I'm facing a strange unrealistic discontinuity of surface salinity in my COAWST (only ROMS activated) attached below.
At the northern part, I found the freshwater flux was likely input at every other grid.
Actually, after the rain stops, the salinity was reasonably diffused by horizontal diffusion and other mixing processes.
At first, I thought that was caused by my low-reso rain data (from NCEP reanalysis II, ~2.5 deg) comparing with my ROMS resolution (21 km).
But after I replaced the rain with GSMAP data（0.1 deg, attached), it still happens.
Here are my settings:
------------------#define ROMS_MODEL
#undef NESTING
#undef WRF_MODEL
#undef SWAN_MODEL
#undef WW3_MODEL
#undef MCT_LIB
#undef MCT_INTERP_OC2AT
#undef MCT_INTERP_WV2AT
#undef MCT_INTERP_OC2WV
#if defined WRF_MODEL && (defined SWAN_MODEL || defined WW3_MODEL)
# define DRAGLIM_DAVIS
# define COARE_TAYLOR_YELLAND
#endif
#ifdef ROMS_MODEL
/* Physics + numerics */
# if defined WW3_MODEL || defined SWAN_MODEL
# define WEC_VF
# define WDISS_WAVEMOD
# define UV_KIRBY
# endif
/* Options associated with momentum equations */
#define UV_ADV
#define UV_COR
#define UV_VIS2
#define UV_LDRAG
/* Model configuration OPTIONS:*/
#define SOLVE3D
#define CURVGRID
#define MASKING
#define AVERAGES
/* OPTIONS associated with tracer equations */
#undef TS_FIXED
#undef TS_MPDATA
#define TS_U3HADVECTION
#define TS_C4VADVECTION
#define TS_DIF2
#define SALINITY
#define NONLIN_EOS
/* OPTIONS Pressure gradient algorithm */
#define DJ_GRADPS
/* OPTION to activate conservative, parabolic spline reconstruction of **
** vertical derivatives. Notice that there also options (see above) for **
** vertical advection of momentum and tracers using splines. **
** See https://www.myroms.org/projects/src/ticket/681 */
#undef SPLINES_VDIFF
#undef SPLINES_VVISC
/* Bottom fluxes */
#define ANA_BSFLUX
#define ANA_BTFLUX
/* bulk fluxes */
# define BULK_FLUXES
# define NL_BULK_FLUXES
# define LONGWAVE_OUT
# undef QCORRECTION
# undef SCORRECTION
# undef SRELAXATION
# define SOLAR_SOURCE
/* Forcing */
# ifdef WRF_MODEL
# undef BULK_FLUXES
# define ATM2OCN_FLUXES
# undef ANA_SSFLUX
# undef LONGWAVE_OUT
# endif
# undef COOL_SKIN
# undef ANA_SMFLUX
# undef ANA_NUDGCOEF
# define ATM_PRESS
# define EMINUSP
/* OPTIONS for shortwave radiation:*/
#undef ALBEDO /* use if albedo equation for shortwave radiation */
#undef DIURNAL_SRFLUX /* use to impose shortwave radiation local diurnal cycle */
/*OPTIONS for horizontal mixing of momentum */
#define VISC_GRID
#define MIX_S_UV
#undef MIX_GEO_UV
/*OPTIONS for horizontal mixing of tracers */
#define MIX_GEO_TS
#define DIFF_GRID
/* never use tracers except for the flat bottom case*/
#undef MIX_S_TS
/* working with Rad boundary condition */
#define RADIATION_2D
# undef SSW_BBL
# ifdef SSW_BBL
# define SSW_CALC_ZNOT
# define ANA_SEDIMENT
/*# else*/
/* # define UV_LOGDRAG */
# endif
# if !defined SWAN_MODEL && defined SSW_BBL
# define ANA_WWAVE
# endif
/* Turbulence closure */
# undef GLS_MIXING
# undef MY25_MIXING
# define LMD_MIXING
/* # define LIMIT_VDIFF */
/*# define LIMIT_VVISC */
/* boundary layer KPP mixing */
# if defined LMD_MIXING
# define LMD_BKPP
# define LMD_CONVEC
# undef LMD_DDMIX
# define LMD_NONLOCAL
# define LMD_RIMIX
# undef LMD_SHAPIRO
# define LMD_SKPP
# endif
# if defined GLS_MIXING || defined MY25_MIXING
# define KANTHA_CLAYSON
# define N2S2_HORAVG
# define RI_SPLINES
# define CRAIG_BANNER
# define CHARNOK
# endif
/* Tidal forcing */
# undef SSH_TIDES
# undef UV_TIDES
# undef RAMP_TIDES
# undef ADD_FSOBC
# undef ADD_M2OBC
# ifdef SSH_TIDES
# undef FSOBC_REDUCED
# define ADD_FSOBC
# endif
# if defined UV_TIDES || defined SSH_TIDES
# define ADD_M2OBC
# endif
/* Output */
# undef DIAGNOSTICS_UV
# undef DIAGNOSTICS_TS
/* NetCDF input/output OPTIONS*/
#undef OUT_DOUBLE
I learned the settings from examples and combined them by myself. There may be strange settings in my head file.
I would appreciate it if you could correct the errors for me.
Best regards,
Ning
By the way, is that possible to use 1-hourly rain while other data remains 6-hourly?
Or I should interpolate other data into 1-hourly?
