13 October 2014

Understanding Pollutant Standards Index And Air Quality

Last year, when Singapore suffered its worst haze ever, the National Environment Agency ("NEA") issued 24-hour PSI readings and 24-hour PM2.5 concentration levels every hour.

Since 1 April 2014, NEA has changed the way the 24-hour PSI is computed. In addition, it publishes the 3-hour PSI and 1-hour PM2.5 concentration levels every hour.

How much additional value is there really in these data?

PSI Concept
In my article Pollution Standards Index — Facts, Opinions and Misperceptions, I highlighted some conceptual issues regarding the 24-hour PSI.

The revised 24-hour PSI is the highest of the six PSI sub-indices (24-hour PM10, 24-hour sulphur dioxide, 8-hour carbon monoxide, 8-hour ozone, 1-hour nitrogen dioxide and 24-hour PM2.5). The PSI is not a composite or an average of the PSI sub-indices.
Reporting the air quality as determined by the level of the single worst pollutant has been criticised. Multiple pollutants affect our health at the same time, and the possibility of joint impact of different air pollutants on our health should not be ignored.

Reporting the air quality by adding all the sub-indices may over-represent the total health effect, as it assumes that the effects of each pollutant are independent of the others. Some pollutants may have synergistic effects, while others may have opposing effects.

Since small particulate matter is the main pollutant in Singapore and small particulate pollution impacts our health even at very low concentrations and there is no threshold below which there is no damage to health, I will focus on small particulate matter of 2.5 microns or smaller (PM2.5) in the remainder of this article.
24-Hour PSI
If a person is considering whether to live in a certain town or (if he lives there) whether to relocate, he will want to know the long-term (e.g., annual) average PSI reading as well as how often the 24-hour PSI readings are in the moderate, unhealthy, very unhealthy and hazardous range and the peak 24-hour PSI readings.

We can see this in the US Environment Protection Agency's National Ambient Air Quality Standards ("NAAQS") and World Health Organisation ("WHO") guidelines.

EPA's NAAQS for PM2.5 is 12 μg/m3 (micrograms per cubic metre, annual mean, averaged over 3 years) and 35 μg/m3 (98th percentile of 24-hour measurements, averaged over 3 years)[1][2].

WHO guidelines for PM2.5 is 10 μg/m3 (annual mean) and 25 μg/m3 (24-hour mean)[3].

NEA considers the 24-hour PM2.5 level to be healthy up to 50 μg/m3[4], which is, or at least appears to be, significantly higher than the levels in WHO guidelines or EPA NAAQS.

Revised 24-Hour PSI
NEA says we may seem to experience more days of "moderate" air quality with the revised 24-hour PSI[5].

It is not clear why this should be so.

Adding the 24-hour PM2.5 sub-index to the other five sub-indices of the 24-hour PSI will likely change the PSI reading but it will not change the occurrence of "moderate" air quality days. This is because the PSI reading now is the worst of the six component indices; previously, health advisories were based on the 24-hour PSI (the worst of its five component sub-indices) and the 24-hour PM2.5, whichever was worse. Whether before or after 1 April 2014, air quality was (and is) based on the worst concentration of the same set of six pollutants.

3-Hour PSI
If a person is spending, or intends to spend, a few hours outdoors and the pollution situation is deteriorating rapidly, the 24-hour PSI is of little use to him.

Since 1 April 2014, NEA has published 3-hour PSI based on PM2.5 concentrations.

This is an improvement but it still is not good enough because:
■ It is a 3-hour average; and
■ It is a national average.
■ It is only an indicative measure
■ NEA does not classify air quality into the different categories (e.g., healthy, moderate, unhealthy etc.).

3-Hour Average and National Average
The following charts illustrate the severe shortcomings of the 24-hour PSI and 3-hour PSI in the evening of 21 September 2014 when it was apparent by sight and smell that the air quality was changing rapidly.

If we want to protect ourselves from pollution, we need to know the 1-hour PSI at or near our location so that we can decide on the type and duration of any outdoor activity.

In reply to a request for 1-hour PSI readings, NEA said that it already provides 1-hour PM2.5 concentrations for each of the five regions[6].

Although we can infer that NEA does not wish to publish the 1-hour PSI readings, such readings are of limited value unless we know how to use it.

To see the difficulty, let's look at the 3-hour PSI.

Indicative Measure Without Air Quality Categories
NEA says that the 3-hour PSI is just an indicative measure[5].

NEA does not say how it calculates the 3-hour PSI, except that it is tied to PM2.5 concentration. Based on data points over the past three weeks, it appears that NEA uses the same relationship between 3-hour PM2.5 concentration and 3-hour PSI and between 24-hour PM2.5 concentration and 24-hour PSI. Thus, for both the 3-hour average and the 24-hour average, a PM2.5 concentration of 12 μg/m3 corresponds to a PSI reading of 50 and a PM2.5 concentration of 55 μg/m3 corresponds to a PSI reading of 100.

Next, NEA does not use the 3-hour PSI readings to classify air quality into the different categories (e.g., healthy, moderate, unhealthy etc.) as these categories are based on the averaging period of the concentration levels of the respective pollutants, which is 24 hours in the case of PM2.5 and PM10[5]. Thus, whilst a 24-hour PSI reading of 120 (for example) means that the air quality is unhealthy, a 3-hour PSI reading of 120 may or may not be unhealthy but we won't know because NEA doesn't and can't tell us.

Despite the above-mentioned deficiency of the 3-hour PSI, it is still better than nothing. It certainly is better than trying to remember the points at which the 3-hour PM2.5 concentration crosses PSI 50, 100, 200 etc. even if these levels do not correspond to the air quality becoming moderate, unhealthy, very unhealthy etc.

Since NEA has decided to publish the 3-hour PSI readings, it should publish the 1-hour PSI readings also.

Health Advisories
The Ministry of Health's health advisory and the Ministry of Manpower's Guidelines For Employers On Protecting Employees From The Effects Of Haze (updated 19 March 2014) continue to be based on the 24-hour PSI, not the 3-hour PSI nor the 1-hour PM2.5 concentration because, according to the Ministry of Health, scientific and epidemiological studies on the health effects of particulate matter have been based on 24-hour PSI measurements and long-term exposure to air pollution, and there is little robust data on the longer-term effects of short-term exposure to haze like the pattern experienced in Singapore[5].

What We Can Do
Students and employed people, other than the self-employed, have no choice but to follow the health advisories based on the 24-hour PSI.

All other people who can decide for themselves whether to go outdoors and what activities they should undertake outdoors can consider the following suggestions:

▪ Assume that the air quality groupings (e.g., good, moderate, unhealthy etc.) for 24-hour PSI apply to 3-hour PSI. Whilst this is not correct (as explained above), it is conservative because a 3-hour PSI reading of 120 (for example) is generally less harmful than a 24-hour PSI reading of 120.

▪ In the absence of 1-hour PSI readings, convert the 1-hour PM2.5 concentration (preferably for your region) to PSI using the relationship between 24-hour PM2.5 concentration and 24-hour PSI to determine the approximate air quality groupings (see previous point). Whilst this is not correct (as explained above), it is conservative because a 1-hour PSI reading of 120 (for example) is generally less harmful than a 24-hour PSI reading of 120.

▪ If the 3-hour PSI or the 1-hour PM2.5 concentration is rising rapidly, it is likely that the "live" PSI or PM2.5 concentration is higher than the latest reading. There are exceptions, such as when the 1-hour PM2.5 concentration has peaked and is falling.


1. ENVIRONMENTAL PROTECTION AGENCY National Ambient Air Quality Standards (NAAQS) updated 11 Jun 2013.

 2. EPA's PM2.5 annual NAAQS was 15 μg/m3 (annual mean, averaged over 3 years) from 1997 to 2012 when it was reduced to the current 12 μg/m3. The 24-hour NAAQS (98th percentile of 24-hour measurements, averaged over 3 years) was 65 μg/m3 from 1997 to 2006, when it was reduced to 35 μg/m3.

3. WORLD HEALTH ORGANISATION Fact Sheet No. 313 Ambient (Outdoor) Air Quality And Health updated March 2014.

4. NATIONAL ENVIRONMENT AGENCY Computation Of The Pollutant Standard Index (PSI) March 2014.

5. NATIONAL ENVIRONMENT AGENCY FAQs On PSI updated 11 August 2014.

6. NATIONAL ENVIRONMENT AGENCY Hourly PM2.5 Readings Available Online TODAY 11 Oct 2014.

No comments:

Post a Comment