#3 Missing/wrong units in output files

[Rasmus Nüß]

In most output files none or even wrong (or at least not generally right) units are given. This is tedious for new users to figure out. Depending on the conversion factors one could come up with a multitude of units that would fit the numbers more or less.

In files that contain emissions (adj_emissions.nc4, emission.nc4, emission_apos.nc4, emission_apri.nc4, optimized_state.nc4) the gridded emission fields have no units whatsoever. Might be kg/gridbox/timestep? The total emissions (per region) do have a unit (kg/timestep).

The observations ( satillite/*, point/*, aggregated_obs_*.nc4, representation_errors.nc4) and resulting mixing ratios (mix/*) give no units but should be mixing ratios as defined by the fscale factor. So ppb for CO, ppm for CO2, ppt for SF6,... The input satellite data (input/satellite/inputfile_*.nc4) on the other hand is given in molecules per square centimeter (which is for once also given in the files).

The station output (stattions/*) gives no units in the _comp file and always ppm in the other two. This is false if the aforementioned unit defined by fscale is not ppm.

The final mass (mass/* and save/*) has no unit given.

[Arjo Segers]

Unit attributes are now added to the emission files in the "cams-ch4" branch.

[raglan_road]

Units also added to the emissions files in the default branch. adj_emissions.nc4 has units of (gridbox sec/kg), which is not the most useful thing to add.