about mask problem

[liangailin]

dear all :
now my version is 452, I use my own grid nc file and define masking, but the result(u,v,temp,salt) even mask_rho,mask_v,mask_u,mask_psi all equal 1E37(model regarded all points as land), but the water depth is the same with my grid file. I use uvp_mask to caculate the mask, so I think it's correct of mask. but why the model can't read the mask correctly?Do some bugs exist in reading mask from grid file?? really confuse me. who can tell me the reasons and give me some suggestions? really thanks a lot

regards

[liangailin]

I just test two cases,
I define ana_mask and define masking, the results all are 1e37;
but I define ana_mask only, the results are normal value;
so the problem must be something about "define masking"
..........

[kate]

When you define ANA_MASK, what do you set for your mask in ana_mask.h? Do you have a cpp name for your case? Did you check to see that the version of ana_mask you set up got into analytical.f90?

[liangailin]

thanks for your reply.
my case is upwelling case;
in my ana_mask, I define all points are sea;
that is :

#elif defined UPWELLING
DO j=Jstr-2,,Jend+2
DO i=Istr-2,Iend+2
mask(i,j)=1.0_r8
END DO
END DO

but if I define ana_mask and define masking together, the model output are all 1e37; if I just define ana_mask only, model results looked reasonable.

[kate]

Have you checked analytical.f90? Next would be a debugger to watch what is happening as the model runs.

[CBian]

but if I define ana_mask and define masking together, the model output are all 1e37; if I just define ana_mask only, model results looked reasonable.

Could we defined ana_mask and define masking together? If MASKING is defined, the model will read mask_rho,mask_u,mask_v,mask_psi from our grid.nc data. So if we define ana_mask together, what happened to mask_rho and so on?

[kate]

If you want to have land masks, define MASKING.

With masking, provide the land mask in either the grid file (no extra defines) or in ana_mask.h and define ANA_MASK.