TKE surface BC

Report or discuss software problems and other woes

Moderators: arango, robertson

Post Reply
Message
Author
andreykoch

TKE surface BC

#1 Unread post by andreykoch »

Hello!

I am looking at the tke in ROMS history output files.

In my_25corstep.F the tke surface boundary conditions are defined for the Nth layer:
! Set surface and bottom boundary conditions.
!
DO i=Istr,Iend
tke(i,j,N(ng),nnew)=my_B1p2o3*0.5_r8*SQRT((sustr(i,j)+sustr(i+1,j))**2+(svstr(i,j)+svstr(i,j+1))**2)
END DO
Does that N=end in the tke(i,j,end) from his* file or N=end-1???

I have Uwind Vwind forcing and compute sustr svstr using BULK FLUX parametrization (bulk_flux.F).

Assuming that the tke(i,j,N) from my_25corstep.F appears in a his* file as tke(i,j,end), I computed tke(i,j,end) with bulk_flux.F algorithm. The tke(i,j,end) taken from his* file is approx. 2 orders larger than that computed from U,Vwind by bulk_flux code. See the plot.

So, the questions are:

1. Does the uppermost layer of tke(i,j,end) correspond to the surface (z_w(i,j,end)+zeta(i,j) m), or to the layer one grid point below the surface?

2. If what I am reading from the his* file tke(i,j,end) is the SURFACE boundary condition tke(i,j,N), why is it so different?

Thanks to those who will try to address my questions!
Attachments
TKE_bc.jpg

andreykoch

Re: TKE surface BC

#2 Unread post by andreykoch »

Let's pretend the tke(i,j,end) I am reading from the history file is the interior value one point below the surface.

The tke variable is defined in w points, w-vertical grid has N+1 levels, and the uppermost level is set to 0 m.

How the uppermost value which is assigned for z_w(i,j,end)=0 m could be interior one?

andreykoch

Re: TKE surface BC

#3 Unread post by andreykoch »

Alright, the problem has been solved. It turned out that I was reading tke values indeed one grid point below the surface, that is why I got so much spatial variability there. Another thing which puzzled me is the big difference between magnitudes of ROMS surface tke and the tke computed by bulk flux code. Apparently, I have to scale it with factor 100 or so which I guess comes from different units in relative humidity.

The attached plot shows curves of surface tke computed from wind stress and taken from history file. They are identical.

Thanks all for your attention to my problem!
Attachments
TKE_bc.jpg

Post Reply