problem about RUN_STARTDATE in B_1850-2000_CN compset

[yaozhixiong@foxmail_com]

Hi,

I am using compset B_1850-2000_CN in cesm1.2.0. There is a topic help me, but I still want to ask hanny some problem. The website of topic is https://bb.cgd.ucar.edu/node/1002273

Question 1:
The compset B_1850-2000_CN has year-to-year varying dataset.
create_newcase -case ../caseoutput/test_b1 -compset B_1850-2000_CN -res f19_g16 -mach szhpc
I want to run an experiment from 1948 to 2009, which is similar as above website.

As default, the setting in env_run.xml are




The model can run successfully.

Then, I modified it to




But now model cannot run with an error hanppend.
*****
POP aborting...
could not find field in binary header file
*****
But there is a binary header file(b40.1850.track1.2deg.003.pop.r.0501-01-01-00000.hdr) in inputdata directory.
Can you tell me how to fix it?



Question 2:
Hanny put 1950 initial conditions on the ftp server:
ftp ftp.cgd.ucar.edu
cd pub/hannay

But it does not exist now. Where can I download it?


Best regards,
yao

 

[klindsay]

I suspect that POP is trying to read CFCs from the restart file, but they are not present in the restart files from the 1850 control run.If the year of your start date is 1930.So you need to start your run from a 20th century run, using files like hannay previously supplied.The CGD ftp site get scrubbed automatically, so the files from 2012 are no longer there.I'm in the process of retrieving the files that hannay previously supplied.I'll post again when they are available on the CGD ftp site.

 

[klindsay]

It took less time than I thought it would.The files you need to do a hybrid run from b40.20th.track1.2deg.001 at 1950-01-01 are nowon ftp.cgd.ucar.edu in the directory /pub/klindsay/b40.20th.track1.2deg.001/1950-01-01-00000

 

[klindsay]

Files for b40.20th.track1.2deg.001 at 1948-01-01 are also on ftp.cgd.ucar.edu, in the directory/pub/klindsay/b40.20th.track1.2deg.001/1948-01-01-00000

 

[yaozhixiong@foxmail_com]

Thanks a lot.I think it is not a good idea to run a hybrid case from restart files of a particular year. For example, I should have initial condition restart files of 1948, which you had given to me, if I want to start a coupled model from 1948. And I should have another different restart files if model start from another time.As we know, initial condition files in b40.1850.track1.2deg.003 are the output of B_1850_CN with cyclical dataset. So, I want to run a 20th control run with cyclical dataset for several years. And then, make the restart file be initial condition for any other 20th control run.(1) which compset should I choose, B_2000 or B_2000_CN?At first, I run a case of B_2000 compset. I try to make the restart files the same as files in ccsm4_init. In order to generate 'hdr' file for POP, I added restart_fmt = 'bin' in user_nl_pop2. I run model for 1 year.Then, the restart files are shown bellow.rpointer.atm rpointer.drvrpointer.icerpointer.lndrpointer.ocn.ovfrpointer.ocn.restartrpointer.roftest_b2.cam.h0.0001-12.nctest_b2.cam.i.0002-01-01-00000.nctest_b2.cam.r.0002-01-01-00000.nctest_b2.cam.rs.0002-01-01-00000.nctest_b2.cice.r.0002-01-01-00000.nctest_b2.clm2.h0.0001-12.nctest_b2.clm2.r.0002-01-01-00000.nctest_b2.clm2.rh0.0002-01-01-00000.nctest_b2.cpl.r.0002-01-01-00000.nctest_b2.pop.r.0002-01-01-00000test_b2.pop.r.0002-01-01-00000.hdrtest_b2.pop.ro.0002-01-01-00000test_b2.rtm.h0.0001-12.nctest_b2.rtm.r.0002-01-01-00000.nctest_b2.rtm.rh0.0002-01-01-00000.nc The restart files of ccsm4_init/b40.20th.track1.2deg.001/1979-01-01 are b40.20th.track1.2deg.001.cam2.h0.1978-12.ncb40.20th.track1.2deg.001.cam2.h1.1979-01-01-00000.ncb40.20th.track1.2deg.001.cam2.i.1979-01-01-00000.ncb40.20th.track1.2deg.001.cam2.r.1979-01-01-00000.ncb40.20th.track1.2deg.001.cam2.rs.1979-01-01-00000.ncb40.20th.track1.2deg.001.cice.r.1979-01-01-00000.ncb40.20th.track1.2deg.001.clm2.h0.1978-12.ncb40.20th.track1.2deg.001.clm2.r.1979-01-01-00000b40.20th.track1.2deg.001.clm2.r.1979-01-01-00000.ncb40.20th.track1.2deg.001.cpl.r.1979-01-01-00000.ncb40.20th.track1.2deg.001.pop.r.1979-01-01-00000b40.20th.track1.2deg.001.pop.r.1979-01-01-00000.hdrb40.20th.track1.2deg.001.pop.ro.1979-01-01-00000rpointer.atmrpointer.drvrpointer.icerpointer.lndrpointer.ocn.ovfrpointer.ocn.restartrpointer.ocn.tavgThey are not all the same as files in ccsm4_init. Some are unnecessary, and some are absence. I think 'cam' and 'cam2'  are different just due to the different version.  (2)So, how to generate a user defined initial condition files for a hybrid case? Here, I want to get restart files of 20th control run with cyclical dataset. And  how many years should I run the model? Do you have such these intitial condition files?Thank you
Yao