problem running test case with gotm-git

106 views
Skip to first unread message

"Christoph Völker"

unread,
Aug 18, 2011, 7:06:23 AM8/18/11
to gotm-...@googlegroups.com
Dear all, 

I have just attempted to run two test cases (one from the web-page, the OWS-Papa setup, the other a set-up that Ying and I have been using for the TENATSO time-series station) with the newest version of the code which I got yesterday via the git repository. 
However, the model instantaneously stops in both cases with the error message 

 FATAL ERROR: I could not read airsea namelist
STOP init_airsea

My first interpretation is that some parameter name in the namelist has changed. I am probably able to find the error myself, but in case anyone knows already what is going on that might save some work. 

I am not using the GUI, and I am doing this on a macbook under Mac Os 10.6.8, using a fairly recent version of gfortran (4.6.0). For inspection, I enclose my Rules.make file. The compilation went well, although I got several warnings about conversions from real*8 to real*4 and unused arguments in subroutine calls. 

Cheers, Christoph
 

Rules.make

Jorn Bruggeman

unread,
Aug 18, 2011, 7:21:09 AM8/18/11
to gotm-...@googlegroups.com
Hi Christoph,

Several airsea variables have indeed changed since 4.0. A quick copy-paste
from gui.py/schemas/scenario/gotm-4.0.0_gotm-4.1.0.converter shows the
following name changes:

<link source="/airsea/airsea/wet_mode"
target="/airsea/airsea/hum_method"/>
<link source="/airsea/airsea/p_e_method"
target="/airsea/airsea/precip_method"/>
<link source="/airsea/airsea/const_p_e"
target="/airsea/airsea/const_precip"/>
<link source="/airsea/airsea/p_e_flux_file"
target="/airsea/airsea/precip_file"/>
<link source="/bio/bio_nml/bio_npar" target="/bio/bio_nml/npar"/>

Additionally, the heatflux input file has been split in separate files for
heat and short-wave radiation. Not sure it that is an issue for OWS-Papa.

Cheers,

Jorn

PS not losing any opportunity for a GUI-advertisement: the GUI can also be
used as a quick tool to converting old namelist-based scenarios to namelists
for the latest developers version - that also takes care of things like
splitting the heatflux file.

Karsten Bolding

unread,
Aug 18, 2011, 7:27:46 AM8/18/11
to gotm-...@googlegroups.com
On Thu, Aug 18, 2011 at 01:06:23 +0200, "Christoph V�lker" wrote:
> Dear all,�
>
> I have just attempted to run two test cases (one from the web-page, the OWS-Papa setup, the other a set-up that Ying and I have been using for the TENATSO time-series station) with the newest version of the code which I got yesterday via the git repository.�

The test cases from the table are not compatible with the code from git.

Code in git is developers version.

Further down are given test cases for developers version.

<snip>

Karsten


--
http://www.gotm.net/

Reply all
Reply to author
Forward
0 new messages