Visualising Future Energy Scenarios
We've been working on a project with Northern Powergrid about Future Energy Scenarios. You can read about the inception of the project elsewhere on our blog. This post covers the visualisation and publishing side of the project.
In September and October we had meetings with Northern Powergrid to establish what they wanted from a visualisation. Northern Powergrid already had a generation availability map and they wanted a similar heat map that showed a variety of parameters across four scenarios for each of the next 30 years. These predictions had been created for each Primary Supply Point by Element Energy. Northern Powergrid also wanted a view aimed at Local Authorities.
The very first task was to create the geographies for the each Primary Supply Point. That was a big job but, once we had the GeoJSON files from this process, we could then proceed with the rest of the visualisation.
The visualisation is based, primarily, around National Grid's four scenarios. These split into two scenarios which meet the 2050 decarbonisation targets ("Two degrees" and "Community renewables") and two which don't ("Consumer evolution" and "Steady progression"). We took the decision to theme the interface with the colours from the National Grid document as you change the scenario. That wasn't something that was necessary but it helps to add some extra visual clues.
For each scenario you can pick one of the following parameters:
- electric vehicles;
- peak demand;
- peak demand with customer flexibility;
- peak utilisation;
- peak utilisation with customer flexibility;
- total consumption with customer flexibility;
- wind generation;
- heat pumps;
- domestic consumption (underlying);
- domestic photovoltaic (i.e. solar panels) generating capacity;
- industrial and commercial consumption (underlying);
- industrial and commercial photovoltaic (i.e. solar panels) generating capacity;
- other generation;
When you select a scenario/parameter combination the appropriate CSV file is loaded and the values shown on the map. The full set of CSV files powering the visualisation are released as open data in the Northern Powergrid Future Energy Scenarios dataset on Data Mill North.
We have three views of the data: Local Authority, Primary Supply Point, and a combination of the two.
The model data are all by Primary Supply Point so this view is the most definitive. However, in creating the geographies we were able to find the fraction (by customers) of each Primary Supply Point within each Local Authority. That way we were able to apportion the data from each Primary Supply Point to Local Authority Districts. For some parameters we can simply add up the contributions from each Primary Supply Point (e.g. electric vehicles) but for others (e.g. peak demand) we find the maximum value within the Local Authority and use that.
As you can see, the different views of the same data can look different and that is largely because of differences in population density. In the Primary Supply Point view, the "hot spots" for electric vehicles (2019) are the southern-most borders of Northern Powergrid's catchment area. However, in the Local Authority view, the hot spot is Leeds. That is because it has a larger number of Primary Supply Points which then add up to a larger number. Both views offer different insights to the data.
When you are in the Local Authority view you can click on a Local Authority and get a breakdown of the Primaries that are in it (fully or not). The barchart makes use of a library we made for previous projects and we use stacked bars to indicate how much of each Primary Supply Point (as a fraction of customers) is in the selected Local Authority.
Publishing in the open
Northern Powergrid are, by necessity, quite a conservative organisation and this collaborative project has been a new approach for them. We've used it to experiment with ways of working and tweak our own workflow.
As is our standard practice at ODI Leeds, we created a repository on Github for the visualisation's code and data. This has a few benefits:
- It provides transparency as others can see how our code works and they can help to improve it;
- It acts as version control and a distributed backup system so people can see the history of changes;
- We get a mechanism that allows chronological discussion about specific issues with the code and/or the data e.g. to get feedback from Local Authorities.
These tools and methods were new for Northern Powergrid so we had a working session where we showed them the basics of the Github web interface so that they could create files, edit files, and use the issue system. They took to it very quickly and we've been able to move discussions that had been spread across our email inboxes into specific issue threads.
The data are stored in the repository so that they are easily accessible by the visualisation but also to give us version control. Northern Powergrid have also created a dataset on Data Mill North which lists all the CSV/XLS files behind the DFES release. Data Mill North is quite agnostic about where files are stored so, where the files already exist on Github, Northern Powergrid have simply added the Github URLs to Data Mill North. That way we avoid having copies of the files in two different places; one just points to the other. That will make is easier to push any updates in future.