This short tutorial explains how to add a new experiment to the metview macros used during the workshop.
Make sure that the monthly mean files have been created from the model ensemble forecast. This is described: in Creating a new experiment: postprocessing.
Transfer the files via ftp to ICTP. They should be placed in directory you are advised to use e.g. 'data_local'.
A small definition needs to be created in order for the Metview scripts to know how to use the new experiment.
In the main Metview window with the openifs_workshop_2017 macros, find the folder 'base'
and open it.
Find the macro labelled 'base_input.mv',
right-click and select 'Edit':
Locate the following text:
#User experiment global expDef_ob_user = ( label: "OBS xx", id: "obxx", dir: "data_local/obxx", fname : "oif_obxx_$P_$D_$T.grib", steps : [2015-12-01,2016-01-01,2016-02-01,2016-03-01,2016-04-01], noCf: 1, member: -1 ) |
You can either edit this or make a copy if you have multiple new experiments. Changes need to be made to the items with 'xx'.
As an example, for a new experiment 'ob01', make a copy of the above as shown below:
# My ob01 experiment global expDef_ob01_user = ( label: "OBS Pacific anomaly", id: "ob01", dir: "data_local/ob01", fname : "oif_ob01_$P_$D_$T.grib", steps : [2015-12-01,2016-01-01,2016-02-01,2016-03-01,2016-04-01], noCf: 1, member: 10 ) |
Changes are:
A final and very important change is to amend this line:
global expDefLst = [expDef_ei, expDef_clim, expDef_ob00, expDef_ob50, expDef_cl50, expDef_ob_user] |
to include the new experiment definition, expDef_ob01_user, to the end:
global expDefLst = [expDef_ei, expDef_clim, expDef_ob00, expDef_ob50, expDef_cl50, expDef_ob_user, expDef_ob01_user] |
The variable name used must match the variable name for the definition as described above.
And that's it! If the metview plotting fails, check that the experiment definition is known and the data is located in the correct directory.