Cleanup of plain old radiance vs spectral radiance vs scaled radiance

[rhoneyager-tomorrow]

The names here are confusing and are not consistent across the converters, ufo transformations, and the IODA conventions. They are also not consistently named within the broader field.

The IODA conventions spec and Wikipedia define radiance following a common convention in radiometry. It is the radiant flux emitted, reflected, transmitted or received by a given surface, per unit solid angle per unit projected area. Units are W m-2 sr-1.

Spectral radiance is the radiance of a surface per unit frequency, or wavelength, or wavenumber. There are also scaled radiances, which represent compressed data that need to be transformed to get regular radiances. These are all distinct variables in the BUFR code tables, and these all have different units.

Here are a few spots where these are confused:

• In bufr_ncep_mtiasi.yaml, it is unclear if a spectral radiance or a regular radiance is being used. The comments say spectral radiance, but SCRA is a plain old scaled radiance. Spectral radiance (SPRD) would be a different mnemonic.

• bufr_iasi.yaml is writing scaled radiance as a regular radiance. These have different units (W m-2 sr-1 vs W m-2 sr-1 m). Comments in yaml do not get carried through to end users, and they end up with a variable that does not contain what they expect.

• In UFO's SatBrightnessTempFromRad transform, the radiance variable is being treated as a spectral radiance variable. This is doubly awkward as a user needs to manually specify if the spectral radiance is expressed in terms of frequency, wavelength, or wavenumber.

@gthompsnJCSDA / @emilyhcliu / @BenjaminRuston would you mind disambiguating these cases? If users are writing spectral radiances and scaled radiances in IODA files, then these variables should be documented in the IODA conventions.

[BenjaminRuston]

@smaticka this is still backlog but may be revisited after converter addressed