Field Data · First-Party Telemetry

One filter change. PM2.5 down 56%.
Measured, not claimed.

97 days of continuous air-quality telemetry from an occupied Dubai residence: 20,640 sensor readings, one routine HVAC filter replacement, and what the data actually shows - method, numbers and limitations included.

See the method WOLKIS IAQ product

TL;DR

Two-panel chart: daily-mean PM2.5 in two zones of a Dubai residence dropping 56% after a filter replacement on 16 July, with a flat CO2 control channel below
Daily-mean PM2.5 in both zones, July 2026. The red line marks the bedroom filter replacement. Bottom panel: CO₂ control channel - flat through the intervention.

1. Method

Instrumentation. Two commercial LoRaWAN indoor air-quality monitors (NDIR CO₂, laser particle counter) in two rooms of an occupied residence in Dubai. Each device reports roughly every 10 minutes - about 140 readings per day. Readings are decoded server-side and appended to a database where every row carries a hash chained to the previous row, so the record is tamper-evident after the fact.

Intervention. A routine HVAC (fan-coil) filter replacement. The bedroom filter was replaced around 16 July; the living-room filter had been replaced roughly a week earlier. Nobody logged the exact hour - which is precisely the realistic condition: the changepoint was recovered from the data, not from a diary.

Attribution controls. Three, stated in advance of the numbers:

2. Results

QuantityWindowValue
Bedroom PM2.5, daily mean - before1-15 July (14 days)25.5 µg/m³
Bedroom PM2.5, daily mean - after17-31 July (15 days)11.3 µg/m³ (−56%)
Bedroom PM2.5, daily mean - sustained17 July - 10 August9.9 µg/m³
Living room PM2.5, daily mean - same period17 July - 10 August2.9 µg/m³
CO₂ rolling 8-h windows above 800 ppm (GU119 v4)full period, both rooms0 of 3,481 (max 577 ppm)
CO₂ hourly mean / maximumfull period, both rooms443-459 ppm / 650 ppm
PM2.5 24-h days above 35 µg/m³ (GU119 v4)full period, 158 sensor-days7 days - see note
Exceedance note - reported, not trimmed

Four bedroom exceedance days cluster around May regional dust events (36-44 µg/m³) plus one pre-filter July day at the limit. Three living-room days are short, extreme spikes (daily means up to 916 µg/m³) consistent with an activity right at the sensor - housekeeping or works - rather than ambient air. A compliance record that never shows an exceedance is a record you should distrust.

3. What this does and does not show

It shows the measurement layer working: detect a change → attribute it against controls → quantify it → keep a tamper-evident record. That chain - not the filter - is the product being demonstrated.

It does not show that filters halve PM2.5 in general (one building, one intervention), and it does not replace laboratory calibration: these are commercial-grade sensors compared against themselves over time, which is exactly the comparison step-change analysis needs, and exactly not a reference measurement.

And the finding that pays for the monitoring: the same filter model is installed in both rooms, yet the bedroom still runs about three times dustier at a slightly lower temperature (comparable fan-coil duty). The remaining particle source is indoors - textiles are the leading candidate - so the next improvement is source control in that room, not more filtration. Without continuous data, that conclusion is unreachable; with it, it is a fifteen-minute read of two charts.

4. Why we publish field data

Compliance platforms usually publish claims; audits and regulators increasingly want evidence. This page is the first in a series where we publish our own first-party telemetry with the method and the limitations attached - the same evidence format our reports produce for clients under Dubai Municipality guidelines (GU119 v4, GU141/EIAQI). Locations are anonymised to the city level; numbers are real; the record behind them is hash-chained.

FAQ

Questions this dataset raises

What exactly was measured and how often?

Two commercial LoRaWAN IAQ monitors (NDIR CO₂, laser particle counter) in two rooms of an occupied Dubai residence, reporting roughly every 10 minutes - about 140 readings per device per day. Between 5 May and 10 August 2026 the system decoded 20,640 readings covering PM2.5, PM10, CO₂, TVOC, temperature and relative humidity.

How do you know the drop came from the filter and not weather or ventilation?

Three checks. CO₂ stayed flat through the change date, so occupancy and fresh-air exchange did not change. The change date was found by a changepoint detector on the between-room difference (t = 7.4), which subtracts outdoor dust events hitting both rooms. And the second room, already on a fresh filter, showed no comparable step on 16 July.

What are the legal CO₂ and PM2.5 limits in Dubai?

For existing buildings, DM-HSD-GU119 v4 (Table 2) sets CO₂ at 800 ppm over 8 hours and PM2.5 at 35 µg/m³ over 24 hours. In this dataset none of the 3,481 rolling 8-hour CO₂ windows exceeded 800 ppm - the highest was 577 ppm.

Were there any PM2.5 exceedances?

Yes - 7 of 158 sensor-days exceeded the 24-hour limit: four bedroom days clustered around May dust events plus one pre-filter July day, and three living-room days that are short extreme spikes consistent with activity right at the sensor. We report them rather than trim them.

Are these laboratory-calibrated reference instruments?

No. These are commercial-grade monitors, not ISO 17025 reference instruments. The dataset demonstrates the monitoring capability - detect, attribute, quantify, record - not laboratory calibration. Step changes and window statistics compare a sensor with itself over time, which is robust to absolute sensor error.

Does one residence prove filters work in general?

No, and the page does not claim that. It is one documented intervention, published with its method. What it shows is the value of measurement: same filter model in both rooms, yet one room runs ~3× dustier - so the next dirham belongs in source control, not more filtration. Without data, nobody knows that.

Can WOLKIS produce this evidence for a commercial building?

Yes - that is the product. The same pipeline (LoRaWAN sensors, hash-chained storage, changepoint and compliance-window analytics) runs for offices, schools, hotels and restaurants in the UAE, mapped to the applicable Dubai Municipality guidelines. See Building IAQ for scope and published pricing.

Want a baseline like this for your building?

We install the sensors, keep the tamper-evident record, and turn it into compliance evidence under GU119 v4 and EIAQI - with numbers you can defend, not claims you have to trust.

Ask about a monitored baseline See IAQ product page

Further reading

Want numbers like these from your own building?

Tell us about the property and we come back with what a monitored baseline would look like for it.

Prefer WhatsApp? +971 56 878 0888. No obligation — we tell you where you stand first.