Hong Kong air pollution: better than it looks?

A few weeks back I asked if our air pollution is getting better or worse, then answered myself: worse over the last 18 years, but possibly getting better slowly in the last one or two years.

Then I went looking for some more numbers to back this up, but found that ... well, see if you find them as surprising as I did.

Visibility

That's the measure I used in the last post - 'hours of reduced visibility'. Here are the totals for recent years:

  Hours of reduced
visibility over the year 
Change
since 1990 
 Change
since 2000
1990  274  100  
 1995  459  168  
 2000  623  227  100
 2005  1502  548  241
 2007  1298  474  208

That matches my own perception, that the air has got considerably smoggier since the 1990s. Then a common assumption is that reduced visibility means lower air quality, so how can we check that?

Pollution concentrations

If you walk around Central, what's in the air that you breathe, and how has it changed?

The EPD have one of their roadside measuring stations in Central [1], and publish its records of pollution levels every hour [2]. I recognised 'Sulphur Dioxide' (SO2) as something unpleasant from school chemistry lessons, so how has that changed over the years. [3] (They only have records back to year 2000 for this station)

 

  Avg. hourly concentration of
SO2 over year, ug/m3 
Change
since 1990 
 Change
since 2000
1990   -  
 1995    -  
 2000 28.1   -  100
 2005 24.2   -  86
 2007 22.0   -  78

That wasn't what I expected to see: if you look back at the first table, there's over twice as much smoggy weather in 2007 compared to 2000. But if you look at what you're actually breathing, levels of one of the main pollutants has dropped over that same time!

Maybe I chose the best pollutant by luck? The station also records values for Carbon Monoxide (CO), Nitrogen Dioxide (NO2), Nitric Oxide (NO), and Respirable Suspended Particulates (RSP). How have they changed at this station over the same time?

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100  100  100 
 2005 107  101  98  104  86 
 2007 84  107  96  103  78 

So it's not just the SO2. Across several pollutants the levels we actually breathe at street level have stayed fairly constant, or even fallen slightly.

Up on the roof

Maybe this only happens at street level? What if we look at the readings from the 'Central & Western' station [4]? This is one of the general measuring stations (the EPD calls them 'roadside' or 'general'). It is on a rooftop 18 metres off the ground, and is away from a main road. It doesn't measure Carbon Monoxide, but it does measure Ozone (O3). A bonus is that we can see measurements back to 1990 for this station.

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100  100   100 
 2005 111  105  106  107  122 
 2007 100  92  114  106  132 

So here the changes are in the bad direction, with concentrations increasing. But still nothing like the doubling we saw with the smog levels. What if we look at the Central & Western figures back to 1990?

   CO, change
since 1990
NO2, change
since 1990 
NO, change
since 1990 
 O3, change
since 1990
RSP, change
since 1990 
SO2, change
since 1990 
1990  100  100  100  100 
1995  108  113  125  114  108 
 2000  106  106  180  96   91 
 2005  117  111  190  103  111 
 2007  106  98  205  102  120 

Remember that over that time the weather got around 4 times smoggier? But if you live on a side street in Western Hong Kong, in say an old 6-storey building, then apart from Ozone, pollution levels in the air you breathe today aren't that much different from the air in 1990.

What am I missing?

Again, that's not what I expected, so am I missing something obvious?

And why do we feel the air quality has got so much worse over the last 10-20 years, when levels of most pollutants have only changed by a few percent?

Is it just that the pollutant that has seen the biggest jump in levels (Ozone) also happens to be one we can see, and so cannot be ignored?

A few more questions spring to mind - let me know if you have any answers, otherwise I'll look at them over the next few weeks:

  • How much of the reduced visibility / increased smog problem is directly linked to rising Ozone levels?
  • Then is reducing the ozone level the main activity to concentrate on if we want to lower the Batgung Smog Index, and bring the blue skies back?
  • Ozone isn't produced directly, instead it is a result of UV radiation acting on NO and CO or Volatile Organic Compounds (VOCs) [5]. Why is the Ozone level rising when the NO level remains steady?

MrB

References:

  1. Central roadside monitoring station
  2. Records of monitoring stations
  3. Calculation of avg hourly concentration for a year:
    • Download figures for chosen year (see [2]).
    • Add up total of all hourly figures for the year.
    • Count how many hours are marked 'N.A.' (not available).
    • Work out how many hours are in the year, remembering to take leap years into account.
    • Average = Total / (hours in year - number of N.A. hours)
  4. Central and Western general monitoring station
  5. Tropospheric Ozone

Comments

Hong Kong Air Pollution

Excellent subject with analysis and discussion! I offer the following amateur opinions:

Modern equipments can sample and analyze chemical compounds in the air hourly or more frequent so their results can't lie; that is, assuming they are calibrated properly and sample each hour year-round. But changes in the surrounding buildings, traffic and industrial activities can alter the trends. The period of record for these parameters are short unfortunately to draw strong conclusions.

The changes, in percentage, of the concentrations of SO2, CO and O3 range from 80% to 130% between the year 2000 and 2007. One may think that the air quality, using these as indicators, has not changed drastically.

Using the year 1990 as the basis of comparison, the number of hours of reduced visibility in subsequently years have been as much as 548%. In other words, the number of hours of reduced visibility in 2005 was five-and-a-half times that in 1990.

That the reduced visibility started to shoot up in the early 1990s should be investigated as to the causes.

This brings me to the question whether the criterion for defining reduced visibility, that is, "visibility below 8 kilometres when there is no fog, mist, or precipitation" has been consistently applied since the 1960s, in terms of locations of the line of sight and equipments used. I wonder if human "eyeballs" were used in the early years.

Finally, to what extent can SO2, CO and O3 affect visibility? Dust blown from construction sites may play a role in this - but this could be a non-issue.

re: HK Air Pollution

Oldtimer, thanks for your feedback. Some comments:

But changes in the surrounding buildings, traffic and industrial activities can alter the trends. The period of record for these parameters are short unfortunately to draw strong conclusions.

Fair point. The conclusion that air pollution levels haven't changed that much are true if you'd been standing next to the monitoring station the whole time. But that effect may be due to some localised changes as you describe. I'll have to take a look at readings from some of the other stations around Hong Kong to see if the same patterns appear.

This brings me to the question whether the criterion for defining reduced visibility, that is, "visibility below 8 kilometres when there is no fog, mist, or precipitation" has been consistently applied since the 1960s, in terms of locations of the line of sight and equipments used. I wonder if human "eyeballs" were used in the early years.

That would be interesting to know, but even if it explains some of the change in the numbers, there's no doubt in my mind that visibility has got worse over the last 10-15 years.

Finally, to what extent can SO2, CO and O3 affect visibility? Dust blown from construction sites may play a role in this - but this could be a non-issue.

Wikipedia has a good article about smog. It explains that the older versions (eg for which London was infamous) were caused by a mixture of smoke and SO2. More recently we face photochemical smog, which it says "is the chemical reaction of sunlight, nitrogen oxides (NOx) and volatile organic compounds (VOCs) in the atmosphere, which leaves airborne particles (called particulate matter) and ground-level ozone (O3)."

CO is also given as one of the ingredients needed to make O3, so although it is colourless itself, the O3 it produces is not.

MrB

HK Air Pollution

Thank you for the response, MrB. No disagreeable on all your explanations.
And may I say from here - "Happy Canada Day". We are also getting more smog days in Toronto but think ours is still more tolerable.

Air quality in the good old days

We've been forwarded this photo from the 1968 HK Standard, and 'Comments from a long time Hong Kong resident':

The CLP power station in Hung Hom used high sulphur fuel oil and had short chimneys because of its proximity to the airport. When the wind blew from the east the smoke rolled across the Kowloon peninsula at very nearly ground level. The sulphur in the air could be so bad that your saliva would become acidic and very painful in contact with tooth fillings. 
 
Similarly HK Electric had a power station complex in the middle of North Point and the government had incinerators at Kennedy Town and Tsing Yi all belching out smoke within residential districts.

Changing pollution levels in Tsuen Wan

Following Old Timer's comment about the effect of local conditions, here are similar calculations for the figures from the measuring station in Tsuen Wan:

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100  100  100   100 
 2005 72  101  95  111  116  134 
 2007 71  105  85  129  119  130 

And

   CO, change
since 1990
NO2, change
since 1990 
NO, change
since 1990 
 O3, change
since 1990
RSP, change
since 1990 
SO2, change
since 1990 
1990  100  100  100  100 
1995  131  177  106  116 
 2000  129  168  118  88   93 
 2005  131  160  130  102  125 
 2007  136  143  152  105  122 

How do they compare to the Central & Western readings?

Similarities:

  • The general trends are mostly similar, eg SO2 rises in mid-90s, falls in 2000, rises again to present day.
  • Ozone (O3) is the pollutant that has changed the most

Differences:

  • Although trends are similar, the magnitude of the trends are quite different. eg O3 increases 50% in Tsuen Wan, but over 100% in Central. This could be a sign of differences due to local effects as OldTimer suggested. We'll need to look at other stations to get a better answer.
  • The increase in NO is much higher in Tsuen Wan, though it has fallen since the peak in 1995. Maybe the 1990 reading for NO in Tusen Wan was unusually low, making other years look high? Or was there really a change that meant NO levels have risen there?
  • For Central & Western we said that 'you live on a side street in Western Hong Kong, in say an old 6-storey building, then apart from Ozone, pollution levels in the air you breathe today aren't that much different from the air in 1990.' That doesn't seem true for Tsuen Wan, as apart from the RSP the pollution concentrations are 20-50% higher than in 1990.

I'll add in figures for more stations as time permits.

MrB

Changing pollution levels in Kwun Tong

Here are similar calculations for the figures from the measuring station in Kwun Tong:

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100  100   100 
 2005 84  72  128  108  101 
 2007 91  75  129  102  104 

And

   CO, change
since 1990
NO2, change
since 1990 
NO, change
since 1990 
 O3, change
since 1990
RSP, change
since 1990 
SO2, change
since 1990 
1990  100  100  100 
1995  95  102  25 
 2000  115  103  46 
 2005  96  74  46 
 2007  105  78  48 

These are more similar to the Central & Western readings, with pollution in 2007 roughly equal to or better than 1990. We might guess that the Kwun Tong air quality would improve after the airport closed in 1998, but that doesn't seem to have made a big impact on any of the figures between 1995 and 2000.

Again, the figure that stands out from the 2000-7 readings is the Ozone level, increasing where others are falling or staying steady.

MrB

Kwun Tong, then and now

About twenty years ago if one stride alone Wai Yip Street or Hung To Road alike, you will be welcomed by a corrosive & acidic stench. The alleys were usually dark and filthy with refuse dripping down from above. At the time, there were still many factories there. Dyeing mills, leather fabrications, metal workshops with electrolysis facilities....... you think of all those awful and smelly stuffs, you have them all. Back then, water pollution around the Kwun Tong waterfront was also very bad. The water near the sewage usually changed colour by the hour.

Situation improved after the Government banished such industries from the city district back in the late 1980's.

T

chemical reaction

Hi,

I was just wondering how HK was, and how the smog was going since I left -- evidentally not too good.
One reason that you don't get an increase across the board is that I believe you get a chemical reaction between NO2 and O3. So if your NO2 goes up, it will keep your O3 down -- I believe this is why there is often less ozone in, say, Kennedy Town than Sai Kung, and why you often get an ozone free period in the morning (NO2 pollution from the cars interacts with the O3 for a short while) Have look here:

http://www.environmental-protection.org.uk/air-quality-and-climate/air-quality/ozone-pollution/

re: chemical reaction

Conrad,

Thanks for the link. It gives a better explanation than the HK EPD's site, though as it says: "The atmospheric chemistry involved in ozone formation is complex". They're not kidding! I'll have to read it a few more times...

Regards, MrB

Changing pollution levels in Tap Mun

Here are the figures from the measuring station in Tap Mun:

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100  100  100   100 
 2005 154  130  152  98  130  203 
 2007 172  136  152  99  138  239 

So if you live in Tap Mun, you are fully justified in saying that the pollution is getting worse!

If you've been to Tap Mun, you'll know it's a quiet island, far from any built-up areas. In that way it can show changes in pollution levels that are not related to changes in local conditions - more traffic, new buildings, etc.

Also remember that these numbers just show changes - whether it's getting better or worse. The absolute readings aren't shown here, but we'll need to come back to them in a future article. They'd show that for most pollutants, though levels have increased in Tap Mun, they are still lower than levels in the built-up areas. The clear exception is Ozone, where levels in Tap Mun are much higher than in the city - as Conrad predicted above.

Hong Kong Air Pollution

Mr. B got me interested on this subject. Nothing to contradict or dispute what Mr. B has said, I took a cursory look at the 2007 July and December hourly data on air pollution index (API). This is a very limited analysis in terms of scope, techniques and data sample, as such, my observations are offered as a matter of general interest only.

First, starting with the July 2007 figures, Monk Kok, Central and Causeway Bay observed consistently higher APIs than others. Monk Kok's air quality was poor and steady through the month. The air quality fluctuated somewhat at the other two stations.

I looked for improvements in the air quality during the weekends when offices and factories are closed and there seem to be such a pattern but not a strong one.

All 14 stations observed lower API values during July 1~4 (Sunday to Wednesday), which suggests that weather and wind conditions may have been a factor for the entire region. Was it a holiday at that time?

Next having high APIs are Kwai Chung and Tsuen Wan, which are located close to each other. They seem to have some weekly cycles having better air quality during the weekends.

Kwun Tong, Shum Shui Po, and Yuen Long are next with high APIs.

Tung Chung, Tap Mun, Central/Western registered short-term and sharp rise in APIs, for example, July 9 through 16, typically in late afternoon. These stations, which at other times register low API's are likely affected by local traffic conditions.

Tai Po, Sha Tin and Eastern would be more preferable to live in judging by their API data, but Eastern is a bit of a surprise.

When the December 2007 data are compared with the July 2007 data, the air pollution was much worse in the winter month. For example at Causeway Bay, the API varied between 40 and 50 during July, and between 60 and 130 in December. What could have caused such a big increase in December, more traffic and shopping activities? More factories working to meet Christmas orders, or colder but polluted northerly air from the border city at the border?

The same (previous paragraph) can be said about what happened at all the other stations, that the December APIs are consistently higher (or worse) in December.

For Tap Mun, which is way out in the country side with little industry and commerce, one can see the late afternoon peaks on many days during rush hour traffic. This type of peaks in December, if any, are not noticeable and masked by the general increases in APIs during December. Something outside Tap Mun was also affecting its air quality.

I can e-mail you my Excel files if you are interested Mr. B.

re: Hong Kong Air Pollution

OldTimer, thanks for taking the time to look at these numbers. Some feedback:

First, starting with the July 2007 figures, Monk Kok, Central and Causeway Bay observed consistently higher APIs than others.

Had you spotted that these are the three 'roadside' stations? So they tend to have the worst figures, reflecting what you breathe when walking next to a busy road.

I looked for improvements in the air quality during the weekends when offices and factories are closed and there seem to be such a pattern but not a strong one.

I hadn't thought of looking for variations across a week, so that's a good idea for another thing to look at in future.

When the December 2007 data are compared with the July 2007 data, the air pollution was much worse in the winter month. For example at Causeway Bay, the API varied between 40 and 50 during July, and between 60 and 130 in December. What could have caused such a big increase in December, more traffic and shopping activities? More factories working to meet Christmas orders, or colder but polluted northerly air from the border city at the border?

You can see a similar trend in the visibility figures - summer has best visibility, winter has worst. I think the wind direction is the main factor. Summer's winds from the south blow in fresh air from the sea, while in winter the wind comes from the north bringing polluted air from the mainland. Summer also tends to be the cloudy, rainy time of year. I haven't seen that mentioned as affecting pollution levels, but I think both cloud and rain will help lower pollution levels.

This report from Civic Exchange has a good discussion about the effects of wind direction on local pollution, but please also read my concerns about how the report's results were presented.

For Tap Mun, which is way out in the country side with little industry and commerce, one can see the late afternoon peaks on many days during rush hour traffic.

It would be worth digging into the detailed readings to see which pollutant caused those peaks. My guess would be Ozone. The API is a coarse measure - a bit like the visibility readings. It shows us the worst figure out of the 8 sub-indices that are calculated each hour. That makes it a good and easy-to-understand indicator to the public of whether pollution levels are high enough to cause problems or not. But that also makes it harder to interpret the causes of problems. For that we need to look at the levels of the individual pollutants.

Also, it's probably something other than rush hour traffic causing the peaks. Tap Mun has a population of around 100, an assortment of boats, but no motor vehicles. Not much of a rush hour in those parts of the world ;-)

Something outside Tap Mun was also affecting its air quality.

Javascript is required to view this map.

This satellite view shows how rural Tap mun is (see red marker on lower right), but also how near the South China industrial miracle is, just to the north. I'll come back to this in a future article, seeing if we can get a better idea of how much pollution is generated within Hong Kong, and how much comes from Southern China.

I can e-mail you my Excel files if you are interested Mr. B.

No need for those, but I'm very interested to keep hearing your conclusions as you think about it some more. It's good to have a second set of eyes looking at the numbers, and keeping me honest!

regards, MrB

Hong Kong Air Pollution

Thank Mr. B for the clarification on roadside stations, population of Tap Mun, APIs, etc. Much appreciated.
With a population of 100, Tap Mun's afternoon peaks need some investigation. Hong Kong's countryside has changed beyond one's imagination one-half century ago; so today no area is immune to local or distant point sources. They may not be considered prestine but in my days there, Yuen Long, Tai Po and Shatin were paradise.
One could keep looking at the statistics like my first attempt here. However, a more systemic approach is to look at the physical/chemical and meteorological process, and social-economic development through time, to identify the causes. And your discussions and analysis have made it a good start. Having looked at the data, Hong Kong's air pollution is indeed getting worse.
My interest continues and will comment after more reading and self-educating.
Regards also from Canada,
OldTimer

Hong Kong Air Pollution

Mr. B., as suggested, I read your comments that appeared on another Batgung page. I haven't read the researchers report or their methods of investigation (I need more time). This is about the Media's headline saying "NEW RESEARCH SHOWS 53% OF THE TIME HONG KONG’S DIRTY AIR CAUSED BY HONG KONG SOURCES OF POLLUTION".

I interpret it saying 53% of the time in a typical year, Hong Kong's dirty air can be primarily attributed to its own pollution sources. Take it to the extreme criterion, for each day or hour, sum up all the pollutants across Hong Kong, when the researchers can differentiate (I wonder if they can) that X tonnes came from outside, X plus (1 or more) tonnes came from within, then this particular day or hour is counted as a part of that 53%.

I don't think they meant to say over half (53%) of HK's air pollution comes from local sources.

I can't get my hand on wind-rose data that show the frequency of winds from the different directions. When I look back into my days flying kites during the summer, most of the time I got S, SE or SW winds. W wind was infrequent. E wind must be up there somewhere but I tend to stay indoor due to inclement weather so don't know how frequent it was. Winds from the northern quadrant was very infrequent in the summer. On this basis and without making any detailed investigation, I would think that during the summer when most of the time the winds are from the southern quadrant, it would be also infrequent to see significant air-borne pollutants from north of the border. During the winter when northerly wind becomes dominant, Hong Kong would see her big share of the pollution coming from the north.

Changing pollution levels at street-level in Causeway Bay

Here are the figures from the roadside measuring station in Causeway Bay:

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100  100   100 
 2005 72  97  82  83  58 
 2007 77  92  73  84  67 

Pollution levels there have clearly dropped since 2000, and more dramatically than the figures above from the Central roadside station. Next I'll do the Mongkok roadside station, so we can see all three.

Hong Kong Air Pollution

Quote from the 2006 annual report: "As a result of the enhanced vehicle emission control programme implemented by the Government since 2000, concentrations of respirable suspended particulates, nitrogen oxides and sulphur dioxide at roadside have been dropping gradually over the past few years." The graphs for the roadside stations show some I would call minor decline in these concentrations. Conditions could have been worse had there been no such government programme. Mr. B., your statistics for the Causeway Bay gives a more comforting picture.

Changing pollution levels at street-level in Mong Kok

Here are the figures from the roadside measuring station in Mong Kok:

   CO, change
since 2001
NO2, change
since 2001 
NO, change
since 2001 
 O3, change
since 2001
RSP, change
since 2001 
SO2, change
since 2001 
 2001 100  100  100  100   100 
 2005 58  94  91  94  121 
 2007 77  103  93  90  131 

Some up, some down, so no clear trend here. Still, none of these readings are even close to doubling in the way that the 'hours have reduced visibility' did over the same period.

OldTimer, thanks for your quote about the changes in vehicle emission control. They list SO2 as one of the pollutants affected, but the suggested reduction hasn't occurred in Mong Kok. As you say, maybe it would be even worse without those improvements in vehicle emissions.

Hong Kong Air Pollution

I grew up in the 1940~50s four blocks from the Monk Kok station. The air was poor then, and the traffic here was busy but not as horrendous as shown in present photos -
http://www.epd-asg.gov.hk/english/backgd/bkg_mongkok08.php
It is great to see the 3-storey buildings survive to this day. It looks like the 4th floor living quarters are add-on's. There was once a gasoline station in front of them where the trees and shrubs are. Car washing was a part of its business.
And Great World Cinema was across the street - saw a movie there which featured B-29 bombers as part of the movie.

Changing pollution levels in Sai Wan Ho

Here are the figures from the measuring station in Eastern (Sai Wan Ho):

   CO, change
since 2001
NO2, change
since 2001 
NO, change
since 2001 
 O3, change
since 2001
RSP, change
since 2001 
SO2, change
since 2001 
 2001 100  100  100   100 
 2005 101  119  116  117 
 2007 102  119  117  141 

So all the measured pollutants have got worse over the last seven years, with SO2 showing the worst change.

I'll do two more (Tung Chung and Yuen Long), then see what conclusions we can draw.

Changing pollution levels in Kwai Chung, Tung Chung, Yuen Long

Here are the figures from the Yuen Long measuring station:

   CO, change
since 2000
NO2, change
since 2000 
NO, change
since 2000 
 O3, change
since 2000
RSP, change
since 2000 
SO2, change
since 2000 
 2000 100  100  100   100 
 2005 101  112  112  146 
 2007 95  127  115  123 

Next the Tung Chung measuring station:

  CO, change
since 2000
NO2, change
since 2000
NO, change
since 2000
O3, change
since 2000
RSP, change
since 2000
SO2, change
since 2000
2000 100 100 100 100 100 100
2005 159 103 110 102 127 147
2007 141 103 104 108 120 158

And finally the Kwai Chung measuring station:

  CO, change
since 2000
NO2, change
since 2000
NO, change
since 2000
O3, change
since 2000
RSP, change
since 2000
SO2, change
since 2000
2000 100 100 100 100 100
2005 91 86 98 114 148
2007 88 87 120 117 143

Funnily enough I started with a measurement of SO2 levels in Central that showed a significant drop over the last 7 years. But looking at SO2 levels for these last three stations, it's clear that the answer to 'Is our air pollution getting better or worse?' depends on what you measure, and where you measure it.

Air quality in the good old days - II

Here's a photo of the Kennedy Town incinerator in action in 1972.

HONG KONG: 1972