Open saeed-moghimi-noaa opened 9 months ago
@pvelissariou1 I am trying to run the case coastal_florence_hsofs_pam2sch
after your PR but getting error like following
4: forrtl: severe (19): invalid reference to variable in NAMELIST input, unit 15, file /work/noaa/stmp/tufuk/stmp/tufuk/FV3_RT/rt_90784/coastal_florence_hsofs_pam2sch_intel/.//param.nml, line 669, position 20
4: Image PC Routine Line Source
4: fv3.exe 0000000001EB2A48 Unknown Unknown Unknown
4: fv3.exe 0000000001EEC4C7 Unknown Unknown Unknown
4: fv3.exe 0000000001BFAADB schism_init_ 508 schism_init.F90
4: fv3.exe 0000000001B9D154 schism_nuopc_cap_ 339 schism_nuopc_cap.F90
I don't think it is related with my change to restructure input files. Do you think I need to update SCHSIM under UFS Coastal? If I check my schism_init.F90
file namelist /OPT/
section, I am not seeing option like iflux_out_format
.
Please update SCHISM in UFS-Coastal. I have updated the param.nml in the test cases to its latest version.
@pvelissariou1 I think mine is more recent since that option is removed with the following commit, https://github.com/schism-dev/schism/commit/376b3a456897d32e8b0baf1801e9bfe538fe4173 (main branch) Do you want me to update again.
@pvelissariou1 Okay. I sync our fork with main and testing again.
@uturuncoglu Thanks, I'll check things out from my side as well after my meetings today.
@uturuncoglu I will update CoastalApp as well and the test cases in the testsuite.
@pvelissariou1 I have still same issue. I think that option needs to be removed.
@uturuncoglu Unless there is a major bug in SCHISM it might be a good idea to freeze SCHISM in UFS-Coastal.
@pvelissariou1 I have also issue with ielm_transport = 0 !1: turn on
. I think this configuration files need to be updated for the recent version of SCHSIM.
@uturuncoglu Ok, thanks for testing. Let me check and update. Please continue with your other works. I'll keep you posted.
@pvelissariou1 Thanks. BTW, false alarm about ielm_transport
. It is fine but I have still issue with iflux_out_format
.
@pvelissariou1 BTW, I think we need to update pam2sch2ww3 to reflect recent changes that I did for atm2ww3 configuration but we could look at that one later.
@uturuncoglu pam2sch2ww3 is not ready yet. I am working on it as well.
@pvelissariou1 Okay. Let's focus two component configurations first. Thanks.
@pvelissariou1 BTW, i comment out that option from case but now it dies with following error,
/work/noaa/stmp/tufuk/stmp/tufuk/FV3_RT/rt_124516/coastal_florence_hsofs_pam2sch_intel/.//drag.gr3
Maybe drag.gr3
is missing in the input directory. I checked and I could not find in your directory (find /work/noaa/nosofs/pvelissa/RT_takis/NEMSfv3gfs/input-data-20221101/SCHISM/. -name "drag.gr3"
).
@uturuncoglu Depending on the configuration in param.nml different files are required. This test case is not configured to use drag.gr3.
@pvelissariou1 Then, we need to adjust the param.nml
since with the current param.nml
it required that file. I am not sure which option need to be adjusted. Do you know?
@pvelissariou1 I think nchi
needs to be set as -1 to use manning.gr3
.
@pvelissariou1 In our case it is set to 0. Based on SCHISM/docs/input-output/friction.md
. It is 0 for drag.gr3
, 1 for rough.gr3
and -1 for manning.gr3
.
@uturuncoglu I thought I had set it up to nchi=-1. Yes because we are using manning.gr3
@pvelissariou1 Okay. I changed and rerun. Now I am getting error from windrot_geo2proj.gr3
. This file again not available for Florence case. Is it used? Probably it is grid dependent. It is also not available under CoastalApp test suite. Any idea? Maybe we could change the configuration too. It seems that it is required for if nws==2
.
@pvelissariou1 Is it possible to create missing windrot_geo2proj.gr3
file?
@uturuncoglu Ufuk I will take care of all SCHISM_PAHM test cases later today. I am in the process of transitioning. I'll let you know.
@pvelissariou1 Okay. Thanks.
The issue is dedicated for PAHM+SCHISM coupling.