AguaTrack-ARCO-SA

Binary Zarr v3 dataset. Use xarray + HfFileSystem to access.

from huggingface_hub import HfFileSystem
import xarray as xr

fs    = HfFileSystem()
store = fs.get_mapper("datasets/NTU-CompHydroMet-Lab/wam2layers/1996.zarr")
ds    = xr.open_zarr(store, consolidated=False)

See code tutorials in GitHub

How to cite

Companion code — archived on Zenodo (DOI: 10.5281/zenodo.20036507):

@software{aguatrack_arco_sa_tutorial_v0_1_0,
  author    = {Lin, Sung Che and Hung, Ho Tin},
  title     = {{AguaTrack-ARCO-SA Tutorial}},
  version   = {v0.1.0},
  date      = {2026-05-05},
  publisher = {Zenodo},
  doi       = {10.5281/zenodo.20036507},
  url       = {https://doi.org/10.5281/zenodo.20036507},
}

Dataset. Cite the AguaTrack-ARCO-SA dataset directly from HuggingFace — the dataset card has its own canonical citation block: AguaTrack/AguaTrack-ARCO-SA (DOI: 10.57967/hf/8650). The monthly and yearly aggregates at AguaTrackSA/AguaTrack-ARCO-SA-Aggregated are derived from the daily store and share the same DOI.

@misc{ho_tin_hung_2026,
    author       = { Ho Tin Hung and Sung Che Lin },
    title        = { AguaTrack-ARCO-SA (Revision a433c0f) },
    year         = 2026,
    url          = { https://huggingface.co/datasets/AguaTrack/AguaTrack-ARCO-SA },
    doi          = { 10.57967/hf/8650 },
    publisher    = { Hugging Face }
}