Characterizing Uncertainty in City-Wide Disaster Recovery through Geospatial Multi-Lifeline Restoration Modeling of Earthquake Impact in the District of North Vancouver

Abstract

Restoring lifeline services to an urban neighborhood impacted by a large disaster is critical to the recovery of the city as a whole. Since cities are comprised of many dependent lifeline systems, the pattern of the restoration of each lifeline system can have an impact on one or more others. Due to the often uncertain and complex interactions between dense lifeline systems and their individual operations at the urban scale, it is typically unclear how different patterns of restoration will impact the overall recovery of lifeline system functioning. A difficulty in addressing this problem is the siloed nature of the knowledge and operations of different types of lifelines. Here, a city-wide, multi-lifeline restoration model and simulation are provided to address this issue. The approach uses the Graph Model for Operational Resilience, a data-driven discrete event simulator that can model the spatial and functional cascade of hazard effects and the pattern of restoration over time. A novel case study model of the District of North Vancouver is constructed and simulated for a reference magnitude 7.3 earthquake. The model comprises municipal water and wastewater, power distribution, and transport systems. The model includes 1725 entities from within these sectors, connected through 6456 dependency relationships. Simulation of the model shows that water distribution and wastewater treatment systems recover more quickly and with less uncertainty than electric power and road networks. Understanding this uncertainty will provide the opportunity to improve data collection, modeling, and collaboration with stakeholders in the future.

Methods for representing regional disaster recovery estimates: modeling approaches and assessment tools for improving emergency planning and preparedness

“…While GMOR has been used in previous recovery assessments of infrastructure systems (Deelstra and Bristow 2020), their scale was limited to a single municipality. The objective of this work, therefore, is to demonstrate the use of GMOR in the development of a recovery model and subsequent assessment of the restoration of service to interconnected infrastructure systems for an entire region with multiple municipalities using relatively accessible infrastructure system information.…”

“…The GMOR model uses an approach that simplifies infrastructure systems within a region by separating them into distinct service zones based on criteria such as population or size. Damage to linear infrastructure systems within each zone is aggregated to produce a total damage level for each system for the zone (Deelstra 2019;Deelstra and Bristow 2020;.…”

“…This approach provides a distinct scenario that can be used for planning and community preparedness purposes. In previous work done using GMOR, outputs from Hazus led to an extensive range of possible outcomes, which is useful to understand the extremes that can be experienced after a disaster but may be too broad for use in planning exercises (Deelstra and Bristow 2020).…”

“…Results from a GMOR model can be processed at varying scales and levels of detail. In addition, GMOR was utilized for earlier studies in the region (Deelstra and Bristow 2020;Deelstra 2019), so its capabilities for supporting the assessment of the considered systems is already established.…”

Evaluation of Improvement Projects in Densely Built-Up Area using a Large Earthquake Disaster Simulator: A Case Study in Kyojima Area, Tokyo

“…Specifically, it provides a realistic environment to discuss disaster prevention among government agencies, landowners, and residents, based on estimated risk (Mota de Sá et al, 2016;Bonacho and Oliveira, 2018). In addition, it is also used to examine evacuation plans for residents (D'Orazio et al, 2014), maintain critical infrastructure (D'Agostino et al, 2019), and create recovery plans for essential lifelines such as water supply, power distribution, and transportation systems at the local government level (Deelstra and Bristow, 2020). Saitoh et al (1999) and Kishimoto and Osaragi (2019) first evaluated disaster damage using a simulation model that describes property damages after a large earthquake.…”