conservation error

[jjakacki@iopan_gda_pl]

Hello,i running a coupled ice-ocean model for our small sea (Baltic). It is exactly pop and cice. Currently after 3 month integration a got an error: area*tracer conserv error
 tracer index =          11
 Initial global area*tracer = -1.892460547955886E-010
 Final global area*tracer = -1.892460547922005E-010
 Fractional error = -1.790331530386112E-011
 atsum_final-atsum_init =  3.388131789017201E-021
 istep1, cat =        1848           3
 transport: conservation error, cat            3
(shr_sys_abort) ERROR: ice remap transport: conservation error
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping Could anybody give me advice how to try to fix it please?Thank you for any help,regards,jaromir jakacki

 

[jjakacki@iopan_gda_pl]

I did some reserach and ice parameter:
xndt_dyn = 2.0
fixed this problem.
thanks,
jaromir

 

[dbailey]

Hi Jaromir,
As you guessed, this is a dynamical instability or a CFL violation. Reducing the dynamic timestep (increasing the number of dynamic steps to thermodynamic steps) often takes care of it. If possible, it is usually better to decrease the overall timestep (dt) for the CICE model. In the CESM, this is tied to the coupling interval and can be changed via ICE_NCPL. However, the system also wants ICE_NCPL = ATM_NCPL = LND_NCPL.
http://www.cesm.ucar.edu/models/cesm1.1/cesm/doc/modelnl/env_run.html#run_start_vars
Dave

 

[subh08@lzu_edu_cn]

I also have run a experiment with a 1120 ppm CO2, I have run a long-term intergration about 3500 model years. After that, I came across the same model collapse porblem. area*tracer conserv error tracer index =          23 Initial global area*tracer =  4.923021510038051E-011 Final global area*tracer =  4.923021510165106E-011 Fractional error =  2.580832560432242E-011 atsum_final-atsum_init =  1.270549420881451E-021 istep1, cat =       10680           1 transport: conservation error, cat            1(shr_sys_abort) ERROR: ice remap transport: conservation error(shr_sys_abort) WARNING: calling shr_mpi_abort() and stoppingWhy? I guess it may results from the negative sea-ice or other error.