`.subset` with da that has time dimension

[Chrismarsh]

Version

2025.11.0

How did you install UXarray?

Source

What happened?

I have a dataset that looks like

>>> da = du.sel(time=["2018-04-28 00:00:00", "2023-03-22 00:00:00"])[['Elevation']]
>>> print(da)

<xarray.UxDataset> Size: 1GB
Dimensions:       (met: 2, time: 2, n_face: 28495325)
Coordinates:
  * time          (time) datetime64[ns] 16B 2018-04-28 2023-03-22
  * met           (met) object 16B 'hrdps' 'capa'
    Mesh2_face_x  (n_face) float64 228MB -123.8 -123.4 -123.9 ... -116.3 -116.3
    Mesh2_face_y  (n_face) float64 228MB 46.49 46.61 46.49 ... 52.56 52.33 52.34
Dimensions without coordinates: n_face
Data variables:
    Elevation     (met, time, n_face) float64 912MB dask.array<chunksize=(1, 1, 1000000), meta=np.ndarray

which I want to clip to a lat lon box:

tmp = da.Elevation.subset.bounding_box(lon, lat)

but it errors as:

IndexError: dimension coordinate 'time' conflicts between indexed and indexing objects:
<xarray.DataArray 'time' (time: 2)> Size: 16B
array(['2018-04-28T00:00:00.000000000', '2023-03-22T00:00:00.000000000'],
      dtype='datetime64[ns]')
Coordinates:
  * time     (time) datetime64[ns] 16B 2018-04-28 2023-03-22
Attributes:
    standard_name:  time
    long_name:      Time
vs.
<xarray.Variable ()> Size: 8B
array('2016-10-01T00:00:00.000000000', dtype='datetime64[ns]')
Attributes:
    standard_name:  time
    long_name:      Time

which I don't totally understand. Weirdly, this code used to work, I want to say 1 or 2 releases ago.

What did you expect to happen?

It to subset across all dim, time included