How to deal with problem "interval was not found for col"

[xpan@cola_iges_org]

Hi, all

I am using CCSM3_0_beta19 to have some runs. But the model crashed (at step 290) after integrated for 2 days. The error message looks like this in cam3 log file :

> -------------------------------------------------------------------
nstep, te 289 3355421958.30737352 -14.2641850471496578 0.142368352173723070E-02 98469.3890214040002
COURLIM: *** Courant limit exceeded at k,lat= 1 23 (estimate = 1.042), solution has been truncated to wavenumber 81 ***
*** Original Courant limit exceeded at k,lat= 1 23 (estimate = 1.042) ***
NSTEP = 289 8.810472866153245E-05 6.543300124892506E-06 253.866 9.84694E+04 2.522928814083643E+01 1.04 0.16
nstep, te 290 3355389069.27055120 -14.4480213717619570 0.144203171884353036E-02 98469.4017615097255
COURLIM: *** Courant limit exceeded at k,lat= 1 23 (estimate = 1.044), solution has been truncated to wavenumber 81 ***
*** Original Courant limit exceeded at k,lat= 1 23 (estimate = 1.044) ***
NSTEP = 290 8.810776386657750E-05 6.715165363693108E-06 253.863 9.84694E+04 2.522950943387390E+01 1.04 0.16
nstep, te 291 -NaNQ -NaNQ NaNQ -NaNQ
interval was not found for col i 1

> ----------------------------------------------------------------------

Do anyone have any experience/suggestion to overcome this problem? Thanks a lots for your time

 

[rneale]

I presume you are running an Eulerian version of the model (is it T42, T85?). The errors you are getting is an indication the zonal winds are getting way to strong at upper levels in high-latitude regions. Are you running a non-standard configuration in any way? If not, my first suggestion would be to reduce the timestep.

 

[forums@ifort_co_in]

In the CCSM3.0 runs that were done with 2xCO2, 4xCO2, and 1% increasing CO2, what are these in reference to? 1990 CO2 values? How many ppmv is 1xCO2?Also, what type of ocean was used in these runs? Dynamical, swamp, data ...? I would like to retrieve the increase in SST values between a control run and an increased CO2 run, I want to make sure this would be possible to do.