Overview
The Air Pollution EXposure (APEX) tool is a web-based application designed to estimate long-term air pollution exposure from a participant’s residential history. It combines satellite-derived air quality data with geospatial modeling to produce annual exposure scores for fine particulate matter (PM2.5) and nitrogen dioxide (NO2).
For PM2.5, APEX uses annual estimates from 1998–2023 generated through a widely used public dataset that integrates satellite-based Aerosol Optical Depth (AOD) retrievals with simulations from the GEOS-Chem chemical transport model. These values are subsequently calibrated to ground-based monitor data using a Geographically Weighted Regression (GWR) approach. APEX currently incorporates version V5.GL.06 at 0.01×0.01° resolution from the Atmospheric Composition Analysis Group.1,2
NO2 exposure is estimated using satellite observations from the OMI and TROPOMI instruments, following methods described in Cooper et al.3. APEX currently uses the publicly available version v1 dataset.4 Annual ground-level NO2 scores are available for 2005–2019.
To produce an exposure estimate for any given residence, APEX averages all available pollutant data within an 11 km radius of the submitted address. Users can enter residential histories for individual participants or upload batch spreadsheets (.xls, .xlsx, .csv) to obtain PM2.5 and NO2 exposure values across multiple years.
Additional methodological details will be provided in the forthcoming manuscript currently under review at JTO Clinical and Research Reports.5
1. Using the Google API
This application requires a Google Maps API key for full functionality. You can obtain one by following the instructions here. The single patient page leverages your Google Maps API key to enable address autocompletion, support the interactive map for selecting locations, and perform geocoding required for the exposure calculation. Please note that all API requests will count against your Google usage quota. Google provides 10,000 free geocode requests and 10,000 address autocompletes each month.
If you prefer to avoid using your API quota, you may pre-geocode addresses independently and use the Batch Upload option instead.
How to Create a Google API Key
- If you do not already have a Google account, please create one at accounts.google.com.
- Visit the Google Cloud Console.
- Create or select an existing project.
- Go to APIs & Services > Credentials.
- Click Create Credentials > API Key.
- Enable these APIs under the Library:
- Maps JavaScript API
- Geocoding API
- Places API (optional for autocomplete)
- Restrict your API key (recommended):
- Set Application restrictions to
HTTP referrersorIP addresses. - Set API restrictions to the APIs above.
- If concerned about API usage, set up budget alerts in your Google Cloud Console.
2. Batch Upload Guidelines
To process multiple participants at once, create a spreadsheet in a supported format (.xls, .xlsx, .csv) according to the provided template below. Please note that addresses for a single participant are grouped together by ID and each row is a unique address.
If you wish to avoid using your API quota, use the Batch Upload form and include both latitude and longitude for each address.
If any row contains only a street address without coordinates, it will be geocoded using your provided API key, which utilizes your API quota.
Data Formatting Options
Upload files must follow one of the two formats below:
Format 1: Split Address Template
| id | start_month | start_year | end_month | end_year | street_address | city | province | postal_code | country | latitude | longitude |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 1 | 1990 | 12 | 2010 | 675 W 10th Ave | Vancouver | BC | V5Z 0B4 | Canada | ||
| 1 | 2011 | 2025 | Rue de Rivoli | Paris | France | ||||||
| 2 | 1 | 1975 | 6 | 2025 | London | UK | 51.5074 | -0.1278 | |||
| 3 | 1980 | 2024 | Shibuya Crossing | Tokyo | Japan | 35.6595 | 139.7006 |
Format 2: Combined Address Template
| id | start_month | start_year | end_month | end_year | street_address | latitude | longitude |
|---|---|---|---|---|---|---|---|
| 1 | 1 | 1990 | 12 | 2010 | 675 W 10th Ave, Vancouver, BC V5Z 0B4, Canada | ||
| 2 | 2011 | 2025 | Rue de Rivoli, Paris, France | ||||
| 2 | 1 | 1975 | 6 | 2025 | London, UK | 51.5074 | -0.1278 |
| 3 | 1980 | 2024 | Shibuya Crossing, Tokyo, Japan | 35.6595 | 139.7006 |
start_month and end_month columns blank if these
values are unknown. If left blank, start_month will default to
January and end_month will default to December.
Batch Upload Processing Time
The processing time varies with file size. Submitting 100 rows may take approximately 10 minutes.
Disclaimer About Calculation
References
-
Aaron van Donkelaar, Melanie S. Hammer, Liam Bindle, Michael Brauer, Jeffery R. Brook, Michael J. Garay, N. Christina Hsu, Olga V. Kalashnikova, Ralph A. Kahn, Colin Lee, Robert C. Levy, Alexei Lyapustin, Andrew M. Sayer and Randall V. Martin (2021). Monthly Global Estimates of Fine Particulate Matter and Their Uncertainty Environmental Science & Technology, 2021, doi:10.1021/acs.est.1c05309
-
Hammer, M. S., van Donkelaar, A., Bindle, L., Sayer, A. M., Lee, J., Hsu, N. C., Levy, R.C., Sawyer, V., Garay, M. J., Kalashnikova, O. V., Kahn, R. A., Lyapustin, A., and Martin, R. V.: Assessment of the impact of discontinuity in satellite instruments and retrievals on global PM2.5 estimates. Remote Sensing of Environment, Volume 294, 2023, 113624, ISSN 0034-4257, doi:10.1016/j.rse.2023.113624
-
Cooper, Matthew J., et al. "Global fine-scale changes in ambient NO2 during COVID-19 lockdowns." Nature 601.7893 (2022): 380-387. https://doi.org/10.1038/s41586-021-04229-0
-
Cooper, M. (2022). Satellite-derived ground level NO2 concentrations, 2005-2019 (Version v1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.5424752
-
Durney, C.H., Tawara, A., Brauer, M., Atkar-Khattra, S., Myers, R., Meza, R., Lam, S. APEX: A Web-Based Tool for Estimating Long-Term PM2.5 Exposure from Residential History - Brief Report (currently under review at JTO Clinical and Research Reports)