ValueError: arguments without labels along dimension 'rows' cannot be aligned because they have different dimension sizes: {2400, 1200}

Question

My objective is to convert the radiance parameter to reflectance from the sentinel three slstr data.

The formula that I am applying is:

relfectance = pi*radiance/solar_radiance/cos(solar_zenith_angle)

The data contains multiple files in netCDF format. It contains S1-S6 radiance and quality data and S7-S9 BT data. It also consists of geometry files that contains solar zenith angles.

I am using xarray to read these files.

For instance, I am using S6 radiance data to store radiance: radiance = S6_radiance['S6_radiance_an']

S6 quality data to store solar irradiance: solar_irradiance = S6_quality['solar_irradiance_an']

cosine_zenith = np.cos(geometry['solar_zenith_tn']

when I substitute the data in the formula: reflectance = pi*radiance/solar_irradiance/cosine_zenith

It gives me an error. It says:

ValueError: arguments without labels along dimension 'rows' cannot be aligned because they have different dimension sizes: {2400, 1200}

The dimensions are:

radiance.dims = ('rows', 'columns')
radiance.shape = (2400, 3000)

solar_irradiance.dims = ('detectors',)
solar_irradiance.shape = (4,)

cosine_zenith.dims = ('rows', 'columns')
cosine_zenith.shape = (1200, 130)