[MITgcm-support] bugs in ctrl pkg?

[Matthew Mazloff]

Hi Charlotte

Yes, this is confusing. I think the idea is to either strictly use the generic controls or not. So if you want to use ALLOW_GENARR3D_CONTROL then you should control atemp using ALLOW_GENTIM2D_CONTROL.

In data.ctrl you would have something like:

 xx_gentim2d_weight(1) = 'EIG_tmp2m_degC_wt_SO3.bin',
 xx_gentim2d_file(1)='xx_atemp_anom',
 xx_gentim2d_startdate1(1)=20080101,
 xx_gentim2d_startdate2(1)=30000,
 xx_gentim2d_period(1)=1209600.0,
 xx_gentim2d_preproc(1,1)='rmcycle',
 xx_gentim2d_preproc_i(1,1)=1,
 xx_gentim2d_preproc(2,1)='smooth',
 mult_gentim2d(1) = 1.0,
#
 xx_gentim2d_weight(2) = 'EIG_tmp2m_degC_wt_SO3.bin',
 xx_gentim2d_file(2)='xx_atemp_mean',
 xx_gentim2d_startdate1(2)=20080101,
 xx_gentim2d_startdate2(2)=30000,
 xx_gentim2d_period(2)=1209600.0,
 xx_gentim2d_preproc(1,2)='docycle',
 xx_gentim2d_preproc_i(1,2)=1,
 xx_gentim2d_preproc(2,2)='smooth',
 mult_gentim2d(2) = 1.0,


Will that work for you?

-Matt

	


> On Aug 30, 2016, at 2:23 AM, Charlotte Breitkreuz <cbreitkreuz at marum.de> wrote:
> 
> Dear MITgcm developers,
> 
> this concerns only the ctrl pkg. I think there are two things that possibly need to be changed in the control pkg - if I'm not wrong. And I also have one question.
> Consider a configuration where the exf fields atemp, aqh, precip, uwind, vwind  are used as control variables and also some 3d generic variables, for example initial tracer fields, are additional control variables. Therefore it is set #define ALLOW_ATEMP_CONTROL,..., and #define ALLOW_GENARR3D_CONTROL, and hence ctrlUseGen = TRUE (see ctrl_readparms.F).
> 
> Two problems arise in this setting:
> In exf_getffields.F (from line 480) it is:
> 
> if (.NOT.ctrlUseGen) then
> 
> #ifdef ALLOW_ATEMP_CONTROL
>       CALL CTRL_GET_GEN (
>      &     xx_atemp_file, xx_atempstartdate, xx_atempperiod, ....
> 
> ....
> endif
> and so on. So ctrl_get_gen is never called for these exf fields (atemp, aqh, precip. swdown, lwdown) if I use a generic controlvariable at the same time. Is this on purpose? Why?
> 
> 
> And the second problem is that again in exf_getffields.F (from line 573)
> 
> CALL CTRL_GET_GEN (
>      &     xx_uwind_file, xx_uwindstartdate, xx_uwindperiod, ....
> 
> is called. Then in ctrl_get_gen (from line 115):
>       if ( (optimcycle .ge. 0).AND.ctrlUseGen ) then
>         ilgen=ilnblnk( xx_gen_file )
>         write(fnamegen(1:80),'(2a,i10.10)')
>      &    xx_gen_file(1:ilgen),'.effective.',optimcycle
>       endif
> 
> ....
> call active_read_xy( fnamegen, ....
> ....
> This will occur as problem at run time, because xx_uwind.effective* was never written. The xx_*effective* file is just written for the generic control variables in ctrl_map_ini_genarr.F. So this problem again only comes up when the exf fields (uwind,vwind,evap,...) are used together with ctrlUseGen=TRUE.
> 
> 
> And then I have one more question. What is this xx*effective*  even for? Why not use the normal xx* name as it is done for not-generic control variables?
> 
> 
> Best wishes,
> 
> Charlotte 
> 
> -- 
> Charlotte Breitkreuz, M. Sc. Technomathematik 
> PhD Student 
> Research Group 'Geosystem Modeling'
> MARUM - Center for Marine Environmental Sciences, University of Bremen 
> FB5 - Department of Geosciences, University of Bremen 
> Klagenfurter Straße 2 
> 28359 Bremen, Germany 
> Room 5420
> 
> Tel: +49(0)421 - 218 65448
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20160831/b80cbb9e/attachment.htm>