in soil including those by the Institute of Occupational Medicine (IOM), the Health and Safety Executive's Contract Research Report No. 83/1996, and the draft MDHS 90 method, in development by the HSL.
However, these are not offered on an accredited basis at present, for example UKAS/MCERTS, or within testing proficiency schemes and historically have technical limitations. The ongoing work by the EA also reflects the practical difficulties of creating a method protocol that can equate varying types, sizes and forms of asbestos, relate these to a body of soil and consider the critical soil properties, such as moisture retention. These issues are critical prior to relating the hazards presented by different asbestos types to levels of risk.
At Woolston, a comparison was made of the visual site inspections and analysis by the HSL to appraise the quantity of ACM in soil and form a judgment of the associated risk. The tests applied included those in development as part of the process to determine a strategy for assessing risk from asbestos in soil for the EA . The testing comprised:
- HSG 248 bulk-product analysis using polarised light microscopy (PLM) and weighing of the associated products.
- Water-based drop tests to quantify fine asbestos fibres in soils (again using PLM/PCM).
- Empirical dustiness tests to measure dust generation from soils, comprising the rotation of dried soils in a sealed chamber with a controlled air flow and filtration rate.
The results of the empirical dustiness testing were entered by the HSL into the Hodgeson and Darton (H&D) risk model to give a numeric judgement of population mortality risk.
Also, an interpretation of the results was completed by the HSL generally following the proposed EA method in draft. The draft method establishes a relationship between types of product, their asbestos content by weight and form, and an established hazard weighting. This later element is then related to risk by considering risk calibrations based upon H&D modelling and dustiness testing.
A comparison of the results and methods is presented on page 30 with final HSL appraisal on the comparison of the methods awaited. What was clear from this work, and is an ongoing topic of research, is the difficulty in relating potential hazard rankings to the actual risk calculated for ACM in soils.
In addition it demonstrates that simple weight-based criteria have the potential to over or understate the risks presented by certain forms of ACMs in soil, as actual risk is associated with the potential for a soil to yield asbestos fibres in air – rather than the overall weight of a given product.
Workplace exposure to asbestos is heavily legislated and a clear regime of guidance exists for contamination on land via the EA's CLR/CLEA publications. But there is no standard remedial level for asbestos in soils or for airborne monitoring for asbestos in the environment.
This creates difficulty when relatively small amounts of asbestos products exist within a large body of soil and a judgment is required as to the "acceptable level" for the purposes of ex-situ remediation.
A standard risk-based threshold is particularly important for remediation practitioners who may wish to consider alternative treatments that could either remove the asbestos source, for example, soil washing or screening, or mitigate pollutant linkages that occur primarily via fibre generation, for example, via stabilisation, capping or vitrification.
Historic work by the Institute of Occupational Medicine identified a threshold of 0.001% weight as an action level (IOM Historical Research Report TM 88/14/1988, historically referred to in ICRCL 64/85). In addition, work by the Netherlands, Ministry of Housing, Spatial Planning and the Environment (VROM), has expressed a target level of 100 mg/kg or 0.01% weight . However, these thresholds are based on specific research projects and not upon a conceptual model of land use used elsewhere for land contamination. Also, a hazard threshold of 0.1% exists for asbestos in waste.
The result is an absence of a clear point of validation and differentiation between levels of contamination that can be retained in a treated soil. This may have limited the adoption of process-based remedial technologies on a consistent risk-based basis, as advised by CLR 11, and led to treatments that focus either on the removal of visible asbestos, capping of asbestos or removal of all affected soils to landfill. These are approaches that are not based on full conceptual risk assessment and do not provide a scientific threshold for ex-situ soils validation.
At the Woolston site both complex sorting and soils washing were trialled for the removal of asbestos from the soil. Treatment efficiencies were focused on the removal of cement-based asbestos products and screening efficiencies were restricted by the desire to keep the processed soils wet throughout.
Analysis by the HSL indicated that the treatment efficiencies results for the trial were not always reliably below the threshold of either 0.1% (for waste) or 0.01% in the event that a bulk material (such as a rope) had passed through the process mechanism, as they did on occasion when mechanical sieving/sorting was employed.
But the testing (box, p30) did highlight the distinction between potential weight content and hazard versus actual risk. While large bulk products were present, reflecting a high hazard, low levels of dispersed asbestos fibres were noted in soils using suspension drop tests (<0.01% to="" 0.001%)="" and="" tolerable="" risks="" were="" theoretically="" identified="" by="" numeric="" (h&d)="" modelling="" even="" in="" their="" untreated="">0.01%>
The value of this work is that it informs the appropriate objective of remediation: principally that remediation processes based solely on weight assessments, requiring the treatment primarily of bulk components may not necessarily present the most cost-effective and risk-based scientific remedial solution.
This is particularly the case where the weight-based target has to account for the worst case scenario and so is prohibitively low on a practical basis.
Remediation based upon empirical tests offers the assurance of a known potential for fibre generation, however, the results may be counter intuitive to the amounts of asbestos seen and undesirable where longer-term betterment is an overriding objective.
The works at Woolston demonstrate that remediation of asbestos in soils should not present an insurmountable technical issue. But because the means of testing and validation are not standard, each project requires specific discussions with regulators and the following challenges:
- Resolution of the issues requires the input of large professional group and where contaminated land skills require the support of occupational health and asbestos regulation specialists. This may be added to by other complexities such as the presence of other contaminants on the site and practical time, space and budgetary issues.
- Multi-point regulatory liaison is required with the local authority, the EA and the Health & Safety Executive (HSE) who also lack the clarity of definition and assessment protocols.
- Because of the associated testing uncertainty it is difficult to address the perception issue of asbestos in soils in a controlled manner as offered for other contaminants using the CLEA model and to benchmark a remediation.
- Until the draft Definition of Waste is an endorsed EA document, treated soils continue to be a waste after treatment and this limits their reuse on a site irrespective of risk.Conclusion
At Woolston it is proposed to offer process-based soil treatments, including ACM bulk removal techniques to lower overall ACM weight, combined with remedial processes to reduce potential fibre release. This should hopefully build on work previously done in this regard, both in the UK (Denner, Langridge and Affleck 1988) and abroad.
However, even with the trials completed, it is proposed that any treated soils will be provided with additional remedial measures, such as placement beneath capping to address perception issues that partly arise because of the issues of technical uncertainty discussed above.
The remediation contract at the Woolston site is ongoing as part of a large civil engineering project incorporating bio-remediation and soils screening as well as methods to remediate and validate soils that contain products with an asbestos content.
The first phase of research is likely to be complete by autumn 2008 and the project will continue into 2009.
Discussions are ongoing with the regulatory authorities, site developer and remedial specialists to finalise the remedial proposals in this regard. A project remediation report will follow the remediation with an emphasis on this issue.
James Clay is an associate with CampbellReith. The views expressed are those of the author and not necessarily those of CampbellReith.
Stephanie Mailer is a senior environmental scientist at CampbellReith.
Davies, L.S.T., Wetherill G.Z., McIntosh, C., McGonagle, C., Addison, J. (1996) Development and validation of an analytical method to determine the amount of asbestos in soils and loose aggregates. HSE contract research report No. 83.
Denner, J.M., Langridge, R.E., Affleck, M.J. (1988) Development of an asbestos-contaminated site – the Faslane Project. Journal of the Institution of Water and Environmental Management, vol 2, no.3, pp. 300-304.
This project has been permitted because of the commitment of SEEDA to obtaining a site-based sustainable treatment approach. It has also been supported by the ongoing technical input of Southampton City Council Environmental Health Department, the Health and Safety Laboratory and the EA.
 Campbell Reith Hill (CampbellReith) acts as consultant to SEEDA.
 This technique is in development still and may be modified prior to issue by the EA. It was followed by the HSL subject to certain limitations and constraints.
 The Netherlands, which has standard analysis methods and an agreed level of asbestos products in treated soils, using 100mg/kg (0.01%) (dry weight) of asbestos as the remedial target assuming activities such as digging, tipping and sifting of soil material are not systematically involved and the (top layer of the) soil is damp for a large part of the year. RIVM report 711701034/2003. Assessment of the risks of soil contamination with asbestos.
Examples of Difference: TP 307/Sample A2. Comments with kind permission of Barry Tylee, HSL.
l Visual observations on site indicated frequent occurrence of ACM materials in the soil profile including fibrous insulation, rope, cloth and cement boarding including chrysotile and crocidolite (notes from sub-contract asbestos surveyor). The soil would appear a major hazard.
l Large debris. Commercial and HSL laboratory HSG 248 identification found Low – High (50%) asbestos content of the large debris materials in the sample. Bulk materials are likely to suggest an exceedance of 0.1% by weight and indicate a high potential hazard.
l Fine soil. HSL analysis of trace asbestos bundles in the soil found a concentration of <5% –="" trace="" of="" asbestos="" (the="" estimations="" are="">5%>
l Water suspension techniques to assess fine amounts of asbestos in a soil using drop tests (sample 307/A1) found amounts of dispersed asbestos were small, ~0.005% by weight.
l The Environment Agency hazard algorithm indicated a high ranking relative to other samples because of the types of asbestos products present in the sample, which included a trace of lagging. Forms of soft products, including crocidolite, attract a high hazard rating as opposed to cement products with other asbestos forms such as chrysotile.
l Empirical dustiness tests were completed on dried (and therefore worst case samples). The analytical result for dustiness testing (displayed as a concentration of fibres/ml air at a respirable dust concentration of 0.2 mg/m3 (a typical dust concentration on construction sites in dry weather) was 0.002034 f/ml/0.2 mg dust/m3. The fibre levels of all the samples tested ranged from 0.001 to 0.03 fibres/ml. (Extract from HSE report.)
l The Hodgeson and Darton risk model identified a worst case risk of <1 death/million/annum="" for="" these="" samples="" –="" before="" treatment="" –="" using="" clea="" time="" exposure="" assumptions="" for="" zero="" to="" six-year-old="" children.="" hse="" believes="" that="" an="" individual="" risk="" of="" death="" of="" one="" in="" a="" million="" per="" annum="" for="" both="" workers="" and="" the="" public="" corresponds="" to="" a="" very="" low="" level="" of="" risk="" and="" should="" be="" used="" as="" a="" guidance="" for="" the="" boundary="" between="" broadly="" acceptable="" and="" tolerable="" regions.="" the="" assessment="" for="" workers="" suggests="" that="" short-term="" exposure="" to="" excavated="" soils="" would="" be="" in="" the="" acceptable="" region="" and="" presents="" a="" very="" low="" risk.="" (extract="" from="" hse="">1>
l There is no environmental airborne asbestos standard or para-occupational standard in the UK, although use has been made of 1/40th of current control level for workers (0.1 fibres/ml), for example, 0.0025 fibres/ml. The dustiness tests produced airborne concentrations well below the control limit of 0.1 f/ml, but around (and greater for one sample) a possible environmental target level of 0.0025 f/ml, but it should be recognised that the frequency of such exposure would be occasional. The sample had an acceptable level of dust generation even with the noted hazard ranking.