← Back to Sessions

Augmenting Near-Source Probabilistic Seismic Hazard Analysis (PSHA) With North American Crustal Stress Field Data

Description: 

Probabilistic seismic hazard assessment (PSHA) provides an assessment of the annual frequency of
exceeding various levels of ground shaking at a location. In locations without detailed characterization of
the seismic hazard associated with individual faults, significant epistemic uncertainties relate to source
characterizations, including the identification of potential rupture characteristics and their likelihoods.
As we show, knowing the state of stress in the Earth’s crust can reduce epistemic uncertainties for PSHA
by providing information about the likelihood of a given fault or fault segment to fail, and the expected
sense of fault slip. The orientation of the maximum horizontal principal stress (S_Hmax ) and the style of
faulting (relative principal stress magnitudes) determine which orientations of pre-existing faults are
more and less likely to fail seismically. The probability that faults of any orientation may be potentially
active may be estimated from stress, fault, and rock property parameters and their corresponding
uncertainty distributions.
Stress maps for North America have been developed that provide both S_Hmax orientations and quantify
the style of faulting, together with uncertainties. We focused on a case study that evaluated the
Meers Fault repeated large magnitude earthquake (RLME) to allow us to assess the potential impact of
the stress information on a well characterized RLME. Stress information was applied to a simplified 2012
CEUS-SSC base model with real and fictional stress scenarios. We observed that stress-informed
epicenter localization and near-fault directivity effects resulted in significant differences in near-fault
ground motion exceedance frequencies.
The use of crustal stress data during the development of fault source parameters for a PSHA may
improve ground motion estimates, particularly in near-source locations, sites sensitive to ground
shaking, or where fault orientations are well characterized but paleoseismic data or other fault hazard
indicators are not.

Session: Opportunities and Challenges in Source Modeling for Seismic Hazard Analysis

Type: Oral

Date: 4/20/2023

Presentation Time: 08:30 AM (local time)

Presenting Author: Constantinos Frantzis

Student Presenter: Yes

Invited Presentation: Yes

Authors