Everything you need to know about navigating and using our Data Library.
Anyone has access to the free public data. However, some data sets require an account.
Public data can be viewed by anyone without the need for an account. However, public data is more limited compared to private data. Private data is more detailed and complex and requires an account.
To create an account, you can click here to redirect to a form to request access. You cannot freely create an account. After you submit a request for an account, and the request has been approved, an account will be created for you.
Our goal is to provide data from any level of complexity to all ranges of users; from students to scientists. Some example use cases are to see how much pollution is in a specific area, or how much UV exposure is in a given area.
We collect our data from multiple reliable sources. We then process the data and make it available for viewing and downloading.
All datasets are currently scoped to the Continental United States (CONUS), using 2020 TIGER/Line census tract boundaries as the standard geographic unit. Hawaii, Alaska, and U.S. territories are not currently supported.
Linked datasets are delivered as CSV or Apache Parquet files, depending on the size and structure of the request. All deliveries include a data dictionary, match rate report, and a set of QA flag columns documenting data quality per variable.
Match rate is the proportion of participants successfully linked to a given exposure dataset for each year. A match rate below 95% is flagged in the QA report, with a breakdown of unmatched records and likely causes — such as address quality, geographic edge cases, or dataset temporal gaps.
Point extraction samples the raster value at the geocoded participant coordinate (their residence). Tract-level extraction returns the mean of all raster pixels intersecting the participant's census tract. Point-level is appropriate for individual-level exposure research; tract-level is used for ecological analyses.
Yes. GREENER supports multi-address longitudinal linkage. Provide a time-stamped address history and each address window will be linked to the corresponding exposure data for that period. Match rates and QA flags are generated per address window.
Coverage varies by dataset. High-resolution PM2.5 (GEEALab) is available through 2024. Harvard air quality datasets cover 2000–2016. NDVI (Landsat) runs through 2023. Social datasets such as SVI and COI are updated annually. See the Data Library for exact coverage per dataset.
Public datasets are available to any registered user. Restricted datasets — including PM2.5 speciation — require PI approval and a brief data use justification. All requests are reviewed by the GREENER team prior to data delivery.
No questions matched your search.