input files for dates after 2005

[raeder]

I'm trying to make some CAM5 forecasts for real dates (gregorian calendar) for 2008.
I chose
/fis/cgd/cseg/csm/inputdata/lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp2.6_simyr1850-2100_c110124.nc
instead of the default
surfdata.pftdyn_0.9x1.25_simyr1850-2005_c091008.nc.

Is that good choice? Is there a better one?

I get an error message:
pftdyn_init mismatch between input pctspec = 96.0000000000000000
and that obtained from surface dataset 100.000000000000000 at g= 311

I can't find anything like pctspec in those input files, so I don't know how to search
for a better file.
I also can't see in the meta data which 'surface dataset' (e.g. fsurdat in clm_inparm?)
was used to make each surfdata... file. So I can't use that to search for a better file.

How can I find a 'surface dataset' which is consistent with the surfdata file I've chosen?

Could I have avoided this problem by specifying some date or dataset in my build
of CAM5?

Thanks,
Kevin

 

[slevis]

Kevin,

Let's discuss this in person.

Sam

 

[slevis]

Kevin and I determined that he needed to use this combination of files:

fpftdyn = '$DIN_LOC_ROOT/lnd/clm2/surfdata/surfdata.pftdyn_0.9x1.25_rcp2.6_simyr1850-2100_c100323.nc'

fsurdat = '$DIN_LOC_ROOT/lnd/clm2/surfdata/surfdata_0.9x1.25_simyr1850_c091006.nc'

We confirmed this by creating a new case:
create_newcase -mach bluefire -compset B_RCP2.6_CN -res f09_g16 -case testing2.6

...then configuring the case and looking at the clm.buildnml.csh

Sam