Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Fast neural network surrogates for very high dimensional physics-based models in computational oceanography.

Rudolph van der Merwe1, Todd K Leen, Zhengdong Lu

  • 1Department of Computer Science and Electrical Engineering, OGI School of Science and Engineering, Oregon Health and Science University, Portland, OR 97006, USA.

Neural Networks : the Official Journal of the International Neural Network Society
|May 23, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Diffusion-Controlled Drug Release from Electrospun Poly(3-hydroxybutyrate) Fibers with Beaded Architecture: An Experimental and Modeling Study.

International journal of molecular sciences·2026
Same author

Quantifying the Acidification-Induced Shift of the Dimerization Equilibrium of PsbS.

The journal of physical chemistry letters·2026
Same author

From Polygenic Risk Score to Bedside: Optimising Sex-Stratified, Multiethnic Atrial Fibrillation Risk Prediction After the UK Biobank Study.

The Canadian journal of cardiology·2025
Same author

Enhancing Valve Selection and Outcome Assessment in Transcatheter Aortic Valve Replacement After Surgical Mitral Replacement: Next-step Strategies.

The Canadian journal of cardiology·2025
Same author

Computing the pH-Dependent Thermodynamics of the Allostery between Dimerization and Palmitate Binding in β-Lactoglobulin.

The journal of physical chemistry. B·2025
Same author

K<sub>D</sub>SAXS: A Tool for Analyzing Binding Equilibria With SAXS Data Using Explicit Models.

Journal of molecular biology·2025
Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Decentralized ADMM for factorization-based Low-rank matrix estimation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

We developed fast neural network surrogates to emulate complex circulation models for the Columbia River and nearby ocean. These models run over 1000x faster, improving weather and oceanography forecasts.

Area of Science:

  • Earth System Science
  • Computational Fluid Dynamics
  • Machine Learning Applications

Background:

  • Large-scale circulation models are crucial for understanding coastal and oceanic dynamics.
  • These models are computationally intensive, limiting their use in real-time applications and ensemble forecasting.
  • The Columbia River, its estuary, and near-ocean regions are complex systems influenced by multiple boundary conditions.

Purpose of the Study:

  • To develop highly accurate and extremely fast surrogate models for a large-scale circulation model.
  • To emulate the dynamics of the coupled Columbia River, estuary, and near-ocean system.
  • To enable significant advances in ensemble forecasting for oceanography and weather.

Main Methods:

  • Development of neural network surrogates.

Related Experiment Videos

  • Emulation of a nonlinear, large-scale circulation model with O(10(7)) degrees of freedom.
  • Validation against the full circulation model for accuracy and speed.

    • Main Results:

    • Neural network surrogates achieve extremely fast emulation speeds, running over 1000 times faster than the original model.
    • The surrogates accurately capture the complex dynamics of the coupled circulation system.
    • Demonstrated the feasibility of using these surrogates for large-scale simulations.

    Conclusions:

    • Fast dynamic surrogates offer a viable solution for computationally expensive circulation models.
    • The developed surrogates can significantly enhance ensemble forecasts in oceanography and weather prediction.
    • This approach paves the way for more efficient and timely environmental modeling.