Singapore's west-region 24-hour PSI read 144 at 8am on 24 October 2026, up from 125 four days earlier (NEA, 2026). All five regions are unhealthy today. The wind shift behind the earlier dip only ever "limited" the smoke, never stopped it, and real relief is not expected before December.
Four days ago, the haze over Singapore looked like it was breaking up. This morning it is thicker again, and the west is not alone this time.
The 24-hour PSI read 144 in the west at 8am on 24 October 2026. Central read 135, east 118, north 117 and south 116. Every region sits above the 101 line that marks unhealthy air (NEA, 2026).
On 20 October, only the west and centre were unhealthy. The other three regions had slipped back to the moderate band. Today none of them have.
Key takeaways
- Singapore's west-region PSI climbed to 144 at 8am on 24 October 2026, up 15% from 125 four days earlier, though still below the 230 peak on 18 October (NEA, 2026).
- All five regions, west, central, east, north and south, are back in the unhealthy band together today, a wider spread than the 20 October dip (NEA, 2026).
- The 20 October improvement was never a resolution. NEA's own forecast said shifting winds would only "limit" Kalimantan smoke, not stop it, and flagged that Sumatra smoke could still drift in (Bernama, 2026).
- Motorcycle riders carried 4 times the PM2.5 of car occupants on the same roads in a Makassar study, so today's island figure is a floor for a rider's dose, not the dose itself (Patel et al., 2016).
What the readings and studies actually measured
The PSI is a 24-hour average, so it moves slowly and always carries yesterday's smoke. The west read 230 at its 18 October peak, fell to 125 by 20 October, and has now climbed back to 144 (NEA, 2026).
That is a real move, not noise. A 15% rise in four days is large on an index built to smooth out bigger swings.
For comparison, the WHO's 24-hour guideline for PM2.5 is 15 µg/m³ (WHO, 2021). Singapore's own regional haze has run well above that most of this season.
Two of the three studies behind this post measured something closer to what a rider actually breathes, not a city average.
Samana and colleagues fitted personal samplers to motorcycle-taxi drivers across six Thai provinces in 2023. Their personal PM2.5 readings moved independently of the nearest fixed monitoring station. In one province, drivers averaged 410.9 µg/m³, far above anything a city-wide index would show (Samana et al., 2025).
Patel and colleagues measured motorcycle, minibus and car trips in Makassar, Indonesia, with handheld particle counters. Riders carried 4 times the PM2.5 and 13 times the PM10 of car occupants on the same roads (Patel et al., 2016).
The third study used Singapore's own hospital data. Ho and colleagues matched all 53,948 heart attacks in the national registry from 2010 to 2015 against the PSI on the days around each one. Each 30-unit PSI rise came with 4% more heart attacks, and days in the unhealthy band had 9% more (Ho et al., 2019).
The wind shift that helped was only ever a limit, not a stop
The 20 October dip followed winds shifting to blow from the south on 18 October. NEA said the change "may limit" smoke transport from Kalimantan, not stop it, and flagged that Sumatra smoke could still drift in regardless (Bernama, 2026).
The fires behind the haze were not shrinking while that wind was helping. Indonesia's Forestry Ministry recorded 17,706 hotspots nationwide in the 24 hours to 18 October, 6,665 more than the period before, with Central Kalimantan the highest single province at 8,338 (Databoks, 2026).
Hotspot counts depend heavily on confidence filters. The ministry's SiPongi system split that same 18 October total into 713 high-confidence, 16,610 medium-confidence and 383 low-confidence detections (Databoks, 2026). A hotspot is a heat signature read by satellite, not a confirmed fire, so treat the count as an activity signal, not a fire tally.
A four-day dip is not a trend. Check the hourly PM2.5 feed the morning you ride, not last week's headline.
A weather and climate expert at Nanyang Technological University's National Institute of Education told the Straits Times she expects real relief only around December, when monsoon rains reach Indonesia and help contain the fires (NTU NIE, 2026).
She cautioned that even then, the respite could be weaker and shorter than usual. Real improvement needs the wind to shift and blow from the north.
A climb on 24 October, while that wind shift is still weeks away, fits the pattern rather than breaking it.
Ride with filtered air
Filters haze PM2.5 on every ride
Easi Breezi clips onto any helmet and feeds you H11 HEPA-filtered air. Bike-powered, no batteries, no charging.
144 is Singapore's floor, not a rider's dose
None of the three studies behind this post measured a rider in Singapore's 2026 haze specifically. Say that plainly rather than imply they did.
What they show is a pattern that should worry a rider more than the headline number. In Thailand, motorcycle-taxi drivers' personal PM2.5 readings moved independently of the nearest station, sometimes reaching a mean of 410.9 µg/m³ in one province, far above anything a city-wide index reports (Samana et al., 2025).
Average motorcycle exposure against cars on the same Makassar roads, in a baseline cleaner than today's haze.
Source: Patel et al., Science of the Total Environment, 2016Singapore's own hospital data shows the city-wide number still matters. Each 30-unit PSI rise came with 4% more heart attacks in the national registry, and the unhealthy band itself carried 9% more (Ho et al., 2019).
That is an association across 53,948 cases, not proof that any single ride caused any single event. It is Singapore's own data, not a borrowed study.
What to do this week
- Check the hourly PM2.5 feed before you ride. A single daily PSI headline hides which hour is worst.
- Plan for more days like this, not fewer. An NTU climate expert expects real relief only around December, and cautioned it may be weaker than usual (NTU NIE, 2026).
- Shorten or combine rides in the west and centre. Both regions sit at 144 and 135 today, the highest in the island.
- Seat a fitted N95 or KN95 properly. In the Makassar study, a loose mask lost most of its benefit (Patel et al., 2016).
- For filtered air that does not depend on mask fit or a window in the wind, look at an active HEPA helmet system.
For the bigger picture, see when the haze will be gone and the live haze map for riders.
Frequently asked questions
What is the PSI in Singapore today?
At 8am on 24 October 2026, the 24-hour PSI was 144 in the west, 135 in the central region, 118 in the east, 117 in the north and 116 in the south, all in the unhealthy band (NEA, 2026).
Why did the haze get worse again after it improved?
NEA's own forecast said the 20 October wind shift would only "limit" Kalimantan smoke, not stop it, and that Sumatra smoke could still drift in. Indonesia's fires were not shrinking at the same time, with 17,706 hotspots nationwide recorded in the 24 hours to 18 October alone (Bernama, 2026; Databoks, 2026).
When will the haze actually clear?
A weather and climate expert at NTU's National Institute of Education expects real relief only around December, when monsoon rains reach Indonesia, and warned the respite could be shorter and weaker than usual (NTU NIE, 2026).
Is a PSI of 144 dangerous for riders?
It sits in the unhealthy band, where NEA advises reducing outdoor activity, but it measures an island average, not a rider's dose. In one study, motorcycle riders carried 4 times the PM2.5 of car occupants on the same roads (Patel et al., 2016).
The research behind this: 3 studies
Peer reviewed work this post is built on. Each card says what was measured, what was found, and what it cannot tell you.
Motorcycle vs car exposure
- What they measured
- Handheld particle counters on motorcycle, minibus and car trips in Makassar, Indonesia, plus tests of surgical masks, bandanas and motorcycle masks.
- What they found
- Against cars, average motorcycle exposure was 4 times higher for PM2.5 and 13 times higher for PM10. Only surgical masks lowered it consistently.
- What it can't tell you
- A non-haze baseline, so the gap in a haze week is probably larger, but that was not measured.
Motorcycle-taxi drivers, personal vs station readings
- What they measured
- Personal PM2.5 and PAH sampling on motorcycle-taxi drivers across six central Thai provinces, January to March 2023.
- What they found
- Pathum Thani drivers carried a mean personal PM2.5 of 410.9 µg/m³, and personal readings moved independently of the nearest fixed monitoring station.
- What it can't tell you
- No individual lung-function outcome was measured, and the province-level health link is ecological, not personal.
PSI and heart attacks
- What they measured
- All 53,948 heart attacks in Singapore's national myocardial infarction registry from 2010 to 2015, compared with PSI on the days around each event.
- What they found
- Each 30-unit PSI rise came with 4% more heart attacks. Days in the unhealthy band had 9% more.
- What it can't tell you
- An association from registry data. It does not prove haze caused any single event, and today's 144 sits at the low end of the unhealthy band.
Sources
- National Environment Agency, Singapore (2026). Real-time 24-hour PSI and PM2.5 feed, data.gov.sg. Pulled 24 October 2026, 8am SGT.
- National Environment Agency, Singapore (2026). Haze portal: PSI bands and descriptors. Viewed 24 October 2026.
- Malay Mail (2026). Singapore's west records season-high PSI of 230 as haze enters 14th day. 18 October 2026.
- World Health Organization (2021). WHO global air quality guidelines: particulate matter, ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide.
- Samana, K., Ito, K., Suthienkul, O., Ketsakorn, A. (2025). Personal exposure to PM2.5 and polycyclic aromatic hydrocarbons among motorcycle-taxi drivers in Thailand. PLoS One.
- Patel, D., Shibata, T., Wilson, J., Maidin, A. (2016). Challenges in evaluating PM concentration levels, commuting exposure, and mask efficacy in reducing PM exposure in growing, urban communities in a developing country. Science of the Total Environment.
- Ho, A. F. W. et al. (2019). Time-stratified case crossover study of the association of outdoor ambient air pollution with the risk of acute myocardial infarction in the context of seasonal exposure to the Southeast Asian haze problem. Journal of the American Heart Association.
- Bernama (2026). Singapore air quality improves as winds shift, but haze persists. 18 October 2026.
- Databoks / Indonesia Ministry of Environment and Forestry (2026). MoEF: Indonesia recorded 17,706 hotspots in last 24 hours. 18 October 2026.
- Nanyang Technological University, National Institute of Education (2026). Heat and haze could linger over Singapore for a longer period due to El Nino, experts say. 2 October 2026.
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.