In earthquake-prone regions, seismic resilience is a critical factor in construction and infrastructure planning.
Image Credit: metamorworks/Shutterstock.com
Designing earthquake-resistant buildings requires a deep understanding of how seismic waves interact with various materials, foundation types, and structural systems. Traditional seismic risk assessments rely on empirical data, but advancements in high-performance computing (HPC) and physics-based simulations are transforming the way construction professionals model, design, and optimize earthquake-resistant infrastructure.
By harnessing cutting-edge computing power, researchers and engineers are gaining deeper insights into how structures respond to seismic events, leading to smarter, more resilient infrastructure designs.1,2
This article explores how supercomputing and advanced seismic modeling are transforming infrastructure design, helping engineers create safer, more earthquake-resistant structures. Specifically, it will:
- Examine how HPC enhances seismic resilience in construction.
- Highlight real-world case studies where seismic simulations have influenced engineering strategies.
- Demonstrate how high-fidelity modeling improves material selection, foundation designs, and structural integrity.
Upgrade your knowledge—grab your free copy for an in-depth look at seismic modeling in infrastructure design!
Supercomputing and Advanced Earthquake Risk Assessment
Accurate earthquake risk assessment has long been a challenge due to the unpredictable nature of seismic events. Magnitude, location, and ground conditions all vary significantly, making it difficult to develop precise predictive models. Traditional methods, heavily reliant on empirical data, often lack the granularity needed to reflect real-world variability. In contrast, physics-based simulations provide a more refined understanding of ground motion, but they demand immense computational resources, extensive geological data, and highly detailed earthquake source modeling—requirements that only supercomputers can meet.
One of the most critical aspects of seismic analysis is understanding how seismic wave frequency affects different structures. Tall buildings are particularly vulnerable to low-frequency ground motion, while smaller structures are more susceptible to high-frequency shaking. Accurately modeling these interactions requires the ability to simulate high-fidelity wave propagation across varying soil and structural compositions. While conventional computers can handle basic computations, supercomputers enable the simulation of large-scale, high-resolution models that capture complex interactions between seismic waves and infrastructure with unprecedented accuracy.
Supercomputers also facilitate real-time scenario testing, allowing researchers to simulate thousands of possible earthquake conditions and analyze structural performance under extreme conditions. This capability is essential for designing infrastructure that can withstand severe seismic events, as it allows engineers to refine designs based on detailed, scenario-specific performance insights. Furthermore, the ability to model the interplay between soil dynamics, material behavior, and structural response in three-dimensional space provides a level of predictive power that was previously unattainable.1-3
Seismic Modeling in Construction Planning and Structural Engineering
Supercomputing-driven seismic simulations are not just theoretical exercises—they are directly influencing construction design, material selection, and engineering standards. Advanced seismic modeling helps engineers and urban planners:
- Optimize building materials and reinforcements: Computational models assess how different materials (e.g., reinforced concrete, steel, engineered wood) react to seismic loads.
- Improve foundation and soil interaction studies: HPC-based finite element method (FEM) models help predict how soil liquefaction, lateral spreading, and ground shaking affect deep foundations, retaining walls, and underground structures.
- Refine performance-based seismic design (PBSD): Instead of using rigid code-based compliance, engineers can now use high-fidelity, scenario-based simulations to design buildings that meet performance objectives for different earthquake intensities.
- Enhance retrofitting strategies for aging structures: Many existing buildings in earthquake-prone zones were not built to modern seismic codes. Supercomputing allows engineers to simulate retrofitting solutions, ensuring cost-effective seismic upgrades.
High-Precision Seismic Modeling with 3D FEM and HPC
Beyond risk assessment, high-precision modeling is essential for understanding soil-structure interactions and optimizing infrastructure designs.
Performance-based design is a fundamental approach in earthquake geotechnical engineering, requiring detailed seismic response analysis. The finite element method (FEM) plays a crucial role in evaluating soil-structure interactions, ground failure mechanisms, and seismic hazards such as soil liquefaction and lateral spreading. High-fidelity three-dimensional (3D) FEM models offer a significant advantage over traditional empirical or two-dimensional approaches by accurately capturing nonlinear soil behavior, wave propagation, and dynamic ground-structure interactions. However, their complexity demands immense computational power.
Historically, the sheer computational demands of large-scale 3D FEM simulations posed a major limitation. However, advancements in HPC now allow full-scale 3D seismic response models to run with high spatial resolution, greatly improving predictive accuracy. These advances help researchers analyze seismic responses in heterogeneous soil conditions, assess multi-hazard interactions, and optimize infrastructure designs to withstand extreme earthquake scenarios.
While two-dimensional (2D) FEM models remain useful for preliminary assessments, they have notable constraints. For example, Japan’s FLIP ROSE program relies on sequential computation, which is inefficient for modern parallel-processing environments. Additionally, its reliance on hexahedral elements limits its ability to model intricate 3D subsurface conditions. The shift to supercomputing-powered 3D FEM models enables more precise seismic hazard assessments and performance-based design optimization for critical infrastructure.4
Case Studies: Real-World Applications of Supercomputing in Seismic Resilience
Supercomputing is playing a pivotal role in enhancing seismic resilience by enabling more precise earthquake simulations and impact assessments. The following case studies illustrate how high-performance computing is being applied in real-world scenarios to improve earthquake preparedness, response, and infrastructure resilience.
Understanding Multi-Fault Earthquakes Through HPC Simulations
Assessing earthquake impact in real-time presents significant technical and scientific challenges. Shake maps, which provide rapid seismic motion distributions immediately after an event, are essential for disaster response. When combined with data on population density and building integrity, these maps become invaluable tools for guiding emergency response efforts.3
A notable advancement in urgent earthquake impact assessment is the Urgent Computing Integrated Services for EarthQuakes (UCIS4EQ) workflow, introduced in the Proceedings of the Platform for Advanced Scientific Computing Conference. This HPC-driven system generates rapid, high-accuracy reports on moderate to large earthquake impacts, supplying civil protection agencies with critical data for swift decision-making.
UCIS4EQ automates data gathering, determines whether 3D simulations are necessary, preprocesses input data, runs simulations on a tier-0 PRACE HPC system, and post-processes results for stakeholders. Leveraging parallel processing, this workflow delivers meaningful earthquake impact assessments in just a few hours—a significant improvement in disaster response capabilities.3
Modeling Multi-Fault Rupture Behavior with Supercomputers
Multi-fault earthquakes, where ruptures propagate across interconnected fault systems, can have far-reaching effects. During the last decade, seismologists have observed several cases of this complicated type of earthquake rupture and are now relying on supercomputers to provide detailed models to better understand the fundamental physical processes that take place during these events, which can have far-reaching effects.5
A study published in the Journal of Geophysical Research: Solid Earth recently examined the interconnected fault system within the Brawley seismic zone (BSZ) in Southern California. This system includes the Imperial Fault (IF), the Southern San Andreas Fault (SSAF), and cross faults within the BSZ, which have the potential to link larger faults in unexpected ways. The researchers employed 3D dynamic finite element spontaneous rupture modeling on supercomputers such as Stampede (Texas Advanced Computing Center) and Comet (San Diego Supercomputer Center) to simulate possible earthquake scenarios.
By using physics-based dynamic rupture models, researchers simulated how multi-fault earthquakes impact urban infrastructure, revealing which foundation types and structural designs are most vulnerable to different fault rupture patterns. These insights help engineers optimize seismic codes for buildings, bridges, and transportation networks in complex fault zones.5
High-Performance Seismic Analysis in Urban Environments
A striking example of supercomputing’s impact comes from a study featured in the Japanese Geotechnical Society Special Publication. The researchers used the "Fugaku" supercomputer to conduct large-scale 3D seismic response analysis of an urban area. They developed T-STRIKE, an extension of STRIKE, a large-scale nonlinear FEM program known for its high computational performance on supercomputers and its automated meshing algorithm with tetrahedral elements. This extended STRIKE (T-STRIKE) was employed as the analysis program, and the constitutive laws of soil and liquefaction that are commonly utilized in seismic design in Japan were implemented in the program.4
T-STRIKE was applied to analyze a 1.4 × 1.3 km urban basin, processing a detailed FEM model with 230 million degrees of freedom (DOF) using 12,288 CPU cores over 48 hours. The results aligned with prior studies, validating T-STRIKE’s effectiveness and demonstrating its ability to handle complex geometries at an unprecedented scale.
Beyond theoretical validation, this research has practical implications for infrastructure planning and retrofitting. Engineers can now leverage these high-resolution models to refine earthquake-resistant designs, ensuring that new structures are optimized for real-world seismic conditions. Moreover, urban planners and policymakers can use the insights from these simulations to prioritize investments in seismic mitigation strategies, such as strengthening vulnerable buildings, improving foundation designs, and optimizing zoning laws to account for ground motion risks.
Want to Learn More?
Supercomputing is revolutionizing seismic modeling for construction professionals, allowing engineers to design safer buildings, optimize retrofitting strategies, and ensure that infrastructure is resilient against future earthquakes. As technology advances, high-fidelity seismic simulations will play an even greater role in shaping urban planning policies, updating building codes, and developing cost-effective earthquake-resistant construction methods.
Want all the details? Grab your free PDF here!
If this article has taken your interest, why not check out some of the below?
References and Further Reading
- Kovner, A. (2023) The Most Advanced Bay Area Earthquake Simulations Will be Publicly Available [Online] Available at https://cs.lbl.gov/news-and-events/news/2023/the-most-advanced-bay-area-earthquake-simulations-will-be-publicly-available/ (Accessed on 17 February 2025)
- Using supercomputers to quantify Bay Area earthquake hazard and risk [Online] Available at https://asc.llnl.gov/article/22801/using-supercomputers-quantify-bay-area-earthquake-hazard-risk (Accessed on 17 February 2025)
- de la Puente, J., Rodriguez, J. E., Monterrubio-Velasco, M., Rojas, O., Folch, A. (2020). Urgent supercomputing of earthquakes: Use case for civil protection. Proceedings of The Platform For Advanced Scientific Computing Conference, 1-8. DOI: 10.1145/3394277.3401853, https://dl.acm.org/doi/abs/10.1145/3394277.3401853
- Otsuka, Y., Tamari, Y., Fujita, K., Ichimura, T. (2024). Large-scale 3D seismic response analysis considering soil liquefaction in an urban area using the supercomputer “Fugaku”. Japanese Geotechnical Society Special Publication, 10(46), 1735-1740. DOI: 10.3208/jgssp.v10.OS-35-06, https://www.jstage.jst.go.jp/article/jgssp/10/46/10_v10.OS-35-06/_article/-char/ja/
- Kyriakopoulos, C. et al. (2019). Dynamic rupture scenarios in the Brawley seismic zone, Salton Trough, Southern California. Journal of Geophysical Research: Solid Earth, 124(4), 3680-3707. DOI: 10.1029/2018JB016795, https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018JB016795
- CyberShake study uses Summit supercomputer to investigate earthquake hazards [Online] Available at https://www.ornl.gov/news/cybershake-study-uses-summit-supercomputer-investigate-earthquake-hazards (Accessed on 17 February 2025)
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.