Reading in Met-Offce Unified Model data

[leifdenby]

@LSaffin and I are going to try and get UM model data read into lagtraj 😄

So far what I know we need is:

• loading of UM netCDF files, this should loaded the relevant netCDF files (by some naming convention, or globbing a directory), rename variables to the ECMWF naming convention and add any derived variables that lagtraj needs
• interpolation from UM height-levels (which are different in each column) to the height levels in lagtraj

[leifdenby]

Name-mapping for ECMWF to UM variable names (taken from https://github.com/EUREC4A-UK/lagtraj/blob/master/lagtraj/forcings/profile_calculation.py#L13)

• u: zonal_velocity in blafile.nc ??
• v
• theta
• q
• clwc
• ciwc
• t
• p_f
• t_l
• cc
• q_t
• r_t
• r_v
• r_l
• r_i
• theta_l

more to follow

[leosaffin]

I think it will be easiest to use CF standard names for UM output. By default people use met office specific file formats for UM output which I can load with iris giving you variables by standard name. If I load UM data output to netCDF with xarray, the variable names are "STASH codes" which aren't very human readable. Standard prognostics are

Standard name	STASH code	Possible ECMWF name
x_wind	STASH_m01s00i002	u
y_wind	STASH_m01s00i003	v
air_potential_temperature	STASH_m01s00i004	theta
specific_humidity	STASH_m01s00i010	q
mass_fraction_of_cloud_ice_in_air	STASH_m01s00i012	ciwc
surface_altitude	STASH_m01s00i033
upward_air_velocity	STASH_m01s00i150
mass_fraction_of_cloud_liquid_water_in_air	STASH_m01s00i254	clwc
cloud_area_fraction_in_atmosphere_layer	STASH_m01s00i265	cc
	STASH_m01s00i389	rho
air_pressure	STASH_m01s00i408	p_f
surface_air_pressure	STASH_m01s00i409

From these, you should be able to derive the rest.

Another thing to watch out for is the grid staggering. u, v, and rho are on half model levels as well as u and v being horizontally staggered.