dynamical-catalog

Load dynamical.org weather datasets in one line.

See dynamical.org/catalog for dataset documentation.

Install

pip install dynamical-catalog

uv add dynamical-catalog

Usage

import dynamical_catalog

# Optional: let us know who you are so we can improve the catalog!
dynamical_catalog.identify("you@example.com")

# Open a dataset as an xarray Dataset via its icechunk repository
ds = dynamical_catalog.open("noaa-gfs-forecast")

# Additional arguments are passed through to `xr.open_zarr`
ds = dynamical_catalog.open("noaa-gfs-forecast", chunks=None)

# Datasets with vertical profiles expose them as groups (e.g. pressure_level)
ds = dynamical_catalog.open("noaa-hrrr-forecast-48-hour-virtual", group="pressure_level")

# Get the underlying Zarr store if you want even more control
store = dynamical_catalog.get_store("noaa-gfs-forecast")
ds = xr.open_zarr(store)

# Get the icechunk repository itself to access its commit history / snapshots
repo = dynamical_catalog.get_repository("noaa-gfs-forecast")
for snapshot in repo.ancestry(branch="main"):
    print(snapshot.written_at, snapshot.message)

# List all available datasets
dynamical_catalog.list()

Set DYNAMICAL_STAC_CATALOG_URL to point the library at a non-production catalog such as https://stac-staging.dynamical.org/catalog.json.