In order to improve the accuracy and reliability of the data reported from lower-cost sensors, IQAir calibrates and validates these readings, using the following methods:

Data calibration: correction for environmental conditions

Laser/optical sensors measure PM2.5 by quantifying the amount of light reflected from a laser beam. Since the light scatter observed from a particle is dependent on parameters such as particle size, shape and density, environmental factors that influence these parameters can contribute to measurement error. High humidity is one such environmental factor that makes particles appear larger and denser than the particulate matter would otherwise be. As a result, measurements taken by laser-based sensors in a highly humid environment may overestimate PM2.5 concentration.

In order to adjust for hyperlocal variations of air, IQAir uses Beta Attenuation Monitors (BAM), such as those widely used by governments, as a local reference for conditions. By calibrating lower cost sensors to their nearest beta attenuation monitors or to beta attenuation monitors in similar environments, it is possible to discover which factors (ex: humidity) may contribute to measurement error, to what degree and when, and adjust accordingly. For example, by using a calibration formula to make values from a lower cost monitor in a specific location more similar to a Beta Attenuation Monitors in the same location, it is possible to find the optimal average correction necessary for a specific environment.

Data validation: identify and prevent sensor anomalies from being published

All sensors, including governmental beta attenuation monitors, can sometimes report anomalies or inaccurate data. Reasons for this may include temporary periods of maintenance or defects, or temporary hyperlocal emissions near the sensor.

IQAir’s cloud-based data validation system immediately identifies potential anomalies published by a station, and cross-checks these against other nearby measurements in order to verify whether the spike is representative, or an anomaly. In the case the data appears to be an anomaly (such as a sudden spike in PM2.5 from 10 ug/m3 to 100 ug/m3, which is not matched by nearby stations) the value will be removed from the data set.

In addition to the data calibration and validation corrections provided to lower-cost sensor readings, it is important to further note that the data format of the IQAir platform differs from the PurpleAir platform. While PurpleAir publishes online data in a 10-minute increment standard, the IQAir platform publishes online data an hour-long increment standard. Here, increments refer to the time frame for which real-time measurements are averaged.

The PurpleAir platform also offers a “1-hour average” provided in its station popup widget. This hourly average is updated in 10-minute increments to reflect the past hour exactly. As a result, this may also differ from IQAir’s standard which always reports data in hourly increments at the start of an hour.

Why does IQAir use hour-long increments to report pollution data?

In order to match the standard set by many air quality experts around the world, IQAir employs the hour-long increment standard to lower-cost sensors. By following this standard, there is greater relevance in comparing data to original data sources as well as official government stations.

Did this answer your question?