Pollution-Associated Risk Geospatial Analysis SITE (PARGASITE)

Map View

Upload dataset to get corresponding pollution estimates

The file returned will have a column for each pollutant; the value will be an average of monthly estimates over the specified time period.

Choose .csv file with Latitude and Longitude columns. A sample input file can be downloaded here.

Download

Greenblatt RE, Himes BE. Facilitating Inclusion of Geocoded Pollution Data into Health Studies. AMIA Jt Summits Transl Sci Proc. 2019;2019:553–561.(PMID: 31259010 ). GITHUB repository.

Map View

Use the download button to get pollution estimates corresponding to each MMSA and pollutant

The file returned will have three columns. The first and second column lists the GEOID and names of all the available MMSA and the third column returns the pollutant value corresponding to the MMSA. Change the input parameters (year and pollutant) to download the data for different years and pollutants.

Download

Map View

Use the download button to get pollution estimates corresponding to each county and pollutant

The file returned will have three columns. The first two columns lists the GOEID and names of all the available counties and the third column returns the pollutant value corresponding to the MMSA. Change the input parameters (year and pollutant) to download the data for different years and pollutants.

Download



Pollution-Associated Risk Geospatial Analysis SITE (PARGASITE) is an online web application and R package that can be used to estimate levels of pollutants in the U.S. for 1997 through 2021 at user-defined geographic locations and time ranges. Measures correspond to monthly and yearly raster files for PM2.5, Ozone, NO2, SO2, and CO covering the contiguous United States and Puerto Rico that were created with U.S. Environmental Protection Agency (EPA) regulatory monitor data. The R package allows users to obtain more customized output and work with the raster layers directly. The complete code for the Shiny app and R package can be accessed at GitHub.


Contributors

Coding and design by Nisha Narayanan, Avantika Diwadkar, Rebecca Greenblatt, Blanca Himes


Data

PARGASITE uses publicly available data from the U.S. Environmental Protection Agency (EPA). We have no affiliation with the EPA. From this data, we generated the monthly and yearly raster files (Jan 1997 to Dec 2021) for PM2.5, Ozone, NO2, SO2, and CO using inverse distance weighted interpolation from the five nearest EPA monitoring stations. Raster files were created separately for 1) the contiguous 48 states and 2) Puerto Rico. The raster layers were not cropped to fit the United States and instead are in the shape of a rectangle covering the corresponding region. When displayed on the app, they are cropped according to the map boundary.


Additional Information

The website and R package are similar, but some of their specific functions include:

1) The website provides visualization of the raster layers as well as the ability to upload a file with geocodes to then download one with PARGASITE estimates without interfacing with R.

2) The unit of measurement is displayed in the Shiny app according to the pollutant selected. To display units with the R package, use the getUnits() function.

3) The first time a raster layer is loaded in a user’s R environment by the R package, it will take some time to retrieve the data online. However, the raster layer will be stored in a temporary file that will improve subsequent loading times.

4) The latest updates to the app allows the user to visualize and download the pollutant data at the MMSA and County level.

5) The latest updates to the R package allows the user to work with the raster layers and download the pollutant values at MMSA, Counties, Census Tracts, and ZIPCODE level. Example use cases can be found here


Reference

Greenblatt RE, Himes BE. Facilitating Inclusion of Geocoded Pollution Data into Health Studies. AMIA Jt Summits Transl Sci Proc. 2019;2019:553–561. (PMID: 31259010).