Reconfiguration

Revision as of 01:30, 18 March 2019 by S.wales (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Template:Pre UM10 This information is only for versions of the UM before 10.0

The reconfiguration is a program run before a UM model run which compiles different data sets into a single initial conditions file ready to be read by the UM.

The UMUI section Atmosphere -> Ancillary and input data -> Start Dump is where you select the input dump for the model, and whether or not to run the reconfiguration on that file.

Interpolation

Additional Data Sources

Some model types add new primary fields to the model, which require their own initial conditions. These can be set in the UMUI window Atmosphere -> STASH -> Initialisation of User Prognostics

User progs.png

For each primary field added by user stashmaster files you must set an option code for how it will be initialised:

  1. Initialise from the start dump: This can be used if the start dump already contains the field, for instance if it was output as a diagnostic in a previous run
  2. Initialise from a user ancillary file. The files to look in are set in Atmosphere -> Ancillary and Input files -> User single/multi-level ancillary files. Stash codes 301-320 are looked for in the single-level ancillary file, codes 321-324 are looked for in the multi-level file. It is possible to update the field from the ancillary file during the model run.
  3. Set the field to zero
  4. Set the field to the missing data index (i.e. the field starts uninitialised)
  5.  ???
  6. Set the field to the value given in the CONSTANT column.
  7. Initialise from an ancillary file named in the NAMED FILE column. The field cannot be updated during the run, however this option can be used for any field, not just stash codes 301-324.

Resources

| Reconfiguration technical paper