problem when run the ESTUARY_TEST case

[leon]

Dear all,
I am a new user of ROMS.

I run the upwelling test, it's ok. But when I run the ESTUARY_TEST, after 'make',I get such problem:

cd Build; /usr/bin/gfortran -c -frepack-arrays -O3 -ffast-math -ffree-form -ffree-line-length-none analytical.f90
analytical.f90:256.32:

BOUNDARY(ng)%zeta_west(j)=9.0E-06_r8*fac
1
wrong： ‘zeta_west’at (1)is not the member of ‘t_boundary’
make: *** [Build/analytical.o] wrong 1


and I use the gfortran complier.I want to know why.

Thank you for your help!

[m.hadfield]

It sounds at bit like the problem reported here:

viewtopic.php?t=1208

There have been some changes in globaldefs.h to the code that sets the *_FSOBC, *_M2OBC, etc preprocessor symbols. It looks like the wrinkles haven't been completely ironed out. It may be that ana_fsobc.h has some redundant code for the ESTUARY_TEST case,

[arango]

Yes, sometimes there is code that it not used in the analytical expressions. The definition and allocation of boundary arrays is very strict now. This problem is now fixed.

Please update

[leon]

That's the piont! It's ok now. You are really cool! Great! Thank you very much!

[leon]

Lm == 200 ! Number of I-direction INTERIOR RHO-points
Mm == 3 ! Number of J-direction INTERIOR RHO-points
N == 20 ! Number of vertical levels


I　think the value of Mm is wrong in the ocean_estuary.in

[m.hadfield]

Wrong in what sense?

[m.hadfield]

The ESTUARY_TEST case runs OK for me with revision 319. There's initially a vertical boundary between fresh and salty water. The fresh water moves seaward, as you'd expect, and the pattern moves back & forth with the tide. I see no sign of Mm being too small.