GOES 16 True Color Notebook Broken

[julienchastang]

• recent python-training conda env
• Run GOES 16 True Color Notebook

I'll assign this issue to myself. Let me know if anyone has any insights here. It used to work so we are either misusing a more recent version of the MetPy API, or something inside MetPy broke.

---------------------------------------------------------------------------
DimensionalityError                       Traceback (most recent call last)
/opt/conda/envs/python-training/lib/python3.7/site-packages/metpy/xarray.py in _fixup_coords(self, var)
    548                 try:
--> 549                     var.coords[coord_name].metpy.convert_units('meters')
    550                 except DimensionalityError:  # Radians!

/opt/conda/envs/python-training/lib/python3.7/site-packages/metpy/xarray.py in convert_units(self, units)
    144         """Convert the data values to different units in-place."""
--> 145         self.unit_array = self.unit_array.to(units)
    146 

/opt/conda/envs/python-training/lib/python3.7/site-packages/pint/quantity.py in to(self, other, *contexts, **ctx_kwargs)
    590 
--> 591         magnitude = self._convert_magnitude_not_inplace(other, *contexts, **ctx_kwargs)
    592 

/opt/conda/envs/python-training/lib/python3.7/site-packages/pint/quantity.py in _convert_magnitude_not_inplace(self, other, *contexts, **ctx_kwargs)
    539 
--> 540         return self._REGISTRY.convert(self._magnitude, self._units, other)
    541 

/opt/conda/envs/python-training/lib/python3.7/site-packages/pint/registry.py in convert(self, value, src, dst, inplace)
    924 
--> 925         return self._convert(value, src, dst, inplace)
    926 

/opt/conda/envs/python-training/lib/python3.7/site-packages/pint/registry.py in _convert(self, value, src, dst, inplace)
   1761 
-> 1762         return super()._convert(value, src, dst, inplace)
   1763 

/opt/conda/envs/python-training/lib/python3.7/site-packages/pint/registry.py in _convert(self, value, src, dst, inplace)
   1375         if not (src_offset_unit or dst_offset_unit):
-> 1376             return super()._convert(value, src, dst, inplace)
   1377 

/opt/conda/envs/python-training/lib/python3.7/site-packages/pint/registry.py in _convert(self, value, src, dst, inplace, check_dimensionality)
    957             if src_dim != dst_dim:
--> 958                 raise DimensionalityError(src, dst, src_dim, dst_dim)
    959 

DimensionalityError: Cannot convert from 'radian' (dimensionless) to 'meter' ([length])

During handling of the above exception, another exception occurred:

ValueError                                Traceback (most recent call last)
<ipython-input-11-d9ca28bf71fb> in <module>
      1 # We'll use the `CMI_C02` variable as a 'hook' to get the CF metadata.
----> 2 dat = C.metpy.parse_cf('CMI_C02')
      3 
      4 geos = dat.metpy.cartopy_crs
      5 

/opt/conda/envs/python-training/lib/python3.7/site-packages/metpy/xarray.py in parse_cf(self, varname, coordinates)
    518                 var.coords['crs'] = CFProjection(proj_var.attrs)
    519 
--> 520         self._fixup_coords(var)
    521 
    522         # Trying to guess whether we should be adding a crs to this variable's coordinates

/opt/conda/envs/python-training/lib/python3.7/site-packages/metpy/xarray.py in _fixup_coords(self, var)
    553                         height = var.coords['crs'].item()['perspective_point_height']
    554                         scaled_vals = new_data_array.metpy.unit_array * (height * units.meters)
--> 555                         new_data_array.metpy.unit_array = scaled_vals.to('meters')
    556                         var.coords[coord_name] = new_data_array
    557 

/opt/conda/envs/python-training/lib/python3.7/site-packages/metpy/xarray.py in unit_array(self, values)
    138     def unit_array(self, values):
    139         """Set data values from a `pint.Quantity`."""
--> 140         self._data_array.values = values.magnitude
    141         self._units = self._data_array.attrs['units'] = str(values.units)
    142 

/opt/conda/envs/python-training/lib/python3.7/site-packages/xarray/core/common.py in __setattr__(self, name, value)
    260         """
    261         try:
--> 262             object.__setattr__(self, name, value)
    263         except AttributeError as e:
    264             # Don't accidentally shadow custom AttributeErrors, e.g.

/opt/conda/envs/python-training/lib/python3.7/site-packages/xarray/core/dataarray.py in values(self, value)
    560     @values.setter
    561     def values(self, value: Any) -> None:
--> 562         self.variable.values = value
    563 
    564     @property

/opt/conda/envs/python-training/lib/python3.7/site-packages/xarray/core/variable.py in values(self, values)
   2113     def values(self, values):
   2114         raise ValueError(
-> 2115             f"Cannot assign to the .values attribute of dimension coordinate a.k.a IndexVariable {self.name!r}. "
   2116             f"Please use DataArray.assign_coords, Dataset.assign_coords or Dataset.assign as appropriate."
   2117         )

ValueError: Cannot assign to the .values attribute of dimension coordinate a.k.a IndexVariable 'y'. Please use DataArray.assign_coords, Dataset.assign_coords or Dataset.assign as appropriate.

[jthielen]

Based on what I've seen, this is a break in MetPy <= 1.0.0rc1 caused by xarray's update to 0.15.1. I have a fix incoming to Unidata/MetPy#1325, but that will only end up applying to the MetPy 1.0 branch due to the other backwards-incompatible changes in that PR.

@dopplershift Since this xarray 0.15.1 issue is showing up everywhere, do we need a bugfix on the MetPy 0.12 branch for this?

[julienchastang]

Thanks @jthielen. Your branch addresses my problem! In the immediate term, that satisfies my concern of having the notebook run now. I would like to demo it in the near future. I look forward to this fix being incorporated into MetPy. Thanks again!