Initial Condition File Issue in HS94-PK02 CESM Setup

[Climatekid]

Hi, I am currently using CESM2.1.5 to set up a HS94-Povalni-Kushner case. The final aim is to place a planetary-scale wave forcing at the bottom to generate sudden warmings in the stratosphere. However, after successfully setting up the original HS94 case and trying to modify the stratospheric settings following the guidance on the CESM website (Running with a different analytical relaxation temperature profile and damping parameters e.g., the Polvani and Kushner (2002) setup | Community Earth System Model), I encountered a problem possibly associated with the IC file. The commands I used are: The commands I used to set up the case are:
./create_newcase --case /data/jiachengye/CESM_exps/ssw1 --compset FTJ16 --res T42z30_T42_mg17 --compiler intel --run-unsupported
./xmlchange STOP_OPTION=ndays,STOP_N=20
./xmlchange --append --file env_build.xml --id CAM_CONFIG_OPTS --val="-nlev=40" (The default setting for PK02)
./case.setup
./case.build
./case.submit.
The content in the user_nl_cam is simply:"

&pkstrat_nl
pkstrat=.True.
vgamma=2.
/
ncdata='/data/jiachengye/CESM_new/dabiic/HS1994.ne16.L40.nc'
"

I placed all modified .f90 files (i.e., runtime_opts.F90, held_suarez.F90, held_suarez_cam.F90) in the directory /data/jiachengye/CESM_exps/ssw1/SourceMods/src.cam/. The namelist file namelist_definition.xml under the directory $CESM/components/cam/bld/namelist_files/ is also correspondingly updated to add the additional pkstrat statement before setting up the case. According to the error report ($OUTPUTDIR/dry_held_suarez/ssw1/run/cesm.log.250311-014419):
"
read_inidat: ERROR: model parameters do not match initial dataset parameters
Model Parameters: plon = 128 plat = 64
Dataset Parameters: dlon = 123435928 dlat = 0
ERROR:
read_inidat: ERROR: model parameters do not match initial dataset parameters
"
I suspect the problem is associated with the IC data file because the resolution (T42z30_T42_mg17) I chose to set up the case is not compatible with the default resolution of the IC data. In the ncl file (/dabiic/makeic.ncl) used to generate initial data, I set
res = "ne16",
which seems to be the error source. However, so far I don't know how to generate the corresponding initial data, as changing from 'ne16' to '64x128' appears to be prohibited. Any guidance would be much appreciated! I have attached the relevant files for reference. I apologize for this seemingly easy question.

 

[Climatekid]

The rest of the log files are attached here.

 

[jedwards]

What dynamical core do you want to use for this experiment? This grid T42z30_T42_mg17 is specifying an Eulerian dynamical core with
128x64 horizontal grid points. I suspect that what you want is a spectral element grid - try ne16_ne16_mg17.

 

[Climatekid]

What dynamical core do you want to use for this experiment? This grid T42z30_T42_mg17 is specifying an Eulerian dynamical core with
128x64 horizontal grid points. I suspect that what you want is a spectral element grid - try ne16_ne16_mg17.

Hi jedwards, thank you for your advice! I originally used the defualt dynamical core for the grid T42z30_T42_mg17. So I just tried using ne16_ne16_mg17 to build the case in which I initially did not add the PK02-related modifications. However, the building process reported an error message with which I am not familiar:

./create_newcase --case /data/jiachengye/CESM_exps/ssw5_nene16 --compset FTJ16 --res ne16_ne16_mg17 --compiler intel --run-unsupported
./xmlchange STOP_OPTION=ndays,STOP_N=50
./xmlchange STOP_OPTION=ndays,STOP_N=300,RESUBMIT=3
./case.setup
./case.build:
"
...
Building wav with output to /data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/wav.bldlog.250312-123437
Building esp with output to /data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/esp.bldlog.250312-123437
socn built in 3.541374 seconds
sglc built in 3.576726 seconds
sice built in 3.596640 seconds
swav built in 3.603435 seconds
slnd built in 3.652305 seconds
sesp built in 3.695128 seconds
srof built in 3.955145 seconds
Component atm build complete with 20 warnings
cam built in 86.156271 seconds
Building cesm with output to /data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/cesm.bldlog.250312-123437
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:44: undefined reference to `vdsqrt_'
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:105: undefined reference to `vdexp_'
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:130: undefined reference to `vddiv_'
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:164: undefined reference to `vdln_'
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:194: undefined reference to `vdsin_'
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:222: undefined reference to `vdcos_'
ERROR: BUILD FAIL: buildexe failed, cat /data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/cesm.bldlog.250312-123437
".
I further took a look at the cesm.bldlog.250312-123437 file:
"
...
-L/data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/lib/ -lwav -L/data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/lib/ -lesp -L../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib -lcsm_share -L../../intel/intelmpi/nodebug/nothreads/lib -lpio -lgptl -lmct -lmpeu -L/software/netcdf-4.9.0+intel-2020/lib -lnetcdff -lnetcdf
../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib/libcsm_share.a(shr_vmath_mod.o): In function `shr_vmath_mod_mp_shr_vmath_sqrt_':
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:44: undefined reference to `vdsqrt_'
../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib/libcsm_share.a(shr_vmath_mod.o): In function `shr_vmath_mod_mp_shr_vmath_exp_':
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:105: undefined reference to `vdexp_'
../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib/libcsm_share.a(shr_vmath_mod.o): In function `shr_vmath_mod_mp_shr_vmath_div_':
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:130: undefined reference to `vddiv_'
../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib/libcsm_share.a(shr_vmath_mod.o): In function `shr_vmath_mod_mp_shr_vmath_log_':
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:164: undefined reference to `vdln_'
../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib/libcsm_share.a(shr_vmath_mod.o): In function `shr_vmath_mod_mp_shr_vmath_sin_':
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:194: undefined reference to `vdsin_'
../../intel/intelmpi/nodebug/nothreads/mct/noesmf/c1a1l1i1o1r1g1w1e1/lib/libcsm_share.a(shr_vmath_mod.o): In function `shr_vmath_mod_mp_shr_vmath_cos_':
/data/jiachengye/CESM_new/my_cesm_sandbox2_sswtest/cime/src/share/util/shr_vmath_mod.F90:222: undefined reference to `vdcos_'
gmake: *** [/data/jiachengye/CESM_exps/ssw5_nene16/Tools/Makefile:878: /data/jiachengye/CESM/test/temp/dry_held_suarez/ssw5_nene16/bld/cesm.exe] Error 1
".
I also tried the default T42z30_T42_mg17 case, which seems to run successfully. Do you have any suggestions regarding the failure to build the ne16_ne16_mg17 case? Any insights would be greatly appreciated!

Thanks,
Jiacheng