Inconsistencies between UM metadata and CMIP6 published data (ESM1.5)

[rbeucher]

I am a bit confused about variables defined on model levels.

For example, CMIP6 Amon.cl (cloud_area_fraction_in_atmosphere_layer) maps to the UM field fld_s02i261. In the ACCESS-ESM1.6 spin-up, the field is defined as:

float fld_s02i261(time,model_level_number,lat,lon) ;
  fld_s02i261:_FillValue = 1.e+20f ;
  fld_s02i261:standard_name = "cloud_area_fraction_in_atmosphere_layer" ;
  fld_s02i261:long_name = "TOTAL CLOUD AMOUNT ON LEVELS" ;
  fld_s02i261:units = "1" ;
  fld_s02i261:um_stash_source = "m01s02i261" ;
  fld_s02i261:missing_value = 1.e+20f ;
  fld_s02i261:cell_methods = "time: mean" ;
  fld_s02i261:grid_mapping = "latitude_longitude" ;
  fld_s02i261:coordinates = "level_pressure sigma surface_air_pressure" ;

So it looks like the vertical coordinate context is pressure-based (via level_pressure, sigma, and surface_air_pressure).

However, the published CMIP6 Amon.cl for ACCESS-ESM1-5 (e.g. under /g/data/fs38/publications/CMIP6/CMIP/CSIRO/ACCESS-ESM1-5/.../Amon/cl/gn/...) has:

float cl(time, lev, lat, lon) ;
  cl:standard_name = "cloud_area_fraction_in_atmosphere_layer" ;
  cl:long_name = "Percentage Cloud Cover" ;
  cl:comment = "Percentage cloud cover, including both large-scale and convective cloud." ;
  cl:units = "%" ;
  cl:original_units = "1" ;
  cl:history = "... altered by CMOR: Converted units from '1' to '%'. ... replaced missing value flag ... with ... (1e+20)." ;
  cl:cell_methods = "area: time: mean" ;
  cl:cell_measures = "area: areacella" ;
  cl:missing_value = 1.e+20f ;
  cl:_FillValue = 1.e+20f ;

…and lev is defined as a hybrid height coordinate:

double lev(lev) ;
  lev:bounds = "lev_bnds" ;
  lev:units = "m" ;
  lev:axis = "Z" ;
  lev:positive = "up" ;
  lev:long_name = "hybrid height coordinate" ;
  lev:standard_name = "atmosphere_hybrid_height_coordinate" ;
  lev:formula = "z = a + b*orog" ;
  lev:formula_terms = "a: lev b: b orog: orog" ;

I would have expected lev here to be atmosphere_hybrid_sigma_pressure_coordinate (or at least a pressure-based hybrid coordinate), given the upstream field metadata.

Is this an inconsistency (or possibly a metadata issue in either the spin-up output or the CMIP6 product), or is the use of a hybrid height coordinate for Amon.cl intentional for ACCESS/UM?

[rbeucher]

@MartinDix

[MartinDix]

This is strange. The cloud is on the model theta levels, but for some reason iris has given it a different vertical coordinate when creating the netCDF.

        float fld_s00i408(time, model_theta_level_number, lat, lon) ;
                fld_s00i408:standard_name = "air_pressure" ;
                fld_s00i408:long_name = "PRESSURE AT THETA LEVELS AFTER TS" ;
                fld_s00i408:coordinates = "sigma_theta surface_altitude theta_level_height" ;

        float fld_s02i261(time, model_level_number, lat, lon) ;
                fld_s02i261:standard_name = "cloud_area_fraction_in_atmosphere_layer" ;
                fld_s02i261:long_name = "TOTAL CLOUD AMOUNT ON LEVELS" ;
                fld_s02i261:coordinates = "level_pressure sigma surface_air_pressure" ;

ACCESS-NRI/access-esm1.6-configs#390

[rbeucher]

OK I am now assuming that

        float fld_s02i261(time, model_theta_level_number, lat, lon) ;
                fld_s02i261:standard_name = "cloud_area_fraction_in_atmosphere_layer" ;
                fld_s02i261:long_name = "TOTAL CLOUD AMOUNT ON LEVELS" ;
                fld_s02i261:coordinates = "sigma_theta surface_altitude theta_level_height" ;

Is the truth