Bad enough that the number you've been reading doesn't describe it. Over seven months in Thailand, researchers put low-cost PM2.5 sensors on ten working motorcycle riders. Fixed monitoring stations recorded levels below 50 µg/m³. The riders' own sensors frequently approached 100 µg/m³ in the same period — roughly double. Six of the ten scored a Hazard Quotient above 1, the threshold at which health authorities consider adverse effects possible. All of this happened while Thailand's national air quality statistics were improving.
Both numbers are real. They are measuring different air.
Thailand's Air Really Has Been Getting Cleaner
Start by giving the good news its due, because it is genuine. According to IQAir's country rankings, Thailand's annual average PM2.5 concentration fell from 26.4 µg/m³ in 2018 to 17.8 µg/m³ in 2025 — a decline of about a third, and the lowest reading in the eight-year series. Thailand now ranks 48th most polluted country in the world.
That decline was not a smooth line. The same dataset shows 21.4 µg/m³ in 2020, 20.2 in 2021, and 18.1 in 2022 — then a sharp reversal to 23.3 µg/m³ in 2023, before falling again to 19.8 in 2024 and 17.8 in 2025. One bad burning season undoes several years of progress. That volatility matters more to a rider than the trend line does, because you don't breathe an eight-year average. You breathe Tuesday.
Thailand's national PM2.5, 2018–2025 — and where the trend points
Annual average, micrograms per cubic metre. Solid blue = measured (IQAir). Dashed orange = projection from the 2020–2025 rate, not a forecast.
- Measured (IQAir)
- Projected from 2020–2025 rate
So Why Did Riders Measure Roughly Double the Station Reading?
Because a regulatory monitoring station is not sitting where you sit. It is set back from the roadway and often elevated, and that siting is deliberate — the station's job is to characterise a region's air, not one traffic queue. If you parked it at the kerb behind a bus, it would report the bus, and the national statistic would become meaningless.
The design is correct for its purpose. It is simply the wrong altitude for a rider's face.
The MobileSense study, presented at the 2025 ACM SIGCAS/SIGCHI Conference on Computing and Sustainable Societies, tracked ten riders across Bangkok and Chiang Mai from 1 November 2023 to 15 May 2024. Its central observation is that the gap between the official number and the human number is not a rounding error. As the authors put it, the platform "captures on-the-ground exposure levels, bridging the gap in accurate human pollution assessment."
A rider's intake sits at roughly the height of the exhaust of the vehicle ahead, inside the plume before it has had any distance to dilute. Three things compound from there:
- You are downstream of the source, not near it. Following distance in dense traffic is a few metres. The concentration gradient across those metres is steep.
- At a standstill, your ventilation stops. Passive helmet vents are ram-air devices. They move air because you move. At zero road speed there is no pressure differential, so the vents stop ventilating and become openings into whatever pocket of air you happen to be sitting in.
- Stopped traffic is where concentrations peak. Idling engines, no dispersion, and you are stationary inside it — which is exactly when the vents have stopped working.
The single highest reading in the study was 133.8 µg/m³, recorded on 30 January 2024. That is one rider, on one day, in a country whose annual average that year was 19.8 µg/m³.
The same seven months, four different numbers
PM2.5 in micrograms per cubic metre. Blue = reference and station readings. Orange = what riders' own sensors recorded.
What Does a Hazard Quotient Above 1 Actually Mean?
A Hazard Quotient is exposure divided by the level at which no adverse effects are expected. The US EPA treats it as a straightforward gate: at 1 or below, adverse effects are unlikely. Above 1, adverse effects are possible, and the potential increases as the number rises — though the metric deliberately doesn't tell you by how much.
Six of the ten riders in the study came out above 1.
This is worth sitting with, because it moves the conversation off opinion. "The air felt bad today" is a vibe. A Hazard Quotient is the same standardised instrument used in occupational risk assessment, applied to people whose workplace happens to be a road. And it was calculated from what the riders actually breathed, not from what the nearest station reported.
Does Riding Less on Bad Days Fix It?
This is the part of the study that almost nobody has reported, and it is the most useful finding in it. The researchers didn't just measure exposure — they modelled the standard public health advice to see how much it would help.
A daily one-hour break reduced Hazard Quotient by up to 8.26% in Bangkok. Taking full days off during peak pollution periods produced reductions ranging from 7.17% to 15.01%, depending on the rider.
Read those numbers as a rider, not as a policy analyst. For someone whose income is the ride, the proposal is to surrender working hours — or whole working days — for something in the region of a tenth off a risk score that started above the safe threshold. It does not get you under 1. Scheduling is not a solution; the study quantified precisely how little it can do.
That is not an argument against taking breaks. It is an argument that avoidance alone was never going to be enough, and now there is a number attached to why.
What Does Rider Exposure Look Like by 2031?
Take the optimistic case. From the 2020 base of 21.4 µg/m³ to the 2025 figure of 17.8 µg/m³, Thailand's national average improved at a compound rate of roughly 3.6% per year. Hold that rate steady for six more years:
17.8 × (1 − 0.036)6 ≈ 14.3 µg/m³ by 2031.
That is a synthesised projection, not a forecast from the source — the arithmetic is above so you can check it. And it assumes uninterrupted improvement, which the 2023 spike shows is not how this behaves.
Even so, the destination is instructive. A national average of 14.3 µg/m³ is still roughly 2.9 times the WHO annual guideline of 5 µg/m³. And it remains a rooftop number. Apply the study's gap to it and the rider's air in 2031 still lands where the rider's air is today: somewhere that a Hazard Quotient calculation would not like.
The regional outlook agrees. The Climate & Clean Air Coalition's assessment for Asia and the Pacific finds that if current policies are enforced, air pollution in 2030 will remain at 2015 levels, leaving over 4 billion people exposed to health-damaging levels. Around 92% of the region's population — about 4 billion people — already are. Their 25 recommended measures could deliver safe air for 1 billion people by 2030, which is real progress and still leaves the majority outside it.
If you ride, waiting for the ambient air to fix this is a plan with a timeline measured in decades.
If You Can't Change the Air, Change What Reaches Your Face
Every finding here points the same direction. The rider's air is local to the rider, so the intervention has to be local to the rider too. A national average can't protect you, and a monitoring station can't move.
Easi Breezi is an active induction system: a powered fan pulls air through an H11 HEPA element rated at ≥95% capture of PM2.5, and delivers it into the breathing zone. The word that matters is powered. Because it doesn't depend on ram air, it keeps moving filtered air when you're stopped in traffic — the exact condition where passive vents stop working and where the study's riders were recording their worst numbers. The unit is IP67 rated and patent pending.
For the wider picture on what riders are breathing, start with our guide to motorcycle air pollution. If this post was about where the official number is measured, our piece on ultrafine particles covers what it leaves out — and we've also compared motorcycle, car and bus exposure directly.
Frequently Asked Questions
How bad is air pollution for motorcycle riders, really?
Worse than ambient figures suggest. In a seven-month Thai study, riders' personal sensors frequently approached 100 µg/m³ while nearby fixed stations read below 50 µg/m³, with a peak of 133.8 µg/m³. Six of ten riders scored a Hazard Quotient above 1, indicating possible adverse health effects.
Is my city's air quality reading accurate for someone on a motorcycle?
It is accurate for what it measures — regional air, from a station set back from traffic and often elevated. It is not a measurement of your breathing zone. Riders sit at exhaust height in an undiluted plume, which is why personal sensors read substantially higher.
What does a Hazard Quotient above 1 mean?
The Hazard Quotient is exposure divided by the level considered safe. Per the US EPA, a value at or below 1 means adverse effects are unlikely; above 1 means adverse effects are possible, with potential rising as the value increases.
Does riding at off-peak times protect you?
Only marginally. The study modelled it: a daily one-hour break cut Hazard Quotient by up to 8.26%, and taking full days off on peak pollution days cut it 7.17% to 15.01%. That is not enough to bring an at-risk rider below the threshold.
Is air quality in Thailand getting better or worse?
Better on the national average — 26.4 µg/m³ in 2018 to 17.8 µg/m³ in 2025, per IQAir — but not smoothly, with a reversal to 23.3 µg/m³ in 2023. And a falling national average has not closed the gap between station readings and what riders actually breathe.
Ready to Breathe Cleaner on Every Ride?
The station on the roof isn't going to move. The filter on your helmet can. Easi Breezi delivers actively filtered air to your breathing zone whether you're at speed or stopped dead in traffic — which is where the numbers get worst and passive vents give up.
Pre-orders are open now at $199.00. Reserve yours here.
Written by Ash — mechanical engineer and founder of Easi Breezi, building an active HEPA filtration system for motorcycle helmets (patent pending). Based between Hong Kong and Bali, riding daily in the traffic this blog writes about.