Benchmarking Multidimensional Ground Response Analyses at the Treasure Island Borehole Array Site Using Different Commercial and Open-Source Software
Over the past decade, numerous studies have examined ground motions recorded at borehole array sites and found that, on average, more than 50% of these sites are poorly modeled using one-dimensional (1D) ground response analyses (GRAs). These discrepancies have been attributed to limitations of conventional 1D GRAs, which inherently disregard complex wave propagation effects resulting from laterally variable subsurface conditions, among other factors. Given these findings, as well as the increasing body of research literature and consulting practice that are implementing multidimensional GRAs, proper modeling of laterally variable subsurface conditions is an important step toward improving seismic hazard assessment and risk mitigation. This presentation will share findings from an ongoing research effort aimed at ground truthing multidimensional GRAs. The study is working toward providing a well-documented and openly accessible set of benchmarking cases that can serve as a resource for practitioners and researchers to calibrate their own analyses. The benchmark cases range in complexity from a single soil column with a single uniform soil layer to a 4000-m wide 2D cross-section with site-specific subsurface spatial variability representing the Treasure Island Downhole Array (TIDA). Collaborators from academia and industry have modeled these cases using different commercial and open-source software (e.g., OpenSees, Sesimo-VLab, FLAC, LS-DYNA). By qualitatively and quantitatively comparing the numerical results across different software platforms and by ground truthing the results against recorded ground motions at TIDA, we share insights on the relative strengths and limitations of each software in terms of accuracy and challenges associated with implementing multidimension GRAs, such as modeling boundary conditions and computational costs.
Session: Ground Truthing Multidimensional Site Response Analyses at Borehole Array Sites
Type: Oral
Room: 209B
Date: 4/20/2023
Presentation Time: 04:30 PM (local time)
Presenting Author: Mohamad M. Hallal
Student Presenter: No
Additional Authors
Mohamad Hallal Presenting Author Corresponding Author mhallal@berkeley.edu University of California, Berkeley |
Brady Cox brady.cox@usu.edu Utah State University |
Kami Mohammadi kami.mohammadi@utah.edu University of Utah |
Christopher de la Torre christopher.delatorre@canterbury.ac.nz University of Canterbury |
Kevin Stanton kevin.stanton@mottmac.com Mott MacDonald |
Dylan Centella dylanmc3@illinois.edu University of Illinois at Urbana-Champaign |
Albert Kottke arkk@pge.com Pacific Gas & Electric |
Brendon Bradley brendon.bradley@canterbury.ac.nz University of Canterbury |
Youssef Hashash hashash@illinois.edu University of Illinois at Urbana-Champaign |
Domniki Asimaki domniki@caltech.edu California Institute of Technology, Pasadena, California, United States |
Jongwon Lee jongwon.lee@mottmac.com Mott MacDonald, San Francisco, California, United States |
Wenyong Rong wenyong.rong@mottmac.com Mott MacDonald, San Francisco, California, United States |
Benchmarking Multidimensional Ground Response Analyses at the Treasure Island Borehole Array Site Using Different Commercial and Open-Source Software
Category
Ground Truthing Multidimensional Site Response Analyses at Borehole Array Sites
Description