Developments in Machine Learning Across Earth System Modeling: Subgrid-Scale Parameterizations, Emulation and Hybrid Modeling

Driscoll, Simon

NG005
Developments in Machine Learning Across Earth System Modeling: Subgrid-Scale Parameterizations, Emulation and Hybrid Modeling

Session ID#: 251817

Session Description:

Machine learning is reshaping the representation of complex physical processes in Earth system models, offering new avenues for parameterization, emulation, and hybrid modeling. This session focuses on the use of machine learning to emulate computationally expensive or unresolved processes, accelerate physical simulations, and improve representation across domains such as convection, turbulence, radiation, hydrology, sea ice, and other components of the Earth system. Topics include (but are not limited to): – Subgrid-scale parameterization via machine learning – Emulators of physical processes, model components, or whole models – Hybrid ML-physics modeling frameworks – Physics-informed neural networks, neural operators, and differentiable programming – Reinforcement learning – Cross-domain applications (atmosphere, ocean, cryosphere, land). By bringing together approaches that integrate data-driven and physically-based methods, this session will provide a critical overview of current progress and emerging directions in the application of machine learning to parameterizations, emulation, and hybrid modeling across Earth system science.

Co-Sponsor(s):

A - Atmospheric Sciences
C - Cryosphere
IN - Informatics
OS - Ocean Sciences

Index Terms:

0545 Modeling [COMPUTATIONAL GEOPHYSICS]
1622 Earth system modeling [GLOBAL CHANGE]
1942 Machine learning [INFORMATICS]
4430 Complex systems [NONLINEAR GEOPHYSICS]

Primary Convener: Simon Driscoll, University of Cambridge, Department of Applied Mathematics and Theoretical Physics, Cambridge, United Kingdom

Conveners: Sara Shamekh, New York University, Center for Atmosphere Ocean Science, Courant Institute of Mathematical Sciences, New York, United States, Akshay Subramaniam, NVIDIA Corporation, Santa Clara, CA, United States and Aakash Sane, Princeton University, Program in Atmospheric and Oceanic Sciences, Princeton, United States

Student/Early Career Convener: Karan Jakhar, Rice University, Department of Mechanical Engineering, Houston, United States; University of Chicago, Department of Geophysical Sciences, Chicago, United States

Probing the Dynamical Response of Ocean Climate Emulators (Invited) (1859610)

Adam Subel, New York University, New York, NY, United States and Laure Zanna, New York University, New York, United States

Machine Learning for sensor placement and understanding the climates of Africa and Asia (1865137)

Simon Driscoll¹, Ligin Joseph², Kieran Hunt³, Emily Black⁴ and Ross Maidment⁴, (1)University of Cambridge, Department of Applied Mathematics and Theoretical Physics, Cambridge, United Kingdom, (2)University of Southampton, School of Ocean and Earth Science, Southampton, United Kingdom, (3)National Centre for Atmospheric Science, Department of Meteorology, University of Reading, Reading, United Kingdom, (4)University of Reading, Reading, United Kingdom

Generalization of Machine-Learned Convection and Cloud Parameterization: Decadal Runs of a Hybrid GCM for Present-Day and +4K SST Climates (1886068)

Yilun Han, Scripps Institution of Oceanography, La Jolla, CA, United States, Guang Jun Zhang, Scripps Institution of Oceanography, La Jolla, United States, Yong Wang, Fudan University, Department of Atmospheric and Ocean Sciences, Shanghai, China and Hui Wan, Pacific Northwest National Laboratory, Richland, WA, United States

FedRAIN-Lite: Federated Reinforcement Algorithms for Improving Idealised Numerical Weather and Climate Models (1887278)

Pritthijit Nath¹, Sebastian Schemm^1,2, Henry Moss¹, Peter Haynes¹, Emily Shuckburgh³ and Mark Webb⁴, (1)University of Cambridge, Department of Applied Mathematics and Theoretical Physics, Cambridge, United Kingdom, (2)ETH Zurich, Institute for Atmospheric and Climate Science, Zurich, Switzerland, (3)University of Cambridge, Department of Computer Science and Technology, Cambridge, United Kingdom, (4)Met Office Hadley Centre, Exeter, United Kingdom

Machine Learning–Based Drag Parameterization in Stokes‑Flow Suspensions (1894623)

Mojtaba Hiteh¹, Xiao Yu¹ and S Balachandar², (1)University of Florida, Department of Civil and Coastal Engineering, Gainesville, United States, (2)University of Florida, Department of Mechanical & Aerospace Engineering, Gainesville, United States

JCM: A Differentiable Testbed for Hybrid Atmospheric Modeling (1895655)

Ellen Davenport¹, Varan Madan², Rebecca Gjini³, Jared Brzenski⁴, Nick Ho², Tien-Yiao Hsu⁵, Yu Liang⁶, Zhixing Liu⁷, Veeramakali Manivannan⁸, Eric Pham⁴, Rohith Vutukuru², Zhiqi Yang⁹, Rose Yu¹⁰, Nicholas Lutsko¹¹, Stephan Hoyer¹² and Duncan Watson-Parris¹⁰, (1)Scripps Institution of Oceanography, La Jolla, CA, United States, (2)University of California San Diego, La Jolla, United States, (3)Scripps Institution of Oceanography, University of California, San Diego, La Jolla, United States, (4)University of California San Diego, Scripps Institution of Oceanography, La Jolla, United States, (5)University of California Irvine, Earth System Science, Irvine, CA, United States, (6)Yale University, New Haven, CT, United States, (7)University of California Santa Barbara, Santa Barbara, United States, (8)UC San Diego, La Jolla, United States, (9)Scripps Institution of Oceanography, University of California San Diego, San Diego, United States, (10)University of California San Diego, La Jolla, CA, United States, (11)UCSD, Scripps Institution of Oceanography, La Jolla, United States, (12)Google, Mountain View, United States

Reducing Coupled Model Biases with Machine Learning Corrections from Ocean Data Assimilation Increments (1896240)

Danni Du, Princeton University, Atmospheric and Oceanic Sciences Program, Princeton, NJ, United States, Feiyu Lu, Princeton University, Program in Atmospheric and Oceanic Sciences, Princeton, United States, Alistair Adcroft, Princeton University, Princeton, United States and Laure Zanna, New York University, New York, United States

Generalizable neural-network parameterization of mesoscale eddies in idealized and global ocean models (1896408)

Pavel Perezhogin, New York University, New York, NY, United States, Alistair Adcroft, Princeton University, Princeton, United States and Laure Zanna, New York University, New York, United States

Incorporating Multivariate Consistency in ML-Based Weather Forecasting with Latent Constraints: A WC-4DVar Perspective (1912336)

Hang Fan, Columbia University, New York, United States; Shanghai Artificial Intelligence Laboratory, Shanghai, China, Ben Fei, The Chinese University of Hong Kong, Hong Kong, China, Yi Xiao, Tsinghua University, Beijing, China, Yongquan Qu, Columbia University of New York, New York, United States and Pierre Gentine, Columbia University, Department of Earth and Environmental Engineering, New York, United States

Differentiable Hybrid Neural-CFD Modeling of Spatiotemporal Dynamics in 3D Wall-Bounded Turbulence (1915286)

Xiantao Fan¹, Meet Hemant Parikh¹, Yi Liu², Meng Wang² and Jian-Xun Wang¹, (1)Cornell University, Ithaca, United States, (2)University of Notre Dame, Notre Dame, United States

Data-Driven Flux Parameterization in the Diabatic Atmospheric Boundary Layer (1931072)

Abed Hammoud, Princeton University, Princeton, NJ, United States, Elie Bou-Zeid, Princeton University, Department of Civil and Environmental Engineering, Princeton, NJ, United States, Marc Calaf, University of Utah, Salt Lake City, UT, United States, Khaled Ghannam, Northeastern University, Civil and Environmental Engineering, Boston, United States, Mitchell Bushuk, NOAA/Geophysical Fluid Dynamics Laboratory, Princeton, NJ, United States and Edriss S. Titi, University of Cambridge, Department of Applied Mathematics and Theoretical Physics, Cambridge, United Kingdom

Impact assessment of a Physics-informed Machine Learning Algorithm for Satellite, In-situ and High-frequency Radar Data Assimilation for West Coast Operational Forecast System using Joint Effort for Data Integration Framework: WCOFS-JEDI-AI (1939248)

Ling Liu, NOAA National Ocean Service, Center for Operational Oceanographic Products and Services (CO-OPS), Silver Spring, United States and Patrick Burke, NOAA National Ocean Service, Center for Operational Oceanographic Products and Services, Silver Spring, United States

CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates (1940186)

Shuchang Liu, Massachusetts Institute of Technology, Cambridge, MA, United States and Paul A O'Gorman, Massachusetts Institute of Technology, Department of Earth, Atmospheric, and Planetary Sciences, Cambridge, United States

Exploring the Dynamical and Physical Encoding of Idealised Baroclinic Waves in a Deep Learning Weather Model (1940963)

Clément Bouvier, University of Helsinki, Institute for Atmospheric and Earth Research / Physics, Helsinki, Finland, Simon Driscoll, University of Cambridge, Department of Applied Mathematics and Theoretical Physics, Cambridge, United Kingdom and Gregory J Hakim, University of Washington, Department of Atmospheric and Climate Science, Seattle, United States

Toward Causally-Constrained, Reduced Stochastic Neural Emulators of the Full Ocean (1945761)

Fabrizio Falasca, New York University, Courant Institute of Mathematical Sciences, New York, NY, United States and Laure Zanna, University of Oxford, Dept of Physics, Oxford, United Kingdom

Towards a Unified Data-Driven Boundary Layer Parameterization for Ocean and Atmosphere (1959465)

Renaud Falga, New York University, New York, NY, United States, Laure Zanna, University of Oxford, Dept of Physics, Oxford, United Kingdom and Sara Shamekh, New York University, Center for Atmosphere Ocean Science, Courant Institute of Mathematical Sciences, New York, United States

Data-driven Models for Predicting Precipitation (1959639)

Ramalingam Saravanan, Texas A&M University College Station, College Station, TX, United States, Chi-Jui Chen, Texas A&M University, College Station, United States, Courtney Schumacher, Texas A&M University, College Station, TX, United States, Istvan Szunyogh, Texas A&M University College Station, Atmospheric Sciences, College Station, United States and Raymond Wong, Texas A&M University College Station, Statistics, College Station, United States

Rewiring Climate Modeling: Integrating Machine Learning Emulators into Earth System Science (1961949)

Peter Van Katwyk, Brown University, Department of Earth, Environmental, and Planetary Sciences and Data Science Institute, Providence, RI, United States, Baylor Fox-Kemper, Brown University, Department of Earth, Environmental and Planetary Sciences, Providence, United States and Karianne J. Bergen, Brown University, Data Science Institute and Department of Earth, Environmental and Planetary Sciences, Providence, United States

Differentiable Land Model Reveals Global Environmental Controls on Latent Ecological Functions (1963541)

Jianing Fang¹, Kevin Bowman², Wenli Zhao¹, Xu Lian¹ and Pierre Gentine¹, (1)Columbia University, Department of Earth and Environmental Engineering, New York, United States, (2)Jet Propulsion Laboratory, California Institute of Technology, Pasadena, United States

Toward Interpretable Convection Parameterization from a Recurrent Neural Network (1967165)

Qiyu Song, Harvard University, Department of Earth and Planetary Sciences, Cambridge, MA, United States and Zhiming Kuang, Harvard Univ, Cambridge, United States

Can AI climate emulators quantify the statistics of unseen weather extremes? (1973648)

Alexander Wikner, University of Chicago, Chicago, United States, Amaury Lancelin, École Normale Supérieure, LMD, Paris, France, Troy Arcomano, Allen Institute for AI, Seattle, United States, Karan Jakhar, Rice University, Department of Mechanical Engineering, Houston, United States; University of Chicago, Department of Geophysical Sciences, Chicago, United States, Dhruvit Patel, University of Chicago, Geophysical Sciences, Chicago, United States, Freddy Bouchet, CNRS, Paris Cedex 16, France; LMD, IPSL, Paris, France and Pedram Hassanzadeh, University of Chicago, Department of the Geophysical Sciences, Chicago, United States

High-Resolution Climate Projections Using Diffusion-Based Downscaling of a Lightweight Climate Emulator (1978353)

Moein Darman¹, Haiwen Guan², Troy Arcomano³, Romit Maulik⁴, Ashesh Kumar Chattopadhyay⁵ and Dibyajyoti Chakraborty², (1)University of California Santa Cruz, Santa Cruz, United States, (2)Pennsylvania State University Main Campus, IST, University Park, United States, (3)Allen Institute for AI, Seattle, United States, (4)Argonne National Laboratory, Lemont, IL, United States, (5)Rice University, Houston, United States

Scalable Differentiable Programming for Sparse and Distributed Jacobians (1980207)

Aleksandra Pachalieva, Los Alamos National Laboratory, Computational Earth Science, Los Alamos, NM, United States, Andrea Liu, University of Michigan, Ann Arbor, United States, Francesc Verdugo Rojano, Vrije Universiteit Amsterdam, Department of Computer Science, Amsterdam, Netherlands and Daniel O'Malley, Los Alamos National Laboratory, Earth and Environmental Sciences Division, Los Alamos, United States

First Coupled gSAM - Neural Network Simulations to Improve Representation of Precipitation in Climate Simulations (1980619)

Griffin Mooers, Massachusetts Institute of Technology, Cambridge, MA, United States, Paul A O'Gorman, Massachusetts Institute of Technology, Department of Earth, Atmospheric, and Planetary Sciences, Cambridge, United States, Marat Khairoutdinov, Stony Brook University, School of Marine & Atmospheric Sciences, Stony Brook, NY, United States and Janni Yuval, Google, Mountain View, United States

A Convolutional Echo State AutoencodeR for High-Resolution Wind Forecasting (1986131)

Stefano Castruccio, University of Notre Dame, Applied and Computational Mathematics and Statistics, Notre Dame, IN, United States, Paola Crippa, University of Notre Dame, Civil and Environmental Engineering and Earth Sciences, Notre Dame, United States, Paolo Giani, Massachusetts Institute of Technology; Department of Earth, Atmospheric and Planetary Sciences, Cambridge, United States and Matthew Bonas, University of Notre Dame, Notre Dame, United States

Data-driven models of a coefficient in a higher-order closure of atmospheric boundary layer turbulence (1988373)

Alex Connolly¹, Benjamin Stephens², Emanuele Silvio Gentile³, Sara Shamekh⁴, Mikolaj Kowalski⁵, Jack Atkinson⁵, Vincent E Larson⁶ and Pierre Gentine⁷, (1)University of California Berkeley, Berkeley, CA, United States, (2)National Center for Atmospheric Research, Boulder, United States, (3)Princeton University, Princeton, NJ, United States, (4)New York University, Center for Atmosphere Ocean Science, Courant Institute of Mathematical Sciences, New York, United States, (5)University of Cambridge, Cambridge, United Kingdom, (6)University of Wisconsin-Milwaukee, Milwaukee, WI, United States, (7)Columbia University, Department of Earth and Environmental Engineering, New York, United States

Subgrid-scale modeling of MHD turbulence (1991258)

Dmitri A Kondrashov, University of California Los Angeles, Atmos. Sci, Los Angeles, CA, United States and Anthony Sciola, Rice University, Department of Physics and Astronomy, Houston, United States

Online learning of a machine learning cloud fraction closure in a hybrid earth system model (2000811)

Julian Schmitt, California Institute of Technology, Pasadena, CA, United States, Tom Beucler, University of Lausanne, Switzerland, Faculty of Geosciences and Environment, Institute of Earth Science Dynamics, Lausanne, Switzerland, Costa Christopoulos, California Institute of Technology, Environmental Science and Engineering, Pasadena, CA, United States and Tapio Schneider, California Institute of Technology, Environmental Science and Engineering, Pasadena, United States

A Differentiable Meshfree Framework for Hybrid ML–Physics Forward and Inverse Modeling of Geomechanics and Geophysical Flow Hazards (2001918)

Honghui Du¹, Binyao Guo², Zihan Lin² and Qizhi He³, (1)University of Minnesota, Department of Civil, Environmental, and Geo- Engineering, Minneapolis, United States, (2)University of Minnesota Twin Cities, Minneapolis, United States, (3)University of Minnesota, Department of Civil, Environmental, and Geo- Engineerin, Minneapolis, United States

Data-Driven Discovery of Constitutive Models for Ice Shelves (Invited) (1848495)

Ching-Yao Lai¹, Yongji Wang¹, David John Prior² and Charlie Cowen-Breen³, (1)Stanford University, Department of Geophysics, Stanford, United States, (2)University of Otago, Geology, Dunedin, New Zealand, (3)Massachusetts Institute of Technology, Mathematics, Cambridge, United States

Long-term Climate Emulation Using Machine Learning: A Comparison of Approaches (1853250)

Xin Ren¹, Dan John Lunt¹ and Fanny M Monteiro², (1)University of Bristol, Bristol, BS8, United Kingdom, (2)University of Bristol, Bristol, United Kingdom

Quantifying Internal Modulation of Convective Organization Using Koopman–VAE (Invited) (1848956)

Sophie Abramian, Columbia University, NYC, NY, UNITED STATES, Olivier Pauluis, New York University, Courant Institute of Mathematical Science, New York, NY, United States and Pierre Gentine, Columbia University, Department of Earth and Environmental Engineering, New York, United States

See more of: Nonlinear Geophysics

<< Previous Session Proposal | Next Session Proposal >>

NG005 Developments in Machine Learning Across Earth System Modeling: Subgrid-Scale Parameterizations, Emulation and Hybrid Modeling

Session Description:

Co-Sponsor(s):

Index Terms:

NG005
Developments in Machine Learning Across Earth System Modeling: Subgrid-Scale Parameterizations, Emulation and Hybrid Modeling