Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Reliability and risk in gas protection design by SA Wilson and GB Card of Card Geotechnics.

PAPER

Abstract

There is large variance in the scope of gas protection measures provided for developments on gassing sites. This is because of:

unclear and ambiguous guidance

subjective interpretation of gas regimes

poor understanding of gas surface emission rates

In response to incidents of methane explosions in the mid-1980s, guidance documents were produced in the UK for both planning authorities and the construction industry. Gas protection measures are now routinely included in engineering designs on any perceived high risk sites, eg former landfills, mine workings, estuarine and fenland and limestone or chalk areas.

Frequently, the scope of protection measures is specified solely on the basis of the maximum measured gas concentration, with no consideration given to the spatial distribution and frequency of the readings, borehole flow rates and estimated surface emission rate or the nature of the gassing source. The lack of a comprehensive and reliable design method has led to inconsistency in the specification of gas protection systems utilising passive ventilation which can, in some circumstances, result in unsafe design.

This paper illustrates the use of a holistic approach to the characterisation of gassing sites based on gas concentrations, borehole flow rates, estimated surface emission rates and the nature of the gassing source. It considers all the recently published guidance by the Construction Industry Research and Information Association (CIRIA) and the Department of the Environment, Transport and the Regions.

Introduction

In response to methane explosions at Loscoe (Williams and Aitkenhead, 1989) and Abbeystead (Health & Safety Executive, 1985), guidance documents on the hazards of methane or landfill gas were quickly prepared by the Government for planning and waste regulation authorities (Department of the Environment, 1989a and 1989b) and later for the construction industry (BRE, 1991). More recently, guidance has been prepared by CIRIA to provide a detailed reference for designers of gas protection systems (Card, 1995) in which an initial attempt was made to characterise gassing sites in terms of the volume of gas rather than just concentrations.

Because of the limited understanding of gas generation and migration of soil gases, the design of gas protection measures is still subject to a great degree of judgement. In particular, passive ventilation is specified and designed solely on the basis of maximum gas concentrations in the ground with no consideration given to the spatial distribution and frequency of concentrations, the surface emission rate (estimated from measured borehole flow rates and gas concentrations) or the nature of the gassing source.

In recognition of this problem, long term monitoring of gas regimes and of passive ventilation systems has been carried out on a number of residential and commercial developments. This paper demonstrates the use of a holistic approach to the specification of gas protection measures, which should give a more consistent approach to the problem, based on the most up to date published guidance. A glossary of terms used is presented at the end of this paper.

Current methods of interpretation and analysis

In line with government policy to develop brownfield sites, a significant proportion of new development in the UK is built on gassing ground, typically in former industrial or urban areas, as well as dock areas, reclaimed estuarine areas and on or adjacent to landfill sites. The source of the gas is variable amounts of organic material within the ground which, as it decays, generates methane and carbon dioxide.

On sites such as former dock areas or reclaimed estuarine sites, gas is generated very slowly from localised sources such as wood fragments and organic deposits and is characterised by low volumes of gas generation and low surface emission rates. Such sites have a long history of redevelopment with no incidents from gassing ground and as such they represent lower risk sites in terms of the presence of methane gas (O'Riordan and Milloy, 1995).

In the case of recent landfills and in mine working areas, methane can be generated rapidly, giving rise to large volumes of gas generation and high surface emission rates. The majority of incidents connected with methane are from these sites and they pose a higher risk to development.

Designers are frequently specifying the scope of protection measures directly from Tables 28 and 29 of CIRIA Report 149 (Card, 1995) with little regard to the nature of the gassing source, borehole flow rates and the estimated surface emission rates. The tables were never intended to be used as a definitive design tool and were only prepared to show the typical scope of measures for gas control that were in current use, at the time the tables were prepared.

Furthermore, Table 28 is based on gas concentrations as the primary means for specifying protection measures. This was because at the time of preparation of the document in the early 1990s, accurate and reliable measurement of borehole flow rates was difficult and not widely practised. Even now, with the availability of sensitive gas flow measuring instruments, there is little consideration of the gas regime beneath a site by either the designers of gas protection systems or the regulatory authorities who approve the measures.

Use of the tables as a design tool can potentially lead to unreliable design of protection systems. For example, rigorous implementation on a site with localised pockets of peat, which may give high concentrations of methane in any monitoring well which happens to pass through one of the pockets, could classify the site as characteristic situation 6. The same scope of protection measures would then be specified as if the development was located on a recent highly gassing landfill site. The latter has a much higher surface gas emission rate and poses a much more serious risk.

Consider also a site with 21% methane and a borehole flow rate of 0.5 litres/h, which would be classified as situation 5, and a site with 5% methane and a borehole flow rate of 35 litres/h, which would be situation 3. Considering surface emission rates, the latter is actually the more onerous case.

In the main, three approaches are used for gas protection design:

1. A rational approach considering frequency and distribution of gas concentrations, borehole flow rate and estimated surface emission rate and the nature of the source of generation. Risk assessment may be used to assess the optimum combination and adequacy of gas protection measures. This method is used by only a very few designers and forward looking local authorities at the present time.

2. Protection measures are specified on the basis of the highest gas concentration measured and use of Table 28 from CIRIA Report 149, Card (1995). This is becoming more frequently used, especially by regulatory authorities.

3. The 'It has worked before so we will use it again,' approach. Little consideration is given to gas concentrations, except that if the limits quoted in Building Regulations approved document C, DoE (1992) and Waste Management Paper 27, DoE (1991) are exceeded, a standard gas protection system is installed. Usually development is limited to commercial use and gas protection limited to a standard active or passive underfloor ventilation system with a gas resistant membrane.

The latter two approaches lead to wide inconsistency in the scope of gas protection measures for sites with similar gas regimes, and most often over-conservatism in the design (Wood and Griffiths, 1994).

Harries et al (1995) identify that interpretation of gas regimes requires an understanding of the gas generation process and the means of gas escape from the ground. Therefore the most important aspect of relating the gas regime below a site to the risks it poses to any development is the surface emission rate, ie how quickly the gas is coming out of the ground. The lower the surface emission rate the lower the risk.

Back analysis from post-construction monitoring

Table 1 summarises the results of long term monitoring of a number of building developments on gassing sites incorporating passively ventilated underfloor systems. For all developments the ventilation capacity of the underfloor void has been designed using the methods described in B55925:1991. In all cases the measured borehole flow rates were used to estimate the surface emission rate using the correlation proposed by Pecksen (1985). Monitoring on a number of other developments is continuing, with no gas detected to date.

As can be seen from Table 1, monitoring of completed passive venting systems invariably indicates the design sub-floor concentrations have been achieved. Indeed, often it is difficult to establish the presence of any soil gas such as methane in the sub-floor void. Wood and Griffiths (1994) report similar findings from the results of monitoring inside buildings and sub-structures on gassing sites.

A holistic approach to site characterisation

A rational method for classifying gassing sites in terms of the risks posed by the presence of gas should consider:

the source of the gas

the generation potential of the source

the location of the source and the presence of any natural geological barriers to migration

the borehole flow rate and estimated surface emission rate

measured gas concentrations

the nature of proposed development

the confidence in the knowledge of the gas regime and any likely future variations in concentration and borehole flow rate

arrangements for long term management of the system

On the basis of the degree of risk, a protection system can be provided with sufficient elements to maintain safety should one element fail. For example, a low risk site with less sensitive development might only require a two component system to reduce the risk to acceptable safe levels. The necessary information and guidance to carry out this approach is provided in the series of reports by CIRIA - Card (1995), Crowhurst and Manchester (1993), Harries et al (1995), O'Riordan and Milloy (1995) and Raybould et al (1995) and also the DETR (1997).

Source and generation potential

O'Riordan and Milloy (1995) provide a risk ranking of gas sources. Natural soils with a low organic content are deemed to a pose the lowest risk in terms of gas generation. This is reasonable as most methane present in peat deposits is trapped in pockets and current generation rates are generally low. Harries et al (1995) quote generation rates of methane from peat bogs and marshes between 0.18 to 43 litres/m2/year.

This view is supported by the absence of any reported problems relating to methane in housing in such areas as the Fenlands and Somerset Levels where houses have been constructed over peat throughout history (Card, 1995 and Hooker and Bannon, 1993).

The highest risks are posed by landfill sites which are relatively recent, ie typically post 1960s. This is because recent waste has a much higher organic content than waste in the past, due to items such as paper from packaging, disposable nappies, etc. In modern domestic landfill sites the rate of generation of methane gas is typically much higher than the other sources mentioned and borehole flow rates of several thousand litres/hour have been recorded on such sites (Harries et al, 1995).

A modified version of the risk ranking suggested by O'Riordan and Milloy (1995) is reproduced in Table 2. O'Riordan and Milloy also placed the presence of carbonate deposits, (eg Chalk) relatively high in terms of risk order. This is presumably as a result of concern over acidic groundwater or rainfall percolation acting on the carbonate to produce carbon dioxide. The authors are not aware of any cases where houses built on such deposits (of which there are many thousands in the UK alone) have been affected by carbon dioxide migrating into the house. Harries et al (1995) provide a calculation to estimate the rate of generation of gas from such a source which shows that generation rates would be extremely low.

Therefore in the revised ranking carbonate deposits are placed at the same level as natural soils with a low peat or organic content.

Skennerton (1997) confirmed the importance of considering surface emission rates and concentrations together. Skennerton has suggested the system of risk ranking in Table 3.

The highest risk ranking of 1 is given to gas regimes where a high concentration exists with a high borehole flow rate. Inspection of the table shows that borehole flow rate is rated more influential on risk than concentration. A low concentration combined with a high flow rate is considered to be a higher risk than a site with high gas concentrations at low flow rates. This is a reasonable assumption as gas must be migrating to the surface to pose a hazard to any development, which is recognised in the Building Regulations (DoE, 1992).

Borehole flow rate and surface emission rate

Waste Management Paper 26A, DoE (1993) gives guidance on borehole gas volume flow rates and the completion of a landfill. When borehole gas volume flow rates have remained below the target values for over two years the licence can be surrendered and the landfill is deemed to no longer pose a risk of causing significant harm to the surrounding environment. The borehole gas volume flow rates are:

methane 15litres/hcarbon dioxide 221itres/h

This does not necessarily mean that such sites are safe for development, but when compared to the values in Table 3, they would define a site as characteristic situation 4, which would appear reasonable if the bulk of the gas generation is com plete and there is sufficient information on site conditions to determine that it is unlikely to increase in the future.

Site characterisation

Harris et al (1995) identify that there is currently inadequate guidance on trigger concentrations for ground gases. The current emphasis on using concentrations for trigger values, particularly in Waste Management Paper 27 and the Building Regulations, should be revised to consider gas pressures, borehole flow rates and estimated surface emission rates.

To achieve more consistent design of protection measures it is more relevant to rewrite Table 28 of CIRIA 149 in terms of borehole gas volume flow rate and gas concentrations, as shown in Table 4. A similar principle is adopted in the recent Partners In Technology report which studied the performance of a variety of passive ventilation systems (DETR, 1997) and used borehole gas volume flow rate as a means of classifying gas regimes.

Based on the preceding discussion, a new method for characterising gassing sites is proposed in Table 5, which takes into account borehole gas volume flow rates in addition to gas concentrations. To facilitate design implementation the limiting values for both methane and carbon dioxide are identical.

This classification system is similar to the gas regimes A to F proposed in the Partners in Technology, Guide to design for passive venting of soil gases beneath buildings (DETR, 1997).

Design of passive ventilation systems

There is wide range of available gas protection methods, each of which has its own advantages and disadvantages. No protective measure is immune from factors unknown or out of control of the designer. Therefore individual measures should be combined to form the gas control system so that the probability of failure of all the parts of a system is reduced. The number and type of the protective measures should be related to the level of risk of failure that can be tolerated for a particular development. A typical scope of measures for each characteristic situation, for two types of development, is provided in Table 6.

Ventilation of the underfloor subspace is the primary method of providing gas protection in most cases, with secondary protection provided by a barrier to gas migration above the subspace. The use of passive ventilation for underfloor subspaces is preferred for characteristic situations 1 to 5, particularly for residential buildings. Card (1995) discusses the advantages and disadvantages of each system and concludes that the use of passive rather than active measures is preferred as they require less maintenance and are less likely to become totally inoperative. This policy is also consistent with the requirement for energy efficiency in building design.

For any development which is provided with gas protection measures it is important that a long term maintenance strategy is adopted. In particular, long term monitoring of underfloor gas concentrations should be carried out on all developments to demonstrate that the measures are performing adequately. This is becoming increasingly acceptable to developers who must demonstrate to other stakeholders (for example occupiers, the general public and local authorities) that a site has been safely developed.

Conclusions

Current guidance on the interpretation of gas regimes is unclear or ambiguous and subject to a great deal of judgement by designers. Long term performance monitoring of installed gas protection systems show there is inconsistency and over-conservatism in current gas protection design methods, particularly where limited volumes of gas are being generated at low rates.

A more comprehensive method of characterising gassing sites that provides a reliable and consistent approach to gas protection design is demonstrated. The methodology has been used successfully on a number of recent developments with the approval of regulatory authorities. Continuing monitoring after construction has demonstrated adequate performance of the gas protection systems installed.

References

BRE (1991). Construction of new buildings on gas-contaminated land. Building Research Establishment Report 212.

Card GB (1995). Protecting development from methane. CIRIA Report 149.

Crowhurst D and Manchester SJ (1993). The measurement of methane and other gases from the ground. CIRIA Report 131.

DoE (1989a). Waste Management Paper No 27, The control of landfill gas. Department of the Environment, HMSO, London.

DoE (1989b). Planning circular 17/89, Landfill Sites: Development control. HMSO, London.

DoE (1991). Waste Management Paper No 27, Landfill gas. Second Edition. Department of the Environment, HMSO, London.

DoE (1992). Building Regulations, Approved Document C. HMSO, London.

DoE (1993). Waste Management Paper No 26A, Landfill completion. Department of the Environment. HMSO London.

Department of the Environment, Transport and the Regions (1997). Passive venting of soil gases beneath buildings, Guide for Design. Volume 1. Department of the Environment, Transport and the Regions, Partners in Technology, prepared by Ove Arup and Partners, September 1997.

Harries CR, Witherington PJ and McEntee JM (1995). Interpreting measurements of gas in the ground. CIRIA Report 151.

Health and Safety Executive (1985). The Abbeystead Disaster. HMSO, London.

Hooker PJ and Bannon MP (1993). Methane and its associated hazards to construction. CIRIA Report 130.

O'Riordan NJ and Milloy CJ (1995). Risk assessment for methane and other gases from the ground. CIRIA Report 152.

Pecksen GN (1985). Methane and the development of derelict land. London Environmental Supplement No 13, Summer 1985.

Raybould JG, Rowan SP and Barry DL (1995). Methane investigation strategies. CIRIA Report 150.

Skennerton S (1997) Landfill gas: Rates vs concentrations. Presentation notes from Remediation of hazardous gases workshop. Regional Group Meeting, Geological Society, 3 June 1997.

Williams GM and Aitkenhead N (1989). The gas explosion at Loscoe, Derbyshire. Methane - facing the problems symposium, Paper 3.6, Nottingham, 26-28 September 1989.

Wood AA and Griffiths CM (1994). Debate: contaminated sites are being over-engineered. Proceedings of the Institution of Civil Engineers. August 1994, 102 pp97-105.

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.