Error while processing Climatology multi-files:

[ljd]

Hi everyone,

I am running the model, the input files has been read by ROMS, and I meet with a problem . can anybody help me? Thanks.

ljd


CHECK_MULTIFILE - Error while processing Climatology multi-files:
data does not include initialization time = 0 00:00:00

1 00:00:00 46 00:00:00 IO_clm_100LO_v3_01.nc
51 00:00:00 96 00:00:00 IO_clm_100LO_v3_02.nc
101 00:00:00 146 00:00:00 IO_clm_100LO_v3_03.nc
151 00:00:00 196 00:00:00 IO_clm_100LO_v3_04.nc
201 00:00:00 246 00:00:00 IO_clm_100LO_v3_05.nc
251 00:00:00 296 00:00:00 IO_clm_100LO_v3_06.nc
301 00:00:00 346 00:00:00 IO_clm_100LO_v3_07.nc
351 00:00:00 361 00:00:00 IO_clm_100LO_v3_08.nc

Elapsed CPU time (seconds):

Node # 0 CPU: 3.192

ROMS/TOMS - Output NetCDF summary for Grid 01:

Analytical header files used:

Node # 1 CPU: 3.192 Node # 3 CPU: 3.192
ROMS/Functionals/ana_nudgcoef.h
Node # 2 CPU: 3.192


Node # 6 CPU: 3.192
ROMS/TOMS - I/O error ............... exit_flag: 4


ERROR: I/O related problem.
Node # 7 CPU: 3.192 Node # 5 CPU: 3.192 Node # 4 CPU: 3.192


real 4.25
user 11.05
sys 1.49
+ exit 0

[rduran]

I wish all errors were as explicit, did you check what the error suggests you check?

Code: Select all

CHECK_MULTIFILE - Error while processing Climatology multi-files: 
data does not include initialization time = 0 00:00:00

[ljd]

Hi rduran, Thanks for your reply! The log file is attached. Hope you can help me to find the problem.thanks!

[guokunlv]

It seems, the clim files don't include data record for time=0.Your clm data(clm_time =1:365?)started from 1 while the model need data start from time=0.
By the way, I didn't see the error in your *.log file except "blow up".

Best regards

[ljd]

Thank you very much for your reply,kunlv! The log file include all the info ROMS output.I am confused of the model blowing up.

[m.hadfield]

Thank you very much for your reply,kunlv! The log file include all the info ROMS output.I am confused of the model blowing up.

So am I, but can we please be clear about the status of the error you originally pointed out? On my reading of the log file you have asked ROMS to read temperature and salinity climatology files, but not to do any interior nudging towards them. ROMS has read in two records from those climatology files, at t=271 days and t=1 days. Are you saying that ROMS is not now producing the "time not found" error you originally reported? And, if it isn't, why not?. Whatever it is doing, I very much doubt that the handling of the climatology fields is what you intended. This may have something to do with the blow-up.

[kate]

The log file shows that your vertical Courant number got large. I don't know the stability limits for it but you might try a shorter timestep. You might also check the restart file to see how and where it blew up. Thirdly, diag.F is the routine reporting the trouble. I have found it useful to add more print statements to it in the case of blowing up. For instance, you get the same blow up for trouble in velocity and density. Which is it and where did it happen?

[shuwen]

Hi, I just want to extend this question more: would it be possible to put several climatological files with different variables? To be more specific, since climatological files require 2D & 3D velocity, temperature, salinity and zeta. However, if the data set is too big to write in one nc file, I am wondering are we allowed to provide 7 nc files, each one contains only one variable?

Thanks for any input for the question!

[kate]

This is not supported in the trunk code. However, I hacked this feature into my branch, available from

Code: Select all

git clone https://github.com/kshedstrom/roms