In his review ‘On Charting Global Ocean Tides’ (Reviews of Geophysics and space physics, Vol 18, No. 1, pages 243-268, 1980) Schwiderski argues that the lateral eddy viscosity is best represented if it is dependent on mesh size and depth.
My present understanding of horizontal viscosity in ROMS is that it is represented by VISC2 or VISC4 and it is defined as a constant by the user. If VISC_GRID option is activated the specified viscosity value is scaled according to grid size.
In other words viscosity can be mesh size dependent but not depth dependent, is this correct?
Can VISC2 be scaled by depth?
Re: Can VISC2 be scaled by depth?
BAN, you will need to do some coding but technically, I don't see why it wouldn't be possible. I envision 2 ways, a) one is by modifiying the code where Hernan dealt with the variable grid size and viscosity, and perhaps importing the bottom topography if that variable is not currently included in that subroutine, or...b) activate a sponge layer which will be actually covering the entire domain and with a user-defined viscosity (Ah) such as Ah=Ah(h(x,y)) where h(x,y) is the bottom topography. I hope this helps. I believe that the latter would be a bit easier. Guillermo.