The objective of these tools are to convert external field data to a format SUHMO can read & output with for the sake of increasing the realism of SUHMO's simulations.
The three parameters that are being evaluated by these tools are bedrock elevation, ice thickness, & geothermal flux.
Bedrock elevation & ice thickness data are derived from NASA MEaSUREs BedMachine. Geothermal flux is from the International Heat Flow Commission (IHFC) 2024.
NASA BedMachine stores Greenland data as a NetCDF (.nc) file. Both parameters are subregioned & interpolated into the same file as BedMachineGreenland-v5.nc.
Geothermal flux is stored as a .xslx (Excel spreadsheet) file. It is converted to .nc utilizing the pandas module.
All three parameters will then be compiled as one .nc file that maps out Greenland's parameters, Greenland_params.nc. This is the file used for the subregion & interpolation.
TODO: Figure out if we combine all three params as one file or subregion & interpolate them separately, then combine as one HDF file at the end.
This README is SOLELY for the preprocessing tools, which can be run without cloning SUHMO. Documentation for SUHMO is available here. The SUHMO governing equations and core algorithm are described in: https://ennadelfen.github.io/SUHMO/Model.
Before starting with preprocessing, ensure CHOMBO_HOME & SUHMO_HOME have exports from the instructions above.
These tools are meant to be automatic, that is, run a singular command with your desired parameters on the size of the plot. The tools take that input, take a subregion of said size & interpolate BedMachineGreenland-v5.nc, & then input it into SUHMO to output a simulation that can be visualized through the VisIt tool.
- pathlib
- scipy
- numpy
- matplotlib
- netCDF4
- h5py
- xarray
- os (for SUHMO inputting)
- ncview (OPTIONAL for easy .nc file viewing)
The easiest method would be to install an Anaconda (or miniconda) environment with these packages. The most important packages are numpy, scipy, netCDF4, & h5py; the rest of them were used in test files & were not integrated (so far) into the main file.
BedMachineGreenland-v5.nc, which can be downloaded via NASA MEaSUREs. Global Heat Flow Database, which can be downloaded via IHFC's website as a .zip file using the 2024 version.
cd SUHMO-Preprocessing-Tools
python preprocess.py
It will then ask you to input bounding box coordinates, as the parameters are such below.
preprocess(x1, y1, x2, y2)IMPORTANT: INPUT x, y values as floats or integers. x1, y1 are the lower-left coordinates & x2, y2 as the upper-right coordinates, forming a rectangle subregion.
BedMachineGreenland-v5.nc's grid size is from x = 10,218 & y = 18,346
The preprocess command should automatically be subregioned & interpolated, then output one hdf5 file as greenland_FINAL.hdf
Specific coordinates for the Isunnguata Sermia glacier approximation: (-2740906, -375354) and (-2845385, -195722)
- Mentors: Dan Martin, Anna Felden, Anjali Sandip
- Dr. Krti Tallam for lending her preprocessing tools
- Stephen for his BISICLES preprocessing tools
- NASA MEaSUREs BedMachine team for their NetCDF Greenland v5 file
- International Heatflow Commission for their Global Heat Database
- NOAA/NESDIS/STAR Aerosols and Atmospheric Composition Science Team