A: The default setting on the PurpleAir map is the US EPA PM2.5 AQI calculated from the raw PM2.5 concentration reported by PurpleAir sensors. You have the option to apply different conversion factors, such as LRAPA or AQandU, to PM2.5 data on the PurpleAir map.
Purple Air uses a laser particle counter to count particulate matter (PM) suspended in the air. This counting process can detect particle sizes from 0.3 μm up to 10 μm. To convert the particle count to a mass concentration (μg/m3), laser particle counters must assume an average particle density. They must use an average because not all PM of a particular size consists of the same stuff. For instance, PM2.5 from wildfire smoke will have a different density than PM2.5 from dust blowing off a gravel pit. This means the mass concentration reported by a PurpleAir sensor varies depending on the specific composition of PM for a given area. The AQ&U conversion factor was developed by the University of Utah to convert the mass concentration data measured by PurpleAir sensors to best fit the PM composition of wintertime air in Salt Lake City.
Salt Lake City is a large urban area with a variety of industrial operations within a mountain valley. Individuals living in comparable environments could consider using the AQ&U conversion factor. Those living in a climate similar to that of Lane County, Oregon, in the Pacific Northwest, could consider using the LRAPA conversion factor. As more research becomes available for different areas, PurpleAir will post additional conversion factors.
Thanks for this article on conversion factors. While the conversion factors are available to use—why isn’t one of them on by default?
Currently, the default reported data is the raw PM2.5 data of PurpleAir sensors, and this raw data feeds directly, without conversion, into other services like IQAir. Based on my use of the PurpleAir Real-time Map applying conversions, it seems that applying the US EPA, AQ&U or LRAPA conversion factors roughly brings in line the EPA-run AirNow and PurpleAir maps, the former of which is focused on normalizing to EPA air quality recommendations. Critically, there seems to be a significant effort by EPA scientists to investigate the accuracy of these sensors’ raw measurements, due the the building public interest in air quality, and the use of these sensors by laypeople. These efforts are described here:
These resources, one of which is peer reviewed, 1) call into question the accuracy of PurpleAir sensors for wild fire smoke, and demonstrate how accuracy varies based on particle density, composition, humidity, etc… and 2) describe how the non-linear US EPA conversion factor addresses these shortcomings and makes those measurements more reliable for public understanding.
Given that PurpleAir sensors feed into IQAir, and other widely used webpages/apps, and ostensibly want to best represent air quality for public understanding and health related decisions (i.e. when to go out, when to stay in), why isn’t the default setting to use one of these conversion factors? Is there another critical element to the conversion aside from what has been described in those articles?
Thanks for you response and the thoughtful answers/articles.
Hi there. The default conversion setting on my map is “no.” I have my sensor under the eaves of my house in a neighborhood in the foothills of the Western Washington Central Cascades. Should I change it to the EPA conversion?
My data layer is set at US EPA PM2.5 AQI. If my data conversion is set at none, I have an AQI of about 2. If I set the conversion to US EPA, which seems the same as my data layer, it jumps to 16. Why such a difference? Thank you.
The US EPA calculation is a weighted average over the last several hours, with higher weight for more recent observations. If the local PM2.5 measurement is declining, then the US EPA conversion will show a higher value than no conversion.