NPZD_franks

Message

feroda · #1 Unread post by **feroda** » Mon May 26, 2008 3:31 pm

hi all,
I am quite doubt about the difference among the files
ad_npzd_Franks.h, rp_npzd_Franks.h, tl_npzd_Franks.h and
npzd_Franks.h.

What I plan to do is coupling NPZD_franks-type ecological model into antother hydrodynamical model(e.g. POM). So, I resort to the ready-made NPZD_franks subroutine here. However, I find the above four subroutines are almost the same.

Would you please give me some advice on the issue?

Thanks for your kind help!

David

bdc · #2 Unread post by **bdc** » Tue May 27, 2008 12:58 am

ad_, tl_, and rp_ are for adjoint, tangent linear, and representer models, which
don't apply to you. Just ignore them.

feroda · #3 Unread post by **feroda** » Tue May 27, 2008 2:36 pm

bdc wrote:ad_, tl_, and rp_ are for adjoint, tangent linear, and representer models, which
don't apply to you. Just ignore them.

Thanks for your replay! I am quite doubt with a mass of the parameters in the npzd_Franks.h. Does it caculates the source and sink terms or advection terms of the biology variables(N,P,Z,D)?

What do the parameters as follow mean?
CALL biology_tile (ng, tile, &
& LBi, UBi, LBj, UBj, N(ng), NT(ng), &
& nstp(ng), nnew(ng), &
#ifdef MASKING
& GRID(ng) % rmask, &
#endif
& GRID(ng) % Hz, &
& GRID(ng) % z_r, &
& GRID(ng) % z_w, &
& OCEAN(ng) % t)

Does the parameter 'ng' equal 4 ??

Thanks for your time and kind help.

kate · #4 Unread post by **kate** » Tue May 27, 2008 5:18 pm

feroda wrote: Thanks for your replay! I am quite doubt with a mass of the parameters in the npzd_Franks.h. Does it caculates the source and sink terms or advection terms of the biology variables(N,P,Z,D)?

The advection and diffusion terms happen elsewhere for all of the tracers. The NPZD code only handles the interactions between the biological components.

What do the parameters as follow mean?
CALL biology_tile (ng, tile, &
& LBi, UBi, LBj, UBj, N(ng), NT(ng), &
& nstp(ng), nnew(ng), &
#ifdef MASKING
& GRID(ng) % rmask, &
#endif
& GRID(ng) % Hz, &
& GRID(ng) % z_r, &
& GRID(ng) % z_w, &
& OCEAN(ng) % t)

Does the parameter 'ng' equal 4 ??

Nope, ng is the number grids in your domain, usually one. You can think of tile as being one as well, with LBi the lower bound i-index, UBI the upper bound i-index, similar for j. N(ng) is the number of vertical levels, NT is the number of tracers where t is dimensioned by (LBi:UBi, LBj:UBj, NT, 3 (timelevels), NT). nstp and nnew are timelevel indices, rmask is the land mask at rho (t) points, and Hz, z_r, z_w relate to the vertical coordinates. Hz is the vertical grid spacing dz.