STASH Code to CF name translations: inconsistent units

[ehogan]

Here is a copy of my upload file:
|STASH(msi)|CFName|units|force_update(y/n)|
|m01s00i010|specific_humidity|1|n|
|m01s03i202|downward_heat_flux_in_soil|W m-2|n|
|m02s00i101|sea_water_potential_temperature|K|n|
|m02s00i130|ocean_barotropic_streamfunction|m3 s-1|n|
|m02s30i201|upward_sea_water_velocity|m s-1|n|
|m02s30i211|northward_ocean_heat_transport|W|n|
|m02s30i320|eastward_sea_water_velocity|m s-1|n|
|m02s30i321|northward_sea_water_velocity|m s-1|n|

Here is the responce:
The file failed to parse; in order to process this file you should consider:
line1: forcing replacement of m01s00i010 -> specific_humidity(kg kg-1) with m01s00i010 -> specific_humidity(1)
line2: forcing replacement of m01s03i202 -> downward_heat_flux_in_soil(unknown) with m01s03i202 -> downward_heat_flux_in_soil(W m-2)
line3: forcing replacement of m02s00i101 -> sea_water_potential_temperature(degC) with m02s00i101 -> sea_water_potential_temperature(K)
line4: forcing replacement of m02s00i130 -> ocean_barotropic_streamfunction(cm3 s-1) with m02s00i130 -> ocean_barotropic_streamfunction(m3 s-1)
line5: forcing replacement of m02s30i201 -> upward_sea_water_velocity(cm s-1) with m02s30i201 -> upward_sea_water_velocity(m s-1)
line6: forcing replacement of m02s30i211 -> northward_ocean_heat_transport(PW) with m02s30i211 -> northward_ocean_heat_transport(W)
line7: forcing replacement of m02s30i320 -> eastward_sea_water_velocity(cm s-1) with m02s30i320 -> eastward_sea_water_velocity(m s-1)
line8: forcing replacement of m02s30i321 -> northward_sea_water_velocity(cm s-1) with m02s30i321 -> northward_sea_water_velocity(m s-1)

The names are consistent but the units are inconsistent.

[marqh]

Hello @ehogan

the first two of these lines are expected.
In the first case, 'kg kg-1' is preferred to 1, although it is calculation equivalent.
In the second case, the unit is unknown, so we'll adopt any improvement

The rest of these are more difficult to manage. There are no longer named code owners for the m02 STASH codes so we are really dependent on finding an authoritative source to resolve the discrepancy.

Using the STASH records, I have found:
http:https://reference.metoffice.gov.uk/um/stash/m02s00i101
describes degC in the title, suggesting the current mapping is valid
http:https://reference.metoffice.gov.uk/um/stash/m02s00i130
describes cm3 s-1 in the title, suggesting the current mapping is valid
http:https://reference.metoffice.gov.uk/um/stash/m02s00i130
describes cm s-1 in the title, suggesting the current mapping is valid
http:https://reference.metoffice.gov.uk/um/stash/m02s30i211
describes W in the title, suggesting this one needs an update
http:https://reference.metoffice.gov.uk/um/stash/m02s30i320
describes cm s**-1 in the title, suggesting the current mapping is valid
http:https://reference.metoffice.gov.uk/um/stash/m02s30i321
describes cm s**-1 in the title, suggesting the current mapping is valid

On this basis, I suggest an upload of

|STASH(msi)|CFName|units|force_update(y/n)|
|m01s03i202|downward_heat_flux_in_soil|W m-2|n|
|m02s30i211|northward_ocean_heat_transport|W|n|

I think to change any of the others, we would want a little more evidence that the change was valid

many thanks
mark