[cam-users] CGD Forums

Fri Feb 25 09:00:00 MST 2005

Dear cam-users, 

As you requested, here is the latest digest of messages posted on CGD Forums forums. Please come and join the discussion!

<<  CCSM3 Smoke test, http://bb.cgd.ucar.edu//viewtopic.php?t=125 >>

PosterDmitry V.ShkurkoPostedFri Feb 25, 2005 7:10 am MST, http://bb.cgd.ucar.edu//viewtopic.php?p=374#374
Message: Dear all,
 
 I am trying to run smoke test SM.01a with resolution T31_gx3v5.
 All components are active (B component set). 
 Everything is going ok except that in log for clm2 I see errors:
 
 surfrd error: CAM fractional land differs from surface dataset at i,j=
 4 19
 
 I use CCSM3.0 and try to port it to Itanium2 cluster.
 
 Could you give me ideas how to resolve this problem?
 May be the cause is in incorrect configuration?
 
 Thank you in advance,
 Dmitry.
 Dmitry

------------------------------

<<  CAM3 Installation problem on Itanium, http://bb.cgd.ucar.edu//viewtopic.php?t=124 >>

PosterrpbPostedThu Feb 24, 2005 9:30 am MST, http://bb.cgd.ucar.edu//viewtopic.php?p=367#367
Message: Hello,
 
 I am trying to install CAM3 on an Itanium cluster, running on Linux. I am having problems with compiling the sources. It fails on BalanceCheckMod.F90 with the following error message:
 
 module BALANCECHECKMOD
 module subroutine BALANCECHECK
 
 use time_manager, only : get_step_size, get_nstep
 
 Error 208 at (63:/home/cas/casnan/rpb/source/cam1/models/lnd/clm2/src/biogeophys/BalanceCheckMod.F90) : No such entity in the module
 
 Error 208 at (63:/home/cas/casnan/rpb/source/cam1/models/lnd/clm2/src/biogeophys/BalanceCheckMod.F90) : No such entity in the module
 
 2 Errors
 compilation aborted for /home/cas/casnan/rpb/source/cam1/models/lnd/clm2/src/biogeophys/BalanceCheckMod.F90 (code 1)
 gmake: *** Error 1
 
 I could only guess the 'missing' entities to be get_step_size and get_nstep. However, when I scroll up the console, I can see that time_manager.F90 has !
 been compiled properly with all the functions and subroutines listed as follows:
 
 module TIME_MANAGER
 module subroutine TIMEMGR_PRESET
 module subroutine TIMEMGR_INIT
 module subroutine TIMEMGR_RESTART
 module subroutine TIMEMGR_PRINT
 module subroutine ADVANCE_TIMESTEP
 module function GET_STEP_SIZE
 module function GET_NSTEP
 module subroutine GET_CURR_DATE
 module subroutine GET_PERP_DATE
 module subroutine GET_PREV_DATE
 module subroutine GET_START_DATE
 module subroutine GET_REF_DATE
 module subroutine GET_CURR_TIME
 module function GET_CURR_CALDAY
 module function IS_END_CURR_DAY
 module function IS_END_CURR_MONTH
 module function IS_FIRST_STEP
 module function IS_FIRST_RESTART_STEP
 module function IS_LAST_STEP
 module function IS_PERPETUAL
 mod!
 ule subroutine TIMEMGR_WRITE_RESTART
 module subrout!
 ine TIME
MGR_READ_RESTART
 module subroutine TIMEMGR_DATEDIFF
 module subroutine CHKRC
 
 1138 Lines Compiled
 
 
 I am using efc &amp; ecc compilers for the Itanium. Kindly help me as to what probably is going wrong.
 
 Thanks,
 rpb

------------------------------
PosterjPPostedFri Feb 25, 2005 4:37 am MST, http://bb.cgd.ucar.edu//viewtopic.php?p=373#373
Message: I had the similar problem. Try to write "use time_manager" instead of "use time_manager, only : get_step_size, get_nstep". It could help.
<br />And by the way, which version of the compiler are you using?

------------------------------
PosterrpbPostedFri Feb 25, 2005 11:33 am MST, http://bb.cgd.ucar.edu//viewtopic.php?p=376#376
Message: Thanks for the reply, jP !!
<br />
<br />But it did not help as much. When I changed as suggested, I started getting a new error mesage:
<br />
<br />This entity has not been given a type ( referring to get_nstep )
<br />
<br />When I somehow managed to bypass the BalanceCheckMod.F90, it went ahead and started giving "Compiler Intercal Error" on some other programs. I am coming to believe that I am using an older version of efc and hence facing the problem.
<br />
<br />Versions:
<br />
<br />efc:
<br />
<br />Intel(R) Fortran Itanium(R) Compiler for Itanium(R)-based applications
<br />Version 7.0, Build 20021028
<br />
<br />Is this the version that was used by jP? Kindly let me know.
<br />
<br />Thanks,
<br />rpb

------------------------------

<<  question about double CO2 exp., http://bb.cgd.ucar.edu//viewtopic.php?t=123 >>

PostermvrPostedThu Feb 24, 2005 5:54 pm MST, http://bb.cgd.ucar.edu//viewtopic.php?p=368#368
Message: hi...
 
 as noted in the user's guide, the co2 volume mixing ratio 
 (CO2VMR) is 3.55e-4 by default...to run at twice that level, simply 
 specify the namelist variable CO2VMR to be 7.1e-4...if you want 
 to ramp the co2 levels up to a cap of 2x, check out the namelist 
 variables: SCENARIO_GHG, RAMP_CO2_START_YMD, 
 RAMP_CO2_CAP, and RAMP_CO2_ANNUAL_RATE for how to 
 do that...
 
 the default value for CO2VMR is a 1990 level (355 ppmv)...
 
 to set the output frequency of the primary history tape to 
 once-per-timestep, set the namelist variable NHTFRQ(1)=1, 
 as described in the user's guide...
 
 i hope that helps, 
 -mathew

------------------------------
Postertz0616PostedThu Feb 24, 2005 9:39 pm MST, http://bb.cgd.ucar.edu//viewtopic.php?p=369#369
Message: Hi, mvr,
<br />
<br />  Thanks very much for your valuable suggestion.
<br />
<br /> 1)
<br />  Listed below is key part of my runscript with modified setting for my purpose.
<br />  I am not sure if this is the exact suggestion you have made for my concern.
<br />  (please look at the last 2 sentences)
<br />...................
<br />## Create the namelist
<br />cd $blddir || echo "cd $blddir failed" &amp;&amp; exit 1
<br />echo "Building the namelist."
<br />$cfgdir/build-namelist -s \
<br />-case $case \
<br />-runtype initial \
<br />-o $rundir/namelist \
<br />-namelist "&amp;camexp nelapse=$nelapse,mss_irt=0,CO2VMR=7.1e-4,NHTFRQ(1)=1 /" \
<br />|| echo "build-namelist failed" &amp;&amp; exit 1
<br />...........................
<br />
<br /> 2) You mean that the default value for CO2VMR is a 1990 level (355 ppmv).
<br />    Does this value come from IPCC or other sources?
<br />
<br />    What's the function of CO2 data under the directory "inputdata/atm/cam2/g!
 gas/"?
<br />    I am not sure if it is used for the run with monthly observed SSTs?
<br />
<br />3) If I want to save the primary history tape with 3 frequncey (timestep, daily,
<br />   monthly) at the same run, How should I set for "NHTFRQ" ?
<br />
<br />  Hope to hear from you,
<br />  Thanks again for your comments.
<br />
<br />  tao

------------------------------
PostermvrPostedFri Feb 25, 2005 12:54 am MST, http://bb.cgd.ucar.edu//viewtopic.php?p=372#372
Message: 1) looks good
<br />2) i know nothing more of the origin of the default value for co2vmr 
<br />other than it is supposed to represent a 1990 level
<br />3) from 3.1 of the user's guide:
<br />"If the user wishes to see a field written at additional time 
<br />frequencies (e.g. daily, hourly), additional history files must be 
<br />declared."
<br />
<br />and from table b13 of the user's guide for entry NHTFRQ:
<br />"Array of write frequencies for each history files series. If 
<br />NHTFRQ(1)=0, the file will be a monthly average. Only the first file 
<br />series may be a monthly average. If NHTFRQ(i) &gt;0, frequency is 
<br />input as number of timesteps. If NHTFRQ(i) &lt;0, frequency is 
<br />input as number of hours. Note that NHTRFQ(1) also controls the 
<br />frequency of restart dataset writes if NREFRQ = 1." 
<br />
<br />so...set nhtfrq(1)=0 nhtfrq(2)=1 nhtfrq(3)=-24 
<br />
<br />and don't forget to add the fields you're interested in to the history !

<br />tapes via the namelist variable FINCL 
<br />
<br />-mathew

------------------------------

Format TEXT
Show Message Text YES
Show My Messages YES
Digest Frequency DAY
Show only new messages since last time I logged in FALSE
Send empty digests NO
Time of day to send digest 9 AM
Maximum characters per message in digest 32000