Train Delays due to Extreme Weather Events in Sweden 2001-2020

Railways are vulnerable to extreme weather events. Some of these are expected to increase in frequency and severity in a warming world, prompting adaptation efforts. To inform these efforts, we have analysed railway operations data from Sweden in the period 2001-2020 to learn more about the past effects of extreme weather events on train delays. We have used a database containing some 1.6 billion records regarding train movements and a total of almost 11 million delay hours. Studying the delay attributions, we find that slightly less than 1% of delay hours are attributed to extreme weather, but that in the most extreme month, this had risen to about 11%. We find that storms are the biggest contributor, closely followed by fire and snow, and that the trend is towards more such delays. Studying the size distributions, we find that the delays are largely driven by the most extreme of the extreme events, following distributions that are mostly exponential but have even more extreme peaks. In future work, we hope to quantify and illustrate how these hazards and delays are likely to evolve under a changing climate, and to propose adaptation measures