Common
We don't use "aqi" or similar pseudo-units. Instead, we use only raw values - micrograms per cubic meter, µg/m³. This is the universal standard for such data. AQI, on the other hand, involves a multitude of standards, calculation methods, and so on. In itself, the term means absolutely nothing, as definitions vary from organization to organization.
Air quality symbols and advisories
We take the overall air quality from table from European Environment Agency:
| Pollutant | Index level (based on pollutant concentrations in µg/m3) | |||||
|---|---|---|---|---|---|---|
| Good | Fair | Moderate | Poor | Very poor | Extremely poor | |
| Particles less than 2.5 µm (PM2.5) | 0-5 | 6-15 | 16-50 | 51-90 | 91-140 | >140 |
| Particles less than 10 µm (PM10) | 0-15 | 16-45 | 46-120 | 121-195 | 196-270 | >270 |
| Ozone (O3) | 0-60 | 61-100 | 101-120 | 121-160 | 161-180 | >180 |
| Nitrogen dioxide (NO2) | 0-10 | 11-25 | 26-60 | 61-100 | 101-150 | >150 |
| Sulphur dioxide (SO2) | 0-20 | 21-40 | 41-125 | 126-190 | 191-275 | >275 |
The descriptions of pollutants and their effects on the body are based on documentation from the WHO and the EEA and represent a summary or quotation of that material.
Data sources
Our data sources are OpenAQ, Sensors.community, TUMO Labs, Air.gov.ge. We will also be adding new ones in the future. We do not use many data sources because, for aggregated sensor data, they do not provide raw values, but only the “AQI,” which means absolutely nothing without real data and an understanding of it; and we cannot honestly show the actual air quality, nor can we query each sensor individually, which makes little sense.
We exclude sensors from the results if their last update was more than 24 hours ago, as well as those where all sensors show zero values. Other than that, we do not perform any independent calculations or modify the data.
Calculating the Average Air Quality for a City
To calculate the current city-wide average, we take the median of all sensors within a 10-km radius of the city center or its district. In our overall assessment (such as “good,” “bad,” etc.), we display the worst value according to the EEA table. The distance is a fairly rough estimate, but it allows us to better average the overall value for the city as a whole.
An important limitation: For our assessment, we consider only the most recent sensor readings, not the average over time (by hour, day, or month). Historical data is important for assessing air quality for long-term living in a given location, but we have not yet implemented this feature.