USACM Technical Thrust Areas

 

Committee: Energy & Earth Systems

Description: The USACM Energy & Earth Systems TTA encompasses a broad range of topics involving the modeling and analysis of Earth systems and energy applications.  Topics relevant to the TTA include: climate and Earth system modeling; climate couplings; geomechanics; wind, solar and nuclear energy; carbon sequestration, capture and storage; oil/gas recovery.  The TTA will promote advancements in both traditional first principles-based modeling and simulation, as well as the development of innovative data-driven model reduction and analysis methods, applied to the aforementioned domains.

Chair: Irina Tezaur, Sandia National Laboratories
Vice-Chair: Steve WaiChing Sun, Columbia University
Members-at-Large: Yuri Bazilevs, Brown University
Clint Dawson, The University of Texas at Austin

 

View Members

 

Webinar Series

June 23, 2026 8AM PDT

Join via Zoom: https://us06web.zoom.us/j/89805920868?pwd=UaxOZeVHZAs9BeKkjqRWbNadZYCIZo.1

Speaker: Shamina Shahrin Hossain-McKenzie, Sandia National Laboratories

Title: griDNA: AI-powered Distributed Anomaly Detection

Abstract: In this talk, the griDNA cyber-physical security technology will be presented as well as field testing results at a utility photovoltaic site. griDNA is designed to protect distributed critical infrastructure systems such as power grids, pipelines, and transportation networks against disruptions caused by cyber-attacks or system failures. The technology continuously gathers and analyzes system data using autoencoder neural networks to fuse together cyber-physical data streams. For detected events, griDNA alerts operators and defenders with early detection of cyber, physical, or combined cyber-physical abnormal behavior, whether it stems from inadvertent equipment failures or malicious cyber attacks.

Bio: Dr. Shamina Hossain-McKenzie is a grid cybersecurity researcher at Sandia National Laboratories focusing on creating cybersecurity tools and analysis frameworks that are uniquely suitable for application in critical infrastructure systems by leveraging cyber-physical modeling and AI-based analytics. She received her B.S. in electrical engineering from Washington State University in 2012 and her M.S. and Ph.D. degrees in electrical engineering from University of Illinois at Urbana-Champaign in 2014 and 2017, respectively. She currently leads projects that focus on incorporating physics-based data into cybersecurity analysis, correlating cyber and physical detected events, coordinating cyber and physical mitigations, and methods to share and visualize cyber-physical alerts for system operators and cyber defenders. This includes the development of distributed cyber-physical situational awareness sensors, cyber-physical federated learning-based SOAR platforms, and cyber-physical interdependency analysis frameworks.

---

Past Webinars

Spring 2026

May 19, 2026: Elizabeth Barnes, Boston University

Title: Explainable AI for Earth System Prediction: From Black Box to Window into the Future

Abstract: The coupled, complex nature of the Earth system makes prediction across timescales, from weeks to decades, incredibly challenging. In this talk, I will discuss our progress in developing and applying AI tools that go beyond black-box prediction to extract scientific insight. First, I will show how explainable AI (XAI) methods applied to climate prediction networks can reveal sources of predictability, expose climate model biases, and constrain future projections using observations. Second, I will demonstrate how deep-learning weather and climate emulators are enabling fundamentally new computational experiments, from reconstructing 500 years of tropical cyclone activity to generating targeted ensembles of rare, high-impact events. Finally, I will argue that understanding how AI models learn is as important as evaluating what they predict, presenting recent work on training dynamics that reveals when and why models acquire or lose key capabilities during training, and how this opens the door to targeted interventions that improve model performance.

Bio: Dr. Elizabeth Barnes, Dalton Family Chair in Environmental Data Science & Sustainability and Professor of Computing & Data Sciences and of Earth & Environment at Boston University.

April 21, 2026: Marianna Maiaru, Columbia University

Title: Enabling Next-Generation Wind Blades Through Process-Aware Composite Design (talk not recorded by request of the speaker)

Abstract: Composite materials are essential to the continued expansion of wind energy, enabling the development of increasingly large and efficient turbine blades. However, as blade lengths surpass 100 meters, significant challenges arise in manufacturing variability, defect control, fatigue reliability, and recyclability. These challenges are largely rooted in the manufacturing process itself, where resin infusion, curing, and thermal gradients introduce microstructural heterogeneity, residual stresses, and defects that ultimately dictate structural performance and lifetime. This seminar presents recent advances in the process modeling of composite materials for wind energy, focusing on a predictive, integrated approach to address these challenges. Particular emphasis is placed on physics-based modeling of thermoset curing for manufacturing and repair, as well as thermoplastic processing, enabling the prediction of residual stresses and their impact on mechanical behavior and fatigue performance. By embedding these capabilities within a multi-scale ICME framework, the work establishes direct links between processing conditions, microstructure, and structural response, providing a pathway to reduce uncertainty and enhance reliability in large-scale blades. Ultimately, this work aims to transition composite blade design from empirical, conservative practices to a predictive, process-driven paradigm, enabling the next generation of high-performance, durable, and sustainable wind energy systems.

Bio: Marianna Maiaru (Associate Professor in Civil Engineering and Engineering Mechanics at Columbia University) is an expert in Integrated Computational Materials Engineering (ICME), process modeling, and computational mechanics. She received her Ph.D. in Aerospace Engineering as a collaboration between Politecnico di Torino in Italy and the University of Michigan. Her research interests include composite structures, damage mechanics, multi-scale analysis, higher-order finite elements, and additive manufacturing. Maiaru has received numerous grants from NASA, NSF, and the Air Force, including the AFOSR Young Investigator Program award in 2020 and the NSF CAREER award in 2022. She received the DEStech Young Researcher Award in 2021 and the AIAA ICME Prize in 2020 and 2022.

March 31, 2026: Dr. Hadi Hajibeygi, Delft University of Technology

Title: Energy Storage in Salt Caverns (talk not recorded by request of the speaker)

Abstract: Salt caverns have been created for decades through solution mining, primarily to extract salt for use specially in the chemical industry. After salt extraction, the empty caverns are either abandoned or repurposed as storage tanks. Currently, more than 2,000 caverns are in use in North America for energy storage. Notably, feedstock hydrogen has been stored in caverns for over five decades. Moreover, for energy transition, the hydrogen economy is expected to significantly scale up the utilization of Caverns. Salt rocks are heterogeneous with complex deformation mechanisms, distinct from porous rocks, specially under cyclic loading. Therefore, investigating their mechanical behavior at different scales is crucial for their safe design, operation, and abandonment. We present our robust method and simulation to characterize their complex elastic-plastic mechanics under cyclic loading. Based on this, stability, performance, and interactions of 3D cavern structures are investigated. Our models and methods are made available to the public through our in-house open-source "SafeInCave" simulator.

Bio: Hadi Hajibeygi is Energi Simulation Chair and professor of Geo-Energy Solid and Fluid Mechanics at TU Delft, where he also leads the Subsurface Storage Theme. He served as the deputy/science lead of the IEA TCP 42 on Underground Hydrogen Storage. His research interests include multiscale modeling and simulation relevant to geoscience and geo-energy applications. Hadi holds PhD with medal from the Institute of Fluid Dynamics of ETH Zurich and was a postdoctoral scholar at Stanford Energy Resources Engineering before joining TU Delft in 2013. 

February 24, 2026: Rebecca Gjini, University of California, San Diego

Title: The Ensemble Kalman Inversion Race

Abstract: Ensemble Kalman methods were initially developed to solve nonlinear data assimilation problems in oceanography, but are now popular in applications far beyond their original use cases. Of particular interest is climate model calibration. As hybrid physics and machine-learning models evolve, the number of parameters and complexity of parameterizations in climate models will continue to grow. Thus, robust calibration of these parameters plays an increasingly important role. We focus on learning climate model parameters from minimizing the misfit between modeled and observed climate statistics in an idealized setting. Ensemble Kalman methods are a natural choice for this problem because they are derivative-free, scalable to high dimensions, and robust to noise caused by statistical observations. Given the many variants of ensemble methods proposed, an important question is: Which ensemble Kalman method should be used for climate model calibration? To answer this question, we perform systematic numerical experiments to explore the relative computational efficiencies of several ensemble Kalman methods. The numerical experiments involve statistical observations of Lorenz-type models of increasing complexity, frequently used to represent simplified atmospheric systems, and some feature neural network parameterizations. For each test problem, several ensemble Kalman methods and a derivative-based method "race" to reach a specified accuracy, and we measure the computational cost required to achieve the desired accuracy. We investigate how prior information and the parameter or data dimensions play a role in choosing the ensemble method variant. The derivative-based method consistently fails to complete the race because it does not adaptively handle the noisy loss landscape.

Bio: Rebecca is a 5th-year PhD candidate in the Institute of Geophysics and Planetary Physics within the Scripps Institution of Oceanography at the University of California, San Diego.  She started her PhD in 2021 after receiving her Bachelor of Science in mathematics from Lehigh University.  Her current work focuses on derivative-free parameter estimation and uncertainty quantification for understanding the climate system under the guidance of her advisor, Matthias Morzfeld.

January 27, 2026: Stefan Henneking, The University of Texas at Austin

Title: 2025 Gordon Bell Winner: A Real-Time Goal-Oriented Bayesian Inversion-Based Digital Twin for Tsunami Early Warning Applied to the Cascadia Subduction Zone

Abstract: We present a digital twin (DT) for tsunami early warning in the Cascadia subduction zone (CSZ). This DT assimilates pressure data from seafloor sensors into an acoustic-gravity wave equation model, solves an inverse problem to infer spatiotemporal seafloor deformation, and forward predicts tsunami wave heights. The entire end-to-end data-to-inference-to-prediction computation is carried out in real time through a Bayesian framework that rigorously accounts for uncertainties. Creating such a DT is challenging due to the enormous size and complexity of both the forward and inverse problems. For example, a discretization of the spatiotemporal seafloor velocity in the CSZ—the parameter field to be inferred—gives rise to a system with one billion parameters. Using current methods, computing the posterior mean alone would require more than 50 years on 512 GPUs. We exploit the shift invariance of the parameter-to-observable map and devise novel parallel algorithms for fast offline-online decomposition. The offline component requires just one adjoint wave propagation per sensor; the PDE solver is implemented with MFEM and exhibits excellent scalability to 43,520 GPUs on LLNL’s El Capitan system. Fast Hessian applications are enabled by an FFT-based algorithm for the resulting block Toeplitz matrices. Using this framework, the Bayesian inverse solution and wave height forecasts are computed in 0.2 seconds, representing a ten-billion-fold speedup over state-of-the-art methods.

Reference:

Stefan Henneking, Sreeram Venkat, Veselin Dobrev, John Camier, Tzanio Kolev, Milinda Fernando, Alice-Agnes Gabriel, and Omar Ghattas. 2025. Real-Time Bayesian Inference at Extreme Scale: A Digital Twin for Tsunami Early Warning Applied to the Cascadia Subduction Zone. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC '25). Association for Computing Machinery, New York, NY, USA, 60–71. https://doi.org/10.1145/3712285.3771787

Bio: Stefan Henneking is a Research Associate at the Oden Institute for Computational Engineering and Sciences at The University of Texas at Austin. His research focuses on theory and HPC-enabled computational methods for large-scale Bayesian inversion-based digital twins, with an emphasis on wave propagation problems, including applications in acoustic sensing, fiber optics, and tsunami early warning. He holds a BS (Computational Engineering) degree from FAU Erlangen-Nuremberg, an MS (Computational Science & Engineering) degree from Georgia Tech, and MS and PhD (Computational Science, Engineering, & Mathematics) degrees from UT Austin. He is a recipient of the 2025 ACM Gordon Bell Prize.

Fall 2025

December 9, 2025: Nicole Aretz, The University of Texas at Austin

Title: Multifidelity uncertainty quantification for ice sheet simulations

Abstract: Simulations of ice mass loss in Greenland and Antarctica have a central role when choosing policies to combat climate change, but suffer from vast parametric uncertainties that propagate into ice mass loss projections. For example, the basal friction field and geothermal heat flux are typically inferred in an ill-posed inverse problem, but their variation has significant long-term effects on ice velocity and temperature simulations. Quantifying the propagated uncertainties in the projections is of utmost importance to enable judicious decision-making, but high-fidelity simulations are too expensive to allow Monte Carlo approximations. Less expensive but also less accurate low-fidelity models are readily available — e.g., coarser meshes, approximated physics, physics-based reduced-order models, machine-learning methods, interpolation, and extrapolation — but the replacement of the entire high-fidelity model for a low-fidelity surrogate introduces model bias. Multifidelity methods, in contrast, keep the high-fidelity model in place but expand the estimator to shift the computational burden onto the low-fidelity models while still guaranteeing an unbiased estimate. Through this exploit of the model hierarchy, the multifidelity estimates guarantee a smaller statistical error than Monte Carlo sampling for the same computational budget. In this talk, we demonstrate and compare three multifidelity estimators — multifidelity Monte Carlo, multilevel Monte Carlo, or the multilevel best linear unbiased estimator technique — on a model of the 2015-2050 Greenland ice mass loss. For a target accuracy equivalent to 1 mm sea level rise contribution at 95 % confidence, the multifidelity estimators achieve computational speedups of two orders of magnitude.

The results presented in this talk are based on:

[1] Nicole Aretz, Max Gunzburger, Mathieu Morlighem, Karen Willcox. Multifidelity uncertainty quantification for ice sheet simulations. Comput Geosci 29, 5 (2025). https://doi.org/10.1007/s10596-024-10329-3

Bio: Nicole Aretz is a postdoctoral fellow at the Oden Institute for Computational Engineering and Sciences at the University of Texas at Austin. She received her Ph.D. in Mathematics in 2022 from RWTH Aachen University. In her work, Nicole combines different uncertainty quantification methods for enabling digital twins. She is particularly interested in inverse problems, multi-fidelity approximations, and optimal experimental design. To achieve computational speed-up at scale, she employs physics-based reduced-order modeling approaches, such as the Operator Inference method. In recent years, Nicole has focused on ice sheet simulations as target application for demonstrating and applying her work.

November 11, 2025: William C. Radünz, Johns Hopkins University

Title: Multiscale simulations of atmosphere–wind farm interactions and the role of spatial gradients on power performance and wake recovery

Abstract: Understanding the multiscale physics of atmosphere–wind farm interactions is one of the central challenges in modern wind energy science. This talk highlights two complementary studies that reveal how the existence or lack of spatial gradients, whether induced by terrain or under-resolved in simulations, govern wind farm performance and wake recovery.

The first part focuses on the American WAKE ExperimeNt (AWAKEN), where multiscale large-eddy simulations with the Weather Research and Forecasting model (WRF–LES) and observations from an onshore wind farm reveal unexpected spatial variability in turbine power. Despite operating in nominally simple terrain, hub-height wind speeds vary by nearly 4 m s⁻¹ over just 5 km during nocturnal low-level jet events, resulting in downwind turbines outperforming upwind ones by 25–50 %. These streamwise gradients arise from terrain-induced vertical displacements of the jet core, revealing how even gentle orography can strongly modulate intra-farm performance.

The second part examines idealized offshore wind farms simulated with the WRF model equipped with the Fitch Wind Farm Parameterization (WFP), benchmarked against large-eddy simulations. Results show that mesoscale models systematically underestimate wake recovery—not due to excessive dissipation, but because coarse grids fail to resolve the spatial wind velocity gradients that sustain turbulence via shear production. This under-resolution of gradients in the near-farm wake propagates biases downstream, limiting predictive skill for cluster-scale effects.

Together, these studies demonstrate that spatial gradients, whether physical or numerical, are fundamental to understanding atmosphere–wind farm interactions. Accurately representing these gradients is essential for capturing the performance variability of onshore wind farms across both simple and complex terrain, as well as for predicting wake effects in future large wind farm clusters.

Bio: William is a postdoctoral researcher at the Whiting School of Engineering’s Ralph O’Connor Sustainable Energy Institute (ROSEI), Johns Hopkins University, with a PhD in Mechanical Engineering from the Federal University of Santa Catarina (UFSC), Brazil. For over 10 years, he has been passionate about advancing wind energy research. His work focuses on understanding the interactions between offshore wind farms and the atmosphere, particularly developing parameterizations for Numerical Weather Prediction (NWP) and climate models. In the past, he explored how terrain influences wind resource variability and impacts wind farm performance. More recently, his research has delved into multiscale large-eddy simulations (LES) of stable boundary layers (SBLs) and low-level jets (LLJs) as part of the American WAKE ExperimeNt (AWAKEN), led by NREL. Looking ahead, he aims to establish a research group dedicated to advancing wind energy technologies and addressing the challenges of wind energy and atmosphere interactions.

October 14, 2025: Rachael Alfant, Sandia National Laboratories

Title: Performant Optimization of Power Systems: Resiliency, Operations, and Planning

Abstract: Both day-to-day and years-long planning are critical to operating and maintaining power grids. The prevalence of extreme weather events, such as wildfires and floods, present uncertainties that complicate these planning efforts. Stochastic programming is a well-suited optimization paradigm for these power grid modeling problems, where the scenarios may represent different realizations of load or renewables generation. This talk presents a progressive hedging algorithm that leverages multi-fidelity modeling and scenario bundling to balance the tradeoff between realism and computational tractability, including a unit commitment exemplar. In addition, this talk proposes several ways to quantify resilience for power grids under compound hazard conditions, comprised of physical threats (such as heatwaves) and deliberate cyber-attacks. 

Bio: Rachael Alfant is a postdoctoral appointee in the Discrete Math and Optimization department at Sandia National Laboratories. Her research focuses on optimization under uncertainty applied to power systems, particularly as it relates to resiliency to extreme weather events, cyber-attacks, and rising electricity demands. She received her PhD in Computational Math and Operations Research from Rice University, and her BA in Mathematics from the University of Chicago. She currently serves as a Mentoring and Outreach Liaison for the INFORMS Forum for Women in Operations Research & Management Sciences; she previously served as the Vice President and President of the Rice University Chapters of the Association of Women in Mathematics and Society for Industrial & Applied Mathematics, respectively. 

September 30, 2025: Xiaojing (Ruby) Fu, California Institute of Technology

Title: Phase-field modeling of microstructure evolution of icy porous media (talk not recorded by request of the speaker)

Abstract: The microstructure of icy geomaterials—including snow, firn, sea ice, and permafrost—governs key geosystem processes by controlling thermal–hydraulic behavior, mechanical strength, and reflectivity. These porous materials undergo continuous microstructural evolution driven by local gradients in temperature, salinity, curvature, and by external fluid flow. While numerous laboratory studies have quantified the evolution of material properties under varying environmental conditions, computational approaches remain limited.

In this talk, I explore the use of phase-field framework for modeling microstructural evolution in icy porous materials. First, I show how the model framework reproduces laboratory observations of snow sintering and captures the role of liquid water in wet snow metamorphism. I then discuss its extension toward permafrost. Second, I introduce a variational phase-field model that incorporates temperature- and salinity-dependent thermodynamics to simulate interactions between ice and salty water. Model fidelity is validated against published experiments. We then conduct mm-scale simulations of sea ice melting in the low-Reynolds regime, capturing key phenomena such as salt release from brine pockets and stratification, providing a basis for connecting microscale processes to large-scale sea ice melt dynamics.

Bio: Xiaojing (Ruby) Fu is an assistant professor in Mechanical and Civil engineering at California Institute of Technology, where she leads the group on Mechanics and Physics of Porous Media Flow. Her group combines mathematical theory, computation and laboratory experiments to advance our predicative capability of field-scale applications in a wide range of geoscience problems in hydrology, subsurface energy and planetary sciences. A primary focus of her current work is on understanding the physics of unsaturated water flow through icy porous media.

Spring 2025

May 20, 2025; Jeremy Bassis, University of Michigan

Title: Stability of ice shelves in a changing climate

Abstract: Ice shelves are freely floating extensions of marine ice sheets that buttress and limit the discharge of grounded ice into the ocean.  Like the lithosphere and other geophysical materials, these ice shelves simultaneously flow through ductile creep and fail through brittle fracture.  Ductile creep is flow and steady, but brittle failure can result in spectacularly rapid changes.  For example, a warming ocean appears to have driven a sequence of calving events resulting in the transition of the western portion of the Thwaites ice shelf, located in West Antarctica, from a largely intact tongue to a completely fragmented remnant ice shelf.  Surprisingly, despite over 75 years of study, significant controversy remains about the interaction of brittle and ductile processes, including basic questions like how wide and deep should crevasses in an ice shelf penetrate?  Here we revisit this question, using numerical models combined with a (very old fashioned) weakly-nonlinear asymptotic expansion.  We show that our numerical simulations are broadly consistent with our weakly-nonlinear analysis, although discrepancies start to emerge at high amplitudes.  Moreover, we find that relatively small amounts of bottom melt can lead to a substantial “roughening” of the ice shelf as systems of crevasses extend deeper and become wider.  Furthermore, our analysis suggests that once basal melt exceeds a critical threshold, ice shelves become unstable and must eventually disintegrate.  However, the time scale for this to occur ranges from decades to centuries. Given current forcing, it is possible that several ice shelves are essentially ghosts that have already transitioned to unstable states with their future demise inevitable unless ocean forcing decreases.  

Bio: Jeremy Bassis is a Professor at the University of Michigan who studies ice sheet dynamics and evolution, past present and future.  Jeremy is also the director of the Climate Impacts and Solutions Program, a Masters of Engineering program that trains physical scientists and engineering students how to work with communities and industry to apply climate information in decision making and design.  

April 22, 2025: Assistant Professor Hannah Lu, The University of Texas at Austin

Title: Data-Driven Modeling for Multiphase Flows in Porous Media

Abstract: Data-driven modeling of complex systems is a rapidly evolving field facilitated by the concurrent rise of data science. To alleviate the prohibitively expensive computational costs of repeated full-model simulations in uncertainty quantification and model calibration, data-driven modeling is often used to describe the behaviors of the complex system by predicting the quantities of interest directly. In this talk, I will present my contributions to this field with an emphasis on (1) improving model performance by using physics-aware machine learning techniques, (2) quantifying uncertainties in the system’s response, and (3) inferring the key parameters of the physics-based models from measured data. Examples of applications will be focused on multiphase flows in porous media. The objective is to develop a convenient computing toolbox to provide more accurate scientific information at lower computational costs for better environmental management and decision-making.

April 8, 2025: Prof. Baskar Ganapathysubramanian, Iowa State University

Title: Engineering the Future of Food (Security): Computational Mechanics for Crop Design

Abstract: Ensuring global food security amidst population growth and climatic variability demands new thinking in crop design. This talk highlights how advances in computational mechanics --- powered by high-performance computing (HPC) and informed by scientific machine learning (SciML) --- can transform the way we model, simulate, and optimize plant architectures, or ideotypes. We present computational approaches that enable large-scale, high-fidelity simulations of plant structures, including implicit geometric representations, octree-based meshes, and the shifted boundary method for embedded domains. These methods allow for rapid generation of diverse canopy realizations, allowing exploration of how variations in leaf shape and arrangement influence photosynthetic efficiency and resource use. By coupling these mechanics-based models with SciML-driven surrogates, we enable fast, data-driven optimization loops that accelerate the discovery of resilient, high-yield ideotypes. Such approaches can inform conventional agriculture, controlled-environment agriculture, and even for growing food in space. Several opportunities exist at the intersection of HPC, computational mechanics, and scientific ML to push the frontiers of plant biology and plant breeding research which can ultimately contribute to feeding a growing global population, while providing food and nutritional security.

Bio: Baskar Ganapathysubramaniam is Anderlik Professor of Engineering at Iowa State University. Baskar received his BTech from IIT Madras, and a PhD from Cornell University. He directs a curiosity driven, computational sustainability group (me.iastate.edu/bglab) with research interests in the areas of scientific computing, applied mathematics, and machine learning with applications in food, energy, and healthcare systems. He is the director of the NSF/USDA funded AI Institute for Resilient Agriculture (aiira.iastate.edu) which is a multi-institutional project focused on use-inspired AI developments.

February 25, 2025: Valerio Lucarini, University of Leicester

Title: Detecting and Attributing Change in Climate and Complex Systems: Foundations, Green's Functions, and Nonlinear Fingerprints

Abstract: Detection and attribution (D&A) studies are cornerstones of climate science, providing crucial evidence for policy decisions. Their goal is to link observed climate change patterns to anthropogenic and natural drivers via the optimal fingerprinting method (OFM). We show that response theory for nonequilibrium systems offers the physical and dynamical basis for OFM, including the concept of causality used for attribution. Our framework clarifies the method's assumptions, advantages, and potential weaknesses. We use our theory to perform D&A for prototypical climate change experiments performed on an energy balance model and on a low-resolution coupled climate model. We also explain the underpinnings of degenerate fingerprinting, which offers early warning indicators for tipping points. Finally, we extend the OFM to the nonlinear response regime. Our analysis shows that OFM has broad applicability across diverse stochastic systems influenced by time-dependent forcings, with potential relevance to ecosystems, quantitative social sciences, and finance, among others.

Key References:
V. Lucarini and M. D. Chekroun, Detecting and Attributing Change in Climate and Complex Systems: Foundations, Green's Functions, and Nonlinear Fingerprints, Phys. Rev. Lett. 133, 244201 (2024)https://doi.org/10.1103/PhysRevLett.133.244201
V. Lucarini and M. D. Chekroun, Theoretical tools for understanding the climate crisis from Hasselmann’s programme and beyond, Nat. Rev. Phys. 5, 744 (2023) https://doi.org/10.1038/s42254-023-00650-8

Bio: Valerio Lucarini is a leading expert in climate dynamics, nonequilibrium statistical mechanics, extreme events, tipping points. He is currently Professor in Applied Mathematics at the University of Leicester, UK. Before that, he was Professor of Statistical Mechanics at the University of Reading, UK, and Professor of Theoretical Meteorology at the University of Hamburg, Germany. He the author of >160 papers and 2 books. He has supervised over 12 PhD students and 15 PostDocs. Recipient of Whitehead Prize from LMS; Arne Richter Award and Richardson Medal from EGU; AGU Lorenz Lecture; SIAM Mathematics of Planet Earth Prize, IUGG Keilis-Borok medal. He is a Fellow of Academia Europaea. He currently chairs the Topical Group Physics of Climate of the American Physical Society. He holds editorial roles for Physical Review Letters and Physical Review E.

January 28, 2025: Dr. Artem Korobenko, University of Calgary

Title: Advances in modeling wind turbines with variational multiscale methods: from blade-resolved simulation to reduced-order modeling

Abstract: Accelerating the deployment and scientific advancement of wind energy systems requires the development of predictive multi-fidelity numerical tools integrated into their design, operation, and management. In this talk, recent advances in variational multiscale (VMS) methods that address various technological and scientific challenges in wind energy applications will be presented. These challenges include high-Reynolds number turbulent flows in complex domains, wake-structure interactions, complex topography, stratification, and more. The numerical framework developed by the CFSMgroup at the University of Calgary (https://www.cfsmgroup.com/) utilizes the incompressible Navier-Stokes equations, along with a transport equation for temperature field (for stratified flows). The VMS method operates as an LES-like approach, eliminating the need for filters or artificial dissipation. The formulation supports both linear finite elements and NURBS discretization. Additionally, the VMS formulation is coupled with the Actuator Line Model (ALM), referred to as ALM-VMS formulation, as a potential tool for modeling wind turbine farms. Finally, the consistent reduced-order model (ROM-VMS) for wind energy applications will be presented. The robustness and accuracy of this framework will be demonstrated through a range of challenging applications, including simulations of horizontal-axis and vertical-axis turbines, at full-scale and full material and geometrical complexity, wind farm modeling, flow over complex terrains, and more.

Bio: Dr. Artem Korobenko is an Associate Professor and Associate Head, Research at the Department of Mechanical and Manufacturing Engineering at the University of Calgary (Canada). He holds a Schulich Research Chair and leads the Computational Fluids and Structural Mechanics Group (CFSMgroup). Dr. Korobenko earned his PhD in 2014, followed by a postdoctoral position (2016), both at the University of California San Diego. His research focuses on the development of multi-fidelity computational methods for the analysis and design of complex systems in aerospace, wind and marine engineering using large-scale computing. A Fulbright Alumni and Alexander von Humboldt Fellowship recipient, Dr. Korobenko is a founding member and current president of the Canadian Association for Computational Science and Engineering, as well as a Member-at-Large of the USACM Technical Thrust Area on Computational Fluid Dynamics and Fluid-Structure Interaction. He is also a founding member and co-director of the University of Calgary Aerospace Network.

Fall 2024

December 17, 2024: Dr. D. Todd Griffith, The University of Texas at Dallas

Title: Multidisciplinary Challenges in the Design and Operation of Offshore Wind Energy Systems

Abstract: Wind energy installations have shown strong growth in recent decades, especially on land, while offshore deployments are also gaining momentum. Wind turbines are the largest rotating structures in the world (e.g., the GE Haliade X wind turbine produces 14MW with blades longer than a football field at 107 meters in length). Despite the progress, new research is needed to develop the next generation of large-scale, reliable wind energy systems, especially in the nascent offshore environment. This presentation will cover multidisciplinary challenges in the design and operation of offshore wind systems. Several recent and ongoing studies will be presented that include design of floating offshore wind systems, new concepts such as floating offshore vertical axis wind turbines, and multi-fidelity wind turbine digital twins for asset management.

Bio: Dr. D. Todd Griffith is a Professor of Mechanical Engineering at the University of Texas at Dallas. Griffith leads a research group focused on developing new technology solutions in wind turbine design and structural dynamics. Prior to joining UT Dallas in Fall of 2017, Griffith was a researcher (Principal Member of the Technical Staff) at Sandia National Laboratories. There, he was the Technical Lead for Sandia’s Offshore Wind Energy Program, responsible for developing and leading national projects for the US Department of Energy. Prior to joining Sandia, he completed PhD work at Texas A&M University in Aerospace Engineering, and BS and MS degrees in Mechanical Engineering from the University of Kentucky. Dr. Griffith is the co-founder and Deputy Director of the UT-Dallas Center for Wind Energy.

November 12, 2024: Prof. Ning Lin, Princeton University

Title: Tropical Cyclone Hazards and Risk in a Changing Climate

Abstract: Tropical cyclones (TCs) cause much damage and loss of life worldwide. The impacts of TCs may worsen in the coming decades due to rapid coastal development coupled with sea-level rise and possibly increasing TC activity due to climate change. Here we discuss about TC hazard projection and risk management in a holistic modeling framework. First, we introduce probabilistic TC models that can be used to generate large numbers of synthetic storms with physically correlated characteristics under projected climate conditions. Second, we discuss about TC wind, rainfall, and surge hazard modeling, and the coupling with the TC models to estimate individual and compound hazard probabilities in a changing climate. Then, we discuss about infrastructure vulnerability modeling, and the coupling with the TC hazard projection to estimate future TC risk and develop risk management strategies. We focus on two examples, namely, TC-blackout-heatwave compound risk and adaptive coastal protection.

Bio: Ning Lin is a Professor of Civil and Environmental Engineering at Princeton University, where she has affiliate appointments with Princeton School for Public and International Affairs, Andlinger Center for Energy and Environment, High Meadows Environmental Institute, and Department of Geosciences. Lin’s research areas include Natural Hazards and Risk Analysis, Wind Engineering, Coastal Engineering, and Climate Change Impact and Adaptation. Her current primary focus is hurricane risk analysis. She integrates science, engineering, and policy to study hurricane-related weather extremes (strong winds, heavy rainfall, and storm surges, and compounding sea level rise and heatwaves), how they change with changing climate, and how their impact on society can be better mitigated. Lin has published in high-impact journals including Science, Nature Climate Change, and Proceedings of the National Academy of Sciences. She is a recipient of CAREER award from National Science Foundation (NSF), Natural Hazards Early Career Award and Global Environmental Change Early Career Award from American Geophysical Union, Huber Research Prize from American Society of Civil Engineers, and The Walter Orr Roberts Lectureship by American Meteorological Society (“for pioneering physics-based weather risk analysis by integrating state-of-the-art weather and risk modeling to understand hurricane hazards under climate change”). Lin has been the lead PI or Co-PI for several large NSF projects, including Interdisciplinary Research in Hazards and Disasters (Hazards SEES), Prediction of and Resilience against Extreme Events (PREEVENTS), and Coastlines and People Hubs for Research and Broadening Participation (CoPe). Lin received her Ph.D. in Civil and Environmental Engineering from Princeton University in 2010. She also received a certificate in Science, Technology and Environmental Policy in 2010 from Princeton. Before rejoining Princeton as an assistant professor in 2012, she conducted research in the Department of Earth, Atmospheric and Planetary Sciences at MIT as a NOAA Climate and Global Change Postdoctoral Fellow.

October 15, 2024: Dr. Alejandro Mota, Sandia National Laboratories

Title: Assessing permafrost demise and infrastructure destabilization using the Arctic Coastal Erosion (ACE) model

Abstract: The Arctic holds one-third of the world's coastline and faces rapid, episodic coastal erosion that current permafrost erosion tools fail to fully explain. In this talk, we introduce the Arctic Coastal Erosion (ACE) model, a novel multi-physics finite element framework designed to simulate permafrost degradation in Arctic coastal regions. The ACE model integrates two key components: A solid mechanics model that calculates 3D stress, strain, and displacement in permafrost, using a plasticity model dependent on frozen water content; and an innovative thermal model that governs 3D heat conduction and the solid-liquid phase transitions within the permafrost.

These components are sequentially coupled through a thermo-mechanical scheme, implemented in the open-source Albany/LCM finite element code. This approach allows us to simulate deformation-induced failures, such as block failure, thermo-denudation, and thermo-abrasion, based on constitutive relationships rather than empirical assumptions. To capture transient erosion events, the model dynamically removes elements from the finite element mesh.

The model's capabilities are demonstrated using a pseudo-realistic scenario of a permafrost slice at Drew Point, Alaska, exposed to actual oceanic and atmospheric conditions from July 2018. This cutting-edge model provides new insights into episodic permafrost erosion processes, offering a more comprehensive understanding of these complex phenomena.

Bio: Alejandro Mota is a Principal Member of the Technical Staff in the Mechanics of Materials Department at Sandia National Laboratories in Livermore, CA. He holds a PhD in Structural Engineering with a concentration in Theoretical and Applied Mechanics from Cornell University.

Dr. Mota's career includes the following contributions to computational solid mechanics. At Caltech, he developed and implemented advanced finite element methods to simulate the fracture and fragmentation of brittle and ductile materials, particularly ceramics and metals, under high-speed impact loads. His research extended to constitutive modeling of ductile metals (porous plasticity) and simulations for medical and material applications, including firearm injury to the human cranium, kidney stone fragmentation (lithotripsy) and traumatic brain injury, as well as the fracture mechanics of steel-polyurea composites.

At Sandia National Laboratories, Dr. Mota has focused on regularization methods for finite element simulations and developed constitutive models for damage, failure, fracture, and fragmentation in elastic and inelastic materials under finite deformations. His work also includes implementing variational mapping schemes for field transfer between finite element discretizations and the development of multiscale and multigrid finite element methods based on the Schwarz alternating method. He is actively involved in projects such as Arctic Coastal Erosion (ACE) and Arctic Critical Infrastructure (ACI), contributing to the understanding of climate change impacts in polar regions.

His research interests include multi-physics simulations, computational methods for fracture and damage mechanics, mesh-free methods, GPU computing, and the integration of variational principles in numerical modeling.

October 1, 2024: Ronaldo Borja, Stanford University

Title: The poromechanics of shale

Abstract: Shale is a clastic sedimentary rock consisting of softer materials such as clay and organics, stiffer minerals such as quartz, feldspar, and pyrite, and void spaces that can range in size from nanometers to millimeters. It is the most common sedimentary rock on Earth, accounting for approximately 75% of rock in sedimentary basins. My talk will explore the mechanical and hydraulic properties of this material that could have some important implications for hydrocarbon extraction and carbon sequestration. Of particular interest are the impacts of fluids in the pore spaces on the evolution of the stiffness, strength, and fluid conductivity properties of this geologic material.

Bio: Ronaldo Borja works in theoretical and computational geomechanics, geotechnical engineering, and geosciences. His research includes the development of mathematical and computational frameworks for multiscale and multi-physical processes in geomechanics and related fields. He is the author of a textbook entitled Plasticity Modeling and Computation published by Springer and serves as executive editor of two journals in his field: the International Journal for Numerical and Analytical Methods in Geomechanics and Acta Geotechnica. Ronaldo Borja is the recipient of the 2016 ASCE Maurice A. Biot Medal for his work in computational poromechanics.

Spring 2024

May 13, 2024: Roger Ghanem, University of Southern California

Title: Modeling Extremes of Powergrid Generation with PLoM

Abstract: I will describe recent efforts at characterizing extreme behaviors of power grid generation associated with rare events such as blackouts. I will introduce a probabilistic learning on manifolds (PLoM) paradigm that permits us to locate these rare events and construct from them meaningful statistical inferences.  In particular, we are interested in power generation profiles, throughout a powergrid network, associated with these extremes.  For this application, PLoM is trained on a dataset consisting of optimal power generation solutions, computed using PYoMo, in response to demand data collected hourly over a one year period. Augmenting this data with a weather component (temperature, wind, cloud cover) permits our current inferences to extend from power generation profiles to weather profiles.

Biography: Roger Ghanem holds the Tryon Chair in Stochastic Methods and Simulation at the University of Southern California where he is Professor in the Departments of Civil \& Environmental Engineering and Mechanical & Aerospace Engineering.  Ghanem obtained his PhD from Rice University in 1989 in Civil Engineering. He held faculty positions at SUNY-Buffalo and Johns Hopkins University before moving to the University of Southern California in 2005. He is an expert in uncertainty quantification (UQ) and scientific machine learning (SciML). He has published over 180 refereed Journal publications and over 180 refereed conference papers. Ghanem's research has been funded by NSF, ONR, AFOSR, DARPA, DOE, NRC, NEUP, Sandia, LLNL, in addition to a number of industries. Ghanem is member of FASTMATH, a US Department of Energy SciDAC Institute. He has organized UQ Summer School at USC from 2010-2024 and the UQ/ML Workshop at USC in 2018 and 2019. Ghanem is President of the International Association for Structural Safety and Reliability, has served as President of the Engineering Mechanics Institute of ASCE, on the Executive Council of USACM, and as Chair of the UQ SIAG of SIAM. He is currently a member iof the US National Committee on Theoretical and Applied Mechanics. Dr. Ghanem is Fellow of AAAS, EMI, SIAM, USACM, IACM, and is a Distinguished Member of ASCE. His research has been recognized by a number of awards from ASCE, USACM, and IASSAR.

April 15, 2024: Prof. Joris Degroote, Ghent University

Title: Fluid-Structure Interaction Simulations in Wind Energy

Abstract: Wind energy is essential in the ongoing transition to a sustainable energy system and the technology in this field is constantly evolving. The horizontal axis wind turbine is increasing in size more rapidly than previously expected, resulting in blades of more than 100m long and an increased importance of aeroelastic effects. Also emerging technologies like airborne wind energy, consisting of a tethered aircraft or kite, are influenced by aeroelasticity. Hence, techniques and models to simulate the fluid-structure interaction (FSI) in these wind energy converters have been developed. This presentation focuses on partitioned simulation of FSI, referring to the coupling of a flow solver with a structural solver, and also the connection with a controller in the case of airborne wind energy. Overset techniques are used to handle the rigid body motion and the deformation in the wind subdomain, as well as to facilitate the meshing process. With these techniques, the deformation of a wind turbine blade during a wind gust and in proximity of the tower has been investigated, and the trajectory and wing deflection of an airborne wind energy system have been simulated.

Biography: Joris Degroote obtained his PhD from Ghent University (Belgium) in 2010. He did research stays of 1 year at the Massachusetts Institute of Technology (USA) as PhD student and 3 months at Technische Universität München (Germany)as post-doctoral researcher. He became associate professor at Ghent University in 2013 and full professor in 2020. His research focuses on simulation of fluid-structure interaction (FSI) and is thus purely numerical. Fundamental aspects of algorithm development and applications in mechanical energy engineering are both investigated. He developed the CoCoNuT coupling software, an object-oriented open-sourcecode for partitioned simulation of coupled problems, containing several quasi-Newton coupling techniques. He is (co)author of 175 journal publications in Web of Science and is the (co)supervisor of 17 completed and 11 ongoing PhDs.

March 18, 2024: Prof. Lou Durlofsky, Stanford University

Title: Data assimilation and optimization frameworks for CO2 storage operations

Abstract: There are many challenges associated with achieving carbon storage at gigatonne scales. In this talk, I will present some of our recent developments in two areas relevant for the reservoir engineering of CCUS projects. A deep-learning-based surrogate model for data assimilation/history matching will be presented. This surrogate model, which involves an extension of a previously developed recurrent-residual U-Net architecture, is extended to treat coupled flow-geomechanics problems. Its ability to model pressure and CO2 plume locations in new aquifer realizations, and displacement at the Earth’s surface, will be demonstrated. The surrogate model is then applied for MCMC-based history matching. Next, I will describe aframework for optimizing CO2 storage operations using derivative-free algorithms. Different objective functions (involving minimization of mobile CO2 and maximization of storage efficiency), along with a range of practical constraints, are treated. A multifidelity optimization approach will be shown to be effective and to provide improved computational efficiency. 

Bio: Louis J. Durlofsky is the Otto N. Miller Professor of Earth Sciences in the Department of Energy Science and Engineering at Stanford University. He codirects the Stanford Smart Fields Consortium and the Stanford Center for Carbon Storage. Earlier in his career, Durlofsky was affiliated with Chevron Energy Technology Company. He holds a BS degree from Penn State, and MS and PhD degrees from MIT, all in chemical engineering. His research interests include subsurface flow simulation and optimization, history matching, uncertainty quantification, and deep-learning-based surrogate modeling.

February 19, 2024: Prof. Lea Jenkins, Clemson University

Title: Water Sustainability: Satisfying the Thirst of Stakeholders

Abstract: Allocation of existing water supplies has become critically important in recent years, as overuse, in conjunction with severe levels of drought, have placed aquifers in jeopardy.  The imbalances in aquifer levels are especially dire in regions whose economies are heavily dependent on agriculture, as irrigation of crops accounts for more than 80% of groundwater resources.

Bio: Dr. Lea Jenkins graduated from NCSU with a Ph.D. in mathematics and is currently a professor in the School of Mathematical and Statistical Sciences at Clemson University.  Her research interests center on mathematical applications; she is particularly motivated by problems which allow her to work in an interdisciplinary environment.  She is a member of a research team whose work on mathematics used to help drought-stricken farmers in California was featured in a PBS NewsHour Science Friday segment, "How Math Is Growing More Strawberries in California", and an NSF Discovery article, "Strawberries With a Thirst".  Her other projects include simulation-based optimization and modeling and for membranes used for protein and virus separation in the pharmaceutical industry and analysis of higher order temporal methods for nonlinear transport problems.

January 22, 2024: Ravindra Duddu, Vanderbilt University

Title: Modeling Glacier and Ice-shelf Flow and Fracture Processes using Computation, Data and Machine Learning
Presentation Slides

Abstract: The dynamic mass loss due to ice flow from the Antarctic and Greenland ice sheets directly into oceans is the greatest source of uncertainty in predicting sea level rise. Fracture propagation in glaciers and ice shelves can accelerate ice flow and cause the detachment of icebergs, thus significantly influencing the mass loss from ice sheets. It has been hypothesized that hydrofracturing of glaciers and ice shelves followed by enhanced basal sliding and ice cliff failure could contribute to rapid sea level rise over the coming centuries. Therefore, it is important that we improve our understanding of the fracture mechanics of ice shelves and glaciers, and better represent their dynamics in Earth system models. In this presentation, I will provide an overview of our recent work on simulating fracture propagation in glaciers and ice shelves using continuum damage mechanics models. First, I will describe phase-field and cohesive fracture models to investigate the propagation of water-filled crevasses in glaciers and ice shelves. Second, I will discuss a two-scale cohesive fracture model that incorporates turbulent fluid flow and accounts for melting/refreezing in fractures to explore the conditions enabling rapid supraglacial lake drainage events. Third, I will present a nonlocal creep damage model incorporated in a shallow shelf approximation to simulate the processes controlling rift paths in ice shelves, which leads to large tabular iceberg calving events. For each modeling study, I will showcase the use of physics-based computation and data from laboratory experiments and field/satellite observations and discuss the important findings. I will end with some remarks on the use of machine learning models in assessment and prediction and future extensions needed to improve the representation of ice sheet fracture over decadal to century timescales in Earth system models.

Bio: Ravindra Duddu got his B. Tech in Civil Engineering from the Indian Institute of Technology Madras. He obtained his M.S. and Ph.D. in Civil and Environmental Engineering from Northwestern University. He worked as postdoctoral researcher at the University of Texas at Austin Institute for Geophysics and Columbia University in the City of New York. Currently, he is an Associate Professor of Civil and Environmental Engineering at Vanderbilt University, with secondary appointments in Mechanical Engineering and Earth and Environmental Sciences. His research interests are in computational solid mechanics with a focus on multi-physics modeling of material damage evolution. in the field of glaciology, he has pioneered the use of continuum damage mechanics models for simulating fracture processes in glaciers and ice sheets. He is a recipient of the US NSF early CAREER award, Fulbright Kalam-Climate Fellowship, UK The Royal Society International Exchanges travel award, and ONR Summer Faculty Fellowship. He is a member of American Society of Civil Engineers (ASCE), American Geophysical Union, United States Association for Computational Mechanics, and American Society of Mechanical Engineers (ASME). He is the Chair of technical committees on Computational Mechanics and Fracture and Failure Mechanics associated with ASCE EMI and ASME IMECE.

Acknowledgements: This work is funded by the NSF Antarctic Glaciology and NASA Cryosphere Science programs.