analytical tides

[edith]

Hi,
I try to add tides from open boundaries for analytical idealized domain river inflow case.

#define EAST_FSCHAPMAN
#define EAST_M2FLATHER
#define EAST_M3RADIATION
#define EAST_TRADIATION
#define ANA_FSOBC
#define ANA_M2OBC

I changed ana_fsobc.h and ana_m2obc.h files and directed in wiki for fjord case (https://www.myroms.org/wiki/index.php/FJORD_TIDAL_CASE). it compiles with debuger but model run fails on first step giving:

NL ROMS/TOMS: started time-stepping: (Grid: 01 TimeSteps: 00000001 - 00019800)

STEP Day HH:MM:SS KINETIC_ENRG POTEN_ENRG TOTAL_ENRG NET_VOLUME

0 0 00:00:00 0.000000E+00 6.201949E+01 6.201949E+01 1.536000E+10
DEF_HIS - creating history file: /home/data/edith/ocean_his_wind_s0.25_tides_0001.nc
WRT_HIS - wrote history fields (Index=1,1) into time record = 0000001
forrtl: error (65): floating invalid
Image PC Routine Line Source
oceanG 0808ABA0 step2d_mod_mp_ste 451 step2d.f90
oceanG 0807EF7A step2d_mod_mp_ste 70 step2d.f90
oceanG 0804D843 main3d_ 237 main3d.f90
oceanG 0804A054 ocean_control_mod 146 ocean_control.f90
oceanG 08049B5B MAIN__ 87 master.f90
oceanG 080499DA Unknown Unknown Unknown
libc.so.6 00152E23 Unknown Unknown Unknown
oceanG 08049911 Unknown Unknown Unknown
Aborted

any recommendations what might be wrong? as much as i understand that type of error means that somewhere in calculations it gets zero or negative values? but it dont give any more info where it might be.

[kate]

Well, you know it's on line 451 of step2d.f90. What is on that line in your copy of the file? Running it in a debugger (or adding judicious print statements) would tell you which i,j it's at and which iteration of the 2D timestepping it's at.

[edith]

thank you for the reply.
looks like that specific error was caused by the attempt to add tides only from one boundary (or over complicating with attempt to modify ana_m2obc.h file, got working and good result with using only zeta on all open boundaries at the end of ana_fsobc.h file).