PyGEM-notebooks

Included in this repository are a series of Jupyter notebooks intended to demonstrate the functionality of the Python Glacier Evolution Model (PyGEM).

Notebooks

Model testing

The following Jupyter notebooks are intended to allow for introduction and testing of PyGEM and may be run using sample data that should have been downloaded during model installation and setup (see here), but can also be downloaded directly here.

• simple_test: PyGEM test run using provided sample data for Khumbu Glacier
  
• advanced_test: a more advanced PyGEM test run, demonstrating Bayesian inference calibration and simulation, using provided sample data for Khumbu Glacier
  
• advanced_test_tw: demonstrates calibration of the frontal ablation parameterization, using provided sample data for LeConte Glacier
  

Supplemental

• simple_test_daily: PyGEM test notebook to demonstrate the ability to run the model at a daily or monthly timestep using provided sample data for Gulkana Glacier
  
• dhdt_processing: demonstrates the ability to process 2d dhdt data into glacierwide mass balance rates and binned elevation change for use in model calibration using provided sample data for Gulkana Glacier
  
• advanced_test_spinup_elev_change_calib: demonstrates the ability to run glacier thickness inverison, dynamical spinup, and calibrate against binned elevation change data using provided sample data for Gulkana Glacier
  

Note: The model testing notebooks listed above will only run on the specified glaciers, unless all the necessary input data for other glaciers of interest are separately downloaded or exist elsewhere on one's computer (in which case, update the root path in ~/PyGEM/config.yaml). See below for futher information.

Further demonstration

The notebooks listed below serve as more advanced demonstration of PyGEM. However, running these notebooks requires more comprehensive datasets, as the sample data utilized in the above notebooks are subset to the two specified test glaciers. Complete PyGEM datasets used to run these notebooks (e.g., climate, reference mass balance, frontal ablation, etc.) can be downloaded here.

• run_calibration: demonstrates how to run PyGEM calibration routines
  
• run_simulation: demonstrates how to run individual or large-scale (regional) glacier simulations
  
• analyze_regional_change: demonstrates how to aggregate individual glacier simulations by region and climate scenario, and analyze the resulting regional change
  
• analyze_mcmc: demonstrates how to analyze the prior and posterior model parameters for an individual glacier following MCMC model calibration
  
• analyze_mcmc_regional: demonstrates how to analyze the performance of MCMC calibration for an entire region
  

Wrappers

Within the wrappers subdirectory, one can find example PyGEM wrappers scripts for executing calibration and/or simulation. These are useful for large-scale model runs.