DataTree.to_netcdf has h5netcdf hardcoded as default

[OriolAbril]

What happened?

Up until the last release, it was possible to do dt.to_netcdf("filename.nc") without specifying the engine independently of which of netcdf4 of h5netcdf were installed. After #10571 engine=None is interpreted as h5netcdf independently of available libraries. This diverges from the other to_netcdf methods and goes against the docstring of DataTree.to_nectdf:

engine ({"netcdf4", "h5netcdf"}, optional) – Engine to use when writing netCDF files. If not provided, the default engine is chosen based on available dependencies, with a preference for “netcdf4” if writing to a file on disk.

What did you expect to happen?

dt.to_netcdf("filename") works even if only netcdf4 is installed and h5netcdf is not.

Minimal Complete Verifiable Example

ds1 = xr.Dataset({"foo": "orange"})
dt = xr.DataTree(name="root", dataset=ds1)

dt.to_netcdf("file.nc")
# full traceback below

dt.to_netcdf("file.nc", engine="netcdf4") # works

MVCE confirmation

• Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
• Complete example — the example is self-contained, including all data and the text of any traceback.
• Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
• New issue — a search of GitHub Issues suggests this is not a duplicate.
• Recent environment — the issue occurs with the latest version of xarray and its dependencies.

Relevant log output

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
    dt.to_netcdf("file.nc")
    ~~~~~~~~~~~~^^^^^^^^^^^
  File "/home/oriol/bin/miniforge3/envs/xarray-minimal/lib/python3.13/site-packages/xarray/core/datatree.py", line 1834, in to_netcdf
    return _datatree_to_netcdf(
        self,
    ...<9 lines>...
        **kwargs,
    )
  File "/home/oriol/bin/miniforge3/envs/xarray-minimal/lib/python3.13/site-packages/xarray/core/datatree_io.py", line 77, in _datatree_to_netcdf
    ds.to_netcdf(
    ~~~~~~~~~~~~^
        target,
        ^^^^^^^
    ...<7 lines>...
        **kwargs,
        ^^^^^^^^^
    )
    ^
  File "/home/oriol/bin/miniforge3/envs/xarray-minimal/lib/python3.13/site-packages/xarray/core/dataset.py", line 2109, in to_netcdf
    return to_netcdf(  # type: ignore[return-value]  # mypy cannot resolve the overloads:(
        self,
    ...<10 lines>...
        auto_complex=auto_complex,
    )
  File "/home/oriol/bin/miniforge3/envs/xarray-minimal/lib/python3.13/site-packages/xarray/backends/api.py", line 2078, in to_netcdf
    store = store_open(target, mode, format, group, **kwargs)
  File "/home/oriol/bin/miniforge3/envs/xarray-minimal/lib/python3.13/site-packages/xarray/backends/h5netcdf_.py", line 169, in open
    import h5netcdf
ModuleNotFoundError: No module named 'h5netcdf'

Anything else we need to know?

I would propose as fix instead of if engine is None: engine="h5netcdf" to call xarray.backends.api._get_default_engine_netcdf() instead, but I also skimmed #10571 and I am not sure if we should remove scipy as an output of the function when neither netcdf4 nor h5netcdf are installed.

Environment

INSTALLED VERSIONS

commit: None
python: 3.13.5 | packaged by conda-forge | (main, Jun 16 2025, 08:27:50) [GCC 13.3.0]
python-bits: 64
OS: Linux
OS-release: 6.14.0-27-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: en_GB.UTF-8
LOCALE: ('en_GB', 'UTF-8')
libhdf5: 1.14.6
libnetcdf: 4.9.2

xarray: 2025.8.0
pandas: 2.3.1
numpy: 2.3.2
scipy: None
netCDF4: 1.7.2
pydap: None
h5netcdf: None
h5py: None
zarr: None
cftime: 1.6.4
nc_time_axis: None
iris: None
bottleneck: None
dask: None
distributed: None
matplotlib: None
cartopy: None
seaborn: None
numbagg: None
fsspec: None
cupy: None
pint: None
sparse: None
flox: None
numpy_groupies: None
setuptools: None
pip: 25.2
conda: None
pytest: None
mypy: None
IPython: None
sphinx: None

[shoyer]

I would propose as fix instead of if engine is None: engine="h5netcdf" to call xarray.backends.api._get_default_engine_netcdf() instead, but I also skimmed #10571 and I am not sure if we should remove scipy as an output of the function when neither netcdf4 nor h5netcdf are installed.

Thanks for the report.

This looks like the right fix to me. SciPy cannot write DataTree objects because it does not support netCDF4, which is required for groups.