Vulnerability and resilience analysis of multimodal public transport networks under extreme rainfall: A case study in Gothenburg

Abstract

Extreme rainfall can disrupt public transport networks by flooding stops, route segments, and transfer connections. This study assesses the vulnerability and structural resilience of Gothenburg’s multimodal public transport network under a climate-adapted 100-year rainfall scenario. It also examines whether walking-scooter transfer links, which represent transfers using walking access and shared e-scooters, can improve network connectivity after flood disruption.
Public transport data from GTFS Sverige 2 were used to construct a baseline network including bus, tram, train, and ferry services. Flood impacts were modelled by overlaying stops and route edges with flood-depth data and applying mode-specific failure thresholds. Two disrupted networks were then constructed: a direct-impact network and a route-impact network. Scooter trip records and road network data were used to identify scooter hotspots and construct walking-only and walking-scooter transfer links. Six network scenarios were converted into undirected weighted NetworkX graphs, and resilience indicators were calculated.
The results show that flood disruption strongly reduced network connectivity. Bus and tram were the most affected modes. For train services, all stops remained active, while all train route edges were classified as directly failed. In the route-impact network, only a small part of the original network remained connected. After adding walking-scooter transfer links, connectivity, local reachability, and robustness improved to different degrees, but the improvement was limited. Overall, shared e-scooters can provide supplementary local connections under flood disruption, but they cannot fully replace disrupted public transport services.

Popular Abstract

Can shared e-scooters help public transport during flooding?
Heavy rainfall can create serious problems for cities. When streets and tracks are flooded, buses, trams, trains, and ferry stops may not work as usual. This can make it harder for people to move around the city, especially when several important stops or route sections are affected at the same time.
This thesis looks at this problem in Gothenburg, Sweden. Gothenburg has a large public transport system with buses, trams, trains, and ferries. The city also has shared e-scooters, which are often used for short trips. The main question of this thesis is whether shared e-scooters can help keep parts of the public transport network connected after extreme rainfall.
To answer this question, public transport data were used to build a network of stops and route connections in Gothenburg. Flood-depth data from a climate-adapted 100-year rainfall scenario were then used to identify which stops and route sections could be affected by flooding. Two flood-disrupted networks were created. The direct-impact network only removed stops and route sections that were directly affected by floodwater. The route-impact network also considered wider route-level effects, where disruption can spread along public transport routes.
The study also used scooter trip data to find areas where shared e-scooters are often available. These areas were used to create possible scooter transfer links between public transport stops. In this way, the study tested whether people could use a short walk and an e-scooter ride to connect stops that became harder to reach after flooding.
The results show that extreme rainfall can strongly reduce the connectivity of Gothenburg’s public transport network. Bus and tram services were the most affected modes. For train services, all stops remained active, but the route sections were disrupted because they crossed flooded areas. In the stricter route-impact network, only a small part of the original network remained connected.
Adding scooter transfer links improved the network in some ways. It increased connectivity, local reachability, and robustness in several scenarios. However, the improvement was limited. Shared e-scooters can help provide extra local connections during flood disruption, but they cannot fully replace disrupted public transport services.
Overall, this thesis shows that shared e-scooters can be useful as a supplementary transport option in flood situations. They may be most helpful near important transfer stops, flood-prone bus and tram corridors, and areas where public transport connections become weak after flooding. However, their contribution depends on scooter availability, safe road conditions, and whether walking and scooter routes remain passable during rainfall.

• author: Yang, Zihui
• supervisor: Pengxiang Zhao
• organization: Department of Earth and Environmental Sciences (MGeo)
• course: NGEM21 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Earth and Environmental Sciences
• keywords: Geographical Information Science, Public transport resilience, Extreme rainfall, Flood disruption, Shared e-scooters, Multimodal network, Gothenburg
• publication/series: Student thesis series INES
• report number: 788
• language: English
• id: 9238109
• date added to LUP: 2026-06-16 12:48:19
• date last changed: 2026-06-16 13:40:56