Inquiry Regarding Use of PM2.5-ATM Data for Research and Policy

1

I hope this message finds you well.

I am currently working on a research project involving PM2.5 data, and I have downloaded raw PM2.5-ATM measurements from the PurpleAir website. I would like to clarify a few points to ensure proper use of the data: whether it is acceptable to use this raw data directly—without applying any correction factors or adjustments—for research purposes.

Could you please clarify if the raw PM2.5-ATM data is appropriate for use in scientific analysis, or if the application of a specific correction (e.g., EPA or LRAPA) is recommended or required for research validity?

Thank you in advance for your time and assistance.

1 Like
2

Your post got truncated, can you edit it and add the missing questions?

1 Like
3

thank you
I hope this message finds you well.

I am currently working on a research project that involves air quality data, and I am using PM2.5-ATM measurements obtained from PurpleAir sensors. I would like to confirm whether it is acceptable to use this raw data directly—without applying any correction factors or adjustments—for research purposes.

Could you please clarify if the raw PM2.5-ATM data is appropriate for use in scientific analysis, or if the application of a specific correction (e.g., EPA or LRAPA) is recommended or required for research validity?

Thank you in advance for your time and assistance.

1 Like
4

Both CF_1 and CF_ATM have been found to overestimate P{M2.5 concentrations by 50-100%. A better estimate is provided by pm2.5 alt. This algorithm was used by two of the largest (thousands of monitors) studies by Wallace 2023 in Indoor Air and Lundenberg 2024 in PNAS.

1 Like
5

For anyone wishing to learn more about how to access PurpleAir data using Lance’s formula via the PurpleAir API, you can find information here: What is the Difference Between CF=1, ATM, and ALT?

In particular, I recommend reviewing the section Retrieving Alt-CF3.4 via the API.