[MISC]: xwrf metpy on wrf data

[fipoucat]

What is your issue?

I am attempting to post-process wrf output using xwrf and metpy. After opening the dataset I am trying to work on wind data but getting an error:
from metpy.calc import wind_speed
wind_speed(ds.U, ds.V)

ds.U.sizes, ds.V.sizes

ds = ds.metpy.quantify()
wind_speed(ds.U.wrf.destagger(), ds.V.wrf.destagger())
Traceback (most recent call last):

Cell In[7], line 2
wind_speed(ds.U, ds.V)

File /opt/anaconda3/lib/python3.12/site-packages/metpy/xarray.py:1330 in wrapper
result = func(*bound_args.args, **bound_args.kwargs)

File /opt/anaconda3/lib/python3.12/site-packages/metpy/units.py:333 in wrapper
return func(*args, **kwargs)

File /opt/anaconda3/lib/python3.12/site-packages/metpy/calc/basic.py:63 in wind_speed
return np.hypot(u, v)

File /opt/anaconda3/lib/python3.12/site-packages/pint/facets/numpy/quantity.py:72 in array_ufunc
return numpy_wrap("ufunc", ufunc, inputs, kwargs, types)

File /opt/anaconda3/lib/python3.12/site-packages/pint/facets/numpy/numpy_func.py:1071 in numpy_wrap
return handled[name](*args, **kwargs)

File /opt/anaconda3/lib/python3.12/site-packages/pint/facets/numpy/numpy_func.py:322 in implementation
result_magnitude = func(*stripped_args, **stripped_kwargs)

File /opt/anaconda3/lib/python3.12/site-packages/dask/array/core.py:1594 in array_ufunc
return elemwise(numpy_ufunc, *inputs, **kwargs)

File /opt/anaconda3/lib/python3.12/site-packages/dask/array/core.py:4795 in elemwise
broadcast_shapes(*shapes)

File /opt/anaconda3/lib/python3.12/site-packages/dask/array/core.py:4723 in broadcast_shapes
raise ValueError(

ValueError: operands could not be broadcast together with shapes (8, 47, 116, 133) (8, 47, 117, 132)

Any hint on what is causing the error and how to fix it?

[lpilz]

The problem is that at the point where you call wind_speed(ds.U, ds.V), the U and V arrays are not destaggered yet. Your second call to wind_speed(ds.U.xwrf.destagger(), ds.V.xwrf.destagger()) (please note the xwrf instead of wrf) should work.

Also, compare with https://xwrf.readthedocs.io/en/stable/tutorials/xgcm.html