PINNICLE

Physics Informed Neural Networks for Ice and CLimatE

A Python library for solving ice sheet modeling problems using a unified framework with Physics Informed Neural Networks

---

NOTE

This project is under active development.

---

Documentation: pinnicle.readthedocs.io

Physics

• Momentum Conservation (stress balance):

  • Shelfy Stream Approximation (SSA)
  • MOno-Layer Higher-Order (MOLHO) ice flow model

• Mass Conservation (mass balance):

  • Thickness evolution

• Coupuling:

  • stress balance + mass balance

Data format

• ISSM model() type, directly saved from ISSM by saveasstruct(md, filename)
• Scattered data

More

• Install and Setup
• An example of stress balance
• An example of mass balance
• An example of coupling stress balance with mass balance