modifying the entrainment coefficient

[salvatore]

I have to modify the convective entrainment coefficient in order to produce an ensemble of climates.
I have been told that the values are in the namelist:
$FAMOUS_MODS/source/params_in_namelist_famous.mod

and also the hand-edit script
/home/famous/bin/he_namelist_new_phase5
(submodelind → postproc → local postproc) which modifies the namelist file input to the UM to include the parameters.

So in the end I have copied the two files and modified them with the new value of ENTCOEF (they
are in /home/n02/n02/swr07sp/bin/params.mod1 and /work/n02/n02/swr07sp/bin/phase5_mod1), included them and excluded the previous ones. But it doesn't work, the job is xdoib and in the leave file I can read:
lib-4001 : UNRECOVERABLE library error

Unable to find error message (check NLSPATH, file lib.cat)

Encountered during a namelist READ from unit 5
Fortran unit 5 is connected to a sequential formatted text file:

"/work/n02/n02/swr07sp/tmp/tmp.nid00007.16457/xdoib.namelists"

lib-4001 : UNRECOVERABLE library error

Unable to find error message (check NLSPATH, file lib.cat)

and things like that. What's been my mistake?

Many thanks,
Salvatore Pascale

[ros]

Hi Salvatore,

The hand-edit file phase5_mod1, which is run by the UMUI, needs to be in a location on PUMA not on HECToR.

I think you'll also find that the modified namelist will need to be in your /work space on HECToR and not /home.

Regards,
Ros.

[salvatore]

Hi, now I am experiencing another kind of problem. When I try to run jobs as xdoil or xdoin, in which I have set the entrainment coefficient to be greater than the tuned value(=3), the jobs fails and particularly in the .leave file I read things like: Model completed with the following :

Error Code : 1 Message : P_TH_ADJ : NEGATIVE PRESSURE VALUE CREATED.

*

Is this relevant to the problem?

In xdoin and xdoil I chose the entrainment coefficient 6 and 9 respectively, foolowing the QUMP study by Murphy et al. in which the pint out the range for such parameters (0.6 —>9), so I don't think to have done something that can cause serious problems to the run.

Many thanks, Salvatore Pascale PhD student