Work

Simulating Cascading Flood Impacts on Urban Infrastructure Systems in Malmö

Year: 2026
Type: preprint
Abstract: Interconnected urban infrastructures are increasingly vulnerable to environmental hazards, where local failures can cascade across systems, disrupting essential services. This study introduces TRIGGER, a novel methodological framework coupling flow-based network modeling with Large Language Models (LLMs) to systematically translate technical cascading failure metrics into socially meaningful information. Furthermore, the framework provides a quantitative approach to evaluate and compare the systemic effectiveness of large-scale nature-based adaptation measures in mitigating indirect service losses. Applied to Malmö, Sweden, TRIGGER simulates interdependencies among electricity, water, emergency services, and port operations under three flood scenarios. Results demonstrate that the integrated LLMs effectively translate complex numerical data into contextualized narratives, successfully scaling qualitative severity to the magnitude of disruptions and abstracting large datasets to reduce cognitive load for decision-makers. The framework also quantifies mitigation potential of nature-based solutions, showing that widespread green roofs could substantially reduce service disruptions. By translating complex failure chains into actionable insights, TRIGGER supports urban planners in advancing from siloed risk management toward proactive, cross-sectoral climate adaptation.
Language: English
Open Access status: green
OpenAlex ID: W7214781254
Cites: 53
Cited by:
Related to: 10