A workflow for seismic imaging with quantified uncertainty

Barbosa, Carlos H. S.; Kunstmann, Liliane; Silva, Rômulo M.; Alves, Charlan D.S.; Silva, Bruno Santana da; Filho, Djalma Manoel Soares; Mattoso, Marta; Rochinha, Fernando A.; Coutinho, Álvaro L. G. A.

doi:10.1016/j.cageo.2020.104615

Cited by 8 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Seismic surveys are the most widespread geophysical tools for detecting brittle‐plastic failures (e.g., faults and folds) and fluid seepage‐related structures (e.g., pockmarks, mounds, and gas seepages) in submarine FTBs and HGs (Cathles et al., 2010; Sultan et al., 2014). However, seismic data acquisition and processing for subsurface layers with the assumption of idealized elastic properties and fracture networks can increase the uncertainties in seismic interpretation results (Barbosa et al., 2020). Thus, several petrogeological studies have adopted a synthetic seismic velocity profile using the Green strain (e.g., volumetric strain) derived from DE fault models (Botter et al., 2014) to enhance the inversion ability of potential seismic data.…”

Section: Discussionmentioning

confidence: 99%

New Particle‐Wise Finite Element Strain Analysis for Discrete Displacement of Fold‐And‐Thrust Belt and Horst‐And‐Graben Models

An,

So,

Kim

2023

Earth and Space Science

View full text Add to dashboard Cite

The granular behaviors of shallow Earth materials are associated with brittle deformations and mass transport along the surface and fault zones. Green strain has been adopted to quantitatively analyze the fold‐and‐thrust belt and horst‐and‐graben discrete element (DE) models driven by contractional and extensional forces, respectively. Deriving Green strain from DE models is challenging because of the need for precise mapping of the discrete particle displacement onto differentiable meshes. Here, we developed a new Green strain calculation methodology, the implicit global‐finite element method (IG‐FEM), which inverts a global matrix to reflect the interconnectivity of all particles. The IG‐FEM obtains strain quantities associated with particles to prevent an information loss, for example, displacement, at particle positions. We validated the accuracy of IG‐FEM compared to the conventional strain analysis scheme for DE model, finite difference method (FDM), by estimating the relative root mean square error from analytical solutions for a given mathematical transformation. On average, the IG‐FEM and the FDM show 2.8% and 17.4% relative root mean square errors, respectively. Furthermore, the strain quantities in the fold‐and‐thrust belt and horst‐and‐graben DE models were computed. The strain patterns of the IG‐FEM are more consistent with the particle spin distribution, which is noninterpolated information of the DE models. We present the potential benefits of the IG‐FEM for assessing strain‐related values such as stress, pore pressure, and seismic velocities within brittle deformation zones. The IG‐FEM is applicable in analog models and geodetic surveys to extract strain from discrete displacement data.

show abstract

Section: Discussionmentioning

confidence: 99%

New Particle‐Wise Finite Element Strain Analysis for Discrete Displacement of Fold‐And‐Thrust Belt and Horst‐And‐Graben Models

An,

So,

Kim

2023

Earth and Space Science

View full text Add to dashboard Cite

show abstract

“…Moreover, it seems promising to try to port this software to vector processor architectures (VCPUs). VCPUs show significant progress in solving vectorizable problems like Markov chain Monte Carlo simulations [2,27,32,41]. In particular, massively parallel NEC SX-Aurora TSUBASA vector processors equipped with high-bandwidth memory (HBM) as well as the newest Intel Xeon processors with AVX-512 vector instructions may provide significant acceleration for cubic lattice Monte Carlo problems [16].…”

Section: Introductionmentioning

confidence: 99%

Efficient Implementation of Liquid Crystal Simulation Software on Modern HPC Platforms

Afanasyev

Lichmanov

Rudyak

et al. 2021

JSFI

View full text Add to dashboard Cite

In this paper we demonstrate the process of efficient porting a software package for Markov chain Monte Carlo (MCMC) simulations on a finite cubic lattice on multiple modern architectures: Pascal, Volta and Turing NVIDIA GPUs, NEC SX-Aurora TSUBASA vector engines and Intel Xeon Gold processors. In the studied software, MCMC methodology is used for simulations of liquid crystal structures, but it can be as well employed in a wide range of problems of mathematical physics and numerical methods. The main goals of this work are to determine the best software optimization strategy for this class of algorithms and to examine the speed and the efficiency of such simulations on modern HPC platforms. We evaluate the effects of various optimizations, such as using more suitable memory access patterns, multitasking for efficient utilization of massive parallelism on the target architectures, improved cache hit-rates, parallel workload balancing, etc. We perform a detailed performance analysis for each target platform using software tools such as nvprof, Ftrace and VTune. On this basis, we evaluate and compare the efficiency of the developed computational kernels on different platforms and subsequently rank these platforms by their performance. The results show that NVIDIA GPU and NEC SX-Aurora TSUBASA platforms, although at first glance seem very different, require similar optimization approaches in many cases due to similarities in data processing principles.

show abstract

Multi-GPU 3-D Reverse Time Migration with Minimum I/O

Barbosa

Coutinho

2022

Communications in Computer and Information Science

View full text Add to dashboard Cite

A workflow for seismic imaging with quantified uncertainty

Cited by 8 publications

References 21 publications

New Particle‐Wise Finite Element Strain Analysis for Discrete Displacement of Fold‐And‐Thrust Belt and Horst‐And‐Graben Models

New Particle‐Wise Finite Element Strain Analysis for Discrete Displacement of Fold‐And‐Thrust Belt and Horst‐And‐Graben Models

Efficient Implementation of Liquid Crystal Simulation Software on Modern HPC Platforms

Multi-GPU 3-D Reverse Time Migration with Minimum I/O

Contact Info

Product

Resources

About