APi & PSI air quality indexes don't tell the whole truth due to very fine particulate matter missing from pollution readings

Figure 1 Jun 29, 2013: People wear face masks on Orchard Road in Singapore (fromAFP)

In our last post we have been discussing about the last Sumatra haze spell in South Peninsular Malaysia and in Simgapore, based on the official readings released by both Malaysia DOE and Singapore NEA. 

We explained that air pollution indexes and sub-indexes in Malaysia or Singapore were established following rules very similar to the US Air Quality Index including 5 only among the US AQI pollutants (PM10, O3, CO, SO2 & NO2), but without the 6^th  PM2.5 pollutant used also in US AQI and EU CAQI indexes.

In Malaysia, it seems that PM2.5 is some time measured but not published, while in Singapore PM2.5 is measured and published along with the PSI index, but not incorporated as a sub-index before air quality index computing. 

Air Quality Index compared computations from EU, US, Malaysia and Singapore

In my post date 14 July, 2013, I have explained how is computed the US AQI Index. You may find the AQI and its sub-indexes definition in « Technical Assistance document for the Reporting of Daily Air Quality the AirQuality Index (AQI)”.

In UE the Common Air Quality Index (CAQI) is an European Development with two indexes « background » and « road-side ». We have selected as a comparison’s reference the background CAQI and its sub-indexes see CAQI from EU Common Air Quality Index European Development.

Because CAQI is aiming mostly light polluted areas, the EU pollutant scale is much more limited than in US AQI and as a consequence the grid values are much higher (for instance CAQI=100  with PM10=100 when US AQI= 100 with PM10=154). So we have adjusted and decreased the CAQI grid in order to meet US AQI at least for PM10.

For Malaysia API: see “API a guide for air pollutant index in Malaysia” and for Singapore PSI  index: see “Computation of the PollutantStandards Index (PSI) “PDF document.

The following five Figures 2-6 are giving, for any value of the measured pollutant concentration located inside two "turning points", the successive linear relationships to calculate the value of the index with a "grid" common understanding between all indexes and sub-indexes. 

Each broken line describes pollutant action on human health, each turning point corresponding to the crossing of a threshold effect on health. The greater the index the more unheathy it is with grid<100 corresponding to good or moderate air quality. 

Figure 2 shows a good relationship between all the various sub-indexes for PM10  sub-indexes; and a slight divergence between EU and US for PM2.5 sub-indexes.

Figure 2: Comparison of air quality sub-indexes for particulate matter pollutants (PM10 & PM2.5), over EU, US and API/PSI South Asia sub-indexes

Figure 3 is showing a good relationship between US, Malaysia and Singapore for CO sub-indexes. Inversely EU index is quite different with a much higher value for the pollutant concentration.

Figure 3 : Comparison of air quality sub-indexes for carbon monoxide (CO) pollutant, over EU, US and API/PSI South Asia sub-indexes

Figures 4, 5 & 6 are showing more divergences between all countries for O3, SO2 and NO2 pollutant sub-indexes.

Figure 4 : Comparison of air quality sub-indexes for Ozone pollutant (O3), over EU, US and API/PSI South Asia sub-indexes.

Figure 5 Comparison of air quality sub-indexes for Sulfur Dioxide pollutant (NO2), over EU, US and API/PSI South Asia sub-indexes

Figure 6 :  Comparison of air quality sub-indexes for Nitrogen Dioxide pollutant (NO2), over EU, US and API/PSI South Asia sub-indexes

The importance of PM2.5 pollutant to assess the air quality 

(See PM2.5 fact from  US EPA)

Particles less than 10 µm in diameter (PM10) pose a health concern because they can be inhaled, accumulated in the lungs and damage the respiratory system.

Particles less than 2.5 µm in diameter (PM2.5) -referred to as "fine" particulate matter- are believed to pose the greatest health risks. Why?  Because of their small size (around 1/30th the width of a human hair), fine particles can be trapped deeply into the lung cells producing various health hazards: asthmas, infection, cancer etc..

Sources of fine particles include ashes and residues from all types of combustion activities (wood burning, power plants, motor vehicles etc.) and certain industrial processes.

Particles with diameters between 2.5 and 10 µm are referred to as "coarse" particulate matter. Sources of coarse particles include crushing or grinding operations, and dust from paved or unpaved roads.

Other particles may be formed in the air from the chemical change of gases. They are indirectly formed when gases from burning fuels react with sunlight and water vapor. These can result from fuel combustion in motor vehicles, at power plants, and in other industrial processes.

A study for Air Pollution Effects published by "The Lancet" in July 2013 assessed air pollution for particulate matter with diameter of less than 10 μm (PM10) & less than 2·5 μm (PM2·5). It shows that exposure of population to particulate matter air pollution even at concentrations below the existing EU air quality limit values for PM10 (40 μg/m³) and PM2·5 (25 μg/m³) might increase the risk for lung cancer.These EU air quality limit values are associated with API/PSI = 50 in Figure 2. 

Roughly one out of three persons is at a higher risk of experiencng PM2.5 related health effects, mostly children with their bodies still developing and the elderly due to their lower resistance. 

People of all age active outdoors during physical activities are at increased risk because with increased air intake PM2.5 penetrate deeper into the more vulnerable part of the lungs.  

Due to their deadly effect on human health it is urgent - as for EU and US- to include PM2.5 in the sub-index standards in Southeast Asia API/PSI readings

As we know that most API/PSI readings are mostly derived from the high value of PM10 sub-index, they have reciprocally the same "turning points" and  it is easy to derive the value of PM10 from the API/PSI readings by using the PSI/API sub-index reciprocal linear relationship.

The following Figure 7 gives both values of PM10 and PM2.5 concentrations respectively computed or read for South Singapore PSI (See: PSIHistorical readings).

We can see that PM10 is mostly composed of 75-80% of PM2.5. This means that the coarse particulate are only 20-25% and - as we can see on Figure 2 above- that the value of the PM2.5 sub-index is much higher than the value of the PM10 sub-index.

Figure 7 : South Singapore PM10 & PM2.5 concentrations computed from PSI or read

The following Figure 8 shows that the “true” PSI - including the PM2.5 sub-index -is around 50-80% higher than the official reading.   

Figure 8 : South Singapore official PSI without PM2.5 and PSI including PM2.5 sub-index  

On 20 -21 June 2013, the “official” PSI as recorded by Singapore NEA was only 246 but the "true value" of the Pollutant Standards Index incorporating the more dangerous PM2.5 pollution was around 344. 

However, it appears that in its communication with the media, Singapore NEA has already made ​​the change, according to an article in the Straits Times referring that Singapore Pollutant Standards Index "soared to 321 on 19 June the worst reading in its history, the previous being 226 in 1997".

If we keep in mind that at the same time in Malaysia the highest readings - 746 in Muar, 487 in Ipoh and 443 in Melaka- were issued without any correction for PM2.5, the general picture on the Southeast Asia pollution was reallly difficult to assess!    

So in order to inform as fairly as possible the population on Air Pollution  health hazards, there is a urgent need to clarify this issue by incorporation-as in EU and US-  the PM2.5 fine particulate matter sub-indexes into API / PSI air quality indexes used in Southeast Asia. 

Sumatra haze becoming a yearly recurring event with growing pollution level?

Figure 1 : On Saturday, June 22, 2013, motorists make their way through a town covered with a thick haze in Muar. FILE PHOTO: AP

In South Peninsular Malaysia and Singapore island, the last Sumatra’s smoke haze spell – 23 days from 10 June to 2 July 2013- was definitely more serious than the last year occurrence, while haze’s pollution could still return anytime during the coming months.  

A huge amount of information was published by various media during this period. But sometimes the true meaning of air pollution information issued and their outcome in terms of public health were difficult to understand and - to say it bluntly- a little hazy!

We would like to return on this period and look at all data issued on air quality by Department of Environment (DOE) in Malaysia, National Environment Agency (EPA) in Singapore and ASEAN Specialised Meteorological Centre (ASMC).

How are established both Air Pollution Index (API) in Malaysia and Pollutant Standards Index (PSI) in Singapore?

All figures concerning air pollution index and sub-indices in Malaysia or Singapore are established following rules very similar to the US Air Quality Index  (see AQI Technical Assistance document for the Reporting of the Air Quality Index AQI).

For the US AQI, first specific sub-indexes are separately computed concerning the 6 following pollutants:

(1)- Particulate matter smaller than 10µm (PM10 ),

(1bis)- Particulate matter smaller than 2.5µm (PM2.5),

(2)- Ozone (O3),

(3)- Carbon monoxide (CO),

(4)- Sulfur dioxide (SO2) ,

(5)- Nitrogen dioxide (NO2) .

For each pollutant a common “grid” of communication relationship is introduced, with a growing scale grid and associated pollution level where: 0-50 is a “good air quality”, 51-100  is a pollution “moderate”, 101-150 a pollution  “unhealthy for specific groups”, 151-200  is a pollution ”Unhealthy”, 201-300  is “very unhealthy”, and 301-500 is “Hazardous”.

Then each sub-index is computed and the highest sub-index value is the US AQI value for the location and the time concerned.

The only difference concerning Malaysia’s API and Singapore’s PSI in relation with US AQI is the PM2.5 pollutant. 

In fact (1)) and (1bis) pollutants are closely interrelated because particulate matter smaller than 2.5µm are also smaller than 10µm. However PM2.5 are also much more unhealthy and as such should have a higher “grid” value!

In the case of Malaysia, PM2.5 seems to be measured but is not published.

In the case of Singapore , PM2.5 is measured and published with the PSI index.

API & PSI values are computed based on the average concentrations of air pollutants PM10, O3, CO, SO2 and NO2 expressed in µg/m3 or in ppm. The average time period is pollutant specific: daily for PM, 8h for CO, hourly O3, SO2 and  NO2.  

During the haze spell in Malaysia the PM10 sub-index value due to the high PM10 concentration is the highest compared with the other pollutants and this determines the API readings.

Nevertheless by the late afternoon or early evening, usually ozone concentration is high and dominates the API readings in some areas. Under the sunlight influence, nitrogen dioxide and volatile organic compounds emitted from motor vehicle exhaust and industry react to form ozone in the earth's surface.    

            

Peninsular Malaysia was more severely affected than Singapore

In both countries the monitored pollutants and sub-indexes are the same and so the comparison is meaningful. 

The following Figures 2 & 3 give the API readings in Malaysia:

Figure 2 : API index from 22 June to 2 July 2013 on 11 locations in Johor, Malaka, N. Sembilan and South Pahang  (see DOE Malaysia)

Johor (Kota Tinggi, Larkin Lama, Muar, Pasir Gudang ), Melaka (Bandaraya, Bukit Rambai) and Negeri Sembilan (Port Dickson)  were especially polluted with the highest API reading  ever seen in Johor Muar (756 API on 23 June), as explained my Malaysian Environment Minister in Asia Pacific Channel News.

Figure 3 : API index reading on 13 locations in Perak, Selangor, Kuala Lumpur  and Terengganu (see DOE Malaysia)

Selangor (Banting , Pelabuhan Kelang) and  Terengganu (Kemaman) were especially polluted with the highest reading in Selangor Pelabuhan Kelang (487  API on 27 June).

The Figure 4 gives the PSI reading in Singapore:

Figure 4 : PSI readings on 5 locations in Singapore (see NEA Singapore) 

Generally speaking, Singapore was less polluted than Malaysia: around 5500 cumulated readings unhealthy for special group on 23 days (at 7am, 12am, 4pm) compared to 7500 cumulated reading in Malaysia during the same period (7am, 11am, 5pm).  

South Singapore was especially polluted, with the reading (PSI=246 on 22 June). The highest points in Peninsular Malaysia being 400-700.

So if we take off the "good readings" and look also at the highest readings, Malaysia was at least "twice more polluted" than Singapore during this haze spell.

Moreover the 2013 haze spell (in June and up to 15 July) was much more severe than 2012 June and July occurrence,  but with a hotspot count very similar

Compared to the 2012 June & July haze in Malaysia: the spell is much stronger and only located on South peninsular Malaysia .

In 2012 there was a 4000 cumulated reading (at 7am, 11am, 5am) and the maximum readings were in Sarawak (API=247 on 27 June 2012), Selangor  (API=147 on 15 June 2012), with Penang having a 5 day haze spell (API=105 on 20 June 2012).

In 2012, the cumulative hotspot count for Sumatra while being among the 2 highest of the last 8 years is in line with last year count progression (see following Figures 5, 6 & 7).  

So the haze in becoming more or less a yearly recurring hazard with pollution growing each year!

Figure 5 : Cumulative hotspot count for Sumatra detected in the ASEAN region for 2006 to 30th June 2013; the counts are based on the hotspots detected by the NOAA-18 Satellite

In Sumatra  the open burning areas seemed to be more located over the Riau Province  in Central Sumatra,  while last year the hot spots were coming up to Kualuh Hilir in North Sumatra.  

This combined by the direction of winds have kept the Northern states of Perak and Pulau Penang  less affected.

The international uproar have been so severe – especially from Singapore- that the Indonesian Government had announced its decision to ratify the“2002 Asean Treaty on Transboundary Haze Pollution signed by all 9 other ASEAN countries.  

Figure 6 : Regional haze map on 21 June 2013 

Figure 7 : Regional haze map on 23 June 2013