Dry_Lapse broadcast for multiple input starting temperature

[gitlinffff]

What went wrong?

p = np.array([1000., 950., 900., 850., 800., 750., 700., 650., 600., 550., 500., 450., 400., 350., 300., 250., 200., 175., 150., 125., 100., 80., 70., 60., 50., 40., 30., 25., 20.]) T = np.array([15, 20]) dry_lapse(p * units.hPa, T * units.degC)

ValueError: operands could not be broadcast together with shapes (2,) (29,)

It should have treated T as multiple starting temperature and return dry adiabatic profile for each starting temperature, instead of requiring p and T to have the same shape. It is also necessary to add a pytest for multiple starting temperature.

Operating System

MacOS

Version

v1.7

Python Version

3.13.3

Code to Reproduce

p = np.array([1000., 950., 900., 850., 800., 750., 700., 650., 600.,
     550., 500., 450., 400., 350., 300., 250., 200.,
     175., 150., 125., 100., 80., 70., 60., 50.,
     40., 30., 25., 20.])
T = np.array([15, 20])
dry_lapse(p * units.hPa, T * units.degC)

Errors, Traceback, and Logs

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[3], line 6
      1 p = np.array([1000., 950., 900., 850., 800., 750., 700., 650., 600.,
      2      550., 500., 450., 400., 350., 300., 250., 200.,
      3      175., 150., 125., 100., 80., 70., 60., 50.,
      4      40., 30., 25., 20.])
      5 T = np.array([15, 20])
----> 6 calc.dry_lapse(p * units.hPa, T * units.degC)

File ~/git_repos/MetPy/src/metpy/xarray.py:1330, in preprocess_and_wrap.<locals>.decorator.<locals>.wrapper(*args, **kwargs)
   1327     _mutate_arguments(bound_args, units.Quantity, lambda arg, _: arg.m)
   1329 # Evaluate inner calculation
-> 1330 result = func(*bound_args.args, **bound_args.kwargs)
   1332 # Wrap output based on match and match_unit
   1333 if match is None:

File ~/git_repos/MetPy/src/metpy/units.py:337, in check_units.<locals>.dec.<locals>.wrapper(*args, **kwargs)
    334 @functools.wraps(func)
    335 def wrapper(*args, **kwargs):
    336     _check_units_inner_helper(func, sig, defaults, dims, *args, **kwargs)
--> 337     return func(*args, **kwargs)

File ~/git_repos/MetPy/src/metpy/calc/thermo.py:513, in dry_lapse(pressure, temperature, reference_pressure, vertical_dim)
    511 if reference_pressure is None:
    512     reference_pressure = pressure[0]
--> 513 return temperature * (pressure / reference_pressure)**mpconsts.kappa

File ~/miniforge3/envs/metpy_dev/lib/python3.13/site-packages/pint/facets/plain/quantity.py:1018, in PlainQuantity.__mul__(self, other)
   1017 def __mul__(self, other):
-> 1018     return self._mul_div(other, operator.mul)

File ~/miniforge3/envs/metpy_dev/lib/python3.13/site-packages/pint/facets/plain/quantity.py:101, in check_implemented.<locals>.wrapped(self, *args, **kwargs)
     99 elif isinstance(other, list) and other and isinstance(other[0], type(self)):
    100     return NotImplemented
--> 101 return f(self, *args, **kwargs)

File ~/miniforge3/envs/metpy_dev/lib/python3.13/site-packages/pint/facets/plain/quantity.py:75, in ireduce_dimensions.<locals>.wrapped(self, *args, **kwargs)
     74 def wrapped(self, *args, **kwargs):
---> 75     result = f(self, *args, **kwargs)
     76     try:
     77         if result._REGISTRY.autoconvert_to_preferred:

File ~/miniforge3/envs/metpy_dev/lib/python3.13/site-packages/pint/facets/plain/quantity.py:1006, in PlainQuantity._mul_div(self, other, magnitude_op, units_op)
   1003 elif no_offset_units_other == len(other._units) == 1:
   1004     other = other.to_root_units()
-> 1006 magnitude = magnitude_op(new_self._magnitude, other._magnitude)
   1007 units = units_op(new_self._units, other._units)
   1009 return self.__class__(magnitude, units)

ValueError: operands could not be broadcast together with shapes (2,) (29,)

[DWesl]

Does this work?

dry_lapse(p.reshape(-1, np.newaxis) * units.hPa, T.reshape(np.newaxis, -1) * units.degC)

[dopplershift]

I agree with @DWesl 's code above, if you pass in arrays so that the math would work following NumPy's broadcasting rules you should get the results you expect.