Closed ghost closed 3 years ago
here are the contents of the hotstart fort.15
:
created on 2021-05-03 18:39 ! RUNDES - 32 CHARACTER ALPHANUMERIC RUN DESCRIPTION
NOMAD mesh v1e MSL ! RUNID - 24 CHARACTER ALPANUMERIC RUN IDENTIFICATION
1 ! NFOVER - NONFATAL ERROR OVERRIDE OPTION
1 ! NABOUT - ABREVIATED OUTPUT OPTION PARAMETER
100 ! NSCREEN - UNIT 6 OUTPUT OPTION PARAMETER
567 ! IHOT - HOT START PARAMETER
2 ! ICS - COORDINATE SYSTEM SELECTION PARAMETER
511113 ! IM - MODEL SELECTION PARAMETER
1 ! NOLIBF - BOTTOM FRICTION TERM SELECTION PARAM; before NWP==1, '2' was used
2 ! NOLIFA - FINITE AMPLITUDE TERM SELECTION PARAMETER
1 ! NOLICA - SPATIAL DERIVATIVE CONVECTIVE SELECTION PARAMETER
1 ! NOLICAT - TIME DERIVATIVE CONVECTIVE TERM SELECTION PARAMETER
5 ! NWP - VARIABLE BOTTOM FRICTION AND LATERAL VISCOSITY OPTION PARAMETER; default 0
mannings_n_at_sea_floor
primitive_weighting_in_continuity_equation
surface_canopy_coefficient
surface_directional_effective_roughness_length
surface_submergence_state
1 ! NCOR - VARIABLE CORIOLIS IN SPACE OPTION PARAMETER
1 ! NTIP - TIDAL POTENTIAL OPTION PARAMETER
520 ! NWS - WIND STRESS AND BAROMETRIC PRESSURE OPTION PARAMETER
8 ! NRAMP - RAMP FUNCTION OPTION
9.81 ! G - ACCELERATION DUE TO GRAVITY - DETERMINES UNITS
-3 ! TAU0 - WEIGHTING FACTOR IN GWCE; original, 0.005
7.000000 ! DTDP - TIME STEP (IN SECONDS)
0 ! STATIM - STARTING TIME (IN DAYS)
0 ! REFTIM - REFERENCE TIME (IN DAYS)
2012 10 22 06 18 0.9 1 2 ! WTIMINC RSTIMINC - meteorological data time increment, wave forcing increment
16.4583 ! RNDAY - TOTAL LENGTH OF SIMULATION (IN DAYS)
12.250 0.000 0.000 0.000 12.250 12.250 1.000 0.000 12.250 ! DRAMP - DURATION OF RAMP FUNCTION (IN DAYS)
0.5 0.5 0 ! A00 B00 C00 - TIME WEIGHTING FACTORS FOR THE GWCE EQUATION
0.01 0 0 0.01 ! H0 NODEDRYMIN NODEWETRMP VELMIN
-77.0775 32.4507 ! SLAM0 SFEA0 - CENTER OF CPP PROJECTION (NOT USED IF ICS=1, NTIP=0, NCOR=0)
0.0025 ! FFACTOR
-0.2 ! smagorinsky coefficient - LATERAL EDDY VISCOSITY COEFFICIENT; IGNORED IF NWP =1
0 ! CORI - CORIOLIS PARAMETER - IGNORED IF NCOR = 1
8 ! NTIF - NUMBER OF TIDAL POTENTIAL CONSTITUENTS BEING FORCED starting 2008082300
110 ! ANGINN - INNER ANGLE THRESHOLD
0 0 0 0 ! NOUTE TOUTSE TOUTFE NSPOOLE - ELEV STATION OUTPUT INFO (UNIT 61)
0 ! NSTAE - TOTAL NUMBER OF ELEVATION RECORDING STATIONS
0 0 0 0 ! NOUTV TOUTSV TOUTFV NSPOOLV - VELOCITY STATION OUTPUT INFO (UNIT 62)
0 ! NSTAV - TOTAL NUMBER OF VELOCITY RECORDING STATIONS
0 0 0 0 ! NOUTM TOUTSM TOUTFM NSPOOLM - METEOROLOGICAL STATION OUTPUT INFO (UNITS 71/72)
0 ! NSTAM - TOTAL NUMBER OF METEOROLOGICAL RECORDING STATIONS
-5 0.000000 16.458333 514 ! NOUTGE TOUTSGE TOUTFGE NSPOOLGE - GLOBAL ELEVATION OUTPUT INFO (UNIT 63)
-5 0.000000 16.458333 514 ! NOUTGV TOUTSGV TOUTFGV NSPOOLGV - GLOBAL VELOCITY OUTPUT INFO (UNIT 64)
0 0.000000 0.000000 0 ! NOUTGM TOUTSGM TOUTFGM NSPOOLGM - GLOBAL METEOROLOGICAL OUTPUT INFO
0 ! NFREQ
0 0 0 0 ! THAS THAF NHAINC FMV - HARMONIC ANALYSIS PARAMETERS
0 0 0 0 ! NHASE NHASV NHAGE NHAGV - CONTROL HARMONIC ANALYSIS AND OUTPUT TO UNITS 51,52,53,54
0 0 ! NHSTAR NHSINC - HOT START FILE GENERATION PARAMETERS
1 0 1E-08 25 ! ITITER ISLDIA CONVCR ITMAX - ALGEBRAIC SOLUTION PARAMETERS
! NCPROJ - PROJECT TITLE
! NCINST - PROJECT INSTITUTION
! NCSOUR - PROJECT SOURCE
! NCHIST - PROJECT HISTORY
! NCREF - PROJECT REFERENCES
! NCCOM - PROJECT COMMENTS
! NCHOST - PROJECT HOST
! NCONV - CONVENTIONS
! NCCONT - CONTACT INFORMATION
2012-10-10 00:00 ! NCDATE - forcing start date
&SWANOutputControl SWAN_OutputHS=False, SWAN_OutputDIR=False, SWAN_OutputTM01=False, SWAN_OutputTPS=False, SWAN_OutputWIND=False, SWAN_OutputTM02=False, SWAN_OutputTMM10=False \
and the difference between the coldstart and hotstart fort.15
s:
diff spinup/fort.15 runs/unperturbed/fort.15
1c1
< created on 2021-05-03 18:38 ! RUNDES - 32 CHARACTER ALPHANUMERIC RUN DESCRIPTION
---
> created on 2021-05-03 18:39 ! RUNDES - 32 CHARACTER ALPHANUMERIC RUN DESCRIPTION
6c6
< 0 ! IHOT - HOT START PARAMETER
---
> 567 ! IHOT - HOT START PARAMETER
21,22c21,22
< 0 ! NWS - WIND STRESS AND BAROMETRIC PRESSURE OPTION PARAMETER
< 1 ! NRAMP - RAMP FUNCTION OPTION
---
> 520 ! NWS - WIND STRESS AND BAROMETRIC PRESSURE OPTION PARAMETER
> 8 ! NRAMP - RAMP FUNCTION OPTION
28,29c28,30
< 12.25 ! RNDAY - TOTAL LENGTH OF SIMULATION (IN DAYS)
< 12.25 ! DRAMP - DURATION OF RAMP FUNCTION (IN DAYS)
---
> 2012 10 22 06 18 0.9 1 2 ! WTIMINC RSTIMINC - meteorological data time increment, wave forcing increment
> 16.4583 ! RNDAY - TOTAL LENGTH OF SIMULATION (IN DAYS)
> 12.250 0.000 0.000 0.000 12.250 12.250 1.000 0.000 12.250 ! DRAMP - DURATION OF RAMP FUNCTION (IN DAYS)
4934,4935c4935,4939
< -5 0.000000 12.250000 514 ! NOUTGE TOUTSGE TOUTFGE NSPOOLGE - GLOBAL ELEVATION OUTPUT INFO (UNIT 63)
< -5 0.000000 0.000000 514 ! NOUTGV TOUTSGV TOUTFGV NSPOOLGV - GLOBAL VELOCITY OUTPUT INFO (UNIT 64)
---
> 0 0 0 0 ! NOUTM TOUTSM TOUTFM NSPOOLM - METEOROLOGICAL STATION OUTPUT INFO (UNITS 71/72)
> 0 ! NSTAM - TOTAL NUMBER OF METEOROLOGICAL RECORDING STATIONS
> -5 0.000000 16.458333 514 ! NOUTGE TOUTSGE TOUTFGE NSPOOLGE - GLOBAL ELEVATION OUTPUT INFO (UNIT 63)
> -5 0.000000 16.458333 514 ! NOUTGV TOUTSGV TOUTFGV NSPOOLGV - GLOBAL VELOCITY OUTPUT INFO (UNIT 64)
> 0 0.000000 0.000000 0 ! NOUTGM TOUTSGM TOUTFGM NSPOOLGM - GLOBAL METEOROLOGICAL OUTPUT INFO
4939c4943
< 5 151200 ! NHSTAR NHSINC - HOT START FILE GENERATION PARAMETERS
---
> 0 0 ! NHSTAR NHSINC - HOT START FILE GENERATION PARAMETERS
4950a4955
> &SWANOutputControl SWAN_OutputHS=False, SWAN_OutputDIR=False, SWAN_OutputTM01=False, SWAN_OutputTPS=False, SWAN_OutputWIND=False, SWAN_OutputTM02=False, SWAN_OutputTMM10=False \
and finally, the contents of configure_adcirc.json
:
{
"adcirc_executable_path": "/scratch2/COASTAL/coastal/save/shared/repositories/ADC-WW3-NWM-NEMS/ADCIRC/work/padcirc",
"adcprep_executable_path": "/scratch2/COASTAL/coastal/save/shared/repositories/ADC-WW3-NWM-NEMS/ADCIRC/work/adcprep",
"modeled_start_time": "2012-10-22 06:00:00",
"modeled_end_time": "2012-10-26 11:00:00",
"modeled_timestep": 7.0,
"fort_13_path": "/scratch2/COASTAL/coastal/save/shared/models/meshes/hsofs/120m/Subsetted_Florence2018_Test/fort.13",
"fort_14_path": "/scratch2/COASTAL/coastal/save/shared/models/meshes/hsofs/120m/Subsetted_Florence2018_Test/fort.14",
"tidal_spinup_duration": 1058400.0,
"tidal_spinup_timestep": 7.0,
"source_filename": "/scratch2/COASTAL/coastal/save/shared/repositories/ADC-WW3-NWM-NEMS/modulefiles/envmodules_intel.hera",
"use_original_mesh": false,
"output_surface": true,
"surface_output_interval": 3600.0,
"output_stations": false,
"stations_file_path": null,
"stations_output_interval": 360.0,
"output_spinup": true,
"output_elevations": true,
"output_velocities": true,
"output_concentrations": false,
"output_meteorological_factors": false,
"processors": 600,
"nems_parameters": {},
"attributes": {
"predictor_corrector": null,
"RUNDES": null,
"_IHOT": null,
"RUNID": null,
"NFOVER": null,
"WarnElev": null,
"iWarnElevDump": null,
"WarnElevDumpLimit": null,
"ErrorElev": null,
"NABOUT": null,
"NSCREEN": null,
"IDEN": null,
"NOLIBF": null,
"NOLICA": null,
"NOLICAT": null,
"NCOR": null,
"NTIP": null,
"G": null,
"TAU0": null,
"DTDP": null,
"STATIM": null,
"REFTIM": null,
"DRAMP": null,
"DRAMPExtFlux": null,
"FluxSettlingTime": null,
"DRAMPIntFlux": null,
"DRAMPElev": null,
"DRAMPTip": null,
"DRAMPMete": null,
"DRAMPWRad": null,
"DUnRampMete": null,
"H0": null,
"NODEDRYMIN": null,
"NODEWETRMP": null,
"VELMIN": null,
"SLAM0": null,
"SFEA0": null,
"FFACTOR": null,
"CF": null,
"HBREAK": null,
"FTHETA": null,
"FGAMMA": null,
"ESLM": null,
"NOUTGE": null,
"TOUTSGE": null,
"TOUTFGE": null,
"NSPOOLGE": null,
"NOUTGV": null,
"TOUTSGV": null,
"TOUTFGV": null,
"NSPOOLGV": null,
"NOUTGM": null,
"TOUTSGM": null,
"TOUTFGM": null,
"NSPOOLGM": null,
"NOUTGC": null,
"TOUTSGC": null,
"TOUTFGC": null,
"NSPOOLGC": null,
"CORI": null,
"ANGINN": null,
"THAS": null,
"THAF": null,
"NHAINC": null,
"FMV": null,
"NHSTAR": null,
"NHSINC": null,
"ITITER": null,
"ISLDIA": null,
"CONVCR": null,
"ITMAX": null,
"NCPROJ": null,
"NCINST": null,
"NCSOUR": null,
"NCHIST": null,
"NCREF": null,
"NCCOM": null,
"NCHOST": null,
"NCCONV": null,
"NCCOUT": null,
"vertical_mode": null,
"lateral_stress_in_gwce": null,
"lateral_stress_in_gwce_is_symmetrical": null,
"adcvection_in_gwce": null,
"lateral_stress_in_momentum": null,
"lateral_stress_in_momentum_is_symmetrical": null,
"lateral_stress_in_momentum_method": null,
"adcvection_in_momentum": null,
"area_integration_in_momentum": null,
"baroclinicity": null,
"gwce_solution_scheme": "semi-implicit",
"passive_scalar_transport": null,
"stress_based_3D": null,
"smagorinsky": true,
"smagorinsky_coefficient": null,
"horizontal_mixing_coefficient": null,
"CFL": null
}
}
So your coldstart was run without SWAN and wind forcing I guess?
If I look at this table for NWS: https://wiki.adcirc.org/wiki/NWS#Extended_NWS_with_Ice_.2B_Waves I don't see any option for NWS equal to "520". Maybe supposed to be 320 (for GAHM + SWAN)?
If I look at this table for NWS: https://wiki.adcirc.org/wiki/NWS#Extended_NWS_with_Ice_.2B_Waves I don't see any option for NWS equal to "520". Maybe supposed to be 320 (for GAHM + SWAN)?
Good catch, I forgot that 520
is reserved for NEMS, and I'm not using NEMS in this specific run. Thanks!
currently running Sandy with gwce_solution_scheme="semi-implicit"
in
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210507_hsofs_120m_subsetflo_besttrack_sandy
and Florence with gwce_solution_scheme="explicit"
in
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210507_hsofs_120m_subsetflo_besttrack_florence
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210507_hsofs_120m_subsetflo_besttrack_sandy/spinup/ADCIRC_SPINUP.err.log
...
...
...
MPI_Abort(159): MPI_Abort(comm=0x18447c30, errorcode=32766) failed
MPI_Abort(97).: Invalid communicator
...
...
...
libc-2.17.so 00002ABC95B6C555 __libc_start_main Unknown Unknown
padcirc 000000000040BDA9 Unknown Unknown Unknown
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
padcirc 00000000006CA8DD Unknown Unknown Unknown
libpthread-2.17.s 00002B6CA37A7630 Unknown Unknown Unknown
libmpi.so.12 00002B6CA2795340 MPI_Abort Unknown Unknown
libmpifort.so.12. 00002B6CA23A9C5D mpi_abort Unknown Unknown
padcirc 000000000044581B msg_abort_ 1524 messenger.F
padcirc 00000000004AF54E sizes_mp_make_dir 332 sizes.F
padcirc 00000000004B91C7 adcirc_mod_mp_adc 180 adcirc.F
padcirc 000000000061A8D2 MAIN__ 18 driver.F
padcirc 000000000040BE9E Unknown Unknown Unknown
libc-2.17.so 00002B6CA3CD8555 __libc_start_main Unknown Unknown
padcirc 000000000040BDA9 Unknown Unknown Unknown
slurmstepd: error: *** STEP 18641261.0 ON h35m02 CANCELLED AT 2021-05-07T19:27:22 ***
srun: Job step aborted: Waiting up to 32 seconds for job step to finish.
slurmstepd: error: _file_write_content: unable to write 6 bytes to cgroup /sys/fs/cgroup/freezer/slurm/uid_20644/job_18641261/step_0/freezer.state: No such device
srun: error: h35m20: tasks 440-444,446-479: Killed
srun: launch/slurm: _step_signal: Terminating StepId=18641261.0
srun: error: h35m05: tasks 80-86,88-119: Killed
srun: error: h35m19: tasks 400-417,419-439: Killed
srun: error: h35m07: tasks 160-180,182-199: Killed
srun: error: h35m06: tasks 120-151,153,155-159: Killed
srun: error: h35m02: tasks 0,2-39: Killed
srun: error: h35m22: tasks 520-529,531-551,553-559: Killed
srun: error: h35m15: tasks 280-294,296-300,302,304-319: Killed
srun: error: h35m04: tasks 40-53,55-57,59-70,72-79: Killed
srun: error: h35m18: tasks 360-364,366-380,382-399: Killed
srun: error: h35m10: tasks 200-215,217-239: Killed
srun: error: h35m24: tasks 560-565,567,569-578,580-599: Killed
srun: error: h35m20: task 445: Exited with exit code 174
srun: error: h35m05: task 87: Exited with exit code 174
srun: error: h35m19: task 418: Exited with exit code 174
srun: error: h35m07: task 181: Exited with exit code 174
srun: error: h35m06: tasks 152,154: Exited with exit code 174
srun: error: h35m02: task 1: Exited with exit code 174
srun: error: h35m22: tasks 530,552: Exited with exit code 174
srun: error: h35m15: tasks 295,301,303: Exited with exit code 174
srun: error: h35m04: tasks 54,58,71: Exited with exit code 174
srun: error: h35m18: tasks 365,381: Exited with exit code 174
srun: error: h35m10: task 216: Exited with exit code 174
srun: error: h35m24: tasks 566,568,579: Exited with exit code 174
srun: error: h35m17: tasks 320-338,340-359: Killed
srun: error: h35m17: task 339: Exited with exit code 174
srun: error: h35m14: tasks 240-279: Killed
srun: error: h35m21: tasks 480-519: Killed
I think this looks like it is related to the prepped PEs not equaling the requested processors for padcirc
alright, I seem to have found the problem. I was submitting the jobs like this
sbatch adcirc.job --dependency=afterok:$setup_jobid:$spinup_jobid
however, the dependency was not actually being applied, so the hotstart would start immediately without waiting for the spinup OR adcprep
to complete. This aligns with the errors that we were seeing in hotstart.
After some testing, I figured out that it works if I write it like this instead:
sbatch --dependency=afterok:$setup_jobid:$spinup_jobid adcirc.job
so I've fixed the script generation and am testing the run now.
it looks like the hotstart run completed,
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210510_hsofs_120m_subsetflo_besttrack_florence/runs/unperturbed
but with a few elevation warnings:
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210510_hsofs_120m_subsetflo_besttrack_florence/runs/unperturbed/ADCIRC_HOTSTART.err.log
TIME STEP = 188082 ITERATIONS = 17 TIME = 0.13165740E+07
ELMAX = 2.0901E+001 AT NODE 46732 SPEEDMAX = 5.9555E+000 AT NODE 47917 ON MYPROC = 4 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 188084 ITERATIONS = 17 TIME = 0.13165880E+07
ELMAX = 2.2019E+001 AT NODE 46732 SPEEDMAX = 5.2908E+000 AT NODE 47917 ON MYPROC = 4 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 188885 ITERATIONS = 17 TIME = 0.13221950E+07
ELMAX = 2.2710E+001 AT NODE 208899 SPEEDMAX = 4.5079E+000 AT NODE -220561 ON MYPROC = 248 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 188886 ITERATIONS = 17 TIME = 0.13222020E+07
ELMAX = 2.1132E+001 AT NODE 208899 SPEEDMAX = 7.0585E+000 AT NODE 207072 ON MYPROC = 248 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 188887 ITERATIONS = 17 TIME = 0.13222090E+07
ELMAX = 2.4033E+001 AT NODE 208899 SPEEDMAX = 7.0544E+000 AT NODE 207072 ON MYPROC = 248 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 188889 ITERATIONS = 17 TIME = 0.13222230E+07
ELMAX = 2.1556E+001 AT NODE 207072 SPEEDMAX = 6.8517E+000 AT NODE 207072 ON MYPROC = 248 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189691 ITERATIONS = 17 TIME = 0.13278370E+07
ELMAX = 2.1822E+001 AT NODE 135311 SPEEDMAX = 1.4178E+000 AT NODE 148766 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189692 ITERATIONS = 17 TIME = 0.13278440E+07
ELMAX = 2.0661E+001 AT NODE 135311 SPEEDMAX = 5.1043E+000 AT NODE 133657 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189693 ITERATIONS = 17 TIME = 0.13278510E+07
ELMAX = 2.5124E+001 AT NODE 135311 SPEEDMAX = 9.9263E-001 AT NODE 152227 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189694 ITERATIONS = 17 TIME = 0.13278580E+07
ELMAX = 2.2977E+001 AT NODE 135311 SPEEDMAX = 5.6609E+000 AT NODE 133657 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189695 ITERATIONS = 17 TIME = 0.13278650E+07
ELMAX = 2.8320E+001 AT NODE 135311 SPEEDMAX = 1.9715E+000 AT NODE 136955 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189696 ITERATIONS = 17 TIME = 0.13278720E+07
ELMAX = 2.5513E+001 AT NODE 135311 SPEEDMAX = 6.2510E+000 AT NODE 133657 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189697 ITERATIONS = 17 TIME = 0.13278790E+07
ELMAX = 3.2227E+001 AT NODE 135311 SPEEDMAX = 1.0183E+000 AT NODE 135310 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189698 ITERATIONS = 17 TIME = 0.13278860E+07
ELMAX = 2.8683E+001 AT NODE 135311 SPEEDMAX = 6.9802E+000 AT NODE 133657 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189699 ITERATIONS = 17 TIME = 0.13278930E+07
ELMAX = 3.6274E+001 AT NODE 135311 SPEEDMAX = 5.0500E+000 AT NODE 136956 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189700 ITERATIONS = 17 TIME = 0.13279000E+07
ELMAX = 2.6005E+001 AT NODE 135311 SPEEDMAX = 5.0487E+000 AT NODE 136956 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189701 ITERATIONS = 17 TIME = 0.13279070E+07
ELMAX = 2.2789E+001 AT NODE 135311 SPEEDMAX = 5.6019E+000 AT NODE 133657 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
TIME STEP = 189702 ITERATIONS = 17 TIME = 0.13279140E+07
ELMAX = 2.3334E+001 AT NODE 135311 SPEEDMAX = 5.5986E+000 AT NODE 133657 ON MYPROC = 259 ** WARNING: Elevation.gt.WarnElev **
here's a run from today, with station data. when running with 15 day spinup, the phase seems to be correct but with shifted amplitude
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210512_hsofs_120m_subsetflo_besttrack_florence/spinup
should the spinup have a starting amplitude?
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210512_hsofs_120m_subsetflo_besttrack_florence/runs/unperturbed
No, that's exactly what the spinup is, we gradually increase the amplitude in order to avoid shocking the system.
alright, I'll try for a longer spinup to see if that makes it reach the correct zeta
alright, I'll try for a longer spinup to see if that makes it reach the correct zeta
That won't be necessary, the bias difference you see will not be resolved by a longer spinup. We need to enable pseudo-pressure loading to fix this.
alright, I'll try for a longer spinup to see if that makes it reach the correct zeta
@zacharyburnettNOAA Please use NAVD88 as input for the plot_fort61
command, these are referenced to MLLW, and that accounts for the largest amount of difference you see.
@zacharyburnettNOAA I would use MSL for the vertical datum of the plots as that should be closest to ADCIRC one. Alternatively for comparison purposes can just offset ADCIRC by adding the mean of the observation over the time period to it.
@zacharyburnettNOAA I would use MSL for the vertical datum of the plots as that should be closest to ADCIRC one. Alternatively for comparison purposes can just offset ADCIRC by adding the mean of the observation over the time period to it.
To be more precise, which one to chose depends on the CRS of the mesh, but MSL is not a CRS. Meshes should be interpolated to the geoid surface. NAVD88 is the closest approximation for the geoid in North America. However some meshes are interpolated @ MSL, such as the HSOFS one, so it's fair to use MSL in such cases. I did not go into this level of detail on my previous posts, because I was thinking we would gradually reach this point of discussion.
That's true the bathymetry should ideally be on a geoid. However, the ADCIRC model we are simulating is a 2D barotropic model and therefore it cannot correctly compute the sea surface height relative to the geoid. Therefore, as an approximation in my opinion for 2D barotropic models it is just better to make the mesh on the "MSL" to account for variations in the sea surface height the model cannot capture. So the HSOFS mesh took the correct route in my opinion, and they provided a correction file to correct back to NAVD88 when needed. Alternatively, if the domain is small enough we provide input boundary condition that accounts for the offset from the geoid to the MSL.
I suppose that another alternative is to apply the psuedo-pressure forcing as you mentioned to account for the discrepancy between the geoid and the MSL (basically this is like accounting for baroclinic pressure gradients missing from 2D barotropic model, see https://doi.org/10.1029/2018JC014682).
In any case, for 2D barotropic model we almost always run into problem of aligning the model with the observations because the model cannot properly sustain any real vertical datum. Therefore, often we just add the mean from observations to the model.
@WPringle I agree with you in general. I think we are on the same page, no issues. @zacharyburnettNOAA Bottom line is we need to plot using MSL for HSOFS.
A relevant reference: https://www.sciencedirect.com/topics/earth-and-planetary-sciences/sea-surface-topography
@zacharyburnettNOAA From https://user-images.githubusercontent.com/52422935/117993696-f584cd00-b30d-11eb-9221-f49bb3a4610f.png, it looks like there's some storm surge going on during the spinup phase. Are you certain that the dates are correct? All storm surge signals should be happening on the hotstart phase. I would also suggest to find stations in the hotstart phase that are showing storm surge signals in order to showcase your results.
EDIT: The signal in Dog River Brigde is far-away from landfall, so that's an independent signal. During further conversation with @zacharyburnettNOAA, the plots will be updated at some point to have a better view of storm surge signals of the modeled event.
here are the hotstart zetas of stations within 150km of the track:
here are the hotstart zetas of stations within 150km of the track: @zacharyburnettNOAA Excellent work!
when running ADCIRC-only with tidal spinup, besttrack, and ww3data forcing on @WPringle's subsetted mesh (in the directory
/scratch2/COASTAL/coastal/save/shared/working/zach/adcirc/run_20210503_hsofs_120m_subsetflo_besttrack
), the spinup phase runs fine but the hotstart fails with the following error: