Nice find! The documentation doesn't explain that at all currently.

But that solution doesn't work for 1d DataArrays. You have to use this kwargs instead: python kwargs = dict(fill_value=(np.min(da.data), np.max(da.data))) So it still isn't as convenient as I would like it to be. The wrapper function can now be simplified to this however:

````python import numpy as np import xarray as xr

def interp(da, coords, extrapolation='clip'): """ Linear interpolation that clips the inputs to the coords min and max value.

Parameters
----------
da : DataArray
    DataArray to interpolate.
coords : dict
    Coordinates for the interpolated value.
"""
if extrapolation == 'clip':
    if len(da.coords) > 1:
        kwargs = dict(fill_value=None)
    else:
        kwargs = dict(fill_value=(np.min(da.data), np.max(da.data)))
return da.interp(coords, kwargs=kwargs)

Create coordinates:

x = np.linspace(1000, 6000, 4) y = np.linspace(100, 1200, 3) z = np.linspace(1, 2, 2)

Create 1D DataArray:

da1 = xr.DataArray(data=2*x, coords=[('x', x)])

Create 2D DataArray:

X = np.meshgrid(*[x, y], indexing='ij') data = X[0] * X[1] da2 = xr.DataArray(data=data, coords=[('x', x), ('y', y)])

Create 3D DataArray:

X = np.meshgrid(*[x, y, z], indexing='ij') data = X[0] * X[1] * X[2] da3 = xr.DataArray(data=data, coords=[('x', x), ('y', y), ('z', z)])

Attempt to extrapolate:

print(interp(da1, {'x': 7000})) print(interp(da2, {'x': 7000, 'y': 375})) print(interp(da3, {'x': 7000, 'y': 375, 'z': 1})) ````