Welcome to the new DiscussCESM forum!
We are still working on the website migration, so you may experience downtime during this process.

Existing users, please reset your password before logging in here: https://xenforo.cgd.ucar.edu/cesm/index.php?lost-password/

Problems running cesm1.2.2 F_2000_CAM5, ERROR dt limit

Hi,    I am using theCESM1.2.2,F_2000_CAM5,where CAM5+CLM4+CICE+DOCN is used,the resolution is f09_g16. Starting simulation on Jan.1, the simulation stops in 15 days after the simulation. The error is exampled as follows: (shr_strdata_advance) ERROR: dt limit1    1126.00000000000   29.5000000000000        1.50000000000000 (shr_strdata_advance) ERROR: dt limit2        10116       43200       10215           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    1126.00000000000   29.5000000000000        1.50000000000000 (shr_strdata_advance) ERROR: dt limit2        10116       43200       10215           0 (shr_strdata_advance) ERROR: dt limit1    1126.00000000000    29.5000000000000        1.50000000000000
 

hannay

Cecile Hannay
AMWG Liaison
Staff member
From the error it seems that you are trying to run for a period for which you don't have data. Cecile 
 
Hi, I have the same problem with CESM1.2.2, compset F20C5TR, resolution f09_f09. I wonder if you have resolved this issue?
In my case I am trying to use custom SSTICE file which I created from previous experiments.
The Error is:
(shr_sys_abort) ERROR: (shr_strdata_advance) ERROR dt limit

(shr_strdata_advance) ERROR: dt limit1 62.0000000000000

(shr_strdata_advance) ERROR: dt limit2 19550201 0 19550301

These are my settings in the env_run.xml:
<!--"Prescribed SST and ice coverage stream name. (char) " --
<entry id="SSTICE_STREAM" value="CAMDATA" />
<!--"Prescribed SST and ice coverage data file name. (char) " -->
<entry id="SSTICE_DATA_FILENAME" value="$DIN_LOC_ROOT/atm/cam/sst/TSICE.e.e122.E_1850_CAM5.f09_g16.t4Ind_stg_co2b_cam5.cam.h0.1-49.nc" />
<!--"Prescribed SST and ice coverage grid file name. (char) " -->
<entry id="SSTICE_GRID_FILENAME" value="$DIN_LOC_ROOT/atm/cam/ocnfrac/domain.camocn.0.9x1.25_gx1v6_090403.nc" />
<!--"The model year that corresponds to SSTICE_YEAR_START on the data file. (integer) " -->
<entry id="SSTICE_YEAR_ALIGN" value="1955" />
<!--"The first year of data to use from SSTICE_DATA_FILENAME. (integer) " -->
<entry id="SSTICE_YEAR_START" value="1" />
<!--"The last year of data to use from SSTICE_DATA_FILENAME. (integer) " -->
<entry id="SSTICE_YEAR_END" value="49" />

and here is my docn_ocn_in:
&shr_strdata_nml
datamode = 'SSTDATA'
domainfile = '/oasis/scratch/comet/divanova/temp_project/cesm1_2_2/inputdata//share/domains/domain.ocn.0.9x1.25_gx1v6_090403.nc'
fillalgo = 'nn'
fillmask = 'nomask'
mapalgo = 'bilinear'
mapmask = 'dstmask'
streams = "docn.streams.txt.prescribed 1955 1 49"
taxmode = 'cycle'
tintalgo = 'linear'
/
 

dbailey

CSEG and Liaisons
Staff member
This usually indicates some missing data. Does your SST/ICE file have monthly means? Are all of the years there from year 1 to 49? The error is indicating that there is a bigger gap in the data than what the model wants to interpolate between. So, for example if the model is expecting monthly mean data and there is a month missing, then this would be a gap of 2 months, which is larger than the dt_limit of 1.5.
 
This usually indicates some missing data. Does your SST/ICE file have monthly means? Are all of the years there from year 1 to 49? The error is indicating that there is a bigger gap in the data than what the model wants to interpolate between. So, for example if the model is expecting monthly mean data and there is a month missing, then this would be a gap of 2 months, which is larger than the dt_limit of 1.5.
Yes, indeed! I figured it out meanwhile. I created my SST file from CAM TS output and concatenated the monthly files in one with nco. In the final file I had time axis which was centered in the end of the month rather than in the middle (I guess something to do with the CAM set up?). After I shifted the time axis to have monthly values in the middle of the month the "dt limit" error was resolved. Thanks!
 
Hi, I am encountering something similar. Perhaps there is some insight into my issue?
-Jonathan


My stream data is daily SST and ICE.

(shr_strdata_advance) ERROR: dt limit2 20102 0 20103 0
(shr_strdata_advance) ERROR: dt limit1 367.000000000000 1.00000000000000 1.50000000000000

> cat docn_ocn_in
&shr_strdata_nml
datamode = 'SSTDATA'
domainfile = '/scratch/snx3000/jbuzan/inputdata/share/domains/domain.ocn.gx1v6.090206.nc'
fillalgo = 'nn'
fillmask = 'nomask'
mapalgo = 'bilinear'
mapmask = 'dstmask'
streams = "docn.streams.txt.prescribed 1 1 34"
taxmode = 'cycle'
tintalgo = 'linear'
/

ncdump
data:
time = 731, 732, 733, 734, 735, 736, 737, 738, 739, 740, 741, 742, 743, 744,
745, 746, 747, 748, 749, 750, 751, 752, 753, 754, 755, 756, 757, 758, ... 1089, 1090, 1091, 1092, 1093, 1094, 1095, 1096, 1097, 1098, 1099, 1100, 1101, 1102,

date = 10101, 10102, 10103, 10104, 10105, 10106, 10107, 10108, 10109, 10110,
10111, 10112, 10113, 10114, 10115, 10116, 10117, 10118, 10119, 10120, ... 11224, 11225, 11226, 11227, 11228, 11229, 11230, 11231, 20101, 20102,
 
I have remapped the time to start from 1, and the date to start from year zero, same problem:

memory_write: model date = 10102 0 memory = 68.32 MB (highwater) -0.00 MB (usage) (pe= 768 comps= OCN)
(shr_strdata_advance) ERROR: dt limit1 2.00000000000000 1.00000000000000 1.50000000000000
(shr_strdata_advance) ERROR: dt limit2 10102 0 10103 0
(shr_strdata_advance) ERROR: dt limit1 2.00000000000000 1.00000000000000 1.50000000000000

Additionally, I constructed a 12 month file based upon the 33 time series, with appropriate 1/2 month time and half month dates. This executed correctly. However, I am trying to maximize the variability imposed upon the atmosphere, which is why I am attempting to use daily SST and ICE.
 
Top