Generalizable Implicit Neural Representations via Parameterized Latent Dynamics for Baroclinic Ocean Forecasting
This work addresses the computationally prohibitive problem of high-resolution ocean forecasting for climate and weather prediction, representing an incremental advance in neural representation methods for domain-specific applications.
The paper tackles the challenge of simulating mesoscale ocean dynamics at high resolution by introducing PINROD, a framework that combines implicit neural representations with parameterized neural ODEs to efficiently model nonlinear oceanic behavior across varying conditions, achieving superior accuracy over baselines and improved computational efficiency compared to standard simulations.
Mesoscale ocean dynamics play a critical role in climate systems, governing heat transport, hurricane genesis, and drought patterns. However, simulating these processes at high resolution remains computationally prohibitive due to their nonlinear, multiscale nature and vast spatiotemporal domains. Implicit neural representations (INRs) reduce the computational costs as resolution-independent surrogates but fail in many-query scenarios (inverse modeling) requiring rapid evaluations across diverse parameters. We present PINROD, a novel framework combining dynamics-aware implicit neural representations with parameterized neural ordinary differential equations to address these limitations. By integrating parametric dependencies into latent dynamics, our method efficiently captures nonlinear oceanic behavior across varying boundary conditions and physical parameters. Experiments on ocean mesoscale activity data show superior accuracy over existing baselines and improved computational efficiency compared to standard numerical simulations.