Environment Canada Weather History

Command line project for parsing hourly historical weather data from Environment Canada given an appropriate stationID. Data can be returned via a CSV or inserted into a MySQL database.

If you only need a month of historical weather, consider using Environment Canada's default ?format=csv option. However, in addition to its SQL capabilities, this script can stitch all observations in a date range together into a single CSV file.

Currently this project is dependent on the MySQL connector being installed in the Python environment, but my goal is to quickly decouple that dependency.

Dependencies

Download and install the MySQL Connector.

Also depends on the pytz module, installed using:

pip install pytz

Setup

MySQL Destination

1. For MySQL destination, run the ./sql/create.sql statements on your target schema.
2. Copy config-example.py to config.py and fill out the appropriate connection details.

CSV Destination

None

Usage

Run the following command for usage details:

python import_history.py --help

The --help documentation gives a brief summary of all required and optional parameters. An example for importing hourly, historical weather information from November 1, 2010 through January 31, 2011 from the Kitchener/Waterloo weather station:

python import_history.py --station_id 48569 --year_start 2010 --month_start 11 --year_end 2011 --month_end 1 --tz_name America/Toronto --dest csv

On notable parameter is the integer passed to --station_id. This corresponds to an Environment Canada weather station. Use form on the "Search by Proximity" tab on Environment Canada's Advanced Search page to find the nearest weather station to your city of interest.

1. Pull up an appropriate city from the "Select City" drop-down and hit "Search".
2. On the listing of nearby weather stations, click the "Go" button of the weather station of your choice that has "Hourly" observations.
3. In the resulting page, grab the number in the &StationID= URL parameter.