What problem does your feature request solve?
There are two ways for point observations to get into CSET - via the Scores operators which take in iris cubes and via METplus which takes in MET ascii or netcdf point obs files.
The same observation data should be usable by either verification system.
Describe the solution you'd like
Add an operator that can convert observation data both ways, cube to MET and MET to cube that can be used in CSET recipes, e.g:
steps:
- operator: misc.combine_cubes_into_cubelist
model:
operator: read.read_cubes
file_paths: $INPUT_PATHS
# ...
obs:
operator: misc.convert_obs_format
file_paths: $METPLUS_OBS_FILES
output_format: point_cube
# ...
- operator: scoreswrapper.scores_rmse
As an enhancement it would be useful to be able to filter out stations using station lists or masks
Describe alternatives you've considered
What problem does your feature request solve?
There are two ways for point observations to get into CSET - via the Scores operators which take in iris cubes and via METplus which takes in MET ascii or netcdf point obs files.
The same observation data should be usable by either verification system.
Describe the solution you'd like
Add an operator that can convert observation data both ways, cube to MET and MET to cube that can be used in CSET recipes, e.g:
As an enhancement it would be useful to be able to filter out stations using station lists or masks
Describe alternatives you've considered