Monitoring    Road sources    Domestic fuel    Part A and B processes    Uncontrolled + fugitive emissions

Third Stage Review & Assessment

66. Where both the First and, if undertaken, Second Stage reviews and assessments indicate that there is a significant risk of the proposed objectives for PM₁₀ not being achieved by 2004, the authorities are advised to undertake a Third Stage review and assessment. This will require a detailed and robust assessment of the potential impacts. This does not necessarily mean that expensive and/or sophisticated modelling and monitoring tools have to be used, but does require that the assumptions within the review and assessment process are considered in depth, and the data which are collected or used are quality-assured to a high standard.
    
67. Due to the potential wide variety of sources and local circumstances that may prevail, it is not possible to be prescriptive for the Third Stage review and assessment. Rather, this document aims to set out appropriate guidelines for the assessment approach. The scope of work will depend upon the nature of the exceedances indicated during the Second Stage review and assessment. For example, where the concern is limited to a clearly defined road or industrial process, then the assessment can be limited to a detailed investigation of these specific sources. However, where there is concern regarding a widespread area of exceedance, for example within a major urban area, then it may prove necessary to undertake more extensive emissions inventories and modelling work.
    
68. The Department of the Environment, Transport and the Regions, the Welsh Office and the Scottish Office, have published detailed guidance documents^3,4,5,6 that provide advice and set appropriate standards for undertaking a Third Stage review and assessment. Authorities are advised to refer to the relevant chapters within these documents before commencing any detailed assessment work.
    
    Monitoring PM10 concentrations
    
69. Due to the uncertainties associated with PM₁₀ source emissions, a programme of PM₁₀ monitoring is likely to form an important element of the Third Stage review and assessment study. Suitable QA/QC procedures for PM₁₀ monitoring are detailed within LAQM.TG1(98).
    
70. A principal area of consideration for authorities considering a monitoring campaign may be whether to use a TEOM (or equivalent) analyser or a gravimetric sampler. The potential discrepancy between the two methods has been described above in Box 2. At this point in time it must be accepted that the two methods will return different results, and that the relationship between them (used in this report) is, as yet, only provisional. It must also be accepted that TEOM analysers and gravimetric samplers have different advantages and disadvantages, both in terms of operation and performance.
    
71. The majority of PM₁₀ monitoring at a national level is currently founded on the use of the TEOM analyser, and these data can be directly compared with the existing prescribed objective. However, a correction factor has to be applied to TEOM data for comparison with the new proposed objectives (which are based on the European gravimetric transfer reference sampler, or equivalent). The DETR are currently undertaking detailed intercomparisons of the TEOM analyser and the European gravimetric transfer reference sampler in order to develop a more robust relationship between the two methods. Pending the results of these investigations, the use of gravimetric samplers clearly offers a more direct approach to collecting monitoring data for comparison against the proposed objectives. However, care should be taken in the selection of gravimetric samplers used for the review and assessment of the proposed objectives, as at this time only a limited number of instruments can demonstrate equivalence to the European reference method.
    
72. However, it is emphasised that it is not the intention to imply that data from TEOM instruments should be disregarded, or that authorities with these instruments should not effectively use them during the Third Stage review and assessment process. It is considered that TEOM data can be used with a reasonable degree of confidence provided that the adjustment factor of 1.3 is applied.
    
73. If gravimetric samplers are used, authorities are reminded that considerable care and attention will need to be given to the processes of filter handling, pre and post-conditioning, and weighing, as these can significantly affect the quality of reported data. Suitable guidance on procedures may be sought from the Monitoring helpline.
    
74. An issue of particular importance to bear in mind with PM₁₀ monitoring is the length of sampling period. This is particularly important when considering the 24 hour objective, which allows 35 days exceedance in a calendar year. In order to accurately assess whether this objective is being exceeded, it is recommended that a full year of monitoring is carried out, and that a minimum of 90% data capture is achieved, with any data gaps evenly spread across the year. In cases where monitoring data extend over a shorter period of time, it is strongly recommended that a comparative assessment is carried out with long-term sites (e.g. national network sites) to assist with the interpretation of findings.
    
    Road traffic sources
    
    
75. The Third Stage review and assessment associated with road traffic emissions is likely to focus upon:
    
    • A compilation of more detailed and accurate road traffic emissions data;
      
    • A more detailed assessment of the road traffic emissions contribution to PM₁₀ concentrations; and
      
    • A more accurate description of existing background levels, possibly supported by roadside monitoring.
      
76. There are a wide variety of dispersion models available to authorities which may be used in addition to, or in support of, the DMRB model. Guidance on the use and selection of these models, and the types of input data that are required, is set out in LAQM.TG3(98), and further advice may be sought from the Modelling Helpline. In many circumstances however, it is considered unlikely that these more sophisticated models will provide much additional information over that gained from the DMRB or other screening models. This may particularly be the case where exceedances of the proposed objectives are only predicted for isolated roads or small groups of roads. Any enhanced accuracy in the predictions is more likely to be achieved from a more detailed appraisal of the traffic input data and the treatment of the background contribution. Where more sophisticated models prove to be either advantageous or necessary, for example in the review and assessment of major urban areas, the treatment of emissions and background concentrations will still remain a fundamental concern.
    
77. The emissions of PM₁₀ are strongly related to vehicle speeds, and it is recommended that more detailed information related to traffic flow and speeds is obtained. For example, it may prove useful to split roads up into much smaller sections, which will then allow a more accurate definition of changing vehicle speeds e.g. close to junctions. It may also prove important to take account of areas where cold-start emissions are particularly important, such as in the vicinity of long-term car parks. PM₁₀ emissions are also known to be affected by engine loading, for example when vehicles are climbing steep hills, and specific speeds and emission factors for these types of areas may need to be considered.
    
78. In addition, the method of defining vehicle types within the DMRB model is relatively simplistic (i.e. vehicles are classified into LGVs and HGVs only). Such a classification may not accurately reflect actual traffic circumstances, and it is therefore recommended that authorities attempt to define the vehicle types more reliably. Types of vehicle classification that are commonly used include:
    
    • cars (petrol and diesel)
      
    • LGV (petrol and diesel)
      
    • HGV (small and large)
      
    • buses
      
    • motorcycles
      
79. Local emissions estimates can be refined using the methods in LAQM.TG2(98), and advice may be sought from the Emissions Inventory Helpline.
    
80. The calculation of the 'roadside enhancement' to the 2004 background is of fundamental importance in predicting compliance with the proposed objectives, and needs to be given more detailed consideration at this stage. The recent APEG report has indicated that the roadside enhancement to PM₁₀ concentrations comprises of roughly equal halves of fine particles (emitted from the vehicle exhausts) and coarse particles (generated by resuspension). Whilst vehicle emissions are confidently expected to decline by 2004, resuspended particle emissions are unlikely to reduce, unless traffic flows change substantially. Predictions of PM₁₀ concentrations in 2004 should attempt to deal with these aspects separately. It should be noted that the resuspension component is not included in the DMRB model.
    
81. Where more accurate vehicle emissions data are used, the following approach is therefore suggested:
    
    • Estimate the current background contribution from 'primary', 'secondary' and 'coarse' particles as described in paras 47 to 50 above;
      
    • Predict the current contribution of road traffic to PM₁₀ concentrations using detailed emissions data and vehicle operating characteristics input to a dispersion model;
      
    • Assume that PM₁₀ from resuspension is equal to PM₁₀ from road traffic exhaust emissions in the current year;
      
    • Estimate the 2004 background contribution from 'primary', 'secondary' and 'coarse' particles as described in para 32 above;
      
    • Calculate the total PM₁₀ concentration in 2004 by adding the total 2004 background contribution to the predicted 2004 road traffic contribution, plus the current year resuspension contribution.
      
82. Due to some of the uncertainties associated with defining vehicle emissions and the resuspension component, it is strongly recommended that the model is calibrated against actual monitoring data. Where suitable roadside monitoring data are not available for the assessment area, then both the model and the assumptions should be tested at a similar roadside site elsewhere, where monitoring does take place.
    
    Domestic solid fuel use
    
    
83. The Third Stage review and assessment associated with domestic solid fuel use is likely to focus upon:
    
    • An assessment of solid fuel use in the area to determine emissions;
      
    • The application of detailed dispersion modelling; and
      
    • Local monitoring to confirm existing concentrations.
      
84. As a first step, it is recommended that authorities should undertake a detailed assessment of solid fuel use within the area of concern, characterising the different fuels and combustion methods used in order to more accurately quantify the emissions. Guidance on emissions calculations is given in LAQM.TG2(98), and additional advice may be obtained from the Emissions Inventory Helpline. These data may then be input to a suitable air quality dispersion model, in order to predict the ground level impact. Guidance on the selection of suitable dispersion models is given in LAQM.TG3(98), and additional advice may be obtained from the Modelling Helpline.
    
85. In the absence of local monitoring data of suitable quality, it is recommended that a monitoring programme is carried out in order to more accurately define current PM₁₀ concentrations in the area of concern. It is suggested that monitoring equipment is sited at relevant locations (see para 12) and ideally, monitoring carried out for a period of 12 months. Where this is not possible, authorities are advised to compare the results of local monitoring with data from national network sites, to assist with the interpretation of findings.
    
    Part A and Part B authorised processes
    
    
86. The Third Stage review and assessment associated with Part A and Part B processes is likely to focus upon:
    
    • The application of detailed dispersion modelling; and
      
    • The use of more detailed local monitoring to confirm existing concentrations.
      
87. There are a variety of dispersion models which can be used to assess the impact of emissions arising from elevated point sources (stacks). Guidance on the selection of suitable dispersion models is given in LAQM.TG3(98), and additional advice may be obtained from the Modelling Helpline. Authorities are advised to consider the following issues:
    
    • Accurate description of PM₁₀ emissions - many processes operate well within their emissions limits, and actual emissions data should be used if at all possible. For certain types of process, both seasonal and daily variations in emissions are significant, and should also be considered wherever possible and input to the model;
      
    • Elevated point sources will have little impact upon the annual mean concentration, and the assessment should focus upon an accurate prediction of the 90th percentile of daily mean concentrations. The modelling approach should therefore seek to predict day-by-day ground level concentrations arising from the stack(s), which will then be added to suitable sequential daily background concentrations for 2004. Whilst the application of single 'uniform' factors to correct current daily background 'primary', 'secondary' and 'coarse' concentrations to 2004 is not ideal (as these will change from day to day) this may form a useful first step. Where a more detailed analysis of the background component is required, authorities are advised to consult the relevant sections of the APEG report, or to contact the Pollutant Specific Guidance Helpline.
    
    
    Uncontrolled and fugitive dust emissions
    
    
88. Due to the uncertainties associated with PM₁₀ emission rates from uncontrolled and fugitive sources, it is likely that the Third Stage review and assessment will need to focus upon a detailed monitoring programme. Whilst emissions data are available for a variety of fugitive sources, for example those published within the Compilation of Air Pollution Emission Factors (USEPA-42), these factors are subject to a variable degree of uncertainty, and frequently require default assumptions to be made. Their principal application lies in allowing predictions to be made for the impact of operations which are currently not in existence, or which are expected to undergo significant change by 2004. If applicable, guidance on emissions data and dispersion models may be sought from the relevant Emissions and Modelling Helplines.
    
89. It is suggested that in many cases a suitable monitoring programme will need to be established to determine the impact of these uncontrolled sources. In addition to those points discussed in paras 69 to 74 above, the following points should be borne in mind:
    
    • Monitoring should be undertaken at relevant locations (see para 12). Consideration should be given to the siting of dust-emitting processes and their location with regard to these locations.
      
    • Monitoring should focus upon those relevant locations where exposure to dust emissions is likely to be highest (eg downwind from the source);
      
    • Ideally monitoring should be carried out for a period of at least 12 months. Where this is not possible, it is recommended that measurements should be carried out over the summer months, when wind-blown resuspension is likely to be highest;
      
    • In all cases, authorities are advised to compare the results of local monitoring programmes with data from national network sites, to assist with the interpretation of findings.
      
90. Where monitoring indicates that the proposed objectives are likely to be exceeded, then it may be helpful to refine the monitoring strategy, in order to more clearly identify the source contributions. In such cases, authorities may find it useful to:
    
    • Undertake monitoring of wind speed and direction to assist with the interpretation of results and any reported exceedances;
      
    • Carry out monitoring at several locations, including an upwind site. This will allow a more accurate assessment of the source emission contribution to the measured data to be carried out. Alternatively, 'directional' monitoring equipment (which allows measurements to be collected only within a pre-defined wind direction) can be employed;
      
    • Consider the use of various speciation and chemical analysis methods, once again to assess the source emission contribution to the measured data.
      
91. Various examples of PM₁₀ monitoring strategies in the vicinity of uncontrolled dust sources are described with The First Phase Air Quality Review and Assessment: A Summary which has been published by the Department of the Environment, Transport and the Regions, the Welsh Office, the Scottish Office and the Department of the Environment (Northern Ireland).
    

Second Stage          References

---

Report prepared by Stanger Science and Environment
Site prepared by the National Environmental Technology Centre, part of AEA Technology, on behalf of the UK Department of the Environment, Transport and the Regions