rip current-run error

[pooran]

My system move to Ubuntu .After a while I can compile new version of COAWST by some changes:
1- changes in linux.gfortran.mk
from
LIBS := $(shell $(NC_CONFIG) –flibs)
to
LIBS := -L$(NETCDF_LIBDIR) -lnetcdf -lnetcdff
2-and in coawst.bash use:
USE_MPI=on, USE_MPIF90=on, USE_NETCDF4=on and
export COMPILERS=${MY_ROMS_SRC}/Compilers

but for running first test (RIPCURRENT) it faced with errors with out any error files:

Code: Select all

mpirun -np 4  ./coawstG coupling_rip_current.in
 Coupled Input File name = coupling_rip_current.in
At line 119 of file read_coawst_par.f90
Fortran runtime error: Index '1' of dimension 1 of array 'wname' above upper bound of 0
At line 119 of file read_coawst_par.f90
Fortran runtime error: Index '1' of dimension 1 of array 'wname' above upper bound of 0
At line 119 of file read_coawst_par.f90
Fortran runtime error: Index '1' of dimension 1 of array 'wname' above upper bound of 0
At line 119 of file read_coawst_par.f90
Fortran runtime error: Index '1' of dimension 1 of array 'wname' above upper bound of 0

===================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   EXIT CODE: 2
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================

I can not find what cause those errors. is it caused by #undef ana_sediment in header file? I have installed swan40.91 .
Would you please help me to solve those errors
all the best
pooran

[kate]

Grepping around in the new coawst, I find:

Code: Select all

SWAN/Src/swan_iounits.F:      allocate ( Wname(NUM_SGRIDS) )

That NUM_SGRIDS comes from:

Code: Select all

Master/get_numswan_grids.F:          read(aline(8:12),'(i5)') NUM_SGRIDS

Did you plunk a new SWAN version into the SWAN directory? That's risky without first checking to see how the COAWST SWAN differs from distribution SWAN.

[jcwarner]

i still have not had the chance to get all the COAWST/Projects test cases up to the new version.
Joe_tc and inlet_test are up to speed, and there are several people working on the other tests. Will send out an email when all the Projects/apps are up and working. For now:

- add to the swan input file
&& KEYWORD for number of nested SWAN grids.
NSGRIDS 1

- if there are future issues with specific Projects tests, these are specific to COAWST and then please post a message here:
http://coawstmodel-trac.sourcerepo.com/ ... el_COAWST/


thanks,
john

[pooran]

Did you plunk a new SWAN version into the SWAN directory?

no
i did not know that just some test cases get up to the new version. I can run inlet-test .
What is SWAN's version in new COAWST? is swan's current version able just to nest in one-way (as old versions) ?
Cheers
pooran

[pooran]

Dear
I am updating my test to COAWST new version.
1- river file need #define UV_PSOURCE and #define TS_PSOURCE in old version. It seems that new version does not need them. Does it need another options?
2- we add river file's name in FRCNAME flag for old version. but it seems its names should be added in SSFNAME not FRCNAME?is not it?
3- would you please tell me that applying river to new version needs additional thing is or not?
4- I have created river file by attached file for COAWST old version. Would you please check it and tell me thats suitable for new version, too?
Sorry for long question but as you know all tests in new version are not updated yet.
Cheers
pooran

[kate]

1- river file need #define UV_PSOURCE and #define TS_PSOURCE in old version. It seems that new version does not need them. Does it need another options?

Indeed, it is now a flag in the ocean.in file.

Code: Select all

! Logical switches (TRUE/FALSE) to activate horizontal momentum transport
394	! point Sources/Sinks (like river runoff transport) and mass point
395	! Sources/Sinks (like volume vertical influx), [1:Ngrids].
396	
397	      LuvSrc == F                          ! horizontal momentum transport
398	       LwSrc == F                          ! volume vertical influx
399	
400	! Logical switches (TRUE/FALSE) to activate tracers point Sources/Sinks
401	! (like river runoff) and to specify which tracer variables to consider:
402	! [1:NAT+NPT,Ngrids].  See glossary below for details.
403	
404	  LtracerSrc == F F                        ! temperature, salinity, inert

2- we add river file's name in FRCNAME flag for old version. but it seems its names should be added in SSFNAME not FRCNAME?is not it?

Correct.

4- I have created river file by attached file for COAWST old version. Would you please check it and tell me thats suitable for new version, too?
Sorry for long question but as you know all tests in new version are not updated yet.
Cheers
pooran

The format of the river input file has not changed.

[pooran]

1-Do you mean: for river we should turn of flags like this?

Code: Select all

LuvSrc == T ! horizontal momentum transport
LwSrc == F ! volume vertical influx
LtracerSrc == T T  ! temperature, salinity, inert

2-I can't completely caught what it “volume vertical influx”.
When should turn it on?

[kate]

1-Do you mean: for river we should turn of flags like this?

Code: Select all

LuvSrc == T ! horizontal momentum transport
LwSrc == F ! volume vertical influx
LtracerSrc == T T  ! temperature, salinity, inert

Yes

2-I can't completely caught what it “volume vertical influx”.
When should turn it on?

It replaces the old Q_PSOURCE, like if you have a hot spring on the bottom of the ocean and you want the vertical flux. I've never used it.