errors running cam5 with user-created prescribed sst file

[hoell@geog_ucsb_edu]

Hello,

I'm trying to run CAM5 with prescribed SST from a .nc file created using the "SST and sea ice dataset tool" found on the CAM3.1 page as a guideline. CAM5 builds properly and runs for 15 days.

The model-configure I'm using is:

/home/hoell/Desktop/cesm/experiments/source/control/models/atm/cam/bld/configure
-dyn fv -hgrid 1.9x2.5 -phys cam5 -ocn dom -spmd -nosmp -ntasks 12
-linker mpif90 -test

The namelist-configure I'm using is:

/home/hoell/Desktop/cesm/experiments/source/control/models/atm/cam/bld/build-namelist
-test -case observed_sst_2010 -runtype startup -ignore_ic_date
-namelist "&camexp start_ymd=20100101 start_tod=0
stream_year_first=2010 stream_year_last=2012 stop_option='nmonths'
stop_n=31 nhtfrq=-24 ndens=1 mfilt=365 sstcyc=.False.
bndtvs='/home/hoell/Desktop/cesm/inputdata/atm/cam/sst/HadISST_200901-201208_observed.nc' empty_htapes=.true.
fincl1='T:A','PS:A','OMEGA:A','SST:A','U:A','V:A','Q:A','Z3:A','TS:A','PRECC:A','PRECL:A','DTCOND:A','SHFLX:A','LHFLX:A','FLNS:A','FSDS:A'
fincl2='T:A','PS:A','OMEGA:A','SST:A','U:A','V:A','Q:A','Z3:A','TS:A','PRECC:A','PRECL:A','DTCOND:A','SHFLX:A','LHFLX:A','FLNS:A','FSDS:A'
fincl3='T:A','PS:A','OMEGA:A','SST:A','U:A','V:A','Q:A','Z3:A','TS:A','PRECC:A','PRECL:A','DTCOND:A','SHFLX:A','LHFLX:A','FLNS:A','FSDS:A'/"

After running for 15 days the model aborts and displays the errors:

SSTINT: Read sst for date (yyyymmdd)      20100215  sec             0
(shr_dmodel_readLBUB) reading file:
/home/hoell/Desktop/cesm/inputdata/atm/cam/sst/HadISST_200901-201208_observed.nc
     14
 (shr_strdata_advance) ERROR: dt limit1     153.0000000000000
 (shr_strdata_advance) ERROR: dt limit1     153.0000000000000
    29.50000000000000         1.500000000000000
 (shr_strdata_advance) ERROR: dt limit1     153.0000000000000
    29.50000000000000         1.500000000000000
 (shr_strdata_advance) ERROR: dt limit2      20100116        43200     20100215
 (shr_strdata_advance) ERROR: dt limit1     153.0000000000000
    29.50000000000000         1.500000000000000
 (shr_strdata_advance) ERROR: dt limit2      20100116        43200     20100215
             0

I think that a date variable (time, date or datesec) in the prescribed SST .nc file isn't written exactly as the model requires, so the model is having difficulty reading and stepping from one time to the next. However, I'm having trouble troubleshooting the problem because I don't know what the variable dt limit 1 means. Lines 635-647 of shr_strdata_mod.F90 contain the portion of code where the errors occur, but I haven't been able to use this information to solve this problem.

Does anyone have any insights or a possible solution?

Thanks,
Andy

 

[eaton]

I would suggest removing "-ocn dom" from the configure command. That is forcing the use of the old data ocean model component rather than using the newer "docn" component which is the default. The namelist variables that you set, stream_year_first and stream_year_last, are recognized by docn, but not by dom. Also, when switching to docn you can get rid of the namelist variable sstcyc which is only recognized by dom.

Next I'd look at one of the default SST datasets used for AMIP runs, for example,
$CSMDATA/atm/cam/sst/sst_HadOIBl_bc_0.9x1.25_1850_2008_c100127.nc, and just make sure that the new sst file contains the same coordinates and metadata.

 

[hoell@geog_ucsb_edu]

Thanks for your suggestions and comments. Unfortunately, the modifications to the configure and namelist didn't solve the problem. On the plus side, at least we've identified the problem as the input SST file. This is still a little bit of a puzzler considering the input SST file I created was written (so I thought) exactly like the data sets that work from the repository. Thanks again and I'll update once I solve the problem.

 

[andrew_kren@colorado_edu]

hoell@,Were you able to figure out the problem with your dt limit errors? I am getting the same ones with my altered sst file. Was wondering if you figured out how to solve the problem?