aqi-process

This is a rudimentary script that pulls the latest AQI data from AirNow.gov.

This script expects that you're in a Python environment (we use Anaconda) and have csvkit installed.

git clone git@github.com:nprapps/aqi-process.git
cd aqi-process
conda create --name aqi-process python=3.9
conda install -c conda-forge csvkit

To run it:

conda activate aqi-process
bash aqi.sh

Data notes

The AQI datafile is a pipe-delimited .dat textfile, updated hourly. Documentation lives here.

Note: Forecast data beyond the current day does not exist for all reporting areas.

Additional data and documentation (API account may be required to access)

What the script does

1. Download the datafile from the AQI site
2. Add column headers
3. Convert to a CSV and trim out extra columns / rows
4. Query results just for a subset of cities
5. Export the filtered set to JSON and copy it over to a dgnext graphic aqi-cities-20230608 in the separate graphics-js repo
   • (Assumption: you work at NPR and already have this repo on your machine.)

Further development

TODO: Perhaps borrowing from the covid data pipeline, upload this to a server and set it on a cron, uploading the result to a folder on apps.npr.org.