WHO/OMS WHO UPDATED GLOBAL DATABASE FOR AIR QUALITY AND HEALTH UNDER STRICT EMBARGO Do not publish, distribute or reproduce before 11 am Geneva time (9 o’clock GMT) Wednesday 7 May 2014. Ne pas publier, distribuer ni reproduire avant 09.00 GMT le mercredi 7 mai 2014 WHO’s Ambient Air Pollution database ‐ Update 2014 Data summary of the AAP database The 2014 version of the Ambient Air Pollution (AAP) database consists mainly of urban air quality data – annual means for PM10 and/or PM2.5 – for about 1 600 cities from 91 countries for the years 2008‐2013. The regional distribution of cities documented in the database, and the number of cities with accessible data by urban inhabitants are described in Table 1 and Figure 1, respectively. Table 1: Total number of cities in AAP database, 2014 version, by region Region Number of cities Number of Total number of countries countries in region Africa (Sub‐Saharan) 16 6 47 America, LMI 535 13 26 America, HI 88 4 9 Eastern Mediterranean, LMI 14 6 15 Eastern Mediterranean, HI 12 5 6 Europe, LMI 109 8 20 Europe, HI 462 29 33 South‐East Asia 167 9 11 Western Pacific, LMI 133 5 21 Western Pacific, HI 93 6 6 World 1 629 91 194 LMI: Low‐and middle‐income; HI: high‐income. Figure 1: Number of cities with accessible PM10 and PM2.5 data in 2014 per urban population 1.8 PM2.5 urban 1.6 PM10 1.4 million 1 1.2 per 1.0 data 0.8 inhabitants PM 0.6 with 0.4 cities of 0.2 No ‐ PM10/2.5: Fine particulate matter of 10/2.5 microns or less; Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia; Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income. PM2.5 measurements can directly be linked to estimates of health risks, and are therefore of particular interest. PM10 measurements first need to be converted to PM2.5 in order to do so. In high‐income countries, PM2.5 measurements are already being widely performed. In low‐ and middle‐income countries, however, while PM2.5 measures are increasingly being developed, they are not yet available in 1 many countries. In low‐and middle‐income countries, annual mean PM2.5 measurements could be accessed in only 69 cities, but PM10 in as many as 512 cities. In high‐income countries, 816 cities with PM2.5 measures could be accessed, against 544 cities with PM10 measurements. An overview of PM10 levels for the WHO regions and selected cities is presented in Figure 2 and 3. Figure 2: PM10 levels by region, for the last available year in the period 2008‐2012. 250 208 200 ] 3 128 150 g/m [ 78 100 87 71 51 PM10 49 26 50 0 PM10: Fine particulate matter of 10 microns or less; Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia; Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income. PM10 values are regional urban population‐weighted. Figure 3: PM10 levels for selected cities by region, for the last available year in the period 2008‐2012. 350 300 250 200 [ug/m3] 10 150 PM 100 50 0 Noi city Alto Jose Paris Sofia Aires Louis Delhi Cairo Doha Seoul Town Accra Roma Dakar Dhabi Beirut Dhaka Lisboa Hà Beijing Manila Ankara Tehran Jeddah Janeiro Muscat Belgrad Caracas Amman Karachi Pretoria San Moscow Bangkok Santiago Colombo Shanghai Auckland Bucuresti Port Singapore Hiroshima de Warszawa Abu Vancouver Cape Melbourne København Mexico Kathmandu Puente Washington Montevideo Ulaanbaatar Buenos Rio Afr Amr LMI Amr HI Emr LMI Emr HI Eur LMI Eur HI Sear Wpr LMI Wpr HI PM10: Fine particulate matter of 10 microns or less; Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia; Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income. 2 Figure 4: Annual mean PM of the assessed urban population compared to the WHO Air Quality Guidelines (AQG)a 100% 80% 60% assessed exceed AQG 40% of population 20% meet AQG % 0% Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia; Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income; AQG: WHO Air Quality Guidelines. a Annual mean PM10: 20 g/m3; Annual mean PM2.5: 10 g/m3. Figure 4 shows the regional and global percentages of the assessed urban population experiencing PM10 or PM2.5 air pollution levels that meet or exceed the WHO Air Quality Guidelines annual mean values of 3 3 1 20 g/m (for PM10) and 10 g/m (for PM2.5) . Globally, according to the currently available data, 12% of the assessed population are exposed to PM10 or PM2.5 annual mean levels complying with AQG levels. 3 3 This figures raises to 27% for the interim target 3 (IT‐3, 30 g/m for PM10 and 15 g/m for PM2.5) of the 3 3 3 AQG, 49% for IT‐2 (50 g/m for PM10 and 25 g/m for PM2.5), and 62% for IT‐1 (70 g/m for PM10 and 3 35 g/m for PM2.5). Comparison of urban air pollution levels in recent years A total of 851 cities in 72 countries are present in both the 2011 and 2014 versions of the database, with air quality data for different years (Table 2). The 2011 version of the database contains data for 2010 or earlier, and the 2014 version for 2012 or earlier. To compare levels of air pollution for an equivalent of a three‐year average for cities present in both versions of the database, a linear interpolation or (rarely) an extrapolation was made. A regional summary is presented by WHO region and income groups (Figure 5). Table 2: Number of cities present in both the 2011 and the 2014 versions of the AAP database, by region Region Number of cities Number of countries Africa (Sub‐Saharan) 4 3 America, LMI 25 10 America, HI 406 4 Eastern Mediterranean, LMI 9 4 Eastern Mediterranean, HI 3 2 Europe, LMI 56 10 Europe, HI 236 24 South‐East Asia 52 8 Western Pacific, LMI 34 3 Western Pacific, HI 26 4 World 851 72 LMI: Low‐ and middle‐income countries; HI: High‐income. 1 For cities with both PM10 and PM22.5 values (337), PM2.5 were used. 3 Globally, annual PM10 levels are estimated to increase by 6% during the recent three‐year periods (2009 to 2012 or earlier period) as assessed in cities present in both databases and weighted by regional urban population. 1 Figure 5: Regional city‐population weighted comparisons in annual mean PM10 for a three‐year period, by region, for cities present in both versions of the database 250 +8% 200 150 +14% [ug/m3] +26% ‐4% 100 +6% PM10 +16% ‐13% 50 ‐2% 0 2009 or earlier 2012 or earlier PM10: Fine particulate matter of 10 microns or less; Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia, Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income. Results are based on 851 cities and are to be interpreted with caution, as yearly variations due for example to climatic changes can be important and 3‐year comparisons do not necessarily represent trends, in particular when changes are limited. 1 The mean for the World is based on weighting by regional urban population. Figure 6 shows the percentage of cities with decreasing levels of annual mean PM10 (in green), increasing levels (in light orange), and levels with changes of ≤3% per year (in blue), by region. The variation in population living in cities with increasing or decreasing population levels is represented in Figure 7. 4 Figure 6: Percentage of cities with increasing and decreasing PM10 annual means, by region. 100% 80% 60% 40% 20% 0% Afr Amr Amr HI Emr Emr HI Eur LMI Eur HI Sear Wpr Wpr HI World* LMI LMI LMI % cities decreasing Limited change (≤3% per year) % cities increasing Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia, Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income. *The world figure is regional population‐weighted. Figure 7: Percentage of city population experiencing increasing and decreasing PM10 annual means, by region. 100% 80% population 60% city of 40% 20% Percentage 0% Afr Amr Amr HI Emr LMI Emr HI Eur LMI Eur HI Sear Wpr Wpr HI World LMI LMI with decreasing AP Limited change (≤3% per year) with increasing AP Afr: Africa; Amr: America; Emr: Eastern Mediterranean; Eur: Europe; Sear: South‐East Asia, Wpr: Western Pacific; LMI: Low‐ and middle‐income; HI: high‐income. 5 Limitations The presented comparison of air pollution levels has a number of limitations: The period of comparison is relatively short. Yearly variations may for example be influenced by the weather and a comparison of three‐year periods may not be sufficient to reflect a longer term trend. Therefore, three‐year running means were used where available. A longer time period of comparison is however required to confirm any trends. The sampling locations may have changed within the period of comparison, and a variation in annual mean PM levels of a city may reflect different sampling locations rather than a trend. Measurement locations are however reasonably stable over time. For further information, please contact: Public Health, Social and Environmental Determinants of Health Department, World Health Organization, 1211 Geneva 27, Switzerland Website: www.who.int/phe © World Health Organization 2014 All rights reserved.