ODINN.jl documentation

Welcome to the ODINN.jl documentation, a large-scale scientific machine learning (SciML) glacier model, leveraging differentiable programming in Julia. This documentation provides the necessary information to understand the ecosystem built around ODINN.jl, its APIs, workflows and some examples of the main usage types of the model(s).

Quick install

ODINN.jl is a registered Julia package, so installing it is as easy as:

julia> using Pkg

julia> Pkg.add("ODINN")

Vision

Rather than focusing on global-scale simulations and sea-level rise contributions, ODINN.jl has, for now, a regional and catchment-scale focus, aiming to exploit the latest remote sensing and in situ observations to capture missing or subgrid glacier processes. In order to do so, ODINN.jl leverages Universal Differential Equations (UDEs), combining PDEs describing ice flow dynamics with data-driven regressors, such as neural networks [1]. For this, ODINN.jl relies heavily on the SciML Julia ecosytem and the native automatic differentiation (AD) support. Therefore ODINN.jl has a two-fold goal:

• To advance the application of scientific machine learning and differentiable programming for large-scale geoscientific modelling.
• To advance the inference of new parametrizations to characterize key missing or subgrid processes of glaciers to improve large-scale glacier simulations.

Architecture

ODINN.jl is a modular model, split into multiple packages, each one handling a specific task:

• ODINN.j is the high-level interace to the whole ODINN ecosystem, containing the SciML functionalities related to automatic differentiation and sensitivity of hybrid models, mixing differential equations and data-driver regressors.
• Huginn.jl is the ice flow dynamics module of ODINN. It contains all the information regarding glacier ice flow models, including the numerical methods to solve the PDEs using OrdinaryDiffEq.jl.
• Muninn.jl is the surface mass balance module of ODINN. It contains all the information regarding glacier interactions with the atmosphere (i.e. surface mass balance processes). For now we support simple temperature-index models, but soon we are planning to incorporate machine learning models coming from the MassBalanceMachine.
• Sleipnir.jl is the core package of ODINN, holding all the basic data structures and functions, common to the whole ecosystem. It directly reads the files provided by Gungnir.
• Gungnir is a Python package, using OGGM to retrieve all the necessary files (i.e. rasters and climate data) for the initial conditions and simulations in all the ODINN ecosystem. The user has the possibility to either store those files locally, or to use the ones we provide in a server. This is work in progress, so we will progressively cover more and more glaciers and regions in the near future.

Developers

ODINN.jl is being developed by Jordi Bolibar (CNRS, IGE), Facundo Sapienza (Stanford University), Alban Gossard (Université Grenoble Alpes, IGE) and Mathieu Le Séac'h (Université Grenoble Alpes, IGE).

Past developers include Lucille Gimenes (Université Grenoble Alpes, IGE) and Vivek Gajadhar (TU Delft).

Citing

If you use ODINN.jl for research, teaching or other activities, please use the following citation from our latest paper:

@article{bolibar_sapienza_universal_2023,
	title = {Universal differential equations for glacier ice flow modelling},
	author = {Bolibar, J. and Sapienza, F. and Maussion, F. and Lguensat, R. and Wouters, B. and P\'erez, F.},
	journal = {Geoscientific Model Development},
	volume = {16},
	year = {2023},
	number = {22},
	pages = {6671--6687},
	url = {https://gmd.copernicus.org/articles/16/6671/2023/},
	doi = {10.5194/gmd-16-6671-2023}
}

Funding

The ODINN project has been funded by an ANR TRACCS early career call, an IRGA fellowship from the Multidisciplinary Institute on Artificial Intelligence (Grenoble, France), the Nederlandse Organisatie voor Wetenschappelijk Onderzoek, Stichting voor de Technische Wetenschappen (Vidi grant 016.Vidi.171.063), the National Science Foundation (EarthCube programme under awards 1928406 and 1928374) and a TU Delft Climate Action grant.