about the new options 'UV_U3DV_SPLIT' and 'TS_U3ADV_SPLIT'

[jiangyuwu]

it is noticed that there are two new options 'UV_U3DV_SPLIT' and 'TS_U3ADV_SPLIT', but when I want to switch the default options in momentum and tracers equations to above options because I focuse on the slope area, the ROMS output of 'KINETIC_ENRG POTEN_ENRG TOTAL_ENRG NET_VOLUME' are "NAN" at first step (without these two options my roms can go on well). can somebody give me some idea about these two new options? thanks a lot.
yuwu jiang

the *.h i used is listed as below:

# define SOLVE3D
# define UV_COR
# define UV_ADV
# define UV_U3ADV_SPLIT
# define UV_QDRAG
# define DJ_GRADPS
# define TS_U3ADV_SPLIT
# define CURVGRID
# define SPHERICAL
# define MASKING
/* Equation of State */
# define SALINITY
# define NONLIN_EOS
# define SPLINES
/* Surface Forcing */

# define BULK_FLUXES /*turn ON or OFF bulk fluxes computation */
# define LONGWAVE_OUT /*Compute ougoing longwave radiation internally = only input longwave down*/
# define EMINUSP /* use rain and tmp to computing E-P */
# define SOLAR_SOURCE /*solar radiation decay with depth*/

/* Bottom Forcing */
# define ANA_BSFLUX
# define ANA_BTFLUX

/* Point Sources - Rivers */
# define TS_PSOURCE
# define UV_PSOURCE


/* Lateral Mixing */
# define MIX_GEO_UV
# define MIX_GEO_TS

/* Vertical Mixing */
# define MY25_MIXING


......

[bzhang]

These two options have not been fully tested and the rotated tracer flux scheme in 4th order mixing might cause unconditional instability of the model. So you may need to wait sometime for hernan to correct this or remove these two options.

[arango]

These options are experimental and we will notify when can be used. It was only released so the beta-tester can start playing with them. As any diffusion problem we need to insure than the numerical scheme is unconditionally stable. Right now, these option become unstable in fine resolution applications with steep topography. We need to do a linear stability analysis to compute the amplitude portrait using finite Fourier series.

[jiangyuwu]

thanks very much, the area with high steep topography always troubles the modeler