A Monte-Carlo based solver for fabric evolution. It is coded in JAX and can be run on GPU, giving ~50x speedup compared to a NumPy implementation. It solves ice fabric evolution equations with:
- a contribution to lattice rotation from D, the deformation tensor
- a contribution to lattice rotation from S, the deviatoric stress tensor subject to some flow law
- A browninan motion term, corresponding to a diffusional process
- Migration recrystallization like Placidi et al. (2010)
A number of rheologies are incoprated, such as Rathmann et al. (2021) and the GOLF rheology of Gillet-Chaulet et al. (2005). A parameter set which reproduces Richards et al. (2021) is also included.
The package requires JAX and JAXlib to be installed. Instructions can be found here. The package also depends on jaxopt if a fully non-linear flow law is used.
The package can be installed by:
git clone https://github.com/dhrichards/mcfab.git
cd mcfab
pip install .import jax.numpy as jnp
TODO-
Gillet-Chaulet et al. (2005): A user-friendly anisotropic flow law for ice-sheet modeling https://doi.org/10.3189/172756505781829584
-
Placidi et al. (2010): Continuum-mechanical, Anisotropic Flow model for polar ice masses, based on an anisotropic Flow Enhancement factor https://doi.org/10.1007/s00161-009-0126-0
-
Rathmann et al. (2021): Effect of an orientation-dependent non-linear grain fluidity on bulk directional enhancement factors https://doi.org/10.1017/jog.2020.117
-
Richards et al. (2021): The evolution of ice fabrics: A continuum modelling approach validated against laboratory experiments https://doi.org/10.1016/j.epsl.2020.116718