Output file from GOTM+FABM

[andic...@gmail.com]

Hi dear GOTM users,

I had installed, compiled and run GOTM with FABM. I'm following the suggestion of first try the couette test-case. So far so good, but in my output file couette.nc I can't see any variable of the NPZD model. I'm calling the corresponding module (gotm_fabm_output.F90). Any idea what I'm missing?

Thanks in advance,

Adriana Huerta-Casas, PhD

Theoretical Ecology and Modelling
Leibniz Center for  Tropical Marine Ecology (ZMT)
Fahrenheitstrasse 8
28359 Bremen
Germany

[Gennadi Lessin]

Hi,

As a first suggestion: do you have file gotm_fabm.nml in your testcase directory and is fabm_calc = .true. there?

I would also suggest to use fabm-users group for FABM - related questions.

Cheers

Gennadi

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Karsten Bolding]

just to add to what Gennadi has already written. It is better you use the nns_annual test case as it has all the forcing for a proper bio-geochemical simulation. The cuette case is an idealised case with e.g. no short wave radiation.

Karsten

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Karsten Bolding

kar...@bolding-bruggeman.com

+4564422058

[Jorn Bruggeman]

Hi Adriana,

 

To activate FABM, you’ll need to provide both gotm_fabm.nml and fabm.yaml (see also section 2.4 on http://fabm.net/wiki). My guess is that you are missing gotm_fabm.nml – you can copy that from the testcases/GOTM directory in the FABM source code.

 

Cheers,

 

Jorn

 

--

[andic...@gmail.com]

Hi Gennadi, Karsten and Jorn,

I do have gotm_fabm.nml and fabm.yaml in the directory of the test case. Now I just run it for the nns_annual test case, as suggested. I still don't see the output of the npzd in the nns_annual.nc file.

I figured out that I don't have a path for NETCDF_FMT, but dunno if I need to do something with it, actually.

It seems that is not reading the npzd model, but don't get why?

Cheers,
Adriana

[Karsten Bolding]

did you follow Jorns suggestions?

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Fenjuan Hu]

Hi Adriana

     Follow Jorn’s suggestion, check out  http://sourceforge.net/p/fabm/wiki/GOTM_FABM/ and copy the whole parameter sets for  gotm_fabm.nml from there.  Try to see what happens. It works for me

 

Best

Fen

[andic...@gmail.com]

Hi!

Indeed, I followed Jorn's suggestion.

Just to see if I understood everything fine, I need to setup GOTM, and in the file of the test cases (mine is : /Users/ahc/GOTMCASES/gotm-cases/nns_annual) I have the corresponding *.nml files, including gotm_fabm.nml. Here I also have fabm.yaml (renemed from fabm-npzd-carbonate.yaml).

In this same directory I have the link to the executable of GOTM and here I run GOTM.

FABM is installed as a library as suggested here: http://sourceforge.net/p/fabm/wiki/cmake/#gotm-fabm in The FABM library.

Well, if all these is correct, there must be something commented out or a call is missing ?!??!

Cheers,
Adriana



On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[Jorn Bruggeman]

Hi Adriana,

 

A last option is that GOTM may have been compiled without FABM support (you are using the developer’s version from the git repository, right?). When you run GOTM, does it mention anything at all related to FABM among the messages you get when the model is running?

 

Cheers,

 

Jorn

 

--

[andic...@gmail.com]

Hi Jorn,

I thought in that, but iIt does write:

...
Compilation options:
 ------------------------------------------------------------------------
    Fortran 95 compilation
    Fortran 2003 compilation
    _FABM_
    SEAGRASS


Is _FABM_ the only thing it needs to write?

I'm trying to debug in the subroutines, i.e. printing out comments, but the ones of npzd.F90 are now written.

Yes, I pull FABM, GOTM, and GOTM cases from git.

cheers,
adriana

On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[andic...@gmail.com]

 Earlier I meant, the .. comments are NOT written... I suspect this routine is not being read.

Adriana

On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[Karsten Bolding]

please post the content of fabm.yaml and gotm_fabm.nml

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[andic...@gmail.com]

gotm_fabm.nml:

!-------------------------------------------------------------------------------
&gotm_fabm_nml
   fabm_calc = .true.,
   cnpar = 1.0,
   w_adv_discr = 6,
   ode_method = 3,
   split_factor = 1,
   bioshade_feedback = .true.,
   salinity_relaxation_to_freshwater_flux = .false.
   save_inputs = .true.
/

On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[Karsten Bolding]

the fabm.yaml file is not a YAML file - just a namelist file you have renamed to fabm.yaml.

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[andic...@gmail.com]

Ah, sorry,  I'm not familiar with .yaml formats. Is the attached file a yaml file? Or, from where can I  download it?

Thanks a lot for your help,
Adriana

On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[Gennadi Lessin]

That is exactly what you need.

You can read about usage and structure of yaml files at FABM wiki Section 2.3:

http://sourceforge.net/p/fabm/wiki/Setting_up_a_simulation/

Does the biogeochemical model produce output now?

Cheers

Gennadi

[andic...@gmail.com]

Unluckilly not :( I'm double checking paths, etc...

cheers,

Adriana


On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[Fenjuan]

Hi Adriana

    The yaml files are under directory "fabm_code_dir"/tesecases

Looking for the files with .ymal.

Good luck. :)

Fen

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

<fabm.yaml>

[andic...@gmail.com]

Hi all,

Thanks for all your help. I needed to do 'make realclean' to GOTM, so now it works!!

Have a nice productive day :)

Adriana


On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote:

[Karsten Bolding]

Hello

You should really use the devel version and the CMake based build system instead of 'make'

Karsten

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[andic...@gmail.com]

Hi Karsten,

Sorry for reply so late.

It may be a bit naive my question, but I followed the steps from here: http://sourceforge.net/p/fabm/wiki/cmake/ (although I'm working in a Mac OS X). So I'm not using the Cmake build system? Is the version from git not devel?  I hope is not rather a basic basic concern.

Cheers,

Adriana




On Tuesday, June 23, 2015 at 1:31:51 PM UTC+2, andic...@gmail.com wrote: