Consequence-Driven Earthquake Scenario Selection

Session: Earthquake Ground Motion and Impacts [Poster]
Type: Poster
Date: 4/30/2020
Time: 08:00 AM
Room: Ballroom
Description: 

Scenario selection for the purpose of emergency management and planning often relies on hypothetical earthquake ruptures on mapped active faults. Mostly, these faults have quantifiable return times that are typically short enough to warrant concern. This strategy is not applicable to many important population centers and critical infrastructure in stable tectonic environments, where active faults are not present or are poorly mapped or constrained. These regions, however, still require scenario earthquakes for emergency planning and mitigation purposes. And, such events do occur, including Charleston, South Carolina (1886), Nahanni, Canada (1985) and Tennant Creek, Australia (1998). Selecting earthquake characteristics in this environment can be difficult and often relies on subjective choices due to the uncertain nature of the long-term seismicity. The challenge is to balance the emergency managers’ needs for a consequential event while avoiding earthquakes that are tectonically implausible. With this motivation in mind, we have optimized the ShakeMap and PAGER codes to efficiently compute point-source scenario ground motions such that we can perform a grid search of candidate earthquake scenarios and loss estimates across a range of locations and magnitudes. From the grid search we can then select the most likely event that meets minimum consequence criteria, which could be based on economic losses, fatalities or both. We demonstrate this procedure in Kansas City, Missouri, which shows that highly consequential events can be realized for relatively moderate magnitudes due to the high building vulnerability that is typical of stable tectonic environments. Selecting consequence-based scenarios offers a solution to risk analyses where existing earthquake hazard studies offer limited constraints on occurrence. Rather than pushing the limited statistical basis for scenario selection, one can ask what impacts are relevant to the stakeholder and then work backwards to determine which events are most likely to cause such consequences.

Presenting Author: Eric M. Thompson

Authors