Bali's projected 2025 population is 4.46 million. The number of KLHK reference-grade PM2.5 monitors serving it on the national ISPU portal is one — and it has not reported in months.
Bali's projected 2025 population, according to BPS Statistics Indonesia, is 4.46 million. The number of KLHK reference-grade PM2.5 monitors serving that population, indexed on the national ISPU portal, is one. That sensor, at Denpasar Lumintang on Jalan Mulawarman, reported sporadically through 2025 and, as of the first quarter of 2026, has not appeared on the portal in months.
Jakarta, for comparison, operates 10 reference-grade sensors and some 110 low-cost community monitors through its DKI programme, according to the 2024 Indonesia Air Quality report from the Centre for Research on Energy and Clean Air. Hong Kong runs 18 fixed stations for 7.5 million residents; London more than 100 continuous sites on its Local Air Quality Network. Bangkok's PCD runs 70 standard stations.
On a per-capita basis — government reference-grade sensors per million residents — Bali runs at roughly 0.2. Jakarta at 11. London at 11. Hong Kong at 2.4. Bangkok at about 4.5.
The CREA report, in its summary of provincial coverage outside of Jakarta, described "a significant coverage gap beyond the city limits into other satellite cities." Bali is one of those gaps. The Denpasar Lumintang station is not, let it be said, a failure of engineering. It is operated by the Environmental Laboratory Unit of the city's Environmental Health and Cleanliness Service under the national ISPUNet programme; when it runs, it publishes hourly AQI and 24-hour PM2.5 readings in the international format.
There is a temptation to treat community sensors (AQICN, Nafas, AirGradient, PurpleAir) as a substitute for government ones. The difficulty with the substitution is not that the private data is bad. It is that public record requires public accountability, and public accountability requires instruments whose calibration, maintenance, and placement are the obligation of an official body, not the citizens it represents.
| City / province | Population | Gov. reference sensors | Per million |
|---|---|---|---|
| Jakarta DKI | ~10.6 m | 10 reference + 110 low-cost | 11.0 |
| London | ~9.0 m | 100+ LAQN continuous | 11.1 |
| Bangkok | ~15.5 m metro | 70 PCD standard | 4.5 |
| Hong Kong | ~7.5 m | 18 EPD fixed | 2.4 |
| Bali | 4.46 m (BPS 2025) | 1 KLHK ISPU — not reporting | 0.2 |
Reference-grade government instruments only. Community and private sensors (AQICN, Nafas, AirGradient, PurpleAir) are not included. Sources in the footer of this dispatch.
"A significant coverage gap beyond the city limits into other satellite cities."Centre for Research on Energy and Clean Air · Indonesia Air Quality 2024
When the Denpasar Lumintang instrument is running, it speaks. When it is not, the island's official record is silent — not because there are no pollutants to measure, but because there is no government instrument to measure them with. The community data, as Dispatch I set out, is the best long-running public record in the absence of the government one. It is an imperfect stand-in.
The proposal — it is not a new one — is modest. A handful of reference-grade monitors at Denpasar, Ubud, Canggu, Seminyak, Nusa Dua and either Singaraja or Amed. A public dashboard on the ISPU portal, updated hourly, with a clean CSV backlog for researchers. An annual uptime report, on the London LAQN model. A budget line in the provincial Dinas Lingkungan Hidup that protects these instruments from the routine budgetary losses that have taken down the Denpasar unit.
The next dispatch, on what has happened since the first of April, is an argument for why that choice matters now and not later.