CESM1.0 FAMIPC5 output problem

[icefog11@yahoo_com]

Hi there,

I have encountered a problem when running CESM1.0 FAMIPC5 component set with f19_f19 resolution. I see that the chemistry input files only support model run up to 2005. But, when I set the model to run to Dec. 2005, the model could not complete the last month and could not generate the last monthly output (XXXX.cam2.h0.2005-12.nc). The error log files prompted to look at .../ccsm-1.0/test/run/cpl.log.XXX file ("Model did not complete - see .../ccsm-1.0/test/run/cpl.log.XXX"). But I didn't see any error message there - just as the following:

tStamp_write: model date = 20051230 0 wall clock = 2010-09-21 22:02:55 avg dt = 42.44 dt = 40.51
memory_write: model date = 20051230 0 memory = 313.55 MB (highwater) 606.36 MB (usage) (pe= 0 comps= cpl ocn atm lnd ice glc)
tStamp_write: model date = 20051231 0 wall clock = 2010-09-21 22:03:36 avg dt = 42.44 dt = 40.57
memory_write: model date = 20051231 0 memory = 313.55 MB (highwater) 606.36 MB (usage) (pe= 0 comps= cpl ocn atm lnd ice glc)

while the atm.log.XXX file does have some error info:
ENDRUN:
chem_surfvals_set_all: error - current date is past the end of valid data

But, I suppose the 2005-12 monthly history file still should have been generated, isn't it? Does someone know the reason please?
Thanks!

 

[icefog11@yahoo_com]

Does anyone have the same problems before??

 

[eaton]

The error from chem_surfvals_set_all occurs when it tries to read past the end of the dataset that is specified by the namelist variable bndtvghg. This variable should be set to use the dataset

bndtvghg = '/fis/cgd/cseg/csm/inputdata/atm/cam/ggas/ghg_hist_1765-2005_c091218.nc'

Where /fis/cgd/cseg/csm/inputdata should be replaced by the location specific to your system. This file contains data through the year 2006 so that interpolation to finish December 2005 is possible. Is that the dataset you're using? This setting is in the atm_in namelist file.

 

[icefog11@yahoo_com]

eaton,

Thanks for your reply! I've checked my 'bndtvghg' setting, and it was ghg_hist_1850-2005_c090419.nc, not ghg_hist_1765-2005_c091218.nc. Anyway, I will try out the data you memtioned.
BTW, are there new data available that can support the model to run beyond 2005? Thanks again!

 

[eaton]

Thanks for checking. This is a bug in the 1850-2005_cam5 use case file. The cam4 version is using the correct bndtvghg file, but the cam5 version is specifying the older file which has the problem you reported. I'll get this fixed in the next release.

At the current time we have only released datasets that go through 2005.

 

[devarajun@gmail_com]

How to download the input-data for 50years, I need to download the data for each year and how to download inputdata for all the years.

Thanks,
Dev