Abnormal blow up

dynamics-numerics-params
,
1

Hello CROCO Users,

I am setting up a configuration for the coastal region of mediterranean sea with CROCO. Upon using the thrifty crocotools box, I manage to see up my model domain etc. However, I notice my initial condition files were filled with NaN values.

Thinking this might be a one off issue that might to be relevant, I set up my croco.in file and other pre-processing files. On running the model, I was greeted with a flat error!

 STEP   time[DAYS] KINETIC_ENRG    POTEN_ENRG    TOTAL_ENRG    NET_VOLUME   trd
       0     0.00000 0.000000000E+00           NaN           NaN 9.0658100E+13  0


 =======================================
 =                                     =
 =   STEP2D:   ABNORMAL JOB END        =
 =                 BLOW UP             =
 =                                     =
 =======================================


  VMAX (M/S) =   NaN
  IMAX JMAX  =    35     9
  IINT IEXT  =     1     1

I assume the error is either in my initial condition which has the NaN values (as the Vmax calculated by the model during runtime shows NaN) or my timestep/configuration in croco.in is wrong.

I will be glad if I can be pointed to ways to solve this issue. Thanks

image descriptioncroco.in(/upfiles/16533218033242173.in)

2

Hi ify,

I am wondering if you have already solved this issue, since I am facing the same error, and I also found my initial file have some NaN values, and trying to solve it now.
Thanks in advance.

Jin

3

Hi,

Yes, the croco_ini.nc file should not contain any NaNs.

Please post your crocotools_param.m file to understand better the error.

4

Hi Andres,

Thank you very much for the help.

But I am not able to attach my file into here, it says ‘Sorry, new users can not upload attachments.’ .

Should I post the raw text here?

Updates :

Just figured out how to attach files here.
Here is my file, thanks in advance.
crocotools_param.m (19.0 KB)

5

Great, thanks.

Seems you are doing a special configuration, forced by a CMIP6 run. I haven´t seen codes for that in the default CROCO_TOOLS scripts

Try to do a climatological (basic) run, 1 month. If all goes well, the problem would by, my guess, in your code to create initial conditions from that CMIP6 run.

Is it available somewhere, Github?

Please try the default run first.

6

Thanks so much for the help.

For make the forcing files, I have actually edited the original ‘make_forcing.m’ file to adjust it for my forcing data. But maybe there are some issuses in there, so I will attach it.
make_forcing.m (6.1 KB)

And also, not only my ‘croco_ini.nc’ have NaNs, but also my ‘croco_clm.nc’ have NaNs too. I don’t know if these two have any relation.

I’m sorry but I am not sure I fully understood this part, but about the ‘make_ini.m’ file for creating the initial condition for my case, I did not edit it at all, so should I also edit it? For instance, replace the data here :

(code from make_ini.m)
% Data climatologies file names:
%
% temp_month_data : monthly temperature climatology
% temp_ann_data : annual temperature climatology
% salt_month_data : monthly salinity climatology
% salt_ann_data : annual salinity climatology
%
temp_month_data = [climato_dir,‘temp_month.cdf’];
temp_ann_data = [climato_dir,‘temp_ann.cdf’];
insitu2pot = 1; %1: transform in-situ temperature to potential temperature
salt_month_data = [climato_dir,‘salt_month.cdf’];
salt_ann_data = [climato_dir,‘salt_ann.cdf’];
%