S5E8

Abstract

Under many natural loading conditions, rocks fail in shear. Characterizing the conditions of failure is crucial for assessing plausible societal threats from natural hazards, such as earthquakes and landslides, as well as the sustainability of human activities such as carbon sequestration and energy resource practices. In this talk, I will discuss recent progress using physics-based computational models of dynamic frictional rupture sequences and slow fault slip to bridge laboratory insight and geophysical field observations of fault failure processes. We study the stress conditions governing the occurrence of simulated earthquake ruptures that spontaneously evolve in numerical simulations of fault models with laboratory-derived rate-and-state friction laws, shear heating, and effects of pore fluids. We find that the critical stress conditions for rupture propagation depend on the size of the rupture and how efficiently the fault shear resistance weakens during fast slip. In particular, larger earthquakes on faults that experience increasingly more efficient weakening during rupture can propagate under systematically lower stress conditions than those required for rupture nucleation. As a result, we find that faults that exhibit efficient weakening can host predominantly large earthquakes at the expense of smaller earthquakes. Our findings illustrate how large earthquakes can occur on faults that appear to be under-stressed compared to expected conditions for rupture nucleation and highlight the importance of finite-fault modeling in relating the local friction behavior determined in the lab to the field scales. Moreover, our findings support a body of work suggesting that the scarcity of small earthquakes on some major mature fault segments, like the central section of the San Andreas Fault, may indicate that they experience substantial weakening during large earthquakes, a consideration that may be particularly useful for earthquake early warning systems.

Introduction of speaker

Dr. Valère Lambert is currently a National Science Foundation Postdoctoral Fellow with the University of California Santa Cruz Department of Earth and Planetary Sciences. He received his PhD from the California Institute of Technology, working with the Caltech Seismological Laboratory and Department of Mechanical and Civil Engineering. His work focuses on developing computational methods to study the multi-scale nature of fracture and frictional failure, with particular implications for geohazards, such as earthquakes and landslides. Valère also co-leads the Southern California Earthquake Center community code-verification project for Advancing Simulations of Earthquakes and Aseismic Slip (SEAS).

Abstract

Structural fire protection is a critical component of fire protection design. Structural fire damage can lead to severe loss of life as well as economic losses such as business disruption. The former requires that the structural fire protection of a building guarantees the life safety of its occupants and first responders during a fire; and the latter, that such protection provides the building with some level of resilience. Regardless of whether the goal is life safety or resilience, however, a framework for comprehensively assessing the fire risk to a building’s structure is indispensable. Previous studies have modified the PEER Performance-based Earthquake Engineering (PBEE) framework for structures on fire. However, they did not consider the multi-physics characteristics of fire damage, and estimated damage mainly based on simplified structural fire analysis and simplified damage classification. We, therefore, propose a probabilistic risk-informed framework that estimates direct fire loss based on explicit analysis of an entire building’s structural response to fire. Also adapted from the PEER PBEE framework, it comprises fire-hazard analysis, response analysis, damage analysis, and loss analysis. This framework takes account of uncertainties pertaining to the occurrence and growth of fires, as well as to buildings’ responses. As part of our transformation of this framework from an earthquake context to a fire context, we define fire-specific engineering demand parameters that are measurable and associated with the damage states of each component in a building. Our classification of damage states addresses gradual levels of repair efforts for those components, in addition to the collapse/no collapse criteria for an entire building. Then, based on this more realistic damage-state classification, we develop fragility functions and consequence functions. When coupled with probabilistic models for describing random variables, and advanced modeling techniques for structures on fire, this framework can output probabilistic risk information at both building-system and component levels. It provides a probabilistic methodology for the risk-informed performance-based design of buildings’ structural fire protection. The detailed risk-assessment results provided by this framework will also facilitate the analysis of a building’s recovery after a fire, which will help expand the scope of structural fire engineering beyond life safety to include resilience.

Introduction of speaker

Dr. Shuna Ni is currently an Assistant Professor in the Department of Fire Protection Engineering at the University of Maryland, College Park. Dr. Before joining UMD, Dr. Shuna Ni worked as an Assistant Professor at Utah State University from 2020 to 2022 and a postdoctoral fellow at Johns Hopkins University from 2018 to 2020. She received her Ph.D. degree at Texas A&M University in 2018 and her Master's degree at Tongji University in 2013. Her group’s overall goal is to increase built environments’ resistance and resilience to fires and fire-related multiple hazards, at both the single-structure and community levels, by bridging the gaps between fire protection engineering and civil engineering.

A Risk-informed Framework for Evaluating the Structural Fire Safety of Buildings