Please note that this article was orginally written in June 2013.
In April 2014, the Singapore NEA has updated the PSI calculation to also include PM_2.5. http://www.nea.gov.sg/anti-pollution-radiation-protection/air-pollution-control/psi

With the recent Southeast Asia Smog from Indonesia affecting Singapore and now Malaysia, we got many questions about why there is a difference between the data which can be read from Singapore NEA website (nea.gov.sg) and the World Air Quality Index project Singapore webpage.

例如，今天可以在 NEA 网站上看到以下内容：

由于某些历史原因，新加坡使用PSI （污染物标准指数）来评估空气质量。在上图中，数字 1 对应于用于 PSI 评估的 PM10 值。值67(/59)可读取为 67 µg/m3，对应于 PSI 59。PSI 被评估为每种污染物的单个 PSI 的最大值：PM ₁₀ 、SO ₂ 、NO ₂ 、O ₃ （臭氧）和CO ₂ 。

PSI(Singapore-North) = max( PSIPM10-based, ... PSIO3-based )= max (59, ..., 77) = 77

尽管如此，有趣的是，在同一个表中，最后一列还提供了 PM2.5 数据（参见图 3）。该数据仅以 μg/m3 为单位提供（参见 2），并且没有转换为 PSI 之类的数据（即从 PM2.5 质量转换为污染或质量指数）。然而，这种转换是存在的，并且是由美国环境保护局定义的。进行转换的最简单方法是使用在线计算器，可在airnow.gov上找到：

If you select the PM2.5 (1), then enter the mass concentration of 48 (2), can click on Calculate, you will obtain the AQI of 132 (3). So, based on the PM2.5 AQI conversion, the PSI that is used for Singapore could be extended (let's call it PSI++) to also take into account the PM2.5 information. In which case, the PSI++ would be the maximum of the regular PSI (based on PM10 only) and the PM2.5 AQI:

PSI++ = max( PSI, AQIPM.25 ) = max( 77, 132 ) = 132
This PSI++, that is commonly referred as AQI (or Air Quality Index), is what is being used on the the World Air Quality Index project, for all the cities (provided PM2.5 is available for the city). And this explains why the values are different between the NEA website and the World Air Quality Index project.

Note: With the PSI update from April 2014, the only remaining difference is that the World Air Quality Index project is using a 1-hour PM_2.5 reading (following the instantcast) while Singpore PSI is based on the 3 hours average.

Moreover, when doing the convertion, make sure you use the 1-hour reading for the PM2.5 concentration rather than the 24-hours averaged value, as shown on the below image:


http://www.haze.gov.sg/haze-updates/pollutant-concentrations/type/PM25-1Hr

If you want to know more about PM10 vs PM2.5, and especially why PM10 is still used, please check the faq entry about why is PM2.5 often higher than PM10? Is PM10 still a relevant measure?


--