Home
GDRIFT provides a unified Python interface for loading geodynamic and seismic datasets used in mantle convection studies. Browse our full collection of datasets and examples in the Data Catalogue.
Community Effort for FAIR Geodynamic Data
This is a community-driven initiative to create a modular, unified package for accessing geodynamic data — usable by anyone, not only G-ADOPT. All datasets presented here have been published elsewhere; we've invested effort to unify them in one place to enable reproducibility and FAIR principles.
If you are a data author who feels there is good reason for your data not to be included here, please reach out to siavash.ghelichkhan@anu.edu.au or visit sia-g.org to contact us.
Datasets are downloaded on first use and cached in the user cache directory
(~/Library/Caches/gdrift on macOS, ~/.cache/gdrift on Linux). All datasets together
need about 7.7 GB. To use a different directory, set GDRIFT_DATA_DIR.
On a cluster, put the cache on a large shared file system. Download the datasets that you
need on a login node before you run a parallel job, for example
gdrift.download_all_datasets(["SLB_21_pyroliteCFMAS", "3d_seismic_REVEAL"]). Without an
argument, the function downloads all datasets. Compute nodes then read the files from the
cache and need no internet access.
The cache is never cleaned. A new version of a dataset is a new file, so old versions stay
until you delete the cache directory. gdrift 0.1.3 and earlier cached the files in
gdrift/data/ inside the installed package. gdrift still reads valid files from there, and
you can delete that directory too.
import gdrift
prem = gdrift.PreliminaryRefEarthModel()
thermo = gdrift.ThermodynamicModel("SLB_21", "pyroliteCFMAS")
seismic = gdrift.SeismicModel("S40RTS")
Zenodo Archival & Dataset Contributions
Reproducible Versions: GDRIFT is integrating with Zenodo to archive every major version. This ensures that if you have a script running against a specific version of GDRIFT, the datasets and functionality remain documented and accessible indefinitely.
Have a Dataset? If you have a geodynamic dataset you'd like to contribute to this community—whether for G-ADOPT or any other application—we'd be happy to help integrate it. Reach out to siavash.ghelichkhan@anu.edu.au or visit sia-g.org.