If you are riding a motorcycle in haze this season, the advice you have been given is aimed at the wrong pollutant. Ride early, avoid rush hour, hang back from the truck. All of that is traffic advice, and haze is not traffic. It is regional, it sits over an entire city for weeks, and there is no gap in it to find. South-East Asia went onto a haze red alert on 26 June 2026, only the second since 2019, with peak risk forecast for August and September. The deeper story is what that alert points to: the source of the PM2.5 a rider breathes is shifting from the tailpipe in front of them to something burning in another province.
Two weeks ago we covered the 2026 haze season alert as news. This is the structural argument underneath it.
Is Air Quality Getting Better or Worse for Riders?
Both, and that is the whole problem. The annual average is improving while the worst days get worse.
Thailand's annual mean PM2.5 has fallen from 20.0 µg/m³ in 2019 to 17.8 µg/m³ in 2025. That looks like progress, and in one sense it is. But the year that broke the trend tells you more than the trend does: 2023 jumped from 18.1 to 23.3 µg/m³, a 29% rise in a single year, and it was a major burning year. The spikes track fire, not traffic.
Riders do not experience annual averages. You ride on a Tuesday. A falling mean with worsening peaks is worse for a daily commuter than a flat mean, because the exposure concentrates into the specific weeks you cannot avoid.
Thailand PM2.5: the average falls, the spike years climb
Annual mean, µg/m³ (IQAir) with a synthesised 2031 spike-year projection
The sources: IQAir's 2021 World Air Quality Report for 2019 to 2021, the 2022 report for 2022, the 2024 report for 2023 and 2024, and IQAir Thailand for 2025. That 2025 figure of 17.8 µg/m³ is still 3.6 times the WHO guideline. In northern Thailand, forest fires drove March and April spikes to 400% above WHO recommended levels, and the Geo-Informatics and Space Technology Development Agency identified 1,060 pollution hotspots across the country's 76 provinces. Northern residents stand to lose 2.7 years of life expectancy at current levels.
Why Doesn't Traffic Advice Work Against Haze?
Because tailpipe exposure is a proximity problem and haze is a regional one.
Exhaust falls off with distance and thins out at certain hours, which is why "ride early, leave a gap" genuinely helps. Haze does neither. It is spatially uniform across a city and persistent for weeks. Riding at 6am on a haze day buys you nothing.
And most of it is not even local. A 2026 WRF-CMAQ source-apportionment study attributes 62.7% of PM2.5 in Thailand's Northern Region to transboundary sources. You cannot ride away from it, and no single country can regulate it away alone.
Is Smoke Different From Exhaust?
Yes, and the difference is what defeats most filters.
Biomass smoke sits in a mono-modal size distribution of roughly 20 to 250 nanometres, with a median near 100 nm, according to Shirman et al. in Atmosphere (2023). That is well below PM2.5's headline size and squarely in the range that deposits deep in the lung. If you have read our piece on ultrafine particles and motorcycle riders, this is the same territory: small enough that the number you see on an AQI app does not describe what is reaching you.
The same study tested how filter media cope. Fiberglass media at MERV 12 to 14 captured 28 to 34% of ~100 nm particles. Electret media, which relies on an electrostatic charge, started better at 35 to 55%, then collapsed to about 2.5% once loaded with smoke. Mechanical fiberglass held its performance. Electret did not.
Here is why that matters on a bike. Most cheap disposable masks rely on exactly that electrostatic charge, and smoke is precisely the load that kills it. The protection degrades fastest at the moment it is needed most.
What Do the Next Five Years Look Like?
Tailpipe emissions per vehicle keep falling across South-East Asia as standards tighten and electrification scales. Fire does the opposite. UNEP projects extreme fire events rising 14% by 2030, 30% by 2050 and 50% by 2100.
Put those two curves together and the conclusion is uncomfortable. By 2031, a growing share of the PM2.5 a South-East Asian rider breathes will come from something burning in another province rather than from the bike in front of them. The tailpipe share shrinks and the smoke share grows to fill it. At 62.7% transboundary in northern Thailand already, the shift is underway rather than hypothetical. Our piece on whether electric motorcycles will reduce air pollution covers the tailpipe half of this curve; this is what replaces it.
There is precedent from a cleaner-air jurisdiction. In the United States, wildfire smoke slowed or reversed PM2.5 improvements in two-thirds of states, eroding 23% of prior gains on average, equal to 3.6 years of progress, and more than 50% in several western states. The researchers make a point worth sitting with: wildfire-driven PM2.5 increases are unregulated under current air pollution law.
The current season is the near-term proof. Hotspot counts across the southern ASEAN region rose 86% in the first half of 2026, the SIIA red alert is the second since 2019, peak risk lands in August and September, and NOAA puts the odds of El Niño persisting into early 2027 at 97%.
What Actually Holds Up Against Smoke?
Two properties matter, and both follow from the filter data above.
The first is mechanical rather than electrostatic filtration. Easi Breezi uses H11 HEPA media rated at ≥95% PM2.5, which captures by fibre geometry rather than by charge, so it does not lose its mechanism as smoke loads it. That is the failure mode that took electret media to 2.5%.
The second is active induction rather than passive vents. Haze does not clear at speed and does not thin out at 6am, so a system that only works while you are moving fast through cleaner air is solving a different problem. Pulling air through a filter and delivering it works the same at a standstill as it does at 80 km/h. For the broader picture on rider exposure, our motorcycle air pollution guide is the place to start.
Frequently Asked Questions
Is riding a motorcycle in haze more dangerous than riding in traffic pollution?
They are different exposures. Traffic pollution is concentrated and avoidable by timing and positioning. Haze is uniform across a region and lasts for weeks, so total exposure over a haze episode can be far higher even when the peak concentration is similar.
Does a normal helmet protect you from haze?
No. A standard helmet has open vents and no filtration media, so smoke particles pass straight through. Helmet vents move air, they do not clean it.
Do disposable masks work against smoke?
Less well than the packaging implies. Masks that rely on an electrostatic charge lose most of their capture efficiency once loaded with smoke, dropping to around 2.5% at ~100 nm in lab testing.
Will electric motorcycles fix this?
They reduce the tailpipe share, which matters. They do nothing about the biomass smoke that is becoming the larger share of what riders breathe.
When is the 2026 haze season expected to peak?
Forecasts point to August and September 2026, following the SIIA red alert issued on 26 June 2026.
Ready to Breathe Cleaner on Every Ride?
The haze peak lands in August and September. Easi Breezi units ship 26 to 28 July, before it arrives. The system uses H11 HEPA media rated ≥95% PM2.5 in an IP67-rated housing, patent pending, and it filters at idle as well as at speed, because haze does not care how fast you are going.
Pre-order is $153.00 with code EB10 (10% off, ends 31 July 2026). See the Easi Breezi unit.
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.