This issue of

Chiang Mai Citylife > Articles > 2018 > 2018 Issue 04 > Explaining Chiang Mai’s Smoke Pollution With Real Data

Explaining Chiang Mai’s Smoke Pollution With Real Data

As the smoke and haze thickens, Chiang Mai’s citizens are beginning to voice concerns, anger and frustrations over what little is being done to alleviate this annual crisis.

Over the years Citylife has published numerous articles about this issue, but in the last year or so we have had little to write about as progress has stagnated and it is obvious to anyone who breathes that we are in a pickle. Little has been done by authorities to combat the issue and in spite of the 51 day burning ban this year (nine days shorter than in 2017), the smoke has increasingly become worse. The city is blanketed in a pall of haze; our beloved Doi Suthep disappearing like a conjuror’s trick before our stinging eyes.

So, with no promising news or solutions being proposed by the government, we turn to the community for some updates and information. Meet a resident expat who has taken it upon himself to set up a detailed network of air pollution sensors that are both low cost and accurate, in a bid to collect a quantatitve amount of pm2.5 data for the first time in Chiang Mai in hopes of raising awareness, informing policy makers and hopefully even pinpointing specific sources of smoke by combining ground data with real time satellite information.

Craig Houston is an aeronautical engineer and volunteer in Chiang Mai, moving here three years ago along with his wife. After his first son was born, he began working on collecting data about air pollution with a team of doctors, professors and activists in the region. Citylife interviewed Houston this smoky season to help clarify the problem and explain what he is trying to do.

Citylife: What’s the situation with air pollution in Chiang Mai?

Houston: Air pollution is basically anything in the air that has a harmful or poisonous effect, such as carbon monoxide, lead, ground level ozone, sulphur, nitrogen dioxide and particulate pollution. In Chiang Mai, the pollutant of most concern is particulate matter — small bits of dust from fires, construction and power plants. Much of what we see in our air over these few months comes from farmers and factories burning huge amounts of agricultural waste.

[ed. According to Warm Heart Foundation, over 200,000 tonnes of particulates and 24 million tonnes of greenhouse gasses are released into the air every burning season.]

These particulates range in size and are referred to as pm10 and pm2.5 — 10 and 2.5 micrometres in width respectively. The worse the air quality is, the more micrograms (μg) of these particles are found in every cubic metre of air — which is the standard measurement used to calculate air quality, written as μg/m3. Breathing in these particles can lead to permanent DNA mutations, heart attacks and premature death.

The problem with pm2.5 is that the particles are so small that they can penetrate deep into the lungs and even enter the bloodstream unfiltered. Roughly 20 pm2.5 particles can fit across the width of one human hair, compared to just five pm10 particles. According to a study by the European Study of Cohorts for Air Pollution Effects of over 300,000 people, every 5μg/m3 increase in pm2.5 increases the risk of lung cancer by an average of 18% over a 12.8 year period.

Citylife: How is AQI measured and what values should we begin to worry about?

Houston: An Air Quality Index (AQI) is a scale that governments use to inform the general public about the level of air pollution and associated health risks. One of the most widely used is the US EPA Scale (from the US Environmental Protection Agency) which is based on a comprehensive review of scientific health studies. Since Thailand does not yet have an AQI for pm2.5, the EPA standard is a good one to follow. When the AQI reaches over 100 for either pm10 or pm2.5 then you should begin to be concerned and take appropriate action.

When it comes to calculating AQI however, the process is not as easy as it seems. Firstly, there are two different AQI scales, one for pm10 and one for pm2.5, so you have to be careful which AQI you are referring to. Secondly, the conversion rate has several ‘breakpoints’ where the ratio to convert particle concentration (μg/m3 over a 24 hour average) to AQI changes. These break points were set based on the scientific evidence of health effects at these different concentrations. The conversion tries to convey that a 20μg increase when the concentration is already high (such as 300μg/m3), is less significant than when the concentration is at say, 10μg/m3.

For example, a 200μg/m3 concentration of pm10 particles equates to a pm10 AQI of 123, which sits in the middle of the orange ‘unhealthy for sensitive groups’. However, a 200μg/m3 concentration of pm2.5 particles equates to a pm2.5 AQI of 250, past the red ‘unhealthy’ level and deep into the pink ‘very unhealthy’ level. Quite simply, for the equivalent concentration, pm2.5 is far more dangerous than pm10. offers a simple to use AQI calculator to save the hassle.

Citylife: How did you get involved with air pollution?

Houston: I used to live in Beijing, China which is significantly worse than Chiang Mai as the smog is all year round. However, during the burning season there are certainly days when Chiang Mai is worse.

I moved to Chiang Mai almost three years ago, but it was only after the birth of our first son here that I started to become more engaged with the smoke problem — I want to protect him as much as I can. Air pollution is democratic, when you are outside it affects everyone equally. I began by placing a HEPA air purifier in my son’s room and measured the indoor air quality with an AirVisual sensor. His room was under 50 AQI (12μg/m3 of pm2.5 particles) while the rest of the house was in the red. Over time we kitted out the whole house with air purifiers to try and make it a safe haven from the smoke outside. What many people don’t realise is that without air purifiers and proper sealing, indoor air can be just as bad as outdoors — these tiny particles can easily enter through very small gaps.

Citylife: Why did you choose an AirVisual sensor?

Houston: Like most people I turned to Google to help me do some online research, and I found a thriving new world of low cost pollution sensors. There were people in China who had compared several low cost air pollution sensors and found their accuracy to be surprisingly good. Some sensors also measure CO2 levels and air humidity, making them more environmental sensors rather than just pollution particle sensors. AirVisual ( caught my eye because the sensor had a good interface and it allowed me to access the data for myself. It also syncs with my smartphone so I can check it wherever I am. Previously, I was using a Gaia sensor, you may have seen them on the AQI websites, but I found the initial setup and user interface to be less intuitive.

You can buy an AirVisual sensor for $270 and then use it for measuring your indoor air quality, or set it up outdoors and join the global network — after a few checks and tests of course. There are many sensors out there too, and some just as good as AirVisual. The industry has reached critical mass with sensors and the lasers that measure the particles have come down in price considerably in the last few years. No longer are we limited to the government sensors which, although generally more accurate, come with a hefty price tag in the tens of thousands of dollars.

Partly because of the high cost, Chiang Mai has only two permanent government air pollution sensors — one at city hall and another in the old city. Despite recording pm2.5 levels, the government has yet to officially report these findings, but they are available online if you can dig your way through the Thai government website.

Thanks to lower costs of these new sensors, I am able to have two AirVisual sensors and a PurpleAir sensor (another low cost air monitor) running outside my home. This way I can check the consistency of my readings and also maintain a level of technological diversity. I guess I am also a bit of a data geek.

Citylife: How did you expand the network of sensors across Chiang Mai and why?

Houston: Around the same time I began monitoring air at my home, a group of people in Chiang Rai began collecting and sharing information online from a number of private air monitors across the province. In Chiang Mai, we wanted to do the same, so myself and a number of keen activists and locals concerned with the smoke problem helped install Gaia sensors at CMIS and PREM international schools and linked them to the AQICN website.

In mid-2017, a group of us purchased a set of AirVisual sensors to begin building a network of sensors across the city. Soon after that we placed another order of three sensors that were funded by the community to be set up and connected with the online public network. Today we have 10 sensors operating across the city in Wat Gate, CMIS, Payap, Hang Dong, Phi Suea, Mae Rim, PREM, APIS Primary (Hang Dong), APIS Main Campus (Samoeng Road), and Lanna International School. Outside of the city we have a sensors in Chiang Dao and Pai. Over the next year we expect to add at several more to the network — right now we have gaps in Mae Hong Son, Mae Chaem and areas further out of the city in the rural areas.

Chiang Rai’s network has grown too but they are still using Gaia sensors and are a separate community, but we would like to connect our data in the future. Ideally, communities like these could be set up in other cities too, creating a large distributed, publicly owned network. This is especially valuable where no monitoring currently exists and installing an expensive government station is unrealistic.

Citylife: What can be done with this information and how will it help Chiang Mai?

Houston: First of all these new sensors are providing the public with real-time data on the most important pollutant for Chaing Mai, which is pm2.5 data. That information hasn’t been available before and in the short-term our most important goal is to help our community be better informed about air quality so they can protect their health. Young children are particularly vulnerable and this is the reason we have been actively working with local schools.

Our network is enabling us, for the first time, to record pm2.5 data simultaneously from a diverse set of locations across the city. Although we have some government pm10 data going back to 1999, it doesn’t help us fully understand the trend of the most dangerous pm2.5 particles over time nor does it give us much geographic granularity to identify problem areas.

This year will be the most closely monitored smoky season ever, and that’s exciting because it opens up new possibilities for our community to really understand the problem and design more targeted and effective solutions. For example, by combining our sensor data with NASA forest fire mapping and other satellite data we could build a smoke model. By triangulating the data we should be able to pinpoint the main sources of the smoke and, if our network is big enough, have real time smog forecasts for the city, as we see the values increase as the smoke rolls in. We want to set up a few sensors around Mae Chaem, where we know large scale burning of corn waste happens. Setting up more sensors in key areas can help us build a more complete picture of this complex problem and identify the biggest contributors to tackle first. Currently, there is a blanket ban but people are still burning, perhaps a location specific ban would be easier to control and have a greater effect in reducing pollutants in the air — as long as we can identify key burning locations that contribute most to the air pollution that is.

We know that hospital admissions for respiratory conditions and other related illnesses spike during the burning season too. This new data could also help us better understand this relationship and provide public health officials and hospitals with up-to-date forecasts for smoke related admissions.

Since I started, I have pulled together a team of people including doctors from the Faculty of Medicine, academics from Research Institute for Health Science and the Faculty of Agriculture, Chiang Mai University, along with a number of local activist such as Sakda Darawan (Raks Mae Ping), Dr. Michael Shafer (Warm Heart Foundation) and Marisa Marchitelli (Smoke: The Documentary). I have also been consulting with Dr. Vijay Limaye, an air pollution scientist trained in environmental epidemiology who has worked for the US EPA and the Natural Resource Defence Council in New York. He specialises in the health impacts of air pollution.

Right now we are developing a history of data, and the more sensors the better. Anyone who puts an AirVisual sensor on the public network will be helping us build a bigger picture of the pollution in Chiang Mai. In the near future we plan to include data from other sources too. If anyone wants to join our network they can reach out to me ([email protected]) or visit

We will make all of our data freely available and my hope is that others will bring their skills to help us better understand our smoke problem and provide actionable advice for local government. Of course, understanding the problem and solving the problem are two different things, but I do believe our efforts to collect and share this data can lead us to more effective, evidence based solutions.