Skip to content

timghill/glads-borehole-calibration

Repository files navigation

GlaDS borehole calibration

Tim Hill, 2025 ([email protected]) https://github.com/timghill/glads-borehole-calibration

Code corresponding to "Emulator-based Bayesian calibration of a subglacial drainage model". This project uses a Gaussian Process emulator and borehole hydraulic head measurements to calibrate 8 uncertain parameters of the Glacier Drainage System (GlaDS) model (Werder et al., 2013). This repository consists of code to run the GlaDS simulation ensembles, run the MCMC sampling to infer the posterior probability distributions, and make figures for the published paper.

Description

The project structure is:

  • issm/: ISSM-GlaDS simulation directories
  • data/: Processed daily borehole flotation-fraction measurements
  • analysis/: Calibration experiments sub-directories
  • utils/: Shared code for consistent simulation settings and MCMC sampling

Each directory has its own README.md file that describes the contents in more detail.

Requirements

Running these scripts requires python 3. The analysis source code has been tested against python 3.12.4. Package requirements are listed in requirements.txt, and it is recommended to use a virtual environment to manage versions. For example

virtualenv --python 3.12.4 pyenv/
source pyenv/bin/activate
pip install -r requirements.txt

Running ISSM requires a working installation of ISSM v4.24.

This code depends on a fork of the SEPIA package (timghill/SEPIA) that that can be installed using pip install -e ..

Installation

The source code in utils/ is installed as an editable python package. To install this code, use pip with the -e option:

pip install -e .

Your python environment and installation can be verified by running test_install.sh. This script should run with no errors and should update two figures in experiments/synthetic/analysis/figures/.

About

Bayesian calibration of a subglacial drainage model using borehole water pressure data

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published