Hong Kong

On December 30, 2013, Hong Kong replaced the Air Pollution Index with a new index called the Air Quality Health Index.^[18] This index, reported by the Environmental Protection Department, is measured on a scale of 1 to 10+ and considers four air pollutants: ozone; nitrogen dioxide; sulfur dioxide and particulate matter (including PM10 and PM2.5). For any given hour the AQHI is calculated from the sum of the percentage excess risk of daily hospital admissions attributable to the 3-hour moving average concentrations of these four pollutants. The AQHIs are grouped into five AQHI health risk categories with health advice provided:^[19]

Each of the health risk categories has advice associated with it. At the low and moderate levels the public are advised that they can continue normal activities. For the high category, children, the elderly and people with heart or respiratory illnesses are advised to reduce outdoor physical exertion. Above this (very high or serious), the general public are likewise advised to reduce or avoid outdoor physical exertion.

Mainland China

China's Ministry of Environmental Protection (MEP) is responsible for measuring the level of air pollution in China. As of January 1, 2013, MEP monitors daily pollution level in 163 of its major cities. The AQI level is based on the level of six atmospheric pollutants, namely sulfur dioxide (SO₂), nitrogen dioxide (NO₂), suspended particulates smaller than 10 μm in aerodynamic diameter (PM₁₀),^[20] suspended particulates smaller than 2.5 μm in aerodynamic diameter (PM_2.5),^[20] carbon monoxide (CO), and ozone (O₃) measured at the monitoring stations throughout each city.^[21]

AQI mechanics

An individual score (Individual Air Quality Index, IAQI) is calculated using breakpoint concentrations below, and using same piecewise linear function to calculate intermediate values as the US AQI scale. and The final AQI value can be calculated either per hour or per 24 hours and is the max of these six scores.^[21]

The score for each pollutant is non-linear, as is the final AQI score. Thus an AQI of 300 does not mean twice the pollution of AQI at 150, nor does it mean the air is twice as harmful. The concentration of a pollutant when its IAQI is 100 does not equal twice its concentration when its IAQI is 50, nor does it mean the pollutant is twice as harmful. While an AQI of 50 from day 1 to 182 and AQI of 100 from day 183 to 365 does provide an annual average of 75, it does not mean the pollution is acceptable even if the benchmark of 100 is deemed safe. Because the benchmark is a 24-hour target, and the annual average must match the annual target, it is entirely possible to have safe air every day of the year but still fail the annual pollution benchmark.^[21]

CAQI

As of 2012^[update], the EU-supported project CiteairII argued that the CAQI had been evaluated on a "large set" of data, and described the CAQI's motivation and definition. CiteairII stated that having an air quality index that would be easy to present to the general public was a major motivation, leaving aside the more complex question of a health-based index, which would require, for example, effects of combined levels of different pollutants. The main aim of the CAQI was to have an index that would encourage wide comparison across the EU, without replacing local indices. CiteairII stated that the "main goal of the CAQI is not to warn people for possible adverse health effects of poor air quality but to attract their attention to urban air pollution and its main source (traffic) and help them decrease their exposure."^[23]

The CAQI is a number on a scale from 1 to 100, where a low value means good air quality and a high value means bad air quality. The index is defined in both hourly and daily versions, and separately near roads (a "roadside" or "traffic" index) or away from roads (a "background" index). As of 2012^[update], the CAQI had two mandatory components for the roadside index, NO₂ and PM₁₀, and three mandatory components for the background index, NO₂, PM₁₀ and O₃. It also included optional pollutants PM_2.5, CO and SO₂. A "sub-index" is calculated for each of the mandatory (and optional if available) components. The CAQI is defined as the sub-index that represents the worst quality among those components.^[23]

Some of the key pollutant concentrations in μg/m³ for the hourly background index, the corresponding sub-indices, and five CAQI ranges and verbal descriptions are as follows.^[23]

Frequently updated CAQI values and maps are shown on www.airqualitynow.eu^[25] and other websites.^[22] A separate Year Average Common Air Quality Index (YACAQI) is also defined, in which different pollutant sub-indices are separately normalised to a value typically near unity. For example, the yearly averages of NO₂, PM₁₀ and PM_2.5 are divided by 40 μg/m³, 40 μg/m³ and 20 μg/m³, respectively. The overall background or traffic YACAQI for a city is the arithmetic mean of a defined subset of these sub-indices.^[23]

Computing the AQI

The air quality index is a piecewise linear function of the pollutant concentration. At the boundary between AQI categories, there is a discontinuous jump of one AQI unit. To convert from concentration to AQI this equation is used:^[43]

${\displaystyle I={\frac {I_{high}-I_{low}}{C_{high}-C_{low}}}(C-C_{low})+I_{low}}$

(If multiple pollutants are measured, the calculated AQI is the highest value calculated from the above equation applied for each pollutant.)

where:

${\displaystyle I}$ = the (Air Quality) index,

${\displaystyle C}$ = the pollutant concentration,

${\displaystyle C_{low}}$= the concentration breakpoint that is ≤ ${\displaystyle C}$,

${\displaystyle C_{high}}$= the concentration breakpoint that is ≥ ${\displaystyle C}$,

${\displaystyle I_{low}}$= the index breakpoint corresponding to ${\displaystyle C_{low}}$,

${\displaystyle I_{high}}$= the index breakpoint corresponding to ${\displaystyle C_{high}}$.

The EPA's table of breakpoints is:^{[44][45][46][47][48]}

Suppose a monitor records a 24-hour average fine particle (PM_2.5) concentration of 26.4 micrograms per cubic meter. The equation above results in an AQI of:

${\displaystyle {\frac {100-51}{35.4-9.1}}(26.4-9.1)+51=83.2}$

which rounds to index value of 83, corresponding to air quality in the "Moderate" range.^[49] To convert an air pollutant concentration to an AQI, EPA has developed a calculator.^[50]

If multiple pollutants are measured at a monitoring site, then the largest or "dominant" AQI value is reported for the location. The ozone AQI between 100 and 300 is computed by selecting the larger of the AQI calculated with a 1-hour ozone value and the AQI computed with the 8-hour ozone value.

Eight-hour ozone averages do not define AQI values greater than 300; AQI values of 301 or greater are calculated with 1-hour ozone concentrations. 1-hour SO₂ values do not define higher AQI values greater than 200. AQI values of 201 or greater are calculated with 24-hour SO₂ concentrations.

Real-time monitoring data from continuous monitors are typically available as 1-hour averages. However, computation of the AQI for some pollutants requires averaging over multiple hours of data. (For example, calculation of the ozone AQI requires computation of an 8-hour average and computation of the PM_2.5 or PM₁₀ AQI requires a 24-hour average.) To accurately reflect the current air quality, the multi-hour average used for the AQI computation should be centered on the current time, but as concentrations of future hours are unknown and are difficult to estimate accurately, EPA uses surrogate concentrations to estimate these multi-hour averages. For reporting the PM_2.5, PM₁₀ and ozone air quality indices, this surrogate concentration is called the NowCast. The Nowcast is a particular type of weighted average that provides more weight to the most recent air quality data when air pollution levels are changing.^[51][52]

Public availability of the AQI

Real time monitoring data and forecasts of air quality that are color-coded in terms of the air quality index are available from EPA's AirNow web site.^[53] Other organizations provide monitoring for members of sensitive groups such as asthmatics, children and adults over the age of 65.^[54] Historical air monitoring data including AQI charts and maps are available at EPA's AirData website.^[55] There is a free email subscription service for New York inhabitants – AirNYC.^[56] Subscribers get notifications about the changes in the AQI values for the selected location (e.g. home address), based on air quality conditions. A detailed map containing current AQI levels and a two-day AQI forecast is available at the Aerostate web site.^[57]

Regulatory Air Monitors and Low Cost Sensors

Historically, EPA has only allowed data from regulatory monitors operated by regulatory or public health professionals to be included in its real time national maps.^[58][59] In the past decade, low cost sensors (LCS's) have become increasingly popular with citizen scientists, and large LCS networks have sprung up in the US and across the globe. Recently, EPA has developed a data correction algorithm^[60][61] for a particular brand of PM_2.5 LCS (the Purple Air monitor) that makes the LCS data comparable to regulatory data for the purpose of computing the AQI. This corrected LCS data currently appears alongside regulatory data on EPA's national fire map.^[62]

History of the AQI

The AQI made its debut in 1968, when the National Air Pollution Control Administration undertook an initiative to develop an air quality index and to apply the methodology to Metropolitan Statistical Areas. The impetus was to draw public attention to the issue of air pollution and indirectly push responsible local public officials to take action to control sources of pollution and enhance air quality within their jurisdictions.

Jack Fensterstock, the head of the National Inventory of Air Pollution Emissions and Control Branch, was tasked to lead the development of the methodology and to compile the air quality and emissions data necessary to test and calibrate resultant indices.^[63]

The initial iteration of the air quality index used standardized ambient pollutant concentrations to yield individual pollutant indices. These indices were then weighted and summed to form a single total air quality index. The overall methodology could use concentrations that are taken from ambient monitoring data or are predicted by means of a diffusion model. The concentrations were then converted into a standard statistical distribution with a preset mean and standard deviation. The resultant individual pollutant indices are assumed to be equally weighted, although values other than unity can be used. Likewise, the index can incorporate any number of pollutants although it was only used to combine SO_x, CO, and TSP^[64] because of a lack of available data for other pollutants.

While the methodology was designed to be robust, the practical application for all metropolitan areas proved to be inconsistent due to the paucity of ambient air quality monitoring data, lack of agreement on weighting factors, and non-uniformity of air quality standards across geographical and political boundaries. Despite these issues, the publication of lists ranking metropolitan areas achieved the public policy objectives and led to the future development of improved indices and their routine application.