This dataset is the estimated long-term (1981-2016) concentrations of ambient fine particulate matter across North America, which combines information from chemical transport modeling, satellite remote sensing, and ground-based monitoring. The estimates included information from updated historical emissions inventories and meteorological data, fine resolution satellite-based estimates of PM2.5, and ground-based measurements of PM2.5, PM10, and total suspended particles (TSP) measurements.

Please contact Jun Meng ( for the dataset usage and further collaborations.


Meng, J., Li, C., Martin, R.V., van Donkelaar, A., Hystad, P., Brauer, M., 2019. Estimated Long-term (1981-2016) Concentrations of Ambient Fine Particulate Matter across North America from Chemical Transport Modeling, Satellite Remote Sensing, and Ground-based Measurements. Environ. Sci. Technol. [Link]

Dataset has been deposited in the Zenodo Digital Repository (DOI: 10.5281/zenodo.2616769). The dataset is annual average PM2.5 estimate data in HDF5 format. Considering the data storage convenience, the PM25 variable is 100 times the value we should use. So you need to read them and divided them by 100 in your code. Monthly data for 1989-2016 are available on request from Jun Meng.

We also have a version of data in NETCDF format. You can also get access to these data from WUSTL BOX [Link]. This version has the real estimated PM2.5 concentrations, so you do not have to divide them by 100 in your data processing code.

Please contact Jun Meng ( if you have more questions.