Air quality expert group - Fine particulate matter (PM2.5) in ... - Defra

More documents

Recommendations

Info

Measuring PM2.5 and its components cut-off for particle sampling is around 4.5 µm (Tang et al., 2009). Denuder and filter samples are analysed at CEH Lancaster using ion chromatography (for the anions), flow injection permeation/conductivity (for ammonium ions) and ICP-OES (for the metals). 2.5 Summary 2.5.1 What does the PM2.5 metric measure? 73. PM2.5 data show, in principle, the mass concentration of airborne particles with an aerodynamic diameter less than 2.5 µm, the size range being based on the high risk respirable convention for the human lung, and the size-selection curve being set by the reference inlet system in EN 14907. In practice, measurement of the mass is complicated by the presence of semi-volatile particles, variations in water content and other factors, such that the PM2.5 metric does not correspond to definite physical or chemical components of the air, but is in effect defined by the measurement method. Within the United Kingdom AURN, the relevant measurement method is that set out in CEN standard EN 14907, the method referred to by the Ambient Air Quality Directive. 74. The difficulties of PM measurement are reflected in the fact that the required measurement uncertainty for PM in the Directive is ±25% with a 95% level of confidence, at concentrations close to the limit value, while for most gaseous pollutants the comparable value is ±15%. 75. Considerable effort is spent ensuring that reported AURN PM2.5 data are both internally consistent and comparable with reference method data. It should be appreciated, however, that the most widely used instrumentation (FDMS) is based on relatively new and complex technology designed to provide an automatic equivalent to the reference method, which is not directly amenable to automation. 2.5.2 How do PM10 and PM2.5 measurement issues compare? 76. The uncertainties in PM2.5 data, expressed as percentages, are inherently larger than for PM10 data. This is because the absolute PM2.5 mass is smaller, making variations in the mass of the filter (required by the reference method) more significant, and also because in general the PM2.5 fraction will contain a larger proportion of semi-volatile and hygroscopic material, which means that the PM2.5 mass is subject to more variation due to environmental conditions during and after sampling. Conclusions drawn from PM2.5 data must therefore be qualified by these inherent measurement limitations. 2.5.3 Do we have robust measurements of PM2.5? 77. There are at least three aspects to this question. First, we need to consider whether the AURN measurements meet the reporting requirements of the Directive, specifically in terms of measurement uncertainty (±25%) and data capture (required to be greater than 90%). Data capture refers to the amount of data meeting the uncertainty requirement compared with the largest achievable set of data for the year. In 2009, the network mean data capture for PM2.5 was 85.8%, with 34 out of 76 sites falling below 90%. This compares with an 39
PM2.5 in the UK 40 average of 93.6%, with 13 out of 81 sites falling below 90% for the relatively simple ozone measurement, and 89.6%, with 34 out of 115 sites falling below 90% for the more complicated NO2 measurements. In 2010, the corresponding numbers were 82.6% for PM2.5 (40/78), 92.7% for ozone (15/80) and 90.5% for NO2 (26/117). The robustness of the data therefore falls short of the Directive requirements, and of that achieved by other pollutants. It must be appreciated, however, that technical discussions on how best to operate automated PM2.5 monitoring networks and assess their uncertainties are currently active at a European level, although they are at a much less advanced stage than for gaseous pollutants. UK representatives are prominent in these discussions. 78. Second, we need to consider whether conclusions about changes smaller than the ±25% uncertainty required by the Directive can be drawn from UK data. Much of the measurement uncertainty is associated with differences from the reference method, which is itself more intrinsically uncertain than those used for gaseous pollutants. Data obtained using the same type of instrument and the same QA/QC procedures (such as FDMS data from individual sites in the AURN) are expected to be comparable with each other such that variations are significantly less than 25%. Variations will be less when longer term averages are taken, removing random variations. However, relevant practical issues in the operation of such relatively complicated instruments are still being discovered and evaluated, and it is difficult to put a precise figure on the relative uncertainties. 79. This aspect is directly relevant to checking compliance with the exposure reduction target in the 2008 Air Quality Directive, based on assessing the national PM2.5 average exposure indicator (AEI) over periods ten years apart. For the United Kingdom, this is expected to mean complying with a 15% reduction target. If there is any significant change in the monitoring instrumentation during the ten-year period, the measured change in the AEI is likely to have a large relative uncertainty, even if attempts are made to correct for the effects of the instrument change. If the same FDMS instruments are used throughout the period, the operational issues may still mean difficulties in quantifying the change in the AEI with sufficient accuracy. Of course, all measurements taken to determine compliance with a limit value or target value have an associated measurement uncertainty, so there is nothing new in the fact that uncertainties can obscure a clear-cut result; however, the nature of the exposure reduction target and the difficulties of PM2.5 measurement conspire to mean that the available data may not be fit for purpose. 80. A further aspect to consider is whether the measurements are robust enough to improve our understanding of the sources of primary PM2.5 and PM2.5 precursors, together with the chemical and other processes involved in PM2.5 formation and evolution, or in other words whether we can use PM measurements effectively to evaluate PM2.5 models and emissions inventories. As discussed in Chapter 5, the uncertainties in PM2.5 measurement data make them far from ideal for this purpose. This situation can be seen as the result of a combination of factors, namely that the PM2.5 metric is defined operationally (making it difficult to model a priori), the relatively large uncertainties that arise in the practical measurement of the metric (as described in this chapter), and the inherent complexities in the formation and evolution of airborne particles.
Page 1 and 2: AIR QUALITY EXPERT GROUP Fine Parti
Page 3 and 4: This is a report from the Air Quali
Page 5 and 6: II Membership Chair Professor Paul
Page 7 and 8: IV Acknowledgements The Air Quality
Page 9 and 10: VI 2.5 Summary 39 2.5.1 What does t
Page 11 and 12: VIII 5.4 Components of PM2.5 and so
Page 13 and 14: PM2.5 in the UK 2 Air Quality Direc
Page 15 and 16: PM2.5 in the UK 4 I.2.1 Concentrati
Page 17 and 18: PM2.5 in the UK 6 • PM2.5 emissio
Page 19 and 20: PM2.5 in the UK 8 I.4.1 Modelling r
Page 21 and 22: PM2.5 in the UK 10 4. In broad term
Page 23 and 24: PM2.5 in the UK Table 1.2: National
Page 25 and 26: PM2.5 in the UK 14 17. Dry depositi
Page 27 and 28: PM2.5 in the UK 16 in ozone in the
Page 29 and 30: PM2.5 in the UK 18
Page 31 and 32: PM2.5 in the UK 20 5. The exact for
Page 33 and 34: PM2.5 in the UK 22 12. These diffic
Page 35 and 36: PM2.5 in the UK 24 2.3 Methods used
Page 37 and 38: PM2.5 in the UK 26 (e) During the
Page 39 and 40: PM2.5 in the UK 28 2.3.3 The Partis
Page 41 and 42: PM2.5 in the UK 30 35. For non-auto
Page 43 and 44: PM2.5 in the UK 32 43. Ratification
Page 45 and 46: PM2.5 in the UK 34 49. Many of the
Page 47 and 48: PM2.5 in the UK 36 57. Black carbon
Page 49: PM2.5 in the UK 38 2.4.4 PM2.5 elem
Page 53 and 54: PM2.5 in the UK Annex 1 - PM2.5 mon
Page 55 and 56: PM2.5 in the UK 44
Page 57 and 58: PM2.5 in the UK 46 3.2.1 Diurnal va
Page 59 and 60: PM2.5 in the UK 48 PM2.5/µgm -3 20
Page 61 and 62: PM2.5 in the UK 50 3.2.4 Seasonal v
Page 63 and 64: PM2.5 in the UK 52 12. Results are
Page 65 and 66: PM2.5 in the UK 54 3.4.2 Relationsh
Page 67 and 68: PM2.5 in the UK 56 Correlation 0.8
Page 69 and 70: PM2.5 in the UK Figure 3.11: Polar
Page 71 and 72: PM2.5 in the UK 60 south. Further a
Page 73 and 74: PM2.5 in the UK 62 3.7 PM2.5 episod
Page 75 and 76: PM2.5 in the UK 64 secondary ammoni
Page 77 and 78: PM2.5 in the UK 66 3.8 PM2.5 concen
Page 79 and 80: PM2.5 in the UK 68 3.9 Composition
Page 81 and 82: PM2.5 in the UK 70 43. The UK has i
Page 83 and 84: PM2.5 in the UK 72 NO 3 6 4 2 2000
Page 85 and 86: PM2.5 in the UK 74 54. PM ammonium
Page 87 and 88: PM2.5 in the UK 76 58. Data from th
Page 89 and 90: PM2.5 in the UK 78 (k) Road traffic
Page 91 and 92: PM2.5 in the UK 80 provided by spec
Page 93 and 94: PM2.5 in the UK 82 14. The London L
Page 95 and 96: PM2.5 in the UK Table 4.1: UK emiss
Page 97 and 98: PM2.5 in the UK 86 24. National inv
Page 99 and 100: PM2.5 in the UK 88 Figure 4.5: Spat
Page 101 and 102:
PM2.5 in the UK 90 either nanoparti
Page 103 and 104:
PM2.5 in the UK 92 37. Further cons
Page 105 and 106:
PM2.5 in the UK Emissions (ktonnes)
Page 107 and 108:
PM2.5 in the UK 96 Figure 4.7: Spat
Page 109 and 110:
PM2.5 in the UK 98 4.4 A critical a
Page 111 and 112:
PM2.5 in the UK 100 60. At a local
Page 113 and 114:
PM2.5 in the UK 102 measurements an
Page 115 and 116:
PM2.5 in the UK 104 4.5.2 Quantifyi
Page 117 and 118:
PM2.5 in the UK 106 77. Early effor
Page 119 and 120:
PM2.5 in the UK 108 4.6.1 Receptor
Page 121 and 122:
PM2.5 in the UK 110 is predicated o
Page 123 and 124:
PM2.5 in the UK 112 PM2.5 contribut
Page 125 and 126:
PM2.5 in the UK 114 inorganic parti
Page 127 and 128:
PM2.5 in the UK 116 93. While there
Page 129 and 130:
PM2.5 in the UK 118 100% 90% 80% 70
Page 131 and 132:
PM2.5 in the UK 120 105. Receptor m
Page 133 and 134:
PM2.5 in the UK 122
Page 135 and 136:
PM2.5 in the UK 124 5. At the regio
Page 137 and 138:
PM2.5 in the UK 126 15. Some of the
Page 139 and 140:
PM2.5 in the UK 128 Figure 5.2: Tot
Page 141 and 142:
PM2.5 in the UK 130 µgm -3 160 140
Page 143 and 144:
PM2.5 in the UK 132 PM component (
Page 145 and 146:
PM2.5 in the UK 134 with larger dev
Page 147 and 148:
PM2.5 in the UK 136 Figure 5.8: UKI
Page 149 and 150:
PM2.5 in the UK Figure 5.9: Compari
Page 151 and 152:
PM2.5 in the UK 140 PM2.5 concentra
Page 153 and 154:
PM2.5 in the UK 142 the North Sea (
Page 155 and 156:
PM2.5 in the UK 144 53. Modelling r
Page 157 and 158:
PM2.5 in the UK Annex 2: PM modelli
Page 159 and 160:
PM2.5 in the UK 148 2. Figure A2.1.
Page 161 and 162:
PM2.5 in the UK 150 6. The non-line
Page 163 and 164:
PM2.5 in the UK 152 Model 50 45 40
Page 165 and 166:
PM2.5 in the UK 154 (a) (b) modPM 2
Page 167 and 168:
PM2.5 in the UK 156 NO3 - concentra
Page 169 and 170:
PM2.5 in the UK 158 26. This was in
Page 171 and 172:
PM2.5 in the UK 160 SIA components
Page 173 and 174:
PM2.5 in the UK 162 x 50 km grid fo
Page 175 and 176:
PM2.5 in the UK 164 A2.6: Photochem
Page 177 and 178:
PM2.5 in the UK 166 PM2.5 Emission
Page 179 and 180:
PM2.5 in the UK 168 A2.8: PCM PM mo
Page 181 and 182:
PM2.5 in the UK 170 illustrated by
Page 183 and 184:
PM2.5 in the UK 172 trajectory mode
Page 185 and 186:
PM2.5 in the UK 174
Page 187 and 188:
PM2.5 in the UK 176 4. The delivery
Page 189 and 190:
PM2.5 in the UK 178 12. With respec
Page 191 and 192:
PM2.5 in the UK 180
Page 193 and 194:
PM2.5 in the UK Chapter 2: Measurin
Page 195 and 196:
PM2.5 in the UK RoTAP (2012). Revie
Page 197 and 198:
PM2.5 in the UK Harrison R.M. and Y
Page 199 and 200:
PM2.5 in the UK Chapter 5 Modelling
Page 201 and 202:
PM2.5 in the UK Jones A.M., Harriso
Page 203:
PB13837 Nobel House 17 Smith Square
show all

Air quality expert group - Fine particulate matter (PM2.5) in ... - Defra

Create successful ePaper yourself

Delete template?

Save as template?