Singapore Haze: How dangerous is PM 2.5?

While the National Environment Agency (NEA) has traditionally reported the Pollutant Standards Index (PSI) which measures the concentration of Particulate Matter less than 10 microns in size (PM 10), experts over the years have been warning that the smaller particles smaller than 2.5 microns or PM 2.5 are actually more dangerous to health.

PM 2.5 from EPA

The physiological reasons are straightforward: at sizes larger than 2.5 microns, the respiratory tract can quite effectively trap and filter away particles. However, at sizes lesser than 2.5 microns, PM 2.5, particles breach these defenses and penetrate deeply into the alveoli of the lungs and may even enter the blood stream. The subsequent effects are still the subject of intense study and a recent article in the Proceedings of the National Academy of Sciences provides a very nice overview. The adverse effects on the lungs are well-known but what is less publicized is the impact on heart disease, stroke and other vascular conditions. The exact mechanism is not clear but one theory is that PM 2.5 triggers an inflammatory response leading to oxidative damage. Similar to the risks from chronic inflammation of any part of the body from any source, a complex cascade of bodily responses then occurs, leading finally to increased plaque formation in the blood vessels.

Thresholds and Dose-Effects

Is there a threshold effect? Is there a dose-effect response? In simple English, these two key concepts in toxicology mean “Is there a minimum level of PM 2.5 below which there is no noticeable physiological effect?” and “Are the harmful effects proportional to the total level of exposure?”

For the former, the answer is unfortunately ‘No’ at the population level. Researchers from the Harvard School of Public Health analyzed data from a study of death rates and PM 2.5 levels in six American cities and concluded there was “no evidence of a threshold down to the lowest levels of PM 2.5”. In fact, the researchers noted that the relationship or curve was quite linear between 0 to 35 micrograms/ m3. Dose Response PM 2.5 Schwarz 2002

It should be noted there is no firm evidence either way at the individual level. A useful but imperfect analogy is that of smoking. At the population level, we know the harmful effects of smoking, but can we predict with certainty for any given individual cancer or chronic lung disease? No.

With regards to dose-effect response, Chen from McGill University in Canada reviewing the scientific literature concludes:

“The latter long-term exposure to PM2.5 increases the risk of non-accidental mortality by 6% per a 10 micrograms/m3 increase, independent of age, gender, and geographic region. Exposure to PM2.5 was also associated with an increased risk of mortality from lung cancer (range: 15% to 21% per a 10 micrograms/m3 increase) and total cardiovascular mortality (range: 12% to 14% per a 10 micrograms/m3 increase).”

In a nutshell: ‘Yes’, the greater the long term exposure, the higher the risk. Again, these numbers reflect observations at the population rather than the individual level.

Safe Levels?

What are the ‘safe’ levels then? Below is the guidance from the American Environmental Protection Agency (EPA):

“With regard to primary (health-based) standards for fine particles (generally referring to particles less than or equal to 2.5 micrometers (mm) in diameter, PM2.5), the EPA is revising the annual PM2.5 standard by lowering the level to 12.0 micrograms per cubic meter so as to provide increased protection against health effects associated with long- and short-term exposures (including premature mortality, increased hospital admissions and emergency department visits, and development of chronic respiratory disease), and to retain the 24-hour PM2.5 standard at a level of 35 micrograms per cubic meter.” [Emphasis added]

In plain English, the EPA recognizes there is an ‘average’ annual as well as a more volatile ’24 hour’ value and both are important in risk assessment.

Singapore’s PM 2.5 levels have in the first 12 days of June been below the EPA 24 hour standard of 35 micrograms/m3. Beginning 13 June, PM 2.5 crossed 35 micrograms/m3 but remained below 100 micrograms/m3 until 18 June when it jumped to 149-169 micrograms/m3, peaking at 304 micrograms/m3 on 20 June and coming down to 71-94 micrograms/m3 yesterday (23 June 2013). Source: NEA

The last 6 days have been above the daily threshold set by the American EPA. This is cause for concern and should warrant precautionary measures as Singaporeans have already taken. Are the numbers cause for alarm? We don’t know but the graph below from Bloomberg which shows Beijing’s PM 2.5 levels in January 2013 help put things into perspective.

Beijing PM 2.5 Jan 2013 Bloomberg

What’s salient to note for Singaporeans living with PM 2.5 values generally in the 100-200 range during this period? Firstly, the average PM 2.5 measured across airport smoking lounges in the USA is 166.6 micrograms/m3. Secondly, Chinese in Beijing live with PM 2.5 levels virtually always above the World Health Organization’s ‘Healthy Level’ and the peak of 886 micrograms/m3 is almost three times Singapore’s highest ever of 304 micrograms/m3.

‘Keep calm and carry on’

What should Singaporeans do then? Various government agencies such as the Ministry of Health have issued guidance which should be heeded. I would encourage adhering to the guidance and not being alarmist. Check the daily PSI and the PM 2.5 levels and decide on what precautionary measures to take for that day. The London Smog of 1952 which saw cows keeling over had PM 2.5 levels as high as 4,500 micrograms/m3 which Singapore is a long way from.

Given the science including the papers cited above, I do wish the MOH would issue advice based on PM 2.5 levels too. The Americans have moved away from PM 10 and instead use PM 2.5 in computing their Air Quality Index (AQI). It should be noted also that PM 2.5 refers only to the size of the particles and not what the particles are made of. The hazardous effects of PM 2.5 would naturally depend also on what particles make up PM 2.5. AQI in addition to PM 2.5 also factors in carbon monoxide, ozone, sulfur dioxide and nitrogen dioxide and it is the composite that makes up AQI. PSI comprises five components (PM 10, SO2, CO, ozone, NO2) but the NEA only reports levels of the individual components without annotation.

An EPA guideline dated May 2006 recommends reporting both the AQI and the responsible pollutant, e.g. AQI: 162, Responsible Pollutant: Ozone. This helps as different pollutants affect the body in different ways- the report lists people with asthma as “the group most at risk from sulfur dioxide” and people with heart disease as “group most at risk from carbon monoxide”. In some of the papers cited above, the particles arising from traffic seem more hazardous than those emanating from say pollen, and it would be wonderful if the relevant government agency could provide the public with some indication on what the ‘haze’ chemically comprises of.

[24 June 2013 7.25 am:

Microns/m3 replaced by micrograms/m3 as unit of measurement for PM 2.5. Apologies for the confusion. A micron is one millionth of a meter

Edited to include component measurements of PSI and EPA recommendations for reporting of AQI]

[25 June 2013 2.40 pm:

Correction: NEA does list in table form without annotation the levels of individual components and has also stated that PM 10 is the dominant pollutant- “During haze, the pollutant that records the highest concentration is PM10.”. It is not clear what chemical substances make up PM 10. The blog post has been amended]



  1. Lim Wee Cheong · · Reply

    I think its microgram/m3, not micron/m3. Occurs several times in your article.

    1. Yes, you are right. My apologies for confusing. Have edited the blog post. Thanks Wee Cheong for pointing it out.

  2. NEA started reporting the levels of the individual components on Saturday evening. You can find them at

    1. Thanks for informing. It’s useful to know. Would be even better if framed in layman terms just what the numbers mean in guiding daily behavior. This example from the EPA 2006 guidance illustrates what the Americans think should be informed:

      “The AIR QUALITY yesterday in Northeast Baltimore was Unhealthy for Sensitive Groups due to ozone. Groups likely to be sensitive to ozone include active children and adults, and people with respiratory disease such as asthma. The Air Quality Index was 110, resulting from an ozone concentration of 0.088 ppm. Elsewhere in Baltimore, the air quality index was 87 or Moderate. Since today’s air quality is expected to be much the same, sensitive groups should consider limiting prolonged or heavy outdoor exertion.”

      1. Hi Jeremy,
        I think that NEA’s health advisories are okay… though they are in a table and not in running prose. 🙂

  3. Hi Jeremy,
    Nice article. Some clarifications. In the London Smog of 1952, it was the smoke levels (all particulate matter), rather than PM2.5, that reached as high as 4,500 microgram/m3.

    Also, the NEA _does_ report the individual pollutant concentrations, at . The NEA also provides such data to external weather-reporting websites. The NEA has also reported that PM10 is the dominant pollutant in PSI.

    1. Thanks Vic. I must confess reading through the literature, I was not too clear about what the 4,500 microgram/m3 actually referred to as there were conflicting reports. Thankfully, directionality and magnitude are unaffected and the take home remains unchanged- We are far from levels where people or cows (as in London) drop dead. Noted on the NEA information as another reader had alerted me. My bad. Traveling these last 2 weeks and didn’t manage to follow every news piece.
      I’m not that clear on what it means “PM 10 is the dominant pollutant in PSI” as PM 10 is a measure of particles and not any specific chemical substance. This is one of the peculiarities of science which you real science guys would be much more familiar with than me that particulate matter standards have been set without specific consideration for what chemical substances make up the particulate matter. My reading is that the risk even at the same PM 10 or PM 2.5 levels can be different depending on what makes up the particles. Please do educate 🙂

      1. Hi Jeremy,
        Right now, PM10 and PM2.5 are catch-alls. Some particles such as pollen can be very allergenic, and others such as heavy metals can be very harmful. Thankfully, our air-quality standards have kept us safe from heavy metals and even from vehicle emissions. As such, most of our PM arises from burning plant matter, and the general health advisories apply.

  4. Thank you for the article, I read from that the Chinese combat cancer causing particles by drinking green tea (tea bags not bottled type) everyday, as green tea has anti-cancer effect. Thought this might be a good tip for people reading this!

  5. Mike Chee · · Reply

    Hi Jeremy,

    Let me rephrase:

    There has been a previous study of the composition of peat forest fires in Sumatra

    Characterization and Source Apportionment of Particulate Matter 2.5 μm in Sumatra, Indonesia, during a Recent Peat Fire Episode Environ. Sci. Technol. 2007, 41, 3488-3494

    This was published from Singapore by NUS researchers. Primarily carbon particles in the fine gunk – at the heart of the sample zone PM 2.5 was 1600 μg/m3….

    One thing readers might benefit from hearing is that life expectancy in London at the time was 40 – 50 years? We don’t have life tables in Riau province but I think they don’t get to 80 years like we do in Singapore….

    1. Thanks Mike for the response. Yes, I think life expectancy in the UK in the 1950s was about 65 years. Looking at the abstract of the paper you referenced, it’s clear that ‘haze’ from peat fires is hazardous and the closer to source, the more hazardous it is. PM 2.5. of 1,600 micrograms/m3 is horrific!
      Guess even as we adapt here in Singapore, we should spare a thought for the Indonesians closer to source who have nothing to do with the fires but have to live with much higher PM 2.5 levels. Don’t have Riau specific life expectancy data but from Wikipedia:
      Life expectancy at birth for Indonesia- Total population: 71.05 years Male: 68.53 years Female: 73.69 years (2010 est.).
      Singapore ranks 4th in the world (82 years overall; 80 for men and 85 for women).

      1. Mike Chee · ·

        Hi Jeremy,

        Thanks for the response and the nice data mining. While you and I know that lots of other factors contributing to life expectancy, this is a nice statistic to remind people about.

        At the higher end of life expectancy – like getting to the finals in a sporting contest, more things must be ‘right’. This is true sans conditions. If nothing, appealing to good old self interest should push people to action.

        Highly educated and wealthy people need to weigh in on particulate pollution and ask for change in their business and social networks if we are going to stand any chance of not collectively committing hare-kiri like other countries in Asia.

        Euston Quah said it right last week. We need to put a dollar value on the haze for the current leaders to do anything.

        We certainly cannot expect Sumatran and Kalimantan farmers to use initiative. This link adds new dimensions to the common phrase ‘lets smoke the audience’….

        Dumb and dumber.

  6. Choy Kah Wai · · Reply

    For whatever it is worth. During those few days when the PIS was in the 300 – 400 range. My dog was sneezing and try to get indoor.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: