I've been working with a coupled ROMS/SWAN application. I had initially tested the application for over a 1 month period forcing ROMS with river, tide, and wind while forcing SWAN with wind at 2hr intervals. This worked fine on 6,12, & 24 cpu's for each model.
I recently moved on to extending the simulation to a 4 year period, and increasing SWAN wind forcing to hourly intervals. The model runs but ROMS quits after 1.5 months stating that the model is blowing up. What's curious to me is that ROMS output looks healthy, as does SWAN, although, throughout, the SWAN does report warnings for the wave height at the boundary being >10% different than the imposed.
Code: Select all
STEP Day HH:MM:SS KINETIC_ENRG POTEN_ENRG TOTAL_ENRG NET_VOLUME
C => (i,j,k) Cu Cv Cw Max Speed
0 0 00:00:00 0.000000E+00 2.566465E+02 2.566465E+02 1.657367E+12
(000,000,00) 0.000000E+00 0.000000E+00 0.000000E+00 0.000000E+00
DEF_HIS - creating history file, Grid 01: Output/ocean_his_0001.nc
WRT_HIS - wrote history fields (Index=1,1) into time record = 0000001
1 0 00:02:30 1.072500E-06 2.714058E+02 2.714058E+02 1.566991E+12
(047,186,10) 1.016723E-02 2.227676E-03 0.000000E+00 5.021956E-02
2 0 00:05:00 3.545184E-06 2.714064E+02 2.714064E+02 1.566992E+12
(062,185,10) 6.898552E-03 1.458758E-03 4.373714E-02 6.960201E-02
...
26355 45 18:07:30 1.090714E-02 2.701372E+02 2.701481E+02 1.554365E+12
(002,256,10) 2.582380E-01 0.000000E+00 1.243842E+03 4.186573E+00
26356 45 18:10:00 1.086319E-02 2.701226E+02 2.701335E+02 1.554063E+12
(002,256,10) 2.498994E-01 0.000000E+00 1.283567E+03 4.076820E+00
26357 45 18:12:30 1.081451E-02 2.701087E+02 2.701195E+02 1.553765E+12
(002,256,10) 2.432082E-01 0.000000E+00 1.315139E+03 3.982541E+00
26358 45 18:15:00 1.076116E-02 2.700955E+02 2.701062E+02 1.553473E+12
(002,256,10) 2.376432E-01 0.000000E+00 1.340432E+03 3.940091E+00
26359 45 18:17:30 1.070618E-02 2.700829E+02 2.700936E+02 1.553185E+12
(002,256,10) 2.329951E-01 0.000000E+00 1.361434E+03 5.304341E+00
26360 45 18:20:00 1.065577E-02 2.700709E+02 2.700816E+02 1.552904E+12
(002,256,10) 2.292045E-01 0.000000E+00 1.378942E+03 6.378915E+00
26361 45 18:22:30 1.062528E-02 2.700597E+02 2.700703E+02 1.552627E+12
(002,256,10) 2.260064E-01 0.000000E+00 1.393181E+03 9.145987E+00
26362 45 18:25:00 1.073788E-02 2.700490E+02 2.700598E+02 1.552357E+12
(002,256,10) 2.231079E-01 0.000000E+00 1.405163E+03 1.773106E+01
26363 45 18:27:30 1.304447E-02 2.700391E+02 2.700521E+02 1.552092E+12
(002,256,10) 2.204189E-01 0.000000E+00 1.416003E+03 6.036142E+01
Blowing-up: Saving latest model state into RESTART file
Node # 5 CPU: 16131.372
Node # 1 CPU: 16131.371
Node # 2 CPU: 16131.371
Node # 3 CPU: 16131.371
Node # 4 CPU: 16131.371
WRT_RST - wrote re-start fields (Index=1,2) into time record = 0000003
Elapsed CPU time (seconds):
Node # 0 CPU: 16131.372
ROMS/TOMS - Output NetCDF summary for Grid 01:
number of time records written in HISTORY file = 00000368
number of time records written in RESTART file = 00000046
Analytical header files used:
ROMS/Functionals/ana_btflux.h
ROMS/Functionals/ana_fsobc.h
ROMS/Functionals/ana_m2obc.h
ROMS/Functionals/ana_nudgcoef.h
ROMS/Functionals/ana_stflux.h
ROMS/TOMS: DONE... Thursday - April 24, 2014 - 5:55:30 PM
Code: Select all
Time of computation -> 20060211.200000 in sec: 3614400.
accuracy OK in 1.41 % of wet grid points ( 98.00 % required)
** WARNING : Differences in wave height at the boundary
Relative difference between input and computation >= 0.10
Hs[m] Hs[m] Hs[-]
ix iy index (input) (computed) (relative)
--------------------------------------------------
150 252 4062 1.00 0.63 0.37
150 251 4061 1.00 0.62 0.37
150 250 4060 1.00 0.62 0.38
150 249 4059 1.00 0.62 0.38
150 248 4058 1.00 0.62 0.38
150 247 4057 1.00 0.62 0.38
150 246 4056 1.00 0.62 0.38
150 245 4055 1.00 0.62 0.38
150 244 4054 1.00 0.62 0.38
150 243 4053 1.00 0.62 0.38
150 242 4052 1.00 0.62 0.38
150 241 4051 1.00 0.62 0.38
150 240 4050 1.00 0.62 0.38
150 239 4049 1.00 0.62 0.38
150 238 4048 1.00 0.62 0.38
150 237 4047 1.00 0.62 0.38
150 236 4046 1.00 0.62 0.38
150 235 4045 1.00 0.62 0.38
150 234 4044 1.00 0.62 0.37
150 233 4043 1.00 0.62 0.37
150 232 4042 1.00 0.63 0.37
150 231 4041 1.00 0.63 0.37
150 230 4040 1.00 0.63 0.37
150 229 4039 1.00 0.63 0.36
150 228 4038 1.00 0.63 0.36
150 227 4037 1.00 0.64 0.36
150 226 4036 1.00 0.64 0.36
150 225 4035 1.00 0.64 0.36
150 224 4034 1.00 0.64 0.36
150 223 4033 1.00 0.64 0.36
150 222 4032 1.00 0.64 0.36
150 221 4031 1.00 0.64 0.35
150 220 4030 1.00 0.64 0.35
150 219 4029 1.00 0.64 0.35
Time of computation -> 20060211.203000 in sec: 3616200.
accuracy OK in 1.52 % of wet grid points ( 98.00 % required)
** WARNING : Differences in wave height at the boundary
Relative difference between input and computation >= 0.10
Hs[m] Hs[m] Hs[-]
ix iy index (input) (computed) (relative)
--------------------------------------------------
150 252 4062 1.00 0.63 0.37
150 251 4061 1.00 0.63 0.37
150 250 4060 1.00 0.62 0.37
150 249 4059 1.00 0.62 0.37
150 248 4058 1.00 0.62 0.38
150 247 4057 1.00 0.62 0.38
150 246 4056 1.00 0.62 0.38
150 245 4055 1.00 0.62 0.38
150 244 4054 1.00 0.62 0.38
150 243 4053 1.00 0.62 0.38
150 242 4052 1.00 0.62 0.38
150 241 4051 1.00 0.62 0.38
150 240 4050 1.00 0.62 0.38
150 239 4049 1.00 0.62 0.38
150 238 4048 1.00 0.62 0.38
150 237 4047 1.00 0.62 0.38
150 236 4046 1.00 0.62 0.37
150 235 4045 1.00 0.62 0.37
150 234 4044 1.00 0.63 0.37
150 233 4043 1.00 0.63 0.37
150 232 4042 1.00 0.63 0.37
150 231 4041 1.00 0.63 0.37
150 230 4040 1.00 0.63 0.36
150 229 4039 1.00 0.64 0.36
150 228 4038 1.00 0.64 0.36
150 227 4037 1.00 0.64 0.36
150 226 4036 1.00 0.64 0.36
150 225 4035 1.00 0.64 0.36
150 224 4034 1.00 0.64 0.35
150 223 4033 1.00 0.64 0.35Any suggestions would be greatly appreciated.
