[Skip to Content]
Banner
Menu
  • Home
  • Submit Abstract
  • Home
  • 2023 Annual Meeting Gallery
  • Evaluating Bias of NGA-East GMMs and Site Factors for Ground Motions From Natural and Potentially Induced Earthquakes in Texas, Oklahoma, and Kansas

← Back to Gallery

Evaluating Bias of NGA-East GMMs and Site Factors for Ground Motions From Natural and Potentially Induced Earthquakes in Texas, Oklahoma, and Kansas

To empirically investigate the bias and linear site amplification of the median NGA-East ground motion model (GMM, Goulet et al. 2021) in predicting ground motions from natural and potentially induced earthquakes in Texas, Oklahoma, and Kansas, we developed a database containing ground motion recordings in the study region with moment magnitude greater than 3.5 and rupture distances smaller than 500 km. The same protocols in Ramos-Sepulveda et al. (2023) are utilized to establish the database. A reference empirical approach (Atkinson 2008) is utilized to adjust the magnitude and distance scaling of the NGA-East median GMM based on the observed data, and a GMM for the region is developed. The site amplification model relative to =760m/s developed in this study is compared to the site amplification models for Central and Eastern North America (CENA) from Parker et al. (2019) and Zalachoris and Rathje (2019). The comparisons suggest similar site amplification effects reflected by the dataset used in this study to those in the other studies. However, the overall bias adjustment factor in the GMM shows discrepancies from the simulation-based model suggested by Stewart et al. (2020). The bias is negative at shorter periods of T<0.1 s and is positive at longer periods, reaching values up to 1.0 (in ln-units) at periods longer than 3.0 s. The negative bias at shorter periods may be associated with the choice of value of reference κ when developing the NGA-East GMM, while the large positive bias at longer periods is difficult to explain physically as a site amplification effect. This large bias may be due to issues with assigning moment magnitudes to much of the data or bias in sampling of the ground motion data, where ground motions of smaller amplitudes are more likely to be dropped because of noise at low frequencies.


Session: USGS National Seismic Hazard Models: 2023 and Beyond

Type: Oral

Room: 208A

Date: 4/18/2023

Presentation Time: 02:15 PM (local time)

Presenting Author: Meibai Li

Student Presenter: No


Additional Authors

Meibai Li

Presenting Author

Corresponding Author

mli@utexas.edu

University of California, Los Angeles

Ellen Rathje

e.rathje@mail.utexas.edu

University of Texas at Austin

Maria Ramos-Sepúlveda

mariaramos1@g.ucla.edu

University of California, Los Angeles

Jonathan Stewart

jstewart@seas.ucla.edu

University of California, Los Angeles

Okan Ilhan

okan.ilhan.uiuc@gmail.com

Ankara Yildirim Beyazit University

Youssef Hashash

hashash@illinois.edu

University of Illinois at Urbana-Champaign

 

Evaluating Bias of NGA-East GMMs and Site Factors for Ground Motions From Natural and Potentially Induced Earthquakes in Texas, Oklahoma, and Kansas

Category

USGS National Seismic Hazard Models: 2023 and Beyond

Description