Bali does not report PSI. Its official air reading is Indonesia's ISPU, an index with its own bands. A PSI figure quoted for Bali is Singapore's index applied to Bali's air. Read the ISPU for the monitoring station nearest your route, and treat any converted number as a rough guide, not a measurement.
Picture the check you run before a ride out of Denpasar: one phone screen, one number, and one word underneath it. Searches for "PSI in Bali" come up often enough that the question deserves a straight answer.
The number may be real, but it is measured on a different scale from the one Bali uses. Before you decide whether the air is safe for a commute in traffic, you need to know which yardstick produced it.
Key takeaways
- Bali's official reading is Indonesia's ISPU, not Singapore's PSI. A PSI figure for Bali is a converted reading of the same smoke (Ministry of Environment and Forestry Regulation No. 14 of 2020).
- In Central Kalimantan's 2023 dry season, outdoor PM2.5 averaged 136 µg/m³, and fires supplied 90 µg/m³ of that (Graham et al., 2024).
- The risk of a respiratory clinic visit was 36.6% higher on haze days and 74.4% higher on fire-haze days in one Central Kalimantan regency (Phung et al., 2025).
Bali reads ISPU, and PSI is a Singapore number
PSI, the Pollutant Standards Index, belongs to Singapore. The National Environment Agency calculates it from six pollutants: sulphur dioxide, PM10, PM2.5, nitrogen dioxide, carbon monoxide and ozone. Each pollutant gets a sub-index from 0 to 500. The reported PSI is the highest of those sub-indices (NEA, Computation of the Pollutant Standards Index). One bad pollutant can set the whole number.
Indonesia uses ISPU, the Indeks Standar Pencemar Udara. Ministry of Environment and Forestry Regulation No. 14 of 2020 sets it out. The public summaries we checked give its bands as 0 to 50 good, 51 to 100 moderate, and 101 to 200 unhealthy. We could not open the regulation text from this run, so check the breakpoints there before you rely on any single band edge.
Bali has a PSI reading, so the PSI number is Bali's air quality.
Bali's official reading is ISPU. A PSI figure for Bali is a Singapore scale applied to Indonesian air.
So when an app shows a PSI number for Bali, it has converted the smoke onto a different scale. The smoke is real. The label is a conversion, and a conversion can put the same air in a different category.
Pick the ISPU station closest to your route. A station in the city centre may read differently from one on the road out to the rice fields.
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.
Why one daily index misses the minute you sit in a queue
Both indices report averages over 24 hours. That suits a public health alert. It does not describe what a rider breathes at exhaust height while stopped at a junction.
A daily figure compresses hours of different air into one number. The sensor study reports a dry-season mean, not the minute-by-minute peak a rider meets on a busy road. The mean is still useful, because it shows how far the air can sit from the WHO guideline.
The WHO 2021 guideline sets an annual mean of 5 µg/m³ for PM2.5 (WHO, 2021). A dry-season mean covers a different period, so the comparison shows the gap in scale. It does not give a compliance verdict on any one day.
Which number to trust before you ride
Use three checks, in this order.
- The station, not the country. Use the ISPU for the monitoring station nearest your route, with its reading time.
- The reading, not the forecast. A reading is what the air did. A forecast is a guess about it. Use readings to decide today's ride.
- The estimate, with its scope. The 87,000 figure is a 2019 nationwide model estimate. The 100,000 figure is a regional burden estimate with a wide range. Both are estimates, and the method changes them (Grosvenor et al., 2024; Phung et al., 2022).
The index tells you whether a day is worth riding. It does not tell you what reaches your face. A seal and a filter change the dose you breathe, and the index cannot.
What the studies actually measured
Four peer-reviewed papers sit behind this post. Two measured the air directly. Two turned health records into estimates. None of the four is a measurement taken in Bali.
Two papers come from Central Kalimantan, a province on Borneo. In 2023, a team ran 13 low-cost PM2.5 sensors across urban and rural sites, indoors and outdoors, from 1 September to 31 October. Outdoor PM2.5 averaged 136 µg/m³ over that dry season, and fires contributed 90 µg/m³ of it (Graham et al., 2024). The houses gave little shelter. The median indoor-to-outdoor ratio in rural areas was 1.01 (Graham et al., 2024).
A 2019 study used a similar sensor network around Palangka Raya and checked a Copernicus air-quality model against it. Its health figures are model estimates: over 1,200 excess deaths in the Palangka Raya region and more than 87,000 nationwide (Grosvenor et al., 2024).
The other two papers look at people. In Pulang Pisau Regency, the risk of a respiratory visit to a local health centre was 36.6% higher on haze days, and 74.4% higher on fire-haze days (Phung et al., 2025). Those are population-level risks, not a personal dose. A 2022 review then pulled the work together. Burden studies put the regional toll near 100,000 deaths, but individual estimates range from 1,064 to 260,000 depending on method (Phung et al., 2022).
Read together, the four papers agree on one point. Haze is a health problem across the region. They disagree on the size of it, and that disagreement is the reason to read the local index with care.
The research behind this (tap to open)
Peer-reviewed studies this post is built on. Each card says what was measured, what was found, and what it cannot tell you.
Catastrophic impact of extreme 2019 Indonesian peatland fires on urban air quality and health
- What they measured
- A low-cost PM2.5 sensor network around Palangka Raya, Central Kalimantan, through the 2019 fire season, checked against a Copernicus air-quality model.
- What they found
- Over 1,200 excess deaths in the Palangka Raya region and more than 87,000 nationwide, attributed to fire PM2.5 (model-based estimates).
- What it can't tell you
- These are modelled health estimates, not counted deaths. The sensors covered one city in one fire season.
Updated Smoke Exposure Estimate for Indonesian Peatland Fires Using a Network of Low-Cost PM2.5 Sensors and a Regional Air Quality Model
- What they measured
- 13 low-cost PM2.5 sensors across urban and rural Central Kalimantan, indoors and outdoors, from 1 September to 31 October 2023.
- What they found
- Mean outdoor PM2.5 of 136 µg/m³, with fires contributing 90 µg/m³. Median indoor-to-outdoor ratio of 1.01 in rural areas.
- What it can't tell you
- One dry season and 13 sensors. The count of exposed people rests on modelled data.
Effects of smoke haze on respiratory clinic visits in Central Kalimantan, Indonesia according to different haze characteristics
- What they measured
- Daily respiratory visits at local primary health centres (puskesmas) in Central Kalimantan, 2015 to 2019, across two major haze episodes.
- What they found
- A 36.6% higher risk on haze days (95% CI 10.2% to 69.3%) and 74.4% on fire-haze days (18.9% to 164.6%), in Pulang Pisau Regency.
- What it can't tell you
- One province and two episodes. The authors call the overall effect inconclusive and most visible where fire haze dominates.
A scoping review on the health effects of smoke haze from vegetation and peatland fires in Southeast Asia
- What they measured
- Published smoke-haze health studies in Southeast Asia, compared by epidemiology, burden estimation and risk assessment.
- What they found
- Burden studies estimate about 100,000 deaths attributable to haze across Southeast Asia. Individual estimates range from 1,064 to 260,000, depending on cause, age group, area and period.
- What it can't tell you
- Health-effect studies come from only several countries, and the three approaches are not directly comparable.
Frequently asked questions
Is there a PSI in Bali?
No official PSI exists for Bali. Indonesian monitoring, including Bali's, reports ISPU under Regulation No. 14 of 2020. A PSI figure for Bali is a converted number, so read the ISPU instead.
What is the PSI in Bali right now?
This post cannot give a live number, because readings change hourly and this page does not carry them. Check the ISPU for your nearest station at the time you ride.
Can I convert PSI to ISPU?
Not with confidence. The two indices use different scales and bands, so a converted figure is only an approximation. Use the ISPU reading from the station near you.
Don't want to keep checking?
What to do this week
- Find your ISPU station. Choose the station closest to your usual route, and note its name.
- Log one reading a day for a week. A pattern across days tells you more than a single number.
- Ride when the station reads lowest. Shift your timing if your route's station shows a clear low window.
- Filter the air you breathe. The EB unit uses an H11 HEPA filter that stops 95% of PM2.5. See the Easi Breezi unit.
The index tells you whether to ride. The filter changes what reaches your face.
Sources
- Grosvenor, M. J., et al. (2024). Catastrophic impact of extreme 2019 Indonesian peatland fires on urban air quality and health. Communications Earth & Environment. doi:10.1038/s43247-024-01813-w
- Graham, A. M., et al. (2024). Updated Smoke Exposure Estimate for Indonesian Peatland Fires Using a Network of Low-Cost PM2.5 Sensors and a Regional Air Quality Model. GeoHealth. doi:10.1029/2024gh001125
- Phung, V. L. H., et al. (2025). Effects of smoke haze on respiratory clinic visits in Central Kalimantan, Indonesia according to different haze characteristics. International Journal of Epidemiology. doi:10.1093/ije/dyaf169
- Phung, V. L. H., et al. (2022). A scoping review on the health effects of smoke haze from vegetation and peatland fires in Southeast Asia. PLoS ONE. doi:10.1371/journal.pone.0274433
- World Health Organization (2021). WHO global air quality guidelines. ncbi.nlm.nih.gov/books/NBK574594
- National Environment Agency Singapore. Computation of the Pollutant Standards Index (PSI). haze.gov.sg
- Ministry of Environment and Forestry, Indonesia (2020). Regulation No. 14 of 2020 on the Air Pollution Standard Index (ISPU). jdih.menlhk.go.id
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.