About minimum depth

Discussion about analysis, visualization, and collaboration tools and techniques

Moderators: arango, robertson

Post Reply
Message
Author
User avatar
susonic
Posts: 167
Joined: Tue Aug 21, 2007 5:44 pm
Location: UST21 / Korea
Contact:

About minimum depth

#1 Unread post by susonic »

Dear ROMS users,

I was wondering why we set minimum depth in ROMS ?
Joonho Lee

User avatar
arango
Site Admin
Posts: 1347
Joined: Wed Feb 26, 2003 4:41 pm
Location: DMCS, Rutgers University
Contact:

Re: About minimum depth

#2 Unread post by arango »

:idea: Look at the governing equations and think about its spatial discretization in numerical models. What would happen if you have a grid cell with zero depth? What are the values of the vertical level thicknesses?

I gave you a good clue... As a homework, you can figure out the rest...

User avatar
susonic
Posts: 167
Joined: Tue Aug 21, 2007 5:44 pm
Location: UST21 / Korea
Contact:

Re: About minimum depth

#3 Unread post by susonic »

Thnank you for your reply, Dr. Arango. Good clue.

So, we have to set non-zero value as a minimum depth in order not to make the vertical level thickness zero.

-JH
Joonho Lee

User avatar
arango
Site Admin
Posts: 1347
Joined: Wed Feb 26, 2003 4:41 pm
Location: DMCS, Rutgers University
Contact:

Re: About minimum depth

#4 Unread post by arango »

Yes, that's part of the answer. But you still have to check the :arrow: governing equations. If the vertical thickness (Hz) is zero, what would happen to the vertical viscosity and diffusion terms? Then if you search for oHz in the code, you will notice that the internal array oHz is in several routines of the numerical kernel. Notice that oHz=1/Hz.

This is the kind of analysis that users with curiosity need to carry out to discover what are the assumptions that we need to make when coding the continuum governing equations in numerical models. This kind of analysis is needed to determine was is the appropriate minimum depth (hmin) for a particular application, for instance. This is going to control the vertical time step limitation (CFL) as a function of your specified number of vertical levels. In terrain-following coordinates, we need to have all the vertical levels to exist at the minimum depth.

So now you know why we cannot have a water column with a total thickness of zero :!:

I added an entry to the FAQ web page of :arrow: WikiROMS and hope that this issue is very clear and never discussed again in the forum.

User avatar
susonic
Posts: 167
Joined: Tue Aug 21, 2007 5:44 pm
Location: UST21 / Korea
Contact:

Re: About minimum depth

#5 Unread post by susonic »

Thank you for your detailed comment and adding an entry of WikiROMS.
That is kind of you to do so.
And I'm sure that users can get this fundamental information via the WikiROMS.

Regards,

-JH
Joonho Lee

Lei_Shi
Posts: 4
Joined: Thu Aug 14, 2003 7:46 pm
Location: School of Marine Science, University of Maine

Re: About minimum depth

#6 Unread post by Lei_Shi »

Considering a wet-dry ROMS, We can have a grid covering a domain without worrying about minimum depth. Just let the water flow around following fluid mechanics. In that sense there is no minimum depth.

Of course, practically wet-dry scheme usually still has a minimum depth for a grid point to be considered a wet point.

lizhicheng
Posts: 6
Joined: Mon Dec 31, 2018 8:42 pm
Location: tianjin university

Re: About minimum depth

#7 Unread post by lizhicheng »

if defind WER_DRY, the Hmin how to be set?

Post Reply