CAM+OML (E compset) run time error for pacemaker experiment: soil balance error

[climatemodelling@gmx_com]

Dear Forum,I am running CAM with an ocean mixed layer (climatological fluxes were obtained from a previous run as stated in the model documentation).Further, I use a mask to restore SSTs in a specified ocean region back to fixed values obtained which are climatological SSTs plus an anomaly component. The model runs without a problem for about 15 days but then crashes due to some soil balance errors in the land model. This confuses me, as I only use fixed SSTs in a region that does not include any land points. Any hints what might cause this issue? I attach the major error messages from the log file below. Thanks!  BalanceCheck: soil balance error nstep = 3 point = 4064 imbalance = -0.000003 W/m2
BalanceCheck: soil balance error nstep = 4 point = 4064 imbalance = -0.000003 W/m2
BalanceCheck: soil balance error nstep = 21 point = 2041 imbalance = -0.000007 W/m2
BalanceCheck: soil balance error nstep = 22 point = 2041 imbalance = -0.000007 W/m2
BalanceCheck: soil balance error nstep = 49 point = 4066 imbalance = -0.000001 W/m2
BalanceCheck: soil balance error nstep = 50 point = 4066 imbalance = -0.000001 W/m2(and so on...)[hpc1069:29162] *** Process received signal ***
[hpc1069:29162] Signal: Segmentation fault (11)
[hpc1069:29162] Signal code: Address not mapped (1)
[hpc1069:29162] Failing at address: (nil)
[hpc1069:29164] *** Process received signal ***
[hpc1069:29164] Signal: Segmentation fault (11)(and so on...)(shr_strdata_advance) ERROR: dt limit1 395.000000000000
32.0000000000000 1.50000000000000
(shr_strdata_advance) ERROR: dt limit2 19580115 0 19580216
0
(shr_sys_abort) ERROR: (shr_strdata_advance) ERROR dt limit
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
(shr_strdata_advance) ERROR: dt limit1 395.000000000000
32.0000000000000 1.50000000000000
(shr_strdata_advance) ERROR: dt limit2 19580115 0 19580216

Furthermore, there seems to be some CFL violation, but I am not sure if that crashes the model (atmosphere model log):COURLIM: *** Courant limit exceeded at k,lat= 1 46 (estimate = 1.013), solution has been truncated to wavenumber 30 ***
*** Original Courant limit exceeded at k,lat= 1 46 (estimate = 1.013) ***
NSTEP = 1 8.826900025062978E-05 6.207719752792190E-06 250.336 9.84502E+04 2.327465191962012E+01 1.01 0.24
nstep, te 2 0.33040663336616769E+10 0.33040393505481038E+10 -0.74823966976817207E-03 0.98450234960668080E+05

 

[hannay]

The error "Courant limit exceeded at k,lat ..." is not crashing teh model either.
     => solution has been truncated to wavenumber 30 ***The error "soil balance error ..." is a warming. This is not the reason why the model is crashing. In which log file do you get:Signal: Segmentation fault (11)Is it possible to attach the log files ?   

 

[climatemodelling@gmx_com]

Hi Hannay, the error occurs in the ccsm log file (please find attached). Thanks!

 

[hannay]

Please could you attach the otehr log files too. Thanks. 

 

[climatemodelling@gmx_com]

Sure. Thanks again!

 

[hannay]

It seems to be a problem with the qfluxes fluxes you created:<br />/auto/ccr-01/ms_632/heatflux_cesm/myfiles/test3_som_forcing1x1_1958_2001_pacemaker1.nc<br /><br />The model run succesfully until Jan 15. Then it looks for the qfluxes for February 15 to start to interpolate between Jan 15 and Feb 15 but it cannot find them (see error)<br />Please check your qfluxes file.<br /><br />  <br />----- <br /><br />shr_dmodel_readLBUB) reading file: /auto/ccr-01/ms_632/heatflux_cesm/myfiles/test3_som_forcing1x1_1958_2001_pacemaker1.nc      14
 (shr_strdata_advance) ERROR: dt limit1    395.000000000000     
   32.0000000000000        1.50000000000000     
 (shr_strdata_advance) ERROR: dt limit2     19580115           0    19580216
           0
(shr_sys_abort) ERROR: (shr_strdata_advance) ERROR dt limit
(shr_sys_abort) WARNING: calling shr_mpi_abort() and stopping
<br /><br /><br /><br />

 

[climatemodelling@gmx_com]

Thanks, that hint solved my issue. The reference date for the time axis unit was off by one year.