A one-page website to answer a very simple question: Did Germany have cleaner electricity than France today?
This code downloads hourly generation data from ENTSO-E and calculates the two countries's respective carbon intensities. The website shows whether Germany was cleaner today (or yesterday, if there is no data for today). It also shows how many hours so far during the current year Germany has had cleaner electricity than France. Year-to-date data is also displayed graphically in a scatter plot.
The Energiewende has been a massive failure. It has not brought cheap power, it has not heralded the demise of coal, and it has not brought down emissions to anything that can be called clean levels. The latter is the most obvious of all, with the internet abounding with examples of the German grid being as much as 10 times dirtier than the French grid. This inevitably brings out accusations of cherry-picking a bad day for Germany. I wrote this program to show that the data consist of nothing but cherries, and the German grid has nothing but bad days. It provides an up-to-date, easy-to-reference counter to such accusations. Simply ask: Is Germany cleaner than France today? Has Germany had a single hour of cleaner energy than France this year?
Every hour, using a systemd timer the program downloads generation data from ENTSO-E.
This data is processed and stored under $XDG_DATA_HOME, which by default is ~./local/share/.
The program then prepares a "report" with the carbon intensities, number of cleaner hours for Germany, etc.
This report is stored as JSON in the same directory.
As part of the report, a scatter plot is also generated and stored.
The results are presented using a simple one-page Flask app, served with uwsgi.
Most functionality is accessible through the command line: try isgermanyclean --help.
Contribute via Github. Design improvements are appreciated.
You can deploy this program on your own server.
The pyproject.toml and requirements.txt should tell you everything about dependencies, except that you will also need a LaTeX installation that matplotlib can find, and uwsgi.
The program expects to find an API key for ENTSO-E in $XDG_CONFIG_HOME/isgermanyclean/APIKEY.
It should be a JSON object with one key, api_key.
You will need to install and enable the systemd units.
Finally, because the systemd timer dowwloads one day's worth of data at most, you will need to manually seed the installation with year-to-date data.