This release contains the code necessary to run the experiments described in "The demise of the world's largest piedmont glacier" by Brinkerhoff et al.
To run the experiments download the data repository found at https://doi.org/10.5281/zenodo.15028490 into examples/malaspina and unzip.
Next, run the following scripts contained in examples/malaspina/preprocessing/ in order:
mesh_boundary.py
build_bed.py
build_surface.py
build_velocity.py
build_beta.py
build_adot.py
Next, navigate to examples/malaspina/inference
and run
inference.py
with desired settings.
Once this is complete, the next step is to produce Hessian-vector product samples by running
sample_hvp.py
Ensembles can be generated with
run_ensemble.py
plots can be produced via the scripts found in examples/malaspina/postprocessing