Physics-Spawned Neural Integration of Rotating Shallow-Water Dynamics
Kaushal Gianchandani, Luka Todorovic, Sai Ravela
Abstract
Equation-derived neural integration can execute prescribed dynamics without training, while informative ensemble learning can estimate uncertain parameters and identify governing interactions without requiring differentiability. Together, these approaches motivate Physics-Spawned Informative Networks (ΦSpIN), a framework for prescribed and adaptive hybrid dynamical systems. As a first step toward flood modeling applications, we present ΦSpIN-SW, its prescribed-physics realization for the rotating shallow-water equations. Compatible C-grid operators on spherical Voronoi meshes define a computational graph that retains non-polynomial operations and advances the state through recurrent neural integration compiled with JAX. Analytical benchmarks characterize numerical errors, and global and coastal experiments closely reproduce reference solutions generated using the Model Prediction Across Scales (MPAS) model. At approximately 14~km global resolution, A100, H200 and B200 GPUs attain 9-, 19- and 34-fold higher timestep throughput than a 96-core MPAS-SW CPU reference under matched workloads. These results extend training-free neural integration beyond polynomial dynamics to complex geophysical simulation, establishing a computational foundation for subsequent adjoint-free ensemble estimation of physical and neural parameters and informative adaptation of operator structure.
Create a lesson
Related papers
A Quantum Genetic Algorithm for Optimization of Benchmark Test Functions using Entangled Qutrits
Muhammad Farooq, Mehrosh Talat
A branch-free three-dimensional ghost-cell immersed boundary method with direction-averaged extrapolation kernels
Itsuki Takarabe, Ryuto Saeki, Koji Fukagata
A Heterogeneous CPU-GPU Framework for Finite Element Micromagnetic Simulations on Tetrahedral Meshes
Zekai Zheng, Zixuan Cui, Guanghui Hu
Scalable Non-Parametric Bayesian Inference of Free Energy Landscapes and Emission Functions from Observations
Eliane Briand, Helmut Grubmüller
The state of nuclear data for fusion activation simulations
Jonathan Shimwell
Mesh Sensitivity of Different Boundary Treatments in High-Order Compact Gas-Kinetic Scheme for Hypersonic Heat-Flux Prediction
Xingyi Jiang, Jiawang Zhang, Junlei Mu et al.