Coupler History File for b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5 case

[Johnny]

I am trying to do a coupler history spin-up simulation.
When I use coupler history files from Globus, everything works fine.

However, when I switched to the BHIST coupler history files that my advisor uploaded for me from /glade/campaign/collections/cmip/CMIP6/cpl_hist/b.e21.BHIST.f09_g17.CMIP6-historical.011/
I got an error in the simulation:
ERROR: (shr_ncread_varDimNum) ERROR inq varid: doma_lat

I found the differences between the two datasets:
In the downloaded b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5.102 case from Globus, I have a variable doma_lat in every January file.

ncdump -h b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5.102.cpl.ha2x1h.2015-01.nc | grep doma_lat
double doma_lat(doma_ny, doma_nx) ;
doma_lat:_FillValue = 0. ;
doma_lat:units = "degrees north" ;
doma_lat:long_name = "latitude" ;
doma_lat:standard_name = "latitude" ;
doma_lat:internal_dname = "doma" ;

However, in the b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5.102 case, variable doma_lat does not exist in the 2005-01 file:
ncdump -h b.e21.BHIST.f09_g17.CMIP6-historical.011.cpl.ha2x1h.2005-01.nc | grep doma_lat
returns nothing.
The variable does show up in the 2006-01 file:
ncdump -h b.e21.BHIST.f09_g17.CMIP6-historical.011.cpl.ha2x1h.2006-01.nc | grep doma_lat
double doma_lat(doma_ny, doma_nx) ;
doma_lat:_FillValue = 0. ;
doma_lat:units = "degrees north" ;
doma_lat:long_name = "latitude" ;
doma_lat:standard_name = "latitude" ;
doma_lat:internal_dname = "doma" ;

After I checked every year between 2005 and 2014, I found the doma_lat variables only exist in the January 2006 and January 2012 files. This variable should be expected in every January file since it appeared in every January file from Globus.

Where can I find coupler history files that have consistent variables for b.e21.BSSP585cmip6.f09_g17.CMIP6-SSP5-8.5 case?

Thanks!

 

[oleson]

There is a script that is used to convert the coupler history data from daily to monthly. The script also handles the issue you mention. So I imagine maybe you need to run the script on the historical coupler history data. I've put the script and an example of how to run it on our ftp site.

ftp://ftp.cgd.ucar.edu/pub/oleson/concat.command
ftp://ftp.cgd.ucar.edu/pub/oleson/concat_cpl_hist_mon.csh

 

[Johnny]

Hi Keith,
Thanks for responding.
I did run two similar commands that my advisor gave to me.

when I examine the csh, I do see
# set dom_vars
set dom_var0 = doma_aream
set dom_vars = doma_lon,doma_lat,doma_mask
if ($aname == "hr2x") then
set dom_var0 = domrb_aream
set dom_vars = domrb_lon,domrb_lat,domrb_mask
endif

and later on,
ncks -A -v $dom_var0,$dom_vars $fname_in0 $fname_out

I will debug and run the script piece by piece to see what went wrong on my part.

Again, thank you! This is really helpful!