The work consists of a map page that displays the change in average temperatures between the 1960s and the 2010s in circa 100.000 European municipalities. After selecting a municipality from the map, the reader receives information on the selected municipality and a contextualization of the data.
This work has two aims: first, to inform about climate change at the local level; second, to encourage local journalists to write stories about their places.
The project has had a great impact so far. Shortly after its publication the map became the most viewed content in the European Data Journalism Network’s website. Moreover, the data has been reused by almost 90 media outlets across Europe, whether they be national or local newspapers.
On top of that, the project has been used for advocacy purposes by a number of civil society organizations calling for stronger initiatives against climate change across Europe, such asStopGlobalWarming, an Initiative of European Citizens that promotes the approval of a carbon tax at the continental level.
The analysis has been carried out using Rstats, and especially its geocomputation libraries. After downloading the dataset from the Copernicus/ECMWF website, which consisted in multiple daily estimates of temperature values for more than a million tiles of 5.5×5.5 km2, we calculated the average temperature variation of every tile in the last half century. In a second step, we matched the values of the tiles with the respective local administrative units’ shapefiles, as well as with NUTS2 and NUTS 3 shapefiles.
After creating the dataset, we worked on the map itself, trying to effectively visualize the data.
From a technical point of view we relied on d3.js and Mapbox to design a granular map that shows 60 years of temperature trends in each country of Europe. From a narrative point of view, we relied instead on a scrollytelling approach to enable readers to understand the data and the temperature trends, comparing a given municipality with others in the same region.
What was the hardest part of this project?
When it comes to the data analysis, the hardest part was matching the municipalities with the dataset tiles. The extension of encompassed municipalities differed largely, sometimes by orders of magnitude.
To obtain a single value for each municipality, we associated the center coordinates of every municipality to the closest grid cell of the available data (having thus a common method for any city). Yet, to calculate the city center, instead of using the simple geometric centroid, we decided to weight the center based on the population grid. This was made in order to have the center of the municipality’s polygons closer to the urban areas. Indeed, when thinking about a city, the average person depicts in its mind the main urban area, generally the most densely populated.
For the Map page development, instead, we had to manage 100.000+ points and preserve a fluid and engaging information experience, so we spent a significant amount of time to find the best coding strategy to balance performance and efficiency.
What can others learn from this project?
This project shows that the analysis of the local impact of global phenomena is both important and successful.
The project moves past a traditional data-journalism project based on a map, offering a scrollytelling experience which brings the reader across the charts and the data. From a data visualization point of view, the point-based design allows a more precise analysis, and discovery of crucial municipalities that are suffering more from climate change compared to others, due to different issues (land use, pollution, population, etc etc). This makes the project something more than an article: a real tool to discovery.
Finally, this project made it available a wealth of data that has been reused by many other organizations, inside and outside the media landscape. This, besides giving to the project a snowball effect, also made it possible to create and nurture a network with other journalists and civil society organizations.