M Perrin and P Valentine, Encia Consulting
A number of documents have been published in the past 18 months providing guidance for hazardous gas (methane and carbon dioxide) assessment in the context of redevelopment on brownfield land.
There are many good, and some not so good, things within this new guidance. The good and the bad are discussed in this article, which also outlines how some ambiguities could have been avoided. It concludes with a suggested means of extending application of the National House-Building Council (NHBC) Traffic Light scheme to all new residential developments.
Recently published documents providing hazardous gas guidance include:
NHBC report 10627-R01, published in March 2007
Construction Industry Research and Information Association (CIRIA) report C659 in December
2006 (now withdrawn)
CIRIA report C665 in December 2007 (revision of C659)
British Standard BS 8485:2007 issued in October 2007.
These are briefly reviewed below.
NHBC report 10627-R01
Unlike previous guidance, which had focused on issues associated with landfill sites, this report assesses risks posed by ground gas in the context of residential redevelopment on brownfield land. It also emphasises the importance of a robust conceptual site model and adopts a risk-based approach in line with the Environment Agency's model procedure (CLR11).
The real significance of this report lies with its introduction of a risk-based methodology for deriving threshold gas screening values (GSVs). It uses GSVs to define four classes (Green, Amber 1, Amber 2 and Red) in Table 14.1, and relates these Traffic Lights to gas protection measures in Table 14.2.
Unlike previous guidance, the derivation of these threshold values is explained in reference to a model low-rise house. This house has a floor plan of 8m by 8m and a sub-floor void height of 150mm, within which, it is conservatively assumed, there is a complete volume change every 24h.
For a given significance (Traffic Light), different GSVs are derived for methane and carbon dioxide – reflecting the different properties and risks associated with each.
If a proposed low-rise housing development differs significantly from the model, an assessor can derive build-specific GSVs, for example, where proposed new dwellings have increased ventilation or a larger floor plan.
It incorporates the Pecksen methodology (1985), in particular the assumption of a 10m2 zone of influence of a standpipe.
The Traffic Light methodology is considered a logical and refreshing development in the field of gas risk assessment although there is some uncertainty regarding the Pecksen methodology, indeed, CIRIA C665 acknowledges that further research is required and recommends sensitivity analysis.
A criticism of the NHBC report is that while it states that site-specific GSVs can be derived, some assumptions used in the low-rise model house, most notably with respect to ventilation (air brick spacing and open area, wind speed) are not provided.
If ventilators are placed in accordance with NHBC standards and building regulations (such as to provide ventilation openings of 1500mm2 per metre run of external wall) we calculate an air flow velocity through the ventilators of 0.009m/s. If this velocity is kept constant, site-specific (really build-specific) GSVs can be derived by adjusting the ventilation.
This means that a site initially classed as Amber 2 could be downgraded to Amber 1 by improving sub-floor ventilation (subject to a sensitivity analysis, to determine the significance of occupiers blocking some of the vents). Build-specific GSVs should also be derived where proposed new dwellings have significantly larger floor plans than the 8m by 8m model house.
It is worth noting that the air flow velocity of 0.009m/s through the ventilators is considerably less than the ventilation velocities that would be calculated using BS 5925 (typically in the range 0.2m/s to 0.3m/s). Calculations in accordance with BS 5925 suggest that the void beneath the NHBC model house undergoes 13.7 air changes per day – confirming the extremely conservative assumption used to derive Traffic Light GSVs.
Having realised the positive impact the Traffic Lights system could have throughout the brownfield redevelopment sector, NHBC duly offered it to CIRIA for inclusion in its chapter on risk assessment.
Like the NHBC report, CIRIA report C665 aims to clarify and update earlier guidance in the context of brownfield redevelopment in line with CLR11.
C665 advocates use of the NHBC Traffic Light methodology for low-rise housing, with a clear sub-floor void (referred to as situation B developments). For all other development, including residential apartment, office, commercial, and industrial (situation A), C665 advocates the use of a modified Wilson and Card (1999) classification.
In addition to threshold GSVs, both methodologies refer to "typical maximum concentrations" when defining the Traffic Light\Characteristic Situation, although it is noted that these can be exceeded in certain circumstances should the conceptual site model indicate it is safe to do so.
This is based on robust data, including monitoring in the worst temporal conditions, and a thorough understanding of the gas regime.
A major plus in C665 is the guidance provided regarding the spacing of monitoring wells (Table 4.2) and particularly the duration and frequency of monitoring (Tables 5.5a and 5.5b).
Importantly, use of these tables requires consideration of the:
a) Conceptual site model;
b) Generation potential of the gas source, and;
c) Sensitivity of the proposed development.
While it is acknowledged that there is no industry consent over the terms used to describe generation potential, it not unreasonable to expect experienced geoenvironmental consultants to be able to justify their selection and, where necessary, revise this in light of monitoring data.
Table 5.5 is considered to be a more logical methodology than that originally advocated by Wilson and Card (1999). They recommended that a site's Characteristic Situation be raised by one class if there was less than 12 months' data and two classes if there was less than six months' data. Furthermore, Table 5.5 provides regulators, the NHBC and others with a means to assess the adequacy of consultants' reports, thereby "levelling the playing field".
On the downside:
1. If the gas regime for a proposed housing development is Green, Box 8.4 states that no gas protection measures are required; a membrane and ventilated sub-floor void are advocated for Amber 1 sites. This implies that a ground-bearing or cast insitu floor slab (with no sub-floor ventilation) would be acceptable on Green sites.
This might be true in some circumstances, and ground bearing slabs can provide reasonable protection. However, the model used to derive threshold GSVs for a classification of Green assumes a ventilated sub-floor void. (This criticism also applies to the NHBC report).
Some minor additional wording from the authors could have resolved this ambiguity. Where Green, we believe the gas regime should also be reviewed against Situation A methodology (which does not assume a ventilated sub-floor), to check whether or not it lies within Characteristic Situation 1 (in which case no measures would be appropriate), or Characteristic Situation 2 (in which case a ventilated sub-floor void and membrane should be required).
2. With respect to methane, gas regime characterisation for low-rise housing (C665 Situation B) is more conservative than that for other developments (Situation A). For example a methane GSV of 1.6 would classify a site as Red (Situation B), but only Characteristic Situation 3 (Situation A).
This may be reasonable and justifiable, but C665 provides a potential loophole that could allow those with problematic residential sites to justify the choice of the less onerous Situation A, on the grounds that their development differs from the NHBC model house (for example, because of the proposed adoption of a void former or ground-bearing slab).
With respect to carbon dioxide, gas regime characterisation for Situation B is less conservative than that for Situation A. For example a carbon dioxide GSV of say 0.75 would classify a site as Green (Situation B), but Characteristic Situation 3 (Situation A). We believe that Situation B methodology could, and probably should, be applied where sub-floor ventilation is achieved by means other than a clear 150mm void.
3. C665 does not explain why residential apartments, for example, are considered to be of lower sensitivity than low-rise housing.
It is understood that this is because apartments are less vulnerable to owner modifications that could interfere with sub-floor ventilation (for the construction of patios and conservatories, for example).
This is quite reasonable – but it is a shame it was not explicitly stated in C665. However, this might be countered by the higher number of occupants and the consequences of explosion within a high-rise building.
Therefore we feel that the more conservative Situation B methodology could, and probably should, be applied to all residential buildings.
4. In Section 7.2, C665 raises the issue of whether peak or steady flows and concentrations should be used. C665 seems to suggest that peak values should be adopted to derive reasonable worst-case GSVs. By definition peak flows are short-lived (typically <30s), so="" their="" contribution="" to="" hazardous="" gas="" concentrations="" within="" a="" large="" sub-floor="" void="" is="" negligible.="" where="" peak="" flows="" are="" maintained="" for="" longer="" than="" 30s,="" they="" should="" generally="" be="" regarded="" as="" steady="">30s),>
Consequently, we believe steady flows should be used to derive GSVs.
5. Also in Section 7.2, the issue of which maximum values (intra-borehole or inter-borehole) should be used is mentioned and reference is made to further discussion in Chapter 8.
Review of Chapter 8 yields no obvious reference to this issue, but in our opinion the highest, steady flow and highest, steady concentration should be used to derive a reasonable worst-case GSV, even if they occur in different boreholes (except where the boreholes are clearly located in areas affected by different gas sources).
Encia Consulting wrote to NHBC in March last year (well before C659 was withdrawn and replaced by C665) requesting clarification of the above and other issues. The response broadly confirmed our thinking.
C665 was issued last December with revisions relating primarily to clarification of a number of relatively minor technical issues, together with a re-edited version of the document.
Given the issues raised by Encia (and similar ones by other practitioners), together with CIRIA's stated concern regarding a lack of clarity in previous guidance, it might be considered unforgivable that the opportunity to clarify matters was not taken. This is compounded by the fact that C659 errata are not listed, some new typographic errors have appeared and the document remains in black and white.
This provides a framework in line with CLR11 and C665 that allows designers to judge the adequacy of ground gas and related site investigation data.
It goes on to relate the gas regime and proposed building to a required gas protection, via a point scoring scheme, outlined in Table 2. Table 3 then provides scores for various mitigation measures, which can be combined to achieve the required gas protection.
While it does not provide explicit advice regarding ventilators (such as spacing and open area of air bricks) the BS scores are considered a major step forward with respect to specifying appropriate protective measures.
Proposed wider application of NHBC Traffic Lights
Encia has adapted Table 3 of BS 8485:2007 to provide more explicit advice to developers, most notably with respect to ventilation. The version applicable to house builders is copied in this article as Table A (similar tables have been derived for other developments). It should be noted that the Green gas regime has been sub-divided in two.
Table A enables application of the NHBC Traffic Light scheme to all new residential developments – low and high rise, regardless of floor construction.
The authors consider that this rectifies a major shortcoming in Chapter 8 of C665 and the NHBC report (where the Traffic Light methodology is restricted to low-rise housing with a clear, ventilated sub-floor void).
There are potential implications of extending the NHBC model – most notably will the proposed sub-floor ventilation system facilitate a complete volume change every 24h?
Within granular vent blankets the volume of air in the sub-floor system is a function of porosity, and the majority of the flow through the system is governed by Darcy's Law. Calculations in accordance with BS 5925 suggest the number of air changes is in the region of 0.3/d to 0.5/d.
Nonetheless, in line with the guidance provided in BS 8485:2007, granular blankets are considered satisfactory for new houses built on low-risk sites (up to and including Amber 1), provided that:
a) The building footprints are less than, say, 8m in cross-ventilation width;
b) Ventilators are placed as per NHBC Standards, Chapter 5.2. That is, the use of granular blankets in situations where the Amber 1 classification has been achieved by adjusting, for example, air brick spacing, is not recommended.
Where the proposed sub-floor ventilation system is achieved by use of a void former, GSVs are identical to those derived for a clear void.
This is because the effect of reducing the void space is cancelled out by an increase in the number of air changes (for a given ventilation rate, for example, 0.4m3/h).
Consequently, it is considered that void formers provide a satisfactory alternative to the adoption of a clear void for all Traffic Lights, up to and including Amber 2.
However, a possible consequence of lessening sub-floor void space is the reduced time taken, during periods of no wind, to reach concentrations of concern, say, 5% v/v methane, in the sub-floor space.
This time should be calculated with reference to recorded site-specific borehole flow rates and concentrations, and compared against the maximum period of still wind referenced in the Partners in Technology Research reports (10h).
When undertaking such calculations, consideration should be given to the initial steady-state concentration within the sub-floor void – such as the equilibrium concentration of methane that has accumulated in the void after a prolonged period of low wind (0.3m/s).
There is some ambiguity within existing guidance, most notably regarding the applicability of the NHBC Traffic Light classification to residential developments that do not concord with the model house, and how to adapt the Traffic Light system in light of build-specific parameters (including the use of peak or steady flows and concentrations).
Boyle, R., Witherington, P., 2007. Guidance on evaluation of development proposals on sites where methane and carbon dioxide are present, incorporating Traffic Lights. Report Ref. 10627-R01-(02). Milton Keynes: National House-Building Council.
British Standards Institution., 2007. BS 8485:2007. Code of practice for the characterization and remediation from ground gas in affected developments. London: BSI.
Construction Industry Research and Information Association C659, 2006. Assessing risks posed by hazardous ground gases to buildings. London: CIRIA.
Construction Industry Research and Information Association C665, 2007. Assessing risks posed by hazardous ground gases to buildings (revised). London: CIRIA.
Office of the Deputy Prime Minister, 2004. The Building Regulations 2000: approved document C – site preparation and resistance to contaminants and moisture. London: Stationery Office.
Partners in Technology for the Department of Environment, Transport and the Regions, 1997.
Passive venting of soil gases beneath buildings research report: guide for design. London: Department of Environment, Transport & the Regions.
Pecksen, G.N., 1985. Methane and the development of derelict land. London Environmental Supplement, Vol 13. London: Greater London Council.
Wilson, S.A., Card G.B., 1999. Reliability and risk in gas protection design. Ground Engineering, 32(2) p33-36.