I tried river_plume2 test case to learn how specify open boundary in ROMS after read paper related to it (P. Marchesiello, J.C. McWilliams and A. Shchepetkin, 2001). But in the end of river_plume2 simulation (40 days) I found some strange flow around southern boundary which also affects interior grid flow. So, my questions are:
1. How could it happen? Am I missing something during setting the test case? I used the default cpp option of it.
2. What should I do to overcome that problem? I already change cpp option using combination of RADIATION_2D, xx_FSRADIATION, xx_M2RADIATION, xx_M3RADIATION, or also specified vbar = -0.05 in ANA_M2OBC for xx_M2FLATHER. But still there are some strange flows which not propagate out of the domain.
3. Are there any ways or combination in defining xx_FS(..), xx_M2(..), and xx_M3(..) in cpp option so that I should not specify any value (or zero value) especially for barotropic velocity and free surface (as in river_plume2 test case) but it still can propagate out the domain?
Code: Select all
default OBC of river_plume2
#define SOUTH_FSCHAPMAN
#define SOUTH_M2GRADIENT
#define SOUTH_M3GRADIENT
#define SOUTH_TGRADIENT
#define NORTH_FSCHAPMAN
#define NORTH_M2GRADIENT
#define NORTH_M3GRADIENT
#define NORTH_TGRADIENT
#define WESTERN_WALL
#define EASTERN_WALL