At the 2014 annual Dispersion Modellers user group meeting guest speaker Christine McHugh spoke on the topic: 'Comparison of Air Quality in World Cities'

1. Comparison of Air Quality in World Cities
Dr. Christine McHugh, Amec Foster Wheeler
Dispersion Modelling Users Group, 2nd December 2014

2. Contents
 The Scope of the Report
 Ranking Method
 About the Monitoring Data
 Type of the monitoring sites
 Siting criteria
 Number of monitoring sites
 About the Air Quality Limits
 Comparison of London with Other Cities
 Summary
1
Disclaimer:This presentation is based on a report prepared by AMEC Environment & Infrastructure UK Limited for the GLA and TfL

3. The Scope of the Report
2

4. The Scope of the Report
 http://www.london.gov.uk/priorities/environment/clearing-londons-
air/useful-documents
 A global comparison - air quality in London is compared with air
quality in cities around the world, based on monitored data
 The global comparison goes beyond previous studies in terms of the
breadth of cities considered, the number of pollutants included and use
of recent data
 To make a wide global comparison (39 cities) using the available
data, only annual averages of pollution (long term) have been used
 The ranking methodology developed is flexible enough that it could be
used to look at short term measures of pollutions for those cities with
suitable data
3

5. The Ranking Method
4

6. Issues to Consider
 Quality and quantity of data
 Number, location and type of monitoring station
 Geographical and meteorological factors
 An index of multiple pollutants
 The role of short term exceedences
5

7. Existing Ranking Methods
 The European Environment Agency - State and Outlook 2010 (EEA)
 Aphekom Project 2008-2011
 Air Pollution at Street Level in European Cities (EEA)
 Soot-free for the Climate!
 Perception of Air Quality 2009 - Urban Audit
 WHO Urban Outdoor Air Pollution Database
 Environment Canada
 Some methods are highly subjective!
6

8. Existing Ranking Methods
 Aphekom Project 2008-2011
 Respondents were asked for
their perception of a wide variety
of issues within their city, these
included air quality and poverty.
 Size of the city seems to matter.
 17 out of the 23 cities where the
majority of respondents thought
that air pollution was not a major
problem have 500,000 or fewer
inhabitants.
 9 out of the 13 cities with the
most unfavourable perception of
air pollution have more than
500,000 inhabitants.
7

9. Existing Ranking Methods
 Soot-free for the Climate!
 Data gathered by questionnaire, some cities even if questionnaire not
returned
 Information supplied by the cities unlikely to be comparable e.g. reduction
success, participation, information
 A measure such as increasing public awareness carried the same weight
as a measure such as the implementation of an LEZ such as the London
LEZ, and the London LEZ that is rigorously enforced could be given the
same weight as an LEZ that has no system of enforcement;
 Differences in governance between cities would account for differences in
responses
 Most measures relate to changes in policy rather than measuring
improvement of present concentrations. High rankings attached to
willingness to improve, regardless of whether air quality actually improves
and regardless of the current levels of pollution
 The only category relating to the reduction of measured concentrations
only considers PM10
8

10. Existing Ranking Methods
 Hairy nose index
 Based on the joke premise
that the more polluted a city’s
air is, the more nasal hair
people will need to be able to
“survive without clean air”.
9
http://cleanairasia.org/hairynose/home

11. Existing Indices Reporting Air Pollution Levels
 UK Air Quality
Index
 CITEAIR
Common Air
Quality Index,
CAQI (right).
Also a year
average index
(YACAQI)
 WHO Air
Quality
Guidelines
10

12. The Ranking Method
 A multi-pollutant weighted index of annual average concentrations
 Concentrations are normalised with respect to an annual average value
such as the EU limit value
 The ranking methodology proposed is flexible
 The index used has a pollutant mixture and weighting suitable for a
mixture of traffic, industrial and fossil-fuelled heating sources
 Citywide - general - NO2: 0.3; PM10: 0.3; SO2:0.3; PM2.5: 0.1;
 Sensitivity tests considered a weighting scheme suitable for cities with
traffic as the dominant source and a weighting that reflects the relative
health impacts of different pollutants:
 Citywide/Traffic Focussed - NO2: 0.4; PM10: 0.4; PM2.5: 0.2;
 Health Impacts - NO2: 0.02; SO2:0.03; PM10: 0.71; PM2.5: 0.24.
11

13. About the Monitoring Data
12

14. About the Monitoring Data: Type
 Monitoring sites can be classified as: Traffic/ Urban
Background/Suburban/Rural/Industrial
 Classifications can vary, in some cities traffic monitoring sites are at
least 10m from the kerb
 Not all countries/cities report the type of the monitoring sites
 Amongst cities reporting site type, London has a high proportion of
traffic sites, as do Brussels, Milan, Munich and Stuttgart (but London
has by far the greatest number of sites)
13

15. About the Monitoring Data: Siting Requirements
 EU Directive – European Cities Monitoring Programmes
 Responsibility of the Member State
 Purpose is for judging compliance – sites must be representative
 UK Local Air Quality Management (LAQM)
 Responsibility of local government
 Purpose – can vary, may investigate hot spots
14

16. About the Monitoring Data: Siting Requirements
 EU Directive – European Cities Monitoring Programmes
 “Sampling points shall in general be sited in such a way as to avoid
measuring very small micro-environments in their immediate vicinity, which
means that a sampling point must be sited in such a way that the air
sampled is representative of air quality for a street segment no less than
100 m length at traffic-orientated sites”;
 “The inlet probe shall not be positioned in the immediate vicinity of sources
in order to avoid the direct intake of emissions unmixed with ambient air”.
 UK Local Air Quality Management (LAQM)
 “The site should not be close to local or point emissions sources, unless
these have been specifically targeted for investigation”;
 “Try to site the monitors as near to the point of public exposure as possible”.
15

17. About the Monitoring Data: Number of Sites
 London has a very large number of automatic monitoring sites: 157
 By comparison, Paris has 32 year-round sites
 In EU cities, there tend to be few monitoring sites other than the official
sites for EU reporting – except in London
 In London there are 17 official sites for EU reporting, the remainder
(139) have been located by local government
 London has almost 2 monitoring stations per 100,000 habitants,
bettered only by Amsterdam and Vancouver
 London has 1 monitoring site per 0.1km2, bettered only by Barcelona,
Brussels and Vancouver
16

18. Considering individual sites reported
to the EU for compliance
 In 2010 the highest NO2 concentration was recorded in Florence,
followed by a site in Stuttgart, then Munich, then Marylebone Road in
London
 In 2011, the sites in Florence and Stuttgart recorded higher
concentrations than Marylebone Road.
 In 2013, one site in Paris recorded higher concentrations than
Marylebone Road
 Concentrations at Marylebone Road have fallen each year from 2009:
107.0 µg/m3 >> 98.3 µg/m3 >> 97.2 µg/m3 >> 94.0 µg/m3 >>
80.6µg/m3
17

19. About the Air Quality Limits
18

20. About the Air Quality Limits
 Limits for annual average NO2 across the world
 EU: 40 µg/m3
 US: 101 µg/m3 (53ppb)
 http://www.epa.gov/air/criteria.html
 Chinese API based on daily averages
 Excellent: up to 80 µg/m3
 Good: 81-120 µg/m3
19
http://www.mep.gov.cn/

21. Comparison of London with Other Cities
20

22. The world’s most polluted cities (PM10)
October 2013, Quartz using WHO data on PM10
 1. Ahwaz, Iran
 2. Ulaanbaatar, Mongolia
 3. Sanandaj, Iran
 4. Ludhiana, India
 5. Quetta, Pakistan
 6. Kermanshah, Iran
 7. Peshawar, Pakistan
 8. Gaborone, Botswana
 9. Yasouj, Iran
 10. Kanpur, India
21
http://science.time.com/2013/10/18/the-10-most-polluted-cities-in-the-world/

23. The world’s most polluted cities (PM10)
http://science.time.com/2013/10/18/the-10-most-polluted-cities-in-the-world/
0
50
100
150
200
250
300
350
400
[1st] Ahwaz [10th] Kampur London
Annual
Average
Concentration
of PM10
(mg/m3)
22

24. Comparison for PM10
 Average of 5 years annual averages (2008-2012)
23
Non-EU City PM10 (mg/m3)
Cairo 140
Beijing 122
Mumbai 103
Shanghai 81
Mexico City 56
Rio de Janeiro 55
Jakarta 53
Istanbul 53
Hong Kong 48
Los Angeles 39
Sao Paulo 37
Singapore 27
London 25
EU City PM10 (mg/m3)
Milan 44
Bucharest 40
Warsaw 36
Budapest 33
Rome 32
Paris 32
Stuttgart 30
Barcelona 30
Brussels 28
Munich 28
Prague 27
Berlin 27
Vienna 27
Amsterdam 26
London 25

25. Comparison for PM10
 Average of 5 years annual averages (2008-2012)
24
0
20
40
60
80
100
120
140
160
Non-EU Cities, PM10 (mg/m3)
0
5
10
15
20
25
30
35
40
45
50
EU Cities, PM10 (mg/m3)

26. Comparison for PM2.5
 Average of 5 years annual averages (2008-2012)
25
Non-EU City PM2.5 (mg/m3)
Cairo 140
Beijing 122
Mumbai 103
Shanghai 81
Mexico City 56
Rio de Janeiro 55
Jakarta 53
Istanbul 53
Hong Kong 48
Los Angeles 39
Sao Paulo 37
Singapore 27
London 25
EU City PM2.5 (mg/m3)
Milan 44
Bucharest 40
Warsaw 36
Budapest 33
Rome 32
Paris 32
Stuttgart 30
Barcelona 30
Brussels 28
Munich 28
Prague 27
Berlin 27
Vienna 27
Amsterdam 26
London 25

27. Comparison for PM2.5
 Average of 5 years annual averages (2008-2012)
26
0
5
10
15
20
25
30
35
40
Non-EU Cities, PM2.5 (mg/m3)
0
5
10
15
20
25
30
35
EU Cities, PM2.5 (mg/m3)

28. Comparison for SO2
 Average of 5 years annual averages (2008-2012)
27
EU City SO2 (mg/m3)
Bucharest 10
Madrid 8
Warsaw 7
Budapest 6
Munich 5
Prague 4
Brussels 4
London 4
Non-EU City SO2 (mg/m3)
Jakarta 52
Shanghai 38
Beijing 34
Cairo 31
Hong Kong 21
Mumbai 18
Mexico City 15
New York 12
Singapore 10
Istanbul 8
Sao Paulo 7
Tokyo 5
Rio de Janeiro 5
Chicago 5
London 4

29. Comparison for SO2
 Average of 5 years annual averages (2008-2012)
28
0
2
4
6
8
10
12
EU Cities, SO2 (mg/m3)
0
10
20
30
40
50
60
Non-EU Cities, SO2 (mg/m3)

30. Comparison for NO2
 Average of 5 years annual averages (2008-2012)
29
EU City NO2 (mg/m3)
Stuttgart 71
Milan 60
Munich 56
London 52
Non-EU City NO2 (mg/m3)
Mexico City 103
Hong Kong 69
Istanbul 66
Beijing 53
Shanghai 53
London 52

31. Comparison for NO2
 Average of 5 years annual averages (2008-2012)
30
0
10
20
30
40
50
60
70
80
EU Cities, NO2 (mg/m3)
0
20
40
60
80
100
120
Non-EU Cities, NO2 (mg/m3)

32. Comparison of NO2 for traffic sites only
 Average of 5 years annual averages (2008-2012)
31
EU City NO2 (mg/m3)
Mexico City* 102
Stuttgart 82
Paris 72
Hong Kong* 70
Munich 69
Rome 68
Milan 66
Istanbul* 65
Bucharest 65
London 63
0
20
40
60
80
100
120
NO2 at traffic sites (mg/m3)
*all monitoring stations, not just traffic sites

33. Summary
32

34. Citywide Index - Ranking
33
City Rank
Vancouver 1
Sydney 2
Stockholm 3
Vienna 4
Berlin 5
Amsterdam 6
Chicago 7
Singapore 8
Prague 9
Frankfurt 10
Brussels 11
Paris 12
Budapest 13
Los Angeles 14
London 15
Barcelona 16
New York 17
Madrid 18
City Rank
Tokyo 19
Warsaw 20
Rome 21
Munich 22
Sao Paulo 23
Moscow 24
Bucharest 25
Stuttgart 26
Rio de Janeiro 27
Milan 28
Istanbul 29
Hong Kong 30
Mexico city 31
Jakarta 32
Mumbai 33
Shanghai 34
Beijing 35
Cairo 36

35. Citywide/Traffic Focussed Index - Ranking
34
City Rank
Vancouver 1
Sydney 2
Singapore 3
Stockholm 4
Chicago 5
Vienna 6
New York 7
Prague 8
Berlin 9
Amsterdam 10
Moscow 11
Madrid 12
Frankfurt 13
Brussels 14
Budapest 15
Warsaw 16
London 17
Paris 18
City Rank
Barcelona 19
Los Angeles 20
Sao Paulo 21
Jakarta 22
Munich 23
Bucharest 24
Rome 25
Stuttgart 26
Tokyo 27
Milan 28
Rio de Janeiro 29
Hong Kong 30
Istanbul 31
Shanghai 32
Mexico city 33
Beijing 34
Cairo 35
Mumbai 36

36. Health Impacts Index - Ranking
35
City Rank
Vancouver 1
Sydney 2
New York 3
Stockholm 4
Chicago 5
Madrid 6
Tokyo 7
Frankfurt 8
London 9
Moscow 10
Amsterdam 11
Prague 12
Vienna 13
Singapore 14
Munich 15
Berlin 16
Brussels 17
Barcelona 18
City Rank
Stuttgart 19
Paris 20
Rome 21
Budapest 22
Sao Paulo 23
Los Angeles 24
Warsaw 25
Bucharest 26
Milan 27
Hong Kong 28
Mexico city 29
Istanbul 30
Rio de Janeiro 31
Jakarta 32
Shanghai 33
Mumbai 34
Beijing 35
Cairo 36

37. Summary
 How does London’s air quality compare with that in cities around the
world?
 There are many cities around the world with exceedingly high levels of particulate
pollution
 Considering the pollutants of concern in the EU, London has low levels of PM10, PM2.5
and SO2
 On a citywide basis the NO2 concentration in London is lower than that in Stuttgart,
Milan and Munich (and cities outside the EU)
 Considering traffic sites only, the NO2 concentration in London is lower than that in
Stuttgart, Paris, Munich, Rome, Milan, Bucharest (and cities outside the EU).
 London has many more automatic monitoring sites than other EU cities with high quality
data and easy, transparent access
 Many of the monitoring sites have been sited by local government investigating hot
spots rather than by Member States looking for representative locations. Note that in
some (non-EU) cities traffic sites are at least 10m from the kerb.
36