Measuring air quality: Simi PSI?

by Yoong Ren Yan

MEASURING air quality is a tricky business. And in these hazy days, everyone thinks they know better. The PSI “feels worse” than 200, for example. You might demand real-time data, but the agencies who have the figures say you don’t need them. They maintain that a longer average is what really matters.

Things get even more complicated across borders. The astronomical index figures from Kalimantan have made the news in both Singapore and Malaysia in recent days. But the media in both countries have assumed that the Indonesian index is the same as theirs. Is there a difference between country indices?

The structure of each index is similar. First, the relevant agency measures five (or six) concentrations of pollutants in the air. These measures of concentration are then converted into an index figure. The maximum of these five (or six) figures is then reported.

As far as we could tell, the Malaysian Indeks Pencemaran Udara (IPU) and Indonesian Indeks Standar Pencemar Udara (ISPU) are identical. The quantitatively inclined can check our cursory calculations here and here.

Singapore’s Pollutant Standards Index (technical details here), as of April last year, includes an additional pollutant, PM2.5 – particulate matter 2.5 microns or smaller. As the National Environment Agency (NEA) puts it: “PM2.5 is the main pollutant of concern during periods of smoke haze.”

PM2.5 was included because of evidence that, in addition to the other pollutants, PM2.5 poses a threat to health. PSI also averages ozone measurements over eight hours instead of one. But aside from these two differences, our PSI appears the same as the indices our neighbours use.

How can we compare them then, even roughly? What does it mean when the ISPU exceeds 2,000 in Kalimantan? We could follow NEA’s lead. PM2.5 is the dominant pollutant that drives the PSI in hazy episodes. Then indices that exclude PM2.5, like those used in Malaysia and Indonesia, would systematically be lower than our PSI. So Kalimantan is even worse off than we thought.

And all this just describes how the 24-hour PSI is calculated. NEA also releases a 3-hour PSI, based only on PM2.5 readings averaged over three hours, and 1-hour PM2.5 concentrations – but not 1-hour PSI. 1-hour figures, according to NEA, are “best used as a guide to adjust your immediate activities”.

Yet enterprising Singaporeans have used concentrations to calculate the 1-hour PSI for themselves. And that might correspond better with what we “feel”. So why hasn’t NEA simply released the figures? It argues that 1-hour figures are “volatile” – but of course, both concentrations and PSI would fluctuate.

Looking at how PM2.5 concentrations are converted, there could be another explanation. Concentrations (in micrograms per cubic metre) are about 50 points less than corresponding PSI figures. For instance, a concentration of 56 is unhealthy (PSI 100), 151 µg/m3 is very unhealthy (PSI 200), and 251 µg/m3 is sufficient to get to a hazardous PSI of 300. Could releasing concentrations, which are consistently lower than PSI figures, be a way to present a more palatable figure?

A longer average might indeed be what matters. But when it comes to the PSI, more – not less – information is what we need to clear the air.