The cheapest reliable PM2.5 sensor capable of contributing to a public network costs about as much as a long weekend in Seminyak. The data it produces — once it joins one of the international platforms — is auditable, timestamped, and impossible to politely walk back. It becomes part of the historical record this site already aggregates.
Bali's monitoring blind spots are not the fault of any one platform. They are the cumulative absence of small decisions to install. North Bali (Lovina, Singaraja), the eastern regencies (Karangasem, Amlapura), and the inland mountain belt (Bedugul, Kintamani) currently appear in our data as silence. A single sensor in any of those places would change that.
What follows is a short comparison of the four platforms most practical for Bali, then a generic install checklist that applies to all four.
Every option below publishes data to a public network this site already reads. Whichever you choose, your readings will appear on the Live Monitor map within one refresh cycle of going online.
A side-by-side of price, refresh cadence, public-data behaviour, and shipping reality for Bali.
| Sensor | Approx. price | Public data | Refresh | Indoor / outdoor | Ships to Bali |
|---|---|---|---|---|---|
Nafas |
Subscription | Default · public | ~15 min | Outdoor | Indonesian company |
PurpleAir Classic |
USD 269 | Default · public | 80 sec | Outdoor | From US · 10–14 days |
IQAir AirVisual Pro |
USD 269 | Opt-in | 10 min | Indoor (Pro) · Outdoor (AVO) | From CH · ~14 days |
AirGradient ONE |
USD 165–249 | Default · public | 10 min | Outdoor | From Thailand · 5–8 days |
The mechanical install is the easy part. Picking the right spot is what determines whether the data is meaningful.
Three to six metres above ground, away from a kitchen exhaust, away from a chimney, away from a busy parking spot. Direct exposure to outdoor air, but sheltered from horizontal rain. A balcony rail or a north-facing eave is usually right; the inside of a kitchen window is wrong.
All four options need 5V USB. Run the cable through a window or a wall pass-through and seal it with silicon. The unit itself mounts with a single screw or a zip-tie; nothing is heavy.
Each platform has its own setup app. PurpleAir and AirGradient use simple captive-portal flows; IQAir and Nafas use first-party apps. All four require 2.4 GHz; some 5 GHz mesh routers will refuse the pairing — temporarily disable 5 GHz or use a guest 2.4 GHz network during setup.
Give the sensor a descriptive name ("Ubud — Jl Raya Goa Gajah") rather than the default hex string. Set its location accurately. Mark it as public when prompted; this is the toggle that lets us — and everyone else — see your data.
The first reading appears on the platform's own map immediately. Within one refresh cycle of this site (5 minutes), it will appear here too. We do not curate or approve; if your sensor is on the public network, it is on this map.
Your sensor's hourly and daily readings become part of the public record. They are downloadable from the source platform as raw CSVs, queryable through this site's archive, and citable in any future report on Bali's air. We do not own the data; you do.
If your sensor goes offline for more than two weeks, it will appear in the Archive of Silence on the homepage rather than disappearing without record. The historical readings remain.