Common wisdom suggests air travel is cheaper and faster than more environmentally-friendly alternatives. But is that always the case? And can we use data to reveal the hidden costs of our travel choices?
We collected thousands of ticket prices, journey times and CO2 emissions figures for trains and planes on six direct routes between European cities: Berlin-Warsaw, Munich-Budapest, London-Amsterdam, London-Marseilles, Paris-Barcelona and Zurich-Milan.
For DW\’s target audience of internationally-minded people around the world, this is a question they might grapple with in everyday life.
Our analysis shows on which routes the train is giving the plane a run for its money — and reveals how the numbers would change if we were made to pay for the environmental consequences of travel.
What makes this project innovative?
Transport, the environment and economics are all journalistic beats in their own right, which might tackle questions around aviation separately. Our project aims to be a comprehensive investigation that brings expertise and data from a range of fields. We used scraping, our own data modeling and data visualisation to produce a unique take on this topic.
What was the impact of your project? How did you measure it?
The project sparked interest on Twitter, with academics, activists and travellers citing the analysis as evidence in an ongoing debate. The dwell time for the article on DW.com was three times that of the average DW article, and scored the highest number of visits among all our data-driven articles.
Source and methodology
You can find a full account of our methodology, as well as the data and code behind the analysis, on our GitHub page (see additional links). We wrote a computer program to look up airline tickets on Google Flights and rail tickets on Trainline. For carbon emissions we used figures from IFEU, an environmental consultancy firm that runs EcoPassenger, a website that estimates the emissions per traveler for any given journey. Our notional carbon tax was based on research published by Ireland's Economic and Social Research Institute. Experts from the University of Southampton and the University of Western Australia commented from a legal and behavioural science perspective.
The analysis and data collection was conducted in Python, using pandas, and documented in a Jupyter Notebook. Visualisations were generated in Matplotlib and adapted for publication in Adobe Illustrator.
Tom Wills, Gianna-Carina Grün