Ice-flow model emulator based on physics-informed deep learning

2023/09/26 by Guillaume Jouvet, Guillaume Cordonnier · 34 citations
Earth and Planetary Sciences · Engineering · Environmental Science · Mathematics · Physics and Astronomy · #Artificial intelligence #Computational science #Computer science #Convolutional neural network #Cryospheric studies and observations #Data assimilation #Deep learning #Engineering #Fidelity #Flow (mathematics) #Geology #Geometry #Geomorphology #Glacier #High fidelity #Landslides and related hazards #Machine learning #Mathematics #Meteorology #Model Reduction and Neural Networks #Simulation #Variety (cybernetics)

paper · pdf · doi:10.1017/jog.2023.73

published in Journal of Glaciology, 1-15 (Cambridge University Press)

openalex publication_date 2023/09/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/02

Abstract

Abstract Convolutional neural networks (CNN) trained from high-order ice-flow model realisations have proven to be outstanding emulators in terms of fidelity and computational performance. However, the dependence on an ensemble of realisations of an instructor model renders this strategy difficult to generalise to a variety of ice-flow regimes found in the nature. To overcome this issue, we adopt the approach of physics-informed deep learning, which fuses traditional numerical solutions by finite differences/elements and deep-learning approaches. Here, we train a CNN to minimise the energy associated with high-order ice-flow equations within the time iterations of a glacier evolution model. As a result, our emulator is a promising alternative to traditional solvers thanks to its high computational efficiency (especially on GPU), its high fidelity to the original model, its simplified training (without requiring any data), its capability to handle a variety of ice-flow regimes and memorise previous solutions, and its relatively simple implementation. Embedded into the ‘Instructed Glacier Model’ (IGM) framework, the potential of the emulator is illustrated with three applications including a large-scale high-resolution (2400x4000) forward glacier evolution model, an inverse modelling case for data assimilation, and an ice shelf.

Citations

• Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Velocity and stress fields in grounded glaciers: a simple algorithm for including deviatoric stress gradients

Cited by

• A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI
• Topographic controls on asynchronous glacial maxima in the European Alps during the Late Pleistocene
• Prediction of Sea Ice Velocity and Concentration in the Arctic Ocean using Physics-informed Neural Network
• Snapshot and time-dependent inversions of basal sliding using automatic generation of adjoint code on graphics processing units
• Modeling Sediment Fluxes From Debris‐Rich Basal Ice Layers
• Physics-constrained generative machine learning-based high-resolution downscaling of Greenland's surface mass balance and surface temperature
• Increased Glacier Melt Across Millennia to Hours Enhances Erosion and Sediment Export Processes
• A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI. [europepmc]
• Topographically-controlled contribution of avalanches to glacier mass balance in the 21st century. [europepmc]
• Global glacier-free topography reveals a large potential for future lakes in presently ice-covered terrain. [europepmc]
• A data-consistent model of the last glaciation in the Alps achieved with physics-driven AI [europepmc]

Related

• Graph Neural Network as Computationally Efficient Emulator of Ice-sheet and Sea-level System Model (ISSM)
• Graph Neural Networks for Emulation of Finite-Element Ice Dynamics in Greenland and Antarctic Ice Sheets
• A physics informed neural network approach to simulating ice dynamics governed by the shallow ice approximation
• KAN-GCN: Combining Kolmogorov-Arnold Network with Graph Convolution Network for an Accurate Ice Sheet Emulator
• A Hybrid Deep Neural Operator/Finite Element Method for Ice-Sheet Modeling