Regional Coastal Ocean Modeling: Tutorial - PowerPoint PPT Presentation

1 / 25
About This Presentation
Title:

Regional Coastal Ocean Modeling: Tutorial

Description:

Data: bathymetry, hydrography, surface fluxes global climatological datasets are ... the nonlinear equation of state in a adiabatic part and a compressible part. ... – PowerPoint PPT presentation

Number of Views:558
Avg rating:3.0/5.0
Slides: 26
Provided by: pmar70
Category:

less

Transcript and Presenter's Notes

Title: Regional Coastal Ocean Modeling: Tutorial


1
Regional Coastal Ocean Modeling Tutorial
  • Roms_tools

2
System requirements
  • F95 (ifort), Matlab
  • Netcdf library for Fortran and Matlab (MexCDF)
  • 2 Gbites of disk space
  • ROMS_AGRIF sources
  • Matlab toolbox for ROMS ROMS_tools
  • Data bathymetry, hydrography, surface fluxes
    global climatological datasets are included

3
Package
4
Pre-processing data
  • cd /Roms_tools/Run
  • matlab
  • gtgt start adds the path of different toolboxes
  • gtgt make_grid
  • gtgt make_forcing
  • gtgt make_clim
  • gtgt make_tides
  • gtgt make_biol
  • gtgt nestgui

5
make_grid.m
Title title'Peru Test Model'
Grid file name grdname'roms_grd.nc'
Grid dimensions lonmin-85 lonmax-75
latmin-15 latmax-7 Grid resolution
degree dl1/3 Minimum depth m
hmin10 Topography netcdf file name
(ETOPO 2) topofile'../Topo/etopo2.nc'
Slope parameter (rgrad(h)/h) maximum value for
topography smoothing rtarget0.2
lon, lat, dx, dy, h
L31 M26
h
6
Smoothing methods
  • r ?h / h is the slope of the logarithm of h
  • One method (ROMS) smoothing ln(h) until r lt rmax

Res 1 km r lt 0.25
Res 5 km r lt 0.25
Senegal Bathymetry Profil
7
Smoothing method and resolution
Bathymetry Smoothing Error off Senegal
Convergence at 4 km resolution
Standard Deviation m
Grid Resolution deg
8
Errors in Bathymetry data compilations
Etopo2 Satellite observations
Gebco1 compilation
Shelf errors (noise)
9
Refine the mask
  • gtgt editmask
  • Interactive matlab tool to modify masking
    according to high resolution coastline data

10
Getting the wind forcing
  • gtgt make_forcing

Title - Grid file name - Forcing file name
title'Forcing (COADS)' grdname'roms_grd.nc
' frcname'roms_frc.nc' Set times and
cycles monthly climatology for all data
time1530345 time cycle360
cycle
Default COADS climatological surface forcing of
Da Silva et al., 1994
11
(No Transcript)
12
Modified Julian dates
  • MJD is a modification of the Julian Date that is
    routinely used by astronomers, geodesists, and
    even some historians.
  • This dating convention, designed to facilitate
    chronological calculations, numbers all days in
    consecutive fashion, beginning so as to precede
    the historical period.
  • Julian Day Number 0 is noon 1 January 4713 B.C.
  • MJD modifies this Julian Date in two ways.
  • The MJD begins at midnight rather than noon, in
    keeping with more standard conventions.
  • Secondly, for simplicity, the first two digits of
    the Julian Date are removed. This is because, for
    some three centuries following 17 November 1858,
    the Julian day lies between 2400000 and 2500000.
    The MJD drops those first "24" digits. Thus, we
    have
  • MJD JD - 2400000.5
  • To convert Julian Dates to Gregorian dates
    (month/day/year) we can use various converters

13
Getting the lateral boundary conditions
Title title'Climatology'
Switches for selecting what to process (1ON)
makeclim1 1 process boundary data makeoa1
1 process oa data makeini1 1 process
initial data Grid file name - Climatology
file name Initial file name - OA file name
grdname 'roms_grd.nc' frcname
'roms_frc.nc' clmname 'roms_clm.nc' ininame
'roms_ini.nc' oaname 'roms_oa.nc'
Vertical grid parameters theta_s7.
theta_b0. hc5. N20 number of vertical
levels (rho)
  • gtgt make_clim

OA (objective analysis) files are intermediate
files where hydrographic data are interpolated
(extrapolated under bathymetry) and stored on a
horizontal grid but on z vertical grid. The
transformation to S-coordinate is done after.
14
obc1 0 1 1 open boundaries (1open , S E N
W) Level of reference for geostrophy
calculation zref-500 Day of
initialization tini15 Set times and
cycles monthly climatology for all data
time1530345 time cycle360 cycle
Data climatologies file names
temp_month_data '../WOA2001/temp_month.cdf'
temp_ann_data '../WOA2001/temp_ann.cdf'
insitu2pot1 transform in-situ
temperature to potential temperature
salt_month_data '../WOA2001/salt_month.cdf'
salt_ann_data '../WOA2001/salt_ann.cdf'

15
(No Transcript)
16
Getting the tides boundary conditions
  • gtgt make_tides

TPXO file name tidename'../Tides/TPXO6.n
c' ROMS file names gname
'roms_grd.nc' fname 'roms_frc.nc'
Number of tides component to process
Ntides10 Set start time of simulation
year 2000 month 1 day
15 hr 0. minute 0. second
0.
This is where tidal information is added
This is the starting time of simulation. A
procedure correct phases and amplitudes (nodal
corrections) for real time runs. It employs parts
of a post-processing code from Egbert and
Erofeeva (2002) TPXO model. Running Real-time
tides requires using modified julian dates as
initial time (roms_ini.nc).
17
Getting child grids for nesting
  • gtgt nestgui

18
Preparing the model
  • vi param.h
  • vi cppdefs.h
  • Define CPP keys that used by the C-preprocessor
    when compiling the model
  • Reduce code to its minimal size fast compilation
  • Avoid fortran logical statements efficient coding

parameter (LLm029, MMm024, N20) !       
Peru Test Case
19
  • PERU Configuration Name, this is used in
    param.h.
  • OPENMP Activate Open-MP parallelization
    protocol.
  • MPI Activate MPI parallelization protocol.
  • AGRIF Activate the nesting capabilities
  • SOLVE3D Define if solving 3D primitive
    equations
  • UV_COR Activate Coriolis terms.
  • UV_ADV Activate advection terms.
  • SSH_TIDES Define for processing sea surface
    elevation tidal data at the model boundaries.
  • UV_TIDES Define for processing ocean current
    tidal data at the model boundaries.
  • VAR_RHO_2D Activate nonuniform density in
    barotropic mode pressure- gradient terms.
  • FLAT_WEIGHTS Use a more dissipative averaging
    for the baroclinic/barotropic coupling.
  • CURVGRID Activate curvilinear coordinate grid
    option.
  • SPHERICAL Activate longitude/latitude grid
    positioning.
  • MASKING Activate land masking in the domain.
  • AVERAGES Define if writing out time-averaged
    data.
  • SALINITY Define if using salinity.
  • NONLIN_EOS Activate the nonlinear equation of
    state.
  • SPLIT_EOS Activate to split the nonlinear
    equation of state in a adiabatic part and a
    compressible part.

20
Input parameter file
title PERU TEST MODEL time_stepping
NTIMES dtsec NDTFAST NINFO
720 1800 45
1 S-coord THETA_S, THETA_B, Hc (m)
7.0d0 0.0d0 5.0d0 grid filename
roms_grd.ncforcing
filename
roms_frc.ncclimatology filename
roms_clm.ncinitial NRREC filename
1
roms_ini.ncrestart NRST, NRPFRST /
filename 720 -1
roms_rst.nchistory LDEFHIS,
NWRT, NRPFHIS / filename T
144 0
roms_his.ncaverages NTSAVG, NAVG, NRPFAVG /
filename 1 144
0
roms_avg.ncprimary_history_fields zeta UBAR
VBAR U V wrtT(1NT)
T F F T T
10Tauxiliary_history_fields rho Omega W
Akv Akt Aks HBL Bostr
F F F F T
F T Fprimary_averages zeta UBAR
VBAR U V wrtT(1NT)
T T T T T
10Tauxiliary_averages rho Omega W Akv Akt
Aks HBL F
T F F T F T
  • vi roms.in

rho0 1025.d0lateral_visc VISC2,
VISC4 m2/sec for all
0. 0.tracer_diff2 TNU2(1NT)
m2/sec for all
100.d0 bottom_drag RDRG m/s, RDRG2,
Zob m, Cdb_min, Cdb_max
3.0d-04 0.d-3 0.d-3
1.d-4 1.d-1 gamma2
1.d0sponge X_SPONGE m, V_SPONGE
m2/sec 150.e3
500.nudg_cof TauT_in, TauT_out,
TauM_in, TauM_out days for all
1. 360. 3.
360.
21
Compiling the model
  • jobcomp
  • Automatic selection of compilation options
    according to the plateform
  • Set library path
  • Use Makefile
  • C-preprocessing file.F ? file.f
  • Compiling file.f ? file.o
  • Links with libraries ? executable roms

22
Preparing AGRIF
  • vi AGRIF_FixedGrids.in

2 20 45 34 59 3 3 3 30 55 70 89 3 3 2 0 1 10 30
20 40 5 3 5 0
1 20 33 34 44 3 3 3
PERU test case
23
Running the model
MAIN started time-steping. STEP timeDAYS
KIN_EN POT_EN TOTAL_EN
NET_VOLUME trd 0 15.00000
0.000000000E00 2.5311945E01 2.5311945E01
2.2335422E15 0 1 15.02083
3.332338732E-06 2.5312150E01 2.5312154E01
2.2335431E15 0 2 15.04167
1.062963402E-05 2.5312316E01 2.5312327E01
2.2335455E15 0 3 15.06250
2.075678260E-05 2.5312446E01 2.5312466E01
2.2335461E15 0 4 15.08333
3.186463589E-05 2.5312543E01 2.5312575E01
2.2335469E15 0 5 15.10417
4.285427484E-05 2.5312627E01 2.5312670E01
2.2335480E15 0 6 15.12500
5.333102059E-05 2.5312691E01 2.5312744E01
2.2335479E15 0 7 15.14583
6.354045596E-05 2.5312719E01 2.5312782E01
2.2335465E15 0 8 15.16667
7.411816854E-05 2.5312701E01 2.5312775E01
2.2335457E15 0 9 15.18750
8.562138804E-05 2.5312650E01 2.5312735E01
2.2335467E15 0 10 15.20833
9.828165268E-05 2.5312569E01 2.5312667E01
2.2335473E15 0 11 15.22917
1.117146701E-04 2.5312465E01 2.5312577E01
2.2335475E15 0 12 15.25000
1.255576462E-04 2.5312345E01 2.5312471E01
2.2335475E15 0 13 15.27083
1.393087941E-04 2.5312135E01 2.5312275E01
2.2335470E15 0 14 15.29167
1.525558114E-04 2.5311800E01 2.5311952E01
2.2335463E15 0 15 15.31250
1.653985076E-04 2.5311350E01 2.5311515E01
2.2335465E15 0 16 15.33333
1.779958127E-04 2.5310792E01 2.5310970E01
2.2335468E15 0 17 15.35417
1.905668926E-04 2.5310134E01 2.5310325E01
2.2335470E15 0 18 15.37500
2.034591092E-04 2.5309385E01 2.5309588E01
2.2335470E15 0 19 15.39583
2.165195050E-04 2.5308554E01 2.5308771E01
2.2335469E15 0 20 15.41667
2.294900067E-04 2.5307653E01 2.5307882E01
2.2335465E15 0 21 15.43750
2.422211112E-04 2.5306695E01 2.5306937E01
2.2335463E15 0 22 15.45833
2.545401621E-04 2.5305693E01 2.5305948E01
2.2335463E15 0 23 15.47917
2.664383353E-04 2.5304665E01 2.5304932E01
2.2335464E15 0 24 15.50000
2.780681955E-04 2.5303629E01 2.5303907E01
2.2335465E15 0 25 15.52083
2.896947197E-04 2.5302600E01 2.5302890E01
2.2335465E15 0 26 15.54167
3.013595537E-04 2.5301595E01 2.5301897E01
2.2335462E15 0
  • roms roms.in

24
Visualizing the results
gtgt roms_gui
25
Analysing the results
  • Make statistics (mean, variance, )
  • Use tracers, compute residence times and
    Lagrangian transport
  • Make budgets (energy, heat, vorticity, momentum,
    )
  • Comparison with available data
Write a Comment
User Comments (0)
About PowerShow.com