Irina Tezaur
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Irina Tezaur has a total of 104 co-authors. Here are the top co-authors by number of shared publications:
Usdoe National Nuclear Security Administration (Nnsa)
14 shared publications- Alejandro Mota
Sandia National Laboratories Livermore California USA
13 shared publications Livermore Sandia National Laboratories (Snl-Ca)
12 shared publicationsAlbuquerque Sandia National Laboratories (Snl-Nm)
9 shared publicationsJoshua Barnett
Cadence Design Systems
8 shared publications- Diana L. Bull
Sandia National Laboratories
7 shared publications
Latest Publications
- Open Access
- 9 September 2026
- Published by Wiley in International Journal for Numerical Methods in Engineering
- Vol. 127 (17)
Hybrid Coupling With Operator Inference and the Overlapping Schwarz Alternating Method
This paper presents a hybrid approach for coupling subdomain‐local, nonintrusive Operator Inference (OpInf) reduced order models (ROMs) with each other and with subdomain‐local, high‐fidelity full order models (FOMs) using the overlapping Schwarz alternating method (O‐SAM). The proposed methodology addresses significant challenges in multiscale modeling and simulation, particularly the long runtime and complex mesh generation requirements associated with traditional high‐fidelity simulations. By leveraging the flexibility of O‐SAM, we enable the seamless integration of disparate models, meshes, and time integration schemes, enhancing computational efficiency while maintaining high accuracy. Our approach is demonstrated through a series of numerical experiments on increasingly complex three‐dimensional solid dynamics problems, showcasing speedups of up to 106 times compared to conventional FOM–FOM couplings. This work paves the way for more efficient simulation workflows in engineering applications, with potential extensions to a wide range of partial differential equations.
- 9 September 2026
- Published by Wiley in International Journal for Numerical Methods in Engineering
- Vol. 127 (17)
Open AccessHybrid Coupling With Operator Inference and the Overlapping Schwarz Alternating Method
This paper presents a hybrid approach for coupling subdomain‐local, nonintrusive Operator Inference (OpInf) reduced order models (ROMs) with each other and with subdomain‐local, high‐fidelity full order models (FOMs) using the overlapping Schwarz alternating method (O‐SAM). The proposed methodology addresses significant challenges in multiscale modeling and simulation, particularly the long runtime and complex mesh generation requirements associated with traditional high‐fidelity simulations. By leveraging the flexibility of O‐SAM, we enable the seamless integration of disparate models, meshes, and time integration schemes, enhancing computational efficiency while maintaining high accuracy. Our approach is demonstrated through a series of numerical experiments on increasingly complex three‐dimensional solid dynamics problems, showcasing speedups of up to 106 times compared to conventional FOM–FOM couplings. This work paves the way for more efficient simulation workflows in engineering applications, with potential extensions to a wide range of partial differential equations.
- Open Access
- 13 March 2026
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 18 (3)
The Arctic Coastal Erosion Model: Overview, Developments, and Calibration at Drew Point, Alaska
Permafrost coastlines are experiencing significant erosion as polar amplification has enhanced the effects of climate change in the Arctic. Warmer temperatures are increasing thermo‐denudation and more energetic oceans are increasing thermo‐abrasion in unlithified, ice‐bonded permafrost coastlines which comprise at least 40% of the circum‐Arctic coastline. Here we present developments to and calibration of the Arctic Coastal Erosion (ACE) model, which couples oceanographic and atmospheric conditions at storm‐resolving time steps with a finite element multi‐physics terrestrial permafrost model. This ice‐bonded unlithified permafrost model unites 3D thermal and mechanical governing equations by allowing heat conduction with solid‐liquid phase change to drive ice saturation, which governs evolution of mechanical stress‐strain fields. Developments to the ACE terrestrial model, including introduction of novel erosion criteria to remove failed elements, reformulation of the mechanical material model, and wave pressure boundary conditions, enable simulation of both slowly advancing thermo‐denudation with permafrost sloughing from the face and highly episodic thermo‐abrasion with niche formation and rapidly advancing block failure. A 2018 summer field campaign at Drew Point, Alaska with observations of thermo‐denudation and thermo‐abrasion, including niche geometry before block failure, enable calibration of the terrestrial model. Detailed compositional and geomechanical characterization of the ice‐bonded sediments enabled advances in the material model representation and calibrated model parameters. We demonstrate a daily root‐mean square error of 0.12 m for thermo‐denudation over the summer and achieve block failure within 2 hr of the observed. The calibrated ACE model is the first step towards simulation of other ice‐bonded unlithified circum‐Arctic coastlines for various applications.
- 13 March 2026
- Published by American Geophysical Union (AGU) in Journal of Advances in Modeling Earth Systems
- Vol. 18 (3)
Open AccessThe Arctic Coastal Erosion Model: Overview, Developments, and Calibration at Drew Point, Alaska
Permafrost coastlines are experiencing significant erosion as polar amplification has enhanced the effects of climate change in the Arctic. Warmer temperatures are increasing thermo‐denudation and more energetic oceans are increasing thermo‐abrasion in unlithified, ice‐bonded permafrost coastlines which comprise at least 40% of the circum‐Arctic coastline. Here we present developments to and calibration of the Arctic Coastal Erosion (ACE) model, which couples oceanographic and atmospheric conditions at storm‐resolving time steps with a finite element multi‐physics terrestrial permafrost model. This ice‐bonded unlithified permafrost model unites 3D thermal and mechanical governing equations by allowing heat conduction with solid‐liquid phase change to drive ice saturation, which governs evolution of mechanical stress‐strain fields. Developments to the ACE terrestrial model, including introduction of novel erosion criteria to remove failed elements, reformulation of the mechanical material model, and wave pressure boundary conditions, enable simulation of both slowly advancing thermo‐denudation with permafrost sloughing from the face and highly episodic thermo‐abrasion with niche formation and rapidly advancing block failure. A 2018 summer field campaign at Drew Point, Alaska with observations of thermo‐denudation and thermo‐abrasion, including niche geometry before block failure, enable calibration of the terrestrial model. Detailed compositional and geomechanical characterization of the ice‐bonded sediments enabled advances in the material model representation and calibrated model parameters. We demonstrate a daily root‐mean square error of 0.12 m for thermo‐denudation over the summer and achieve block failure within 2 hr of the observed. The calibrated ACE model is the first step towards simulation of other ice‐bonded unlithified circum‐Arctic coastlines for various applications.
- 22 December 2025
- Published by Wiley in International Journal for Numerical Methods in Engineering
- Vol. 126 (24)
Kernel Manifolds: Nonlinear‐Augmentation Dimensionality Reduction Using Reproducing Kernel Hilbert Spaces
This paper generalizes recent advances on quadratic manifold (QM) dimensionality reduction by developing kernel methods‐based nonlinear‐augmentation dimensionality reduction. QMs, and more generally feature map‐based nonlinear corrections, augment linear dimensionality reduction with a nonlinear correction term in the reconstruction map to overcome approximation accuracy limitations of purely linear approaches. While feature map‐based approaches typically learn a least squares optimal polynomial correction term, we generalize this approach by learning an optimal nonlinear correction from a user‐defined reproducing kernel Hilbert space. Our approach allows one to impose arbitrary nonlinear structure on the correction term, including polynomial structure, and includes feature map and radial basis function‐based corrections as special cases. Furthermore, our method has relatively low training cost and has monotonically decreasing error as the latent space dimension increases. We compare our approach to proper orthogonal decomposition and several recent QM approaches on data from several example problems.
- 22 December 2025
- Published by Wiley in International Journal for Numerical Methods in Engineering
- Vol. 126 (24)
Kernel Manifolds: Nonlinear‐Augmentation Dimensionality Reduction Using Reproducing Kernel Hilbert Spaces
This paper generalizes recent advances on quadratic manifold (QM) dimensionality reduction by developing kernel methods‐based nonlinear‐augmentation dimensionality reduction. QMs, and more generally feature map‐based nonlinear corrections, augment linear dimensionality reduction with a nonlinear correction term in the reconstruction map to overcome approximation accuracy limitations of purely linear approaches. While feature map‐based approaches typically learn a least squares optimal polynomial correction term, we generalize this approach by learning an optimal nonlinear correction from a user‐defined reproducing kernel Hilbert space. Our approach allows one to impose arbitrary nonlinear structure on the correction term, including polynomial structure, and includes feature map and radial basis function‐based corrections as special cases. Furthermore, our method has relatively low training cost and has monotonically decreasing error as the latent space dimension increases. We compare our approach to proper orthogonal decomposition and several recent QM approaches on data from several example problems.
- 22 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computing in Science & Engineering
- Vol. 27 (3) , 5-6
European Seminar on Computing 2024
At the ninth European Seminar on Computing (ESCO 2024) (https://esco2024.femhub.com/), an interdisciplinary conference dedicated to modern methods and practices of scientific computing, more than 200 participants from 17 countries gathered in Pilsen, Czech Republic on 10–14 June 2024. The conference began in 2008 and recurs every two years. Its primary significance is that it brings together, in a relatively small and intimate setting, cutting-edge computational science and real-world applications. Under the overarching theme of “Modern Methods and Practices of Scientific Computing,” the 2024 installment of ESCO convened researchers in advanced computational methods across a diverse range of scientific domains. The program featured seven minisymposia, four software workshops, and a diverse set of thematic areas, including computational electromagnetics, fluid–structure interaction, multiphase flows, computational chemistry and quantum physics, computational civil and structural engineering, computational biology and bioinformatics, computational geometry and topology, mathematics education and outreach, petascale and exascale computing, hydrology, wave propagation, climate and weather modeling, GPU and cloud computing, uncertainty quantification, applied statistics, and stochastic models in machine learning.
- 22 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computing in Science & Engineering
- Vol. 27 (3) , 5-6
European Seminar on Computing 2024
At the ninth European Seminar on Computing (ESCO 2024) (https://esco2024.femhub.com/), an interdisciplinary conference dedicated to modern methods and practices of scientific computing, more than 200 participants from 17 countries gathered in Pilsen, Czech Republic on 10–14 June 2024. The conference began in 2008 and recurs every two years. Its primary significance is that it brings together, in a relatively small and intimate setting, cutting-edge computational science and real-world applications. Under the overarching theme of “Modern Methods and Practices of Scientific Computing,” the 2024 installment of ESCO convened researchers in advanced computational methods across a diverse range of scientific domains. The program featured seven minisymposia, four software workshops, and a diverse set of thematic areas, including computational electromagnetics, fluid–structure interaction, multiphase flows, computational chemistry and quantum physics, computational civil and structural engineering, computational biology and bioinformatics, computational geometry and topology, mathematics education and outreach, petascale and exascale computing, hydrology, wave propagation, climate and weather modeling, GPU and cloud computing, uncertainty quantification, applied statistics, and stochastic models in machine learning.