Modelling risk systems, in which natural hazards and exposure elements are intricately intertwined, poses a significant challenge, especially over large spatial and temporal scales. To address this issue, this study introduces the use of hypergraphs as a modelling framework for dynamic multi-hazard systems. Hypergraphs have found applications across disciplines for effectively capturing complexities in various systems.The study demonstrates the suitability of hypergraphs to multihazard risk assessment through a case study of the 2015 Gorkha earthquake in Nepal and its subsequent coseismic landslides. The initial test case is followed by the generation of cascading scenarios initiated by thirty high-magnitude simulated earthquakes across Nepal and analysis of the subsequent cascading impacts arising from landsliding on buildings and roads. The modelling is being developed to provide scientific evidence to inform preparedness planning at a range of scales.Our results show that this approach is effective, offering several key advantages. First, the easy compatibility with spatial data enables a more accurate representation of real-world scenarios. Second, the proposed method is hazard-agnostic, allowing it to accommodate various types of natural hazards. Third, the high computational efficiency of the hypergraph-based model enables the use of large scenario ensembles. Finally, the capability to handle complex interactions between hazard processes and exposure elements streamlines the risk assessment process.We emphasise that the adoption of hypergraphs as a modelling framework has the potential to substantially enhance multi-hazard risk assessment in natural systems. By providing a comprehensive and flexible approach, this method offers a promising avenue for improving risk management strategies and bolstering preparedness measures to mitigate the impacts of environmental disasters.
原文链接
分享
引用
