Maximum permitted CFL number for ROMS

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
lanerolle
Posts: 157
Joined: Mon Apr 28, 2003 5:12 pm
Location: NOAA

Maximum permitted CFL number for ROMS

#1 Post by lanerolle » Mon Mar 09, 2009 9:56 pm

I usually run the ROMS with the following CPP options for advection, etc :

#define SOLVE3D

#define PERFECT_RESTART

#define CURVGRID

#define UV_ADV
#define UV_SADVECTION

#define UV_COR

#define UV_QDRAG

#define TS_SVADVECTION
#define TS_U3HADVECTION

#define DJ_GRADPS
#define SPLINES

#define RADIATION_2D

Hence, I do not use any explicit horizontal diffusion/viscosity but rely on the upstream-biased advection to get rid of the numerical oscillations in space for momentum and tracers.

For a set-up such as the above, what is the maximum CFL number (ie. CFL=|u|*dt/dx + |v|*dt/dy + |w|*dt/dz) permitted to ensure numerical stability (plus consistency & convergence also)? Is it 1.0 or a value smaller than 1.0?

Post Reply