pydata
`Dataset.interp()` silently dropping time-like data arrays
What happened?
Interpolation of data arrays with time informatively errors:
# /// script
# requires-python = ">=3.11"
# dependencies = [
# "xarray[complete]@git+https://github.com/pydata/xarray.git@main",
# ]
# ///
import numpy as np
import xarray as xr
da = xr.DataArray(
data=(np.array([1, 4, 2]).astype("datetime64[ns]")),
dims=("x",),
coords={"x": [0, 1, 2]},
)
da_interp = da.interp(x=[0, 0.75, 1.25, 1.75])
Traceback (most recent call last):
File "/Users/Hodgs004/coding/repos/parcels/xarray-repro/min-rep-da.py", line 18, in <module>
da_interp = da.interp(x=[0, 0.75, 1.25, 1.75])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/Hodgs004/.cache/uv/environments-v2/min-rep-da-138593a3f0628b04/lib/python3.12/site-packages/xarray/core/dataarray.py", line 2411, in interp
raise TypeError(
TypeError: interp only works for a numeric type array. Given datetime64[ns].
However when doing it with a Dataset, the time array is silently dropped with no warnings.
cc @michaeldenes
What did you expect to happen?
If it's expected behaviour for it to be dropped, then it would be nice for it to emit a warning or something
Minimal Complete Verifiable Example
# /// script
# requires-python = ">=3.11"
# dependencies = [
# "xarray[complete]@git+https://github.com/pydata/xarray.git@main",
# ]
# ///
#
# This script automatically imports the development branch of xarray to check for issues.
# Please delete this header if you have _not_ tested this script with `uv run`!
import numpy as np
import xarray as xr
ds = xr.Dataset.from_dict(
{
"coords": {},
"dims": {"trajectory": 10, "obs": 10},
"data_vars": {
"something": {
"data": np.random.rand(10, 10).astype("float64"),
"dims": ["trajectory", "obs"],
},
"time": {
"data": np.random.rand(10, 10).astype("datetime64[ns]"),
"dims": ["trajectory", "obs"],
},
},
}
)
print("Original dataset:\n", ds)
assert "time" in ds.data_vars
# following https://docs.xarray.dev/en/stable/user-guide/interpolation.html#advanced-interpolation
# for Advanced Interpolation and indexing
trajectory = ds.trajectory.rename({"trajectory": "z"})
obs = ds.obs.rename({"obs": "z"})
ds_interp = ds.interp(trajectory=trajectory, obs=obs)
print("Interped dataset:\n", ds_interp)
assert "something" in ds_interp.data_vars
assert "time" in ds_interp.data_vars, "Why is 'time' missing from the variables?"
Steps to reproduce
Run script. MCVE output:
Original dataset:
<xarray.Dataset> Size: 2kB
Dimensions: (trajectory: 10, obs: 10)
Dimensions without coordinates: trajectory, obs
Data variables:
something (trajectory, obs) float64 800B 0.05063 0.03253 ... 0.8666 0.4773
time (trajectory, obs) datetime64[ns] 800B 1970-01-01 ... 1970-01-01
Interped dataset:
<xarray.Dataset> Size: 240B
Dimensions: (z: 10)
Dimensions without coordinates: z
Data variables:
something (z) float64 80B 0.05063 0.9336 0.6782 ... 0.2084 0.466 0.4773
trajectory (z) int64 80B 0 1 2 3 4 5 6 7 8 9
obs (z) int64 80B 0 1 2 3 4 5 6 7 8 9
Traceback (most recent call last):
File "/Users/Hodgs004/coding/repos/parcels/xarray-repro/min-rep.py", line 42, in <module>
assert "time" in ds_interp.data_vars, "Why is 'time' missing from the variables?"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AssertionError: Why is 'time' missing from the variables?
MVCE confirmation
- [x] Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
- [x] Complete example — the example is self-contained, including all data and the text of any traceback.
- [x] Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
- [x] New issue — a search of GitHub Issues suggests this is not a duplicate.
- [x] Recent environment — the issue occurs with the latest version of xarray and its dependencies.
Relevant log output
Anything else we need to know?
No response
Environment
commit: None python: 3.12.10 | packaged by conda-forge | (main, Apr 10 2025, 22:19:24) [Clang 18.1.8 ] python-bits: 64 OS: Darwin OS-release: 24.6.0 machine: arm64 processor: arm byteorder: little LC_ALL: None LANG: en_US.UTF-8 LOCALE: (None, 'UTF-8') libhdf5: 1.14.6 libnetcdf: 4.9.3
xarray: 2025.10.2.dev33+g6da8c2c92 pandas: 2.3.3 numpy: 2.3.4 scipy: 1.16.3 netCDF4: 1.7.3 pydap: 3.5.8 h5netcdf: 1.7.3 h5py: 3.15.1 zarr: 3.1.3 cftime: 1.6.5 nc_time_axis: 1.4.1 iris: None bottleneck: 1.6.0 dask: 2025.11.0 distributed: 2025.11.0 matplotlib: 3.10.7 cartopy: 0.25.0 seaborn: 0.13.2 numbagg: 0.9.3 fsspec: 2025.10.0 cupy: None pint: None sparse: 0.17.0 flox: 0.10.7 numpy_groupies: 0.11.3 setuptools: None pip: None conda: None pytest: None mypy: None IPython: None sphinx: None None
Also, I'm not 100% sure why a data array for trajectory is added after interpolation in the MCVE:
Original dataset:
<xarray.Dataset> Size: 2kB
Dimensions: (trajectory: 10, obs: 10)
Dimensions without coordinates: trajectory, obs
Data variables:
something (trajectory, obs) float64 800B 0.05063 0.03253 ... 0.8666 0.4773
time (trajectory, obs) datetime64[ns] 800B 1970-01-01 ... 1970-01-01
Interped dataset:
<xarray.Dataset> Size: 240B
Dimensions: (z: 10)
Dimensions without coordinates: z
Data variables:
something (z) float64 80B 0.05063 0.9336 0.6782 ... 0.2084 0.466 0.4773
+ trajectory (z) int64 80B 0 1 2 3 4 5 6 7 8 9
obs (z) int64 80B 0 1 2 3 4 5 6 7 8 9
Traceback (most recent call last):
File "/Users/Hodgs004/coding/repos/parcels/xarray-repro/min-rep.py", line 42, in <module>
assert "time" in ds_interp.data_vars, "Why is 'time' missing from the variables?"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AssertionError: Why is 'time' missing from the variables?
is this expected?
EDIT: Actually, nevermind - I see one is added for obs and trajectory, makes sense
This should be an easy fix with a .view(np.int64) and undoing that at the end.
