Chapter 14 Control parameters
Contents
1. 0 1 weirs barriers module iopt thndam Disables enables thin dam module pcm iopt weibar Disables enables 14 4 12 Time integration iopt cor impl Time integration of the Coriolis term 1 0 explicit 1 semi implicit 2 implicit iopt vadv impl Time integration for vertical advection 1 0 explicit 1 semi implicit 2 implicit iopt vdif impl Time integration for vertical diffusion 2 0 explicit 1 semi implicit 2 implicit 14 4 13 Open boundary conditions iopt obc advflux Type of open boundary condition for the cross stream 2 D and 3 D advective fluxes see Section 5 3 16 2 1 zero gradient condition 2 quasi upwind scheme 546 iopt_obc_advrlx iopt_obc_bio iopt_obc_int iopt_obc_invbar iopt_obc_relax iopt_obc_sal iopt_obc_sed lopt obc temp iopt_obc_2D CHAPTER 14 CONTROL PARAMETERS Disables enables 0 1 the relaxation scheme for horizontal momentum advection see Section 5 3 16 2 0 relaxation scheme disabled default 1 relaxation scheme enabled In that case the parameter distrlx obc representing the parameter dmax must be de fined by the user in usrdef_mod_params or in the CIF General type of open boundary conditions for biological variables 0 Currently not implemented 0 default conditions at all open boundaries 1 non default conditions for at least one open boundary point Disables enables 0 1 momentum advection adjacent to open boundaries 02. Disables enables 0 1 inverse barometric effect at open boun daries 0 Disables enables 0 1 open boundary relaxation as discussed in Section 0 General type of open boundary conditions for salinity 0 0 default conditions at all open boundaries 1 non default conditions for at least one open boundary point General type of open boundary conditions for sediments 0 0 default conditions at all open boundaries 1 non default conditions for at least one open boundary point General type of open boundary conditions for temperature 0 0 default conditions at all open boundaries 1 non default conditions for at least one open boundary point General type of open boundary conditions for the 2 D mode 0 14 4 MODEL SWITCHES 547 0 default conditions at all open boundaries 1 non default conditions for at least one open boundary point iopt_obc_3D General type of open boundary conditions for the 3 D cur rents 0 0 default conditions at all open boundaries 1 non default conditions for at least one open boundary point Note that the open boundary conditions automatically reduce to their de faults see Section 4 10 and input of open boundary data is disabled if the appropriate switch is not set 14 4 14 Tides iopt_astro_anal Disables enables 0 1 the use of astronomical arguments for harmonic analysis if iopt_astro_pars gt 0 and iopt_out_anal 1 0 iopt_astro_pars Enbable
3. Parameter b in the Song amp Haidvogel 1994 vertical grid transformation 0 1 Charnock s constant a used in Charnock s relation 4 291 0 014 Constant surface drag coefficient Cy when iopt_sflux_cds 0 0 0013 0 0013 0 0013 von Karman s constant amp 0 4 Critical water depth deri used in the drying wetting algo rithm m 0 1 Constant water depth used to set up a default bathymetry m 0 0 Data flag marking land points in the bathymetry m 0 0 Maximum distance dar from the open boundaries used in the relaxation factor 5 292 for momentum advection Reference latitude to be used for the Coriolis frequency in the case of a Cartesian grid decimal degrees 0 0 Reference longitude to be used for solar irradiance in the case of a Cartesian grid decimal degrees 0 0 If iopt_astro_pars gt 0 harmonically analysed phases are taken with respect to the astronomical argument for this refer ence longitude at the central time decimal degrees positive East 0 0 If iopt astro pars gt 0 phases at open boundaries are as sumed to be taken with respect to the astronomical argu ment at this reference value decimal degrees If zero the reference longitude is taken at Greenwich 0 0 Parameter d in the Burchard amp Bolding 2002 vertical grid transformation 4 26 1 5 Minimum water depth dmin used in the drying wetting al gorithm m 0 02 Constan
4. amp amp modfiles Only the underlined parameters can be defined by the user the others are used internally in the program e g iunit giving the FORTRAN file unit number An element of the array modfiles can be generically represented as mod files idesc ifil iotype where idesc is the file descriptor ifil the file number and iotype represents input output data if 1 2 The meaning of the third index iotype is as follows Almost all forcing data except nesting are input data i e represented by an element of modfiles with iotype 1 By defining a corresponding output file with iotype 2 one has the possibility to re write the same input data now in a COHERENS standard format This file can be used as input within a subsequent run The user then needs to change only the status atrtribute from N to R see below In case of nested output iotype must take the value of 2 Input data can be spread over multiple files for a given descriptor by specifying different file numbers This is further discussed below The ma ximum value of ifil is given by the system parameter MaxlOFiles defined in syspars f90 14 7 1 File descriptors The following key ids are available as file descriptors io mppmod parallel decomposition ifil 1 io_inicon initial conditions for the physical ifil ics phys and sediment model ifil ics sed 14 7 ATTRIBUTES OF FORCING FILES io modgrd io_metgrd io_sstgrd Io wavgrd io_nstgrd
5. lopt turb tke bcc lopt turb tke sbc 1 constant value 4 198 2 Munk Anderson form 4 199 3 from RANS model as explained in Section 4 4 3 3 Selects level for stability functions if iopt turb stab form 3 1 1 quasi equilibrium method Section 4 4 3 3 2 non equilibrium method Section 4 4 3 3 Selects type of closure RANS model 4 MYS2 model KC94 model BB95 model HR82 model CA01 model CA02 model Formulation for the turbulent diffusion coefficient v or stability coefficient S of turbulent energy 2 1 constant value for S as given by equation 4 200 2 S is taken as proportional to momentum stability function S as given by 4 201 3 using the formulation of Daly amp Harlow 1970 as given by 4 185 or 4 191 depending on the value of iopt turb stab lev Type of bottom boundary condition for turbulence energy 2 1 Neumann condition 4 352 2 Dirichlet condition 4 351 Type of surface boundary condition for turbulence energy 2 1 Neumann condition 4 283 2 Dirichlet condition 4 281 14 4 MODEL SWITCHES 545 14 4 10 Drying wetting scheme iopt fld Selects the type of drying wetting scheme 0 0 Drying wetting disabled 1 Drying wetting algorithm without dynamic masks 2 Drying wetting algorithm using dynamic masks 14 4 11 Structures iopt_dischr Disables enables 0 1 discharge module iopt drycel Disables enables 0 1 dry cell module 0 1
6. LU PCLU 11 Cholesky PCCHOLESKY 12 No preconditioning PCNONE Oo ABN nD oF W N FR Type of solver used by PETSc 5 For details see the PETSc User Manual 1 Richardson KSPRICHARDSON 2 Chebychev KSPCHEBYCHEV 552 CHAPTER 14 CONTROL PARAMETERS 3 Conjugate Gradient KSPCG 4 Biconjugate Gradient KSPBICG 5 Generalised Minimal Residual KSPGMRES 6 BiCGSTAB KSPBCGS 7 Conjugate Gradient Squared KSPCGS 8 Transpose Free Quasi Minimal Residual 1 KSPTFQMR 9 Transpose Free Quasi Minimal Residual 2 KSPTCQMR 10 Conjugate Residual KSPCR 11 Least Squares Method KSPLSQR 12 Shell for no KSP method KSPPREONLY 14 4 21 User output harmonic output 0 time averaged output 0 1 time series output 1 time series output 1 14 4 22 NetCDF iopt CDF abort 0 If an error is detected in a netCDF routine an error mes sage will be written but the program will not abort im mediately 1 If an error is detected in a netCDF routine an error mes sage will be written and the program will abort immedi ately afterwards iopt CDF fill Disables enables 0 1 the use of fill values 0 iopt CDF format Selects the type netCDF file format 1 1 classic format 2 64 bit offset format The different netCDF file formats are discussed in the netCDF User Manual 14 5 Model parameters All parameters in this section are defined in usrdef mod params 14 5 MODEL
7. default value no log file will be written The inilog file only contains information about model initialisation and is closed as soon as the program enters the time loop In parallel mode the size npworld of the vector array equals the number of pro cesses initially defined within MPl COMM WORLD or equals 1 in the serial case INTEGER levprocs run npworld Determines the level of tracing of the runlog file for each process Different levels can be defined for different files If 0 default value no log file will be written The runlog file traces program execution during the time loop CHARACTER LEN leniofile inilog file Name of the inilog file Default is TRIM runtitle inilogA In parallel mode the name is appended with the process id number CHARACTER LEN leniofile runlog file Name of the runlog file Default is TRIM runtitle runlogA In parallel mode the name is ap pended with the process id number LOGICAL exitlog Writes an exit statement of the form num R where num is the program level in the log file on exit of a routine call if TRUE TRUE INTEGER runlog count Sets the number of time steps after which the runlog file is overwritten Default is the total number of time steps i e information is written at all time steps and the file is never over written 14 2 3 Error files INTEGER maxerrors Maximum allowed number of error messages within the errlog file Default is MaxEr
8. lambda ad parameter A in equation 4 146 0 0 m omegal ad parameter w in equation 4 151 1074 s71 riccrit iw critical Richardson number Rig in the 1994 back ground mixing scheme 4 227 0 7 r1 ad parameter r in equation 4 143 1 0 r2 ad parameter r9 in equation 4 143 1 0 sigma k parameter c used to define S in 4 201 1 0 skeps neutral value 5 6 of the stability coefficient S in the k e model see equation 4 200 0 09 sq my parameter S used to determine Spo in the Mellor Yamada model see equation 4 202 0 2 tkelim background limit kum for k see equation 4 226 107 J kg tkemin numerical lower limit kmin for k 10 4 J kg vbmom pp 5 mE T the Pacanowski amp Philander 1981 scheme oan vbscal pp me 7 in the Pacanowski amp Philander 1981 scheme TEE EE vdifmom iw em wave me mos coefficient vro for PLA in the Large et al Leere unt 4 background mixing scheme 4 227 1074 m s vdifscal_iw internal wave aT 5H diffusion coefficient Aro for MO in the Large et al Large et al 1994 background mixing scheme 4 227 5x10 m s vdifshear_iw maximum mixing due to unresolved vertical shear vj in the Large et al 1994 background mixing scheme 4 227 0 005 m s vmaxmom ma parameter Vmax in the Munk amp Anderson 1948 scheme 4 136 4 139 3 0 vmaxscalma parameter Amaz in the Munk amp Anderson 1948 scheme 4 136 4 139 4 0
9. rotated grids Remarks e delxdat and delydat are only needed for uniform rectangular grids e The meaning of the reference location for a rectangular grid is explained in Section 4 1 2 e n case of a fully curvilinear grid the model grid is defined in routine usrdef grid The only attributes which may be supplied are the coordi nates of the reference location used for avoiding rounding errors For details see Section 15 1 e Grid rotation is only allowed for rectangular grids 14 7 Attributes of forcing files Model forcing requires the definition of pamameters and input data They can be directly defined by the user in one of the usrdef routines below or by reading them for some external file Before these data can be obtained a series of file attributes needs to be set by the user to inform the program 564 CHAPTER 14 CONTROL PARAMETERS which parameters data are needed and how they are accessed These at tributes are stored in the 3 D array modfiles of DERIVED TYPE FileParams defined by TYPE FileParams LOGICAL defined info opened time regular CHARACTER LEN 1 form status CHARACTER LEN leniofile filename pathname CHARACTER LEN lendesc filedesc INTEGER endfile header type iostat iunit lenrec amp amp maxrecs nocoords nodim novars timeid amp amp timerec tskips varid zetaid INTEGER DIMENSION 3 tlims END TYPE FileParams TYPE FileParams DIMENSION MaxIOTypes MaxIOFiles 2
10. D time steps int fill ntobcrlx The relaxation period 7 divided by the 2 D time step delt2d optionally used to define the relaxation factor o t defined by 4 356 for the 2 D mode at open boundaries For details see Section 0 idmaster Process id of the master process 0 Must be between 0 and nprocs 1 index obc 1 nconobc Key ids of the tidal constituents used for the tidal for cing at open boundaries index astro 1 nconastro Key ids of the tidal constituents for the astronomical tidal forcing e nowaitsecs and maxwaitsecs are used in connection to the endfile at tribute discussed in Section 14 7 2 e norestarts must not exceed the value of the system parameter MaxRestarts defined in syspars f90 e Key ids for tidal constituents are defined in tide f90 14 5 4 Physical model parameters The defaults of parameters marked with a can be generally applied and should in priciple not be changed atmpres ref Reference atmospheric pressure P Pa 101325 0 bdragcoef cst Constant bottom drag coefficient C when iopt_bstres_drag 1 0 0 556 bdraglin B SH ccharno cds cst ces cst chs_cst ckar dcrit_fld depmean cst depmean flag distrlx obc dlat ref dlon ref dlon ref anal dlon ref obc dl BB dmin fld CHAPTER 14 CONTROL PARAMETERS Bottom friction velocity kun used in the linear bottom fric tion law if iopt bstres form 1 m s 0 0
11. Other integer model parameters fld mask nofldmasks Enables 1 or disables 0 a specific mask criterium for the drying flooding algorithm as given by equations 5 372 5 383 Default values are fld mask 1 1 fld mask 2 0 The number of available criteria given by nofldmask equals 11 in the current implementation and cannot be changed by the user maxitsimp Largest allowed iteration number for the outer loop in the free surface corrector method implicit scheme 1 nconastro number of constituents for the astronomical tidal forcing used when iopt_astro_tide 1 0 nconobc number of constituents for the open boundary tidal forcing if iopt grid nodim 1 or for the surface forcing if iopt grid nodim 1 0 norlxzones number of relaxation zones used when iopt obc relax 1 0 nonestsets number of nested sub grids used when iopt nests 1 0 numdis number of discharge locations 14 5 MODEL PARAMETERS 955 numdry number of dry cells numthinu number of thin dams at U nodes numthinv number of thin dams at V nodes numwbaru number of weirs barriers at U nodes numwbarv number of weirs barriers at V nodes nowaitsecs number of seconds to wait between two read attempts s 0 maxwaitsecs maximum allowed time spent in wait calls s 3600 norestarts number of restart times 1 ntrestart 1 norestarts Restart time indices for writing of initial conditions If a value equals int fill it will be replaced by the total number of 2
12. PARAMETERS 14 4 4 Density lopt dens lopt dens grad iopt sal iopt sal sbc lopt temp Evaluation of the density and expansion coefficients 0 0 uniform density zero expansion coefficients 1 density from the linear equation of state 4 108 expansion coefficients are uniform 2 from the McDougall et al 2003 general equation of state 4 103 4 107 without pressure effects 3 from the McDougall et al 2003 general equation of state 4 103 4 107 with pressure effects included Selects numerical algorithm for discretisation of the baroclinic pressure gradient 1 0 gradient set to zero 1 traditional o coordinate second order method 2 z level method 3 method of Blchcperkin amp MeWiliams Salinity update 0 0 uniform space and time salinity field 1 salinity field initialised but not updated in time 2 salinity field initialised and updated in time Type of surface boundary condition for salinity 0 0 zero surface flux 1 surface flux given by Temperature update 0 0 uniform space and time temperature field 1 temperature field initialised but not updated in time 2 temperature field initialised and updated in time iopt temp optic Disables enables 0 1 the optical module 1 0 all solar radiation is assumed to be absorbed at the surface i e the water column is considered as opaque 1 solar radiation is absorbed within the water column using specified values f
13. PARAMETERS 553 14 5 1 Date and time parameters CStartDateTime Start date in string format yyyy mm dd hh mm ss mmm of 23 characters If the last 4 characters are omitted they are set to 000 by default CEndDateTime End date in string format If the last 4 characters are omit ted they are set to 000 by default delt2d Barotropic 2 D time step mode splitting scheme or time step for all 2 D 3 D transport equations implicit scheme ls 7 ic3d number of 2 D time steps within one 3 D time step 1 If iopt hydro impl 1 or iopt grid nodim 1 or 2 ic3d is always 1 icnodal Time step measured in units of delt2d for an update of the nodal tidal factors and astronomical arguments if iopt astro pars 70 If zero nodal corrections amplitudes and phases are evaluated at the initial time only 0 time zone Time zone i e the difference of the local time with respect to GMT hours Difference is positive negative eastwards westwards from Greenwich 0 Remarks e If the 2 D time step is lower than 1000 seconds its precision is 1 millisecond and decimal numbers from the fourth position after the decimal point will be discarded If the time step is larger than 1000 seconds its precision is 1 second and its decimal part is ignored If iopt hydro impl 0 the 2 D time step is limited by the CFL condition for surface gravity waves The maximum allowed 2 D time step is written to the log file e T
14. W myciffile In the example the first line initiates the run conesA without CIF the second one reads the setup from the file conesA cifmodA the third writes the CIF data to the file myciffile Lines can be commented if the first character is a This replaces for compatibility with the CIF syntax below the character used in previous versions The procedure is used to combine multiple simulations within one run 1 The program opens the file at the start 2 The first line is read 3 A simulation is started with the given title 4 When the simulation ends a next line is read giving a new title and a next simulation initiates 5 When there are no more lines to be read the file is closed and the program terminates 14 2 Parameters for monitoring This section describes the parameters used to set up the monitoring and a few other general parameters They are defined in usrdef init params The routine is called by all processes 14 2 1 Cold start LOGICAL cold start If TRUE the program executes model initialisation and finalisation but does not enter the time loop FALSE If defruns contains multiple lines a cold start is per formed for each simulation The option is useful for debugging 14 2 PARAMETERS FOR MONITORING 533 14 2 2 Log files INTEGER levprocs ini npworld Determines the level of tracing of the inilog file for each process Different levels can be defined for different files If 0
15. in the Pacanowski amp Philander scheme ETA 5 0 ns Bm in thelMunk amp Anderson 1948 scheme 4 136 4 139 3 33 attenuation factor 6 in the ping Davies Xing amp Davies mixing length formulation 4 215 2 0 parameter C in equation n 2 0 constant Cie in the shear production term of the equation 4 205 1 44 constant co in the dissipation term of the e equation 4 205 1 92 constant c3 in the buoyancy sink term of the e equation 4 205 in case of stable stratification IN gt 0 0 2 constant ca in the buoyancy source term of the amp equation 4 205 in case of unstable stratification N lt 0 1 0 Daly Harlow parameter c in 4 177 0 15 parameter 6 in equation 4 143 0 0 parameter in equation 4 143 0 0 numerical lower limit for 10 W kg parameter n in the Munk amp Anderson 1948 scheme 4 136 4 139 0 5 H2 ny in the Pacanowski amp Philander scheme 2139 2 0 ee n in the Munk Munk amp Anderson Anderson 1948 scheme 4 136 4 139 1 5 constant E in the shear production term of the kl equation 4 209 1 8 constant Ey in the wall proximity term 4 210 of the kl equation 4 209 1 33 constant Es in the buocancy source sink term of the kl equation 4 209 1 0 560 CHAPTER 14 CONTROL PARAMETERS k1 ad parameter A in equations 4 148 and 4 150 0 0025 k2 ad parameter K in equation 4 149 2x 10
16. vmax_pp Ew Vmax in the Pacanowski amp Philander 1981 scheme 134 3 0 vOdif_ma o Vom in the Munk Munk amp Andersoni Anderson 1948 scheme 4 136 4 139 0 06 m s 14 6 PARAMETERS FOR SURFACE DATA GRIDS 561 vOdif_pp parameter Vop in the Pacanowski amp Philander 1981 scheme 4 132 4 134 0 01 m 2 e aor wfltke surface wave factor c used in the surface flux condition 4 283 for turbulent energy 0 0 zlmixmin numerical lower limit min for 1 7x10 9 m zrough bot bottom roughness length zo in the mixing length formulation 4 212 0 0 m zrough sur surface roughness length zo in the mixing length formulation 1 212 0 0 m 14 6 Parameters for surface data grids Surface data grids are external grids where e g meteorological data are defined for the surface forcing The parameters characterising a surface grid are stored into the 2 D array surfacegrids of DERIVED TYPE GridParams defined by TYPE GridParams LOGICAL rotated INTEGER nhtype nidat n2dat REAL delxdat delydat gridangle xOdat yOdat yOrot END TYPE GridParams TYPE GridParams DIMENSION MaxGridTypes 2 surfacegrids An element of the array surfacegrids can be generically represented as sur facegrids igrd ifil where igrd is a key id called the grid descriptor and ifil the file number The file number can take the value of 1 for external data intended for input and 2 for data written by the m
17. Chapter 14 Control parameters The parameters discussed in this chapter except the first section are de fined in the following routines of the file Usrdef Model f90 e usrdef init params setup of monitoring parameters Section 14 2 e usrdef mod params switches model parameters and attributes of for cing files Sections 14 3414 8 e usrdef MPI partition user defined domain decomposition Section 14 9 14 1 File defruns The program open this file at the start of the simulation s and is read line wise Each line represents a separate run and contains the definitions of three parameters defined separated by a The general systax is runtitle status filename where runtitle status filename the title of the simulation stored in the model parameter runtitle the status of the CIF 0 The CIF utility is switched off both for reading and writing This is the default condition R Model setup parameters are read from a CIF W Model setup parameters are written to a CIF Name of the CIF file If not given the default name TRIM runtitle cifmodA is taken This parameter is obviously not used if status equals 0 531 532 CHAPTER 14 CONTROL PARAMETERS Defaults are taken except for runtitle which must always be given when the value is an empty string one blank or several blanks All blanks are ignored on the input line Consider the following example conesA conesA R conesA
18. defined in usrdef partition This routine is called in parallel mode by reader processes if iopt MPl_partit 2 and mod files io_mppmod 1 1 status N nclprocs nprocs global X index of lower upper left cell of the process do mains nc2procs nprocs global X index of lower upper right cell of the process do mains nrlprocs nprocs global Y index of lower upper left cell of the process do mains nr2procs nprocs global Y index of lower upper right cell of the process do mains 570 CHAPTER 14 CONTROL PARAMETERS
19. defined uniform value kinvisc cst 1 TTC 1978 relation 7 24 iopt vdif coef Selects the general type of vertical diffusion scheme 3 0 vertical diffusion disabled 1 uniform diffusion coefficient 2 algebraic formulation as described in Section 4 4 2 2 3 second order turbulence closure as described in Section 4 4 3 Remarks e If horizontal diffusion is enabled the Smagorinsky formulation taken from LES modelling is a more robust scheme compared to a constant diffusion coefficient e Horizontal diffusion of scalars may be potentially dangerous since it introduces spurious diapycnal mixing e Horizontal diffusion of turbulence variables is only introduced for his torical reasons and compatibility with COHERENS V1 but has no real physical basis 14 4 9 Turbulence schemes iopt_turb_alg Type of algebraic scheme if iopt_vdif_coef 2 1 1 Pacanowski Philander formulation 4 132 4 135 2 Munk Anderson formulation 4 136 4 140 3 flow dependent formulation as described in Section 4 4 2 2 with a given by 4 148 4 flow dependent formulation as described in Section 4 4 2 2 with a given by 4 149 5 flow dependent formulation as described in Section 4 4 2 2 with a given by 4 150 6 parabolic profile 4 154 14 4 MODEL SWITCHES 543 lopt turb dis bbc lopt turb dis sbc iopt turb iwlim iopt_turb_kinvisc iopt_turb_Imix iopt_turb_ntrans iopt_turb_param Type of bottom boundary condi
20. e as given by the parameter ces_cst or chs_cst see below 1 equation from Large amp Pond 2 equation from Anderson amp Smith 3 4 equation 4 295 from Wu 1980 equation 4 294 from Kondo 1975 iopt_sflux_strat Selects dependence of surface drag and exchange coefficients on atmospheric stratification effects 0 0 1 2 no dependence using the 1975 parameterisation Section 4 8 2 using Monin Obukhov similarity theory Section 4 8 3 14 4 18 Nesting iopt_nests Disables enables 0 1 the writing of open boundary data for nested sub grids 0 550 CHAPTER 14 CONTROL PARAMETERS 14 4 19 MPI mode iopt_MPI_abort 0 If an error is detected in a MPI routine an error message will be written but the program will not abort immediately If an error is detected in a MPI routine an error message will be written and the program will abort immediately afterwards iopt MPl comm all Communication type for all to all operations 2 1 2 3 4 blocking standard send blocking synchronous send non blocking standard send non blocking synchronous send iopt MPI comm coll Disables enables 0 1 the use of MPI collective calls 0 iopt MPl comm exch Communication type for exchange operations 2 1 blocking standard send 2 blocking synchronous send 3 4 5 non blocking standard send non blocking synchronous send send receive blocking calls io
21. er report 1 1 seconds 2 minutes 3 hours 4 days 14 3 Dimensions of the process domain grid The parameters below are used to setup a domain decomposition and are de fined in usrdef_mod_params The routine is called if ciffiles icif_model Yostatus 0 or W 14 4 MODEL SWITCHES 535 nprocs the actual number of processes to be used 1 nprocsx X dimension of the decomposed domain 0 nprocsy Y dimension of the decomposed domain 0 e nprocsx and nprocsy are needed by the program for making a simple domain decomposition when the switch iopt MPI partit 1 Otherwise if iopt_MPI_partit 2 these parameters are determined by the program e nprocs must be defined if the decomposition is obtained from a data file or defined in usrdef partition In that case its value must match the size of the arrays nclprocs nc2procs nrlprocs nr2procs In case of a simple decomposition each but not all of these three parameters may be zero However their values must be between 0 and npworld which is the number of processes in the MPI communicator MPl comm world or equivalently the number of processes defined in the script launching the program The program follows the following procedures 1 nprocsx and nprocsy are non zero nprocs is set to nprocsx x nprocsy 2 both nprocsx and nprocsy are zero both values are set internally so that nprocsxxnprocsy nprocs and nprocsx nprocsy is minimal 3 nprocsx is non zero whi
22. he parameter time zone is of type REAL and must be between 12 0 and 12 0 and is used to reset the start and end dates to GMT where necessary A time zone must be given for the calculation of solar radiance and the astronomical Greenwich argument at the local time when the start and end dates are not expressed in GMT e It is clear that ic3d only needs to be defined for 3 D applications iopt grid nodim 3 Note that the 3 D time step is limited by the constraints 5 5 5 6 554 CHAPTER 14 CONTROL PARAMETERS 14 5 2 Grid parameters nc number of grid cells in the X direction including an extra column along the eastern edge nr number of grid cells in the Y direction including an extra column along the northern edge nz number of grid cells in the vertical direction nosbu number of open sea boundaries at West East U nodes 0 nosbv number of open sea boundaries at South North V nodes 0 nrvbu number of river boundaries at West East U nodes 0 nrvbv number of river boundaries at South North V nodes 0 e nc and nr must be positive and are automatically re set to 3 for water column applications iopt grid nodim 1 e nz must be positive and is automatically re set to 1 for 2 D applications iopt grid nodim 2 e The last row and the last column of the computational domain rep resent dummy land points The phyical horizontal dimension of the domain is therefore nc 1 x nr 1 14 5 3
23. io_sedspc io_luvsur io 2uvobc io 3uvobc io salobc io tmpobc io sedobc io rlxobc lo nstspc lo 2uvnst io 3uvnst io salnst io_tmpnst io_sednst io_metsur io_sstsur io wavsur io drycel io thndam 565 model grid ifil 1 surface meteorological grid ifil 1 sea surface temperature grid ifil 1 surface waves grid ifil 1 nested sub grids one file per sub grid specific arrays for the sediment module see Section 19 1 2 3 specifiers for 1 D surface forcing if ifil 1 forcing data if ifil 2 specifiers for 2 D mode open boundary forcing if ifil 1 open boundary data if ifil gt 1 specifiers for 3 D mode baroclinic currents open boundary for cing if ifil 1 open boundary data if ifil gt 1 specifiers for salinity open boundary forcing if ifil 1 open boun dary data if ifil gt 1 specifiers for temperature open boundary forcing if ifil 1 open boundary data if ifil gt 1 specifiers for sediment open boundary forcing if ifil 1 open boun dary data if ifil gt 1 definitions of relaxation zones ifil 1 specifiers for sub grid nesting ifil 1 2 D open boundary data for nested sub grids one file per sub grid 3 D baroclinic current open boundary data for nested sub grids one file per sub grid salinity open boundary data for nested sub grids one file per sub grid temperature open boundary data for nested sub grids one file per sub grid sediment open boundary data for
24. le is open and file pointer is located at the start or before the end of the file 2 file pointer is located at the end of the file i e an EOF condition will occur on a next read 3 an end of file condition did occur 14 8 Parameters for user defined output A few general parameters need to be specified in usrdef mod params for user defined output They need to be defined in usrdef mod params All other specifiers for user defined output are to be defined in other Usrdef files For more details about the meaning of the parameters below see Section nosetstsr number of time series file sets if iopt out tsers 1 0 nostatstsr number of time series output stations if iopt out tsers 1 0 novarstsr number of time series variables if iopt out tsers 1 0 nosetsavr number of time averaged file sets if iopt out avrgd 1 0 nostatsavr number of time averaged output stations if iopt out avrgd 1 0 novarsavr number of time averaged variables if iopt_out_avrgd 1 0 nosetsanal number of harmonic file sets if iopt out anal 1 0 nofreqsanal number of harmonic frequencies if iopt out anal 1 0 nostatsanal number of harmonic output stations if iopt out anal 1 0 novarsanal number of harmonic variables if iopt out anal 1 0 intitle title used to create names of model forcing files outtitle title used to create names of user output files 14 9 DOMAIN DECOMPOSITION 569 14 9 Domain decomposition The domain decomposition is
25. le nprocsy is zero nprocsy nprocs nprocsx 4 nprocsy is non zero while nprocsx is zero nprocsx nprocs nprocsy Remarks e Cases 2 4 If nprocs is zero its value is set to npworld e Case 3 4 If no integer division is possible an error is issued 14 4 Model switches A total of 83 switches is implemented They are defined in usrdef mod params 536 CHAPTER 14 CONTROL PARAMETERS 14 4 1 Model grid lopt grid htype lopt grid nodim iopt_grid_sph iopt_grid_vtype Type of horizontal grid 1 1 uniform rectangular grid 2 non uniform rectangular grid 3 curvilinear grid Grid dimension 3 1 1 dimensional grid water column model 2 2 dimensional grid depth averaged model without vertical structure 3 3 dimensional grid Type of coordinates 0 0 Cartesian coordinates 1 spherical coordinates Type of vertical grid 1 1 uniform o grid 2 horizontally uniform and vertically non uniform grid 3 horizontally and vertically non uniform o grid iopt grid vtype transf Type of vertical grid transformation 0 0 uniform vertical grid iopt grid vtype 1 or user defined 11 log transformation 4 23 at the bottom following 1991 if iopt grid vtype 2 12 log transformation 4 24 at the surface following 1991 if iopt grid vtype 2 13 transformation with enhanced resolution near the bottom and or the bottom as defined in amp Bolding 2002 21 Song amp Haidv
26. m 0 067 Long wave fraction R of surface solar radiance as used in 4 59 0 54 Relative tolerance used by PETSc for solving the linear sys tem The parameters atol dtol maxits used by PETSc in the solution procedure are set to the PETSc defaults 1077 Mean radius of the Earth R m 6371000 0 Air mass density pq kg m 1 2 Reference salinity 5 used if iopt sal 0 or in the linear equation of state 4 108 or as default initial condition PSU 33 0 Parameter o in the 1991 vertical grid transformations 4 23 and 4 24 0 0 Parameter og in the 1991 vertical grid transformations 4 23 and 4 24 0 1 558 CHAPTER 14 CONTROL PARAMETERS smag coef mom Smagorinsky coefficient Cm for horizontal diffusion of mo smag coef scal specheat temp min temp ref theta cor theta SH theta vadv theta vdif vdifmom cst vdifscal cst wbarrlxu wbarrlxv zbzozOlim zrough cst mentum 0 1 Smagorinsky coefficient C for horizontal diffusion of scalars 0 1 Specific heat of seawater c at constant pressure J kg degC 3987 5 Minimum temperature If set to real fill the minimum is taken as the freezing point of sea water see equation 4 49 which is a function of salinity deg C 0 0 Reference temperature Tef used if iopt temp 0 or in the linear equation of state 4 108 or as default initial condi tion deg C 12 0 Implicity factor 0 for the Co
27. me for the advection of scalar quantities 3 iopt_adv_tvd iopt_adv_turb iopt_adv_2D iopt_adv_3D 0 1 2 3 advection disabled upwind scheme Lax Wendroff explicit in the horizontal central semi implicit in the vertical TVD scheme Type of limiting function for TVD scheme 1 1 2 superbee limiter monotone limiter Type of scheme for the advection of turbulence quantities 0 0 1 2 3 advection disabled upwind scheme Lax Wendroff explicit in the horizontal central semi implicit in the vertical TVD scheme Type of scheme for the advection of 2 D transports 1 3 advection disabled upwind scheme Lax Wendroff explicit in the horizontal central semi implicit in the vertical TVD scheme Type of scheme for the advection of 3 D currents 1 3 advection disabled upwind scheme Lax Wendroff explicit in the horizontal central semi implicit in the vertical TVD scheme iopt scal depos Discretisation for the deposition vertical advective flux at 0 the sea bed of particulate matter 1 Deposition flux is set to zero 14 4 MODEL SWITCHES 541 1 first order upwind scheme 2 second order scheme using extrapolation Remarks e The Lax Wendroff central scheme is non monotone and should not be selected This is illustrated with the cases cones and front see Sec tions and 23 2 e The TVD scheme has the ability to retain
28. nested sub grids one file per sub grid meteorological data ifil 1 SST data ifil 1 wave data ifil 1 dry cell locations thin dam locations 566 io weibar Io disspc io disloc io disvol Io discur io dissal io distmp 14 7 2 status form filename tlims info endfile CHAPTER 14 CONTROL PARAMETERS weirs barriers locations and parameters discharge specifiers discharge locations volume discharges momentum discharges salinity discharge temperature discharges File parameters for input forcing iotype 1 Status of the data file 0 0 zero not defined N user defined R COHERENS standard file File format A ASCII portable sequential U unformatted binary non portable sequential N netCDF format portable non sequential File name including file path if needed Start end step time indices i e times measured in units of delt2d These parameters are not directly used for reading the data but to make updates after tlims 3 x delt2d seconds If tlims 3 gt 0 time interpolation will be performed see below An info file with all header information will be created if TRUE FALSE Switch to decide what action needs to be taken when an end of file conditions occurs 0 0 The program aborts with an error message 1 The program continues no further attempt will be made to read data 2 The program continues a next attempt t
29. o read the data will be made after nowaitsecs seconds e Important to note that the status attribute equals 0 by default which means that the corresponding usrdef routine is not called by the pro gram 14 7 14 7 status form ATTRIBUTES OF FORCING FILES 567 The meaning of tlims is illustrated as follows for the case of meteorolog ical forcing data These data are used to evaluate the surface fluxes of momentum heat and salinity and for the atmospheric pressure gradient in the momentum equations All these quantities will be updated from time tlims 1 upto time tlims 2 at time intervals given by tlims 3 The data are read into the program with a date time stamp which is saved If tlims 3 gt 0 which is usally shorter than the time interval between two input dates the meteo data are first linearly interpolated in time between their values at the most recent date earlier than the current program time and the earliest date later than the current time Since these dates are stored in memory the program knows automatically when new data need to be read If tlims 3 lt 0 the method is the same but without time interpolation i e the data at the current program time are set to their values at the most recent date earlier than or equal to the program time Although it is not absolutely necessary it is rec ommended that tlims 3 is smaller than the time interval between two consecutive inputs Note that if an element of the vecto
30. odel to the external grid The latter is intended for future applications and currently not implemented All parameters of this section are defined in usrfdef mod params 14 6 1 Grid descriptors The grid descriptor may take in the current version the following values igrd model model grid igrd meteo meteorological external grid igrd sst sea surface temperature external grid igrd waves surface wave external grid 562 CHAPTER 14 CONTROL PARAMETERS Identifying the model grid as an external grid seems rather strange at first sight The intention is to provide the possibility to define a uniform rectan gular grid with the parameters below 14 6 2 Grid parameters In the case of an external meteorological surface temperature surface wave grid the following attributes must or may be defined nhtype Type of the surface data grid 0 1 2 3 4 single grid point uniform rectangular grid non uniform rectangular grid non rectangular curvilinear or non structured the same as the model grid nldat X dimension of the surface grid n2dat Y dimension of the surface grid delxdat grid spacing in the X direction meters or degrees longitude when nhtype 1 delydat grid spacing in the Y direction meters or degrees latitude when nhtype 1 xOdat X coordinate meters or degrees longitude of the lower left corner when nhtype 1 yOdat Y coordinate meters or degrees latitude of the lower left corner when nhty
31. ogel 1994 transformation given by 4 33 and 4 35 if iopt grid vtype 3 14 4 MODEL SWITCHES 537 14 4 2 Interpolation iopt_arrint_hreg Disables enables 0 1 the use of non uniform weighted aver ages for interpolation in the horizontal of arrays on the model grid 0 iopt_arrint_vreg Disables enables 0 1 the use of non uniform weighted av erages for interpolation in the vertical of arrays on the model grid 0 iopt arrint 3D Selects dimension of mask or weight factor in some array interpolations 0 2 D masks or weights 1 3 D masks or weights It is recommended to set the first two of these switches only for grids with highly irregular grid spacings 14 4 3 Hydrodynamics iopt_curr Type of current fields 2 0 Currents and elevations are set to their default zero va lues and are not updated 1 Currents and elevations are initialised but not updated in time 2 Currents and elevations are initialised and updated in time iopt_curr_wfall Type of formulation for the settling of particulate matter 1 1 settling enabled without correction terms 2 settling enabled with the correction terms 7 117 7 118 included iopt hydro impl Disables enables the implicit scheme 0 01 The momentum equations are solved with the explicit mode splitting scheme default 11 The momentum equations are solved using the implicit algorithm The compiler option DPETSC must be set 538 CHAPTER 14 CONTROL
32. or the attenuation depths 14 4 MODEL SWITCHES 539 iopt temp sbc Type of surface boundary condition for temperature 1 1 Neumann condition using the model s surface heat flux for mulations Dirichlet using prescribed surface temperatures taken at the first grid point below the surface Dirichlet using prescribed surface temperature taken at the surface itself 14 4 5 External modules iopt_biolgy Disables enables 0 1 the activation of an external biological module not available 0 Since no biological module is imple mented in the current version this default cannot be changed iopt sed Disables enables 0 1 the activation of an external sediment module 0 14 4 6 Bottom boundary conditions lopt bstres drag lopt bstres form lopt bstres nodim Formulation for the bottom drag coefficient C 3 not used spatially uniform value spatially non uniform obtained from a data file using a spatially uniform roughness length Ae N e c using a spatially non uniform roughness length Type of formulation for the bottom stress 2 0 bottom stress set to zero 1 linear bottom stress law 4 338 or 4 339 2 quadratic bottom stress 4 340 or 4 339 Type of currents in the linear or quadratic bottom stress formulation 3 2 depth mean currents 3 3 D current taken at the bottom grid cell 540 CHAPTER 14 CONTROL PARAMETERS 14 4 7 Advection iopt_adv_scal Type of sche
33. pe 1 Remarks e If nhtype 1 all parameters need to be defined e If nhtype 2 3 only nldat and n2dat need to be defined e If nhtype 4 then nldat nc and n2dat nr and no further definitions need to be made e The corner coordinates xOdat yOdat and the grid spacings delxdat delydat are given in meters or degrees longitude and latitude depending on whether iopt grid sph equals 0 or 1 14 7 ATTRIBUTES OF FORCING FILES 563 In the case of a model grid grid descriptor igrd model the attribute nhtype equals the value of iopt grid htype between 1 and 3 and nldat n2dat are given by the previously defined grid sizes nc nr delxdat grid spacing in the X direction meters or degrees longitude when iopt_grid_htype 1 In the case of a rotated grid the spacing is given in transformed coordinates delydat grid spacing in the Y direction meters or degrees latitude when lopt grid htype 1 In the case of a rotated grid the spacing is given in transformed coordinates xOdat X coordinate meters or degrees longitude of the reference location Ly or A yOdat Y coordinate meters or degrees longitude of the reference location Yr Or Oy rotated must be set to TRUE in case of a rotated grid Default is FALSE gridangle grid rotation angle a see Section 4 1 3 decimal degrees Must be between 0 and 180 yOrot transformed latitude of the reference location in case of a rotated grid decimal degrees Only used for spherical
34. pt MPl comm gath Communication type for all to one gather combine operations 2 1 2 3 4 blocking standard send blocking synchronous send non blocking standard send non blocking synchronous send iopt MPl comm scat Communication type for one to all scatter distribute and copy operations 2 1 blocking standard send 2 blocking synchronous send 3 4 non blocking synchronous send non blocking standard send 14 4 MODEL SWITCHES 551 iopt MPI partit iopt MPI sync Remarks Selects the method for domain decomposition 1 1 simple partition based on the values of nprocsx and nprocsy 2 decomposition obtained from an external data file or defined in usrdef partition Disables enables 0 1 synchronisation calls at the end of a series of blocking or non blocking operations 0 e The non blocking options are not yet tested and should not be used in the current version of COHERENS e Synchronisation of communication calls may lower the CPU perfor mance 14 4 20 PETSc iopt petsc precond Type of preconditioner used by PETSc 5 For details lopt petsc solver see the PETSc User Manual Jacobi PCJACOBI Block Jacobi PCBJACOBI SOR and SSOR PCSOR SOR with Eisenstat trick PCEISENSTAT Incomplete Cholesky PCICC Incomplete LU PCILU Additive Schwarz PCASM Linear solver PCKSP Combination of preconditioners PCCOMPOSITE 10
35. r tlims is set to the undefined value int_fill this value will be automatically replaced by the total number of 2 D time steps in the simulation nstep which means that the corresponding time is set to the end date of the run If endfile equals 2 and an end of file condition occurs during a read the program waits for nowaitsecs seconds before make a next attempt The total waiting time is given by maxwaitsecs after which the program aborts with an error message The procedure is intended for making simulations in interactive mode For example assume that a main grid writes the open boundary data for a nested sub grid If the main and sub grid are launched together and the former runs slower than the latter the nested grid will wait for input from the main grid 3 File parameters for output forcing iotype 2 Status of the data file 0 0 not defined W a COHERENS standard file will be created File format A ASCII sequential U unformatted binary machine dependent sequential N netCDF format portable non sequential 568 CHAPTER 14 CONTROL PARAMETERS filename File name including file path if needed 14 7 4 Other forcing attributes Other relevant parameter components not defined in usrdef mod params but used internally are iunit File unit This parameter is set internally and cannot be defined by the user iostat File I O status 1 open error occurred 0 file not opened 1 fi
36. rMsgs defined in syspars f90 INTEGER levprocs err npworld Level of error checking for each process 0 0 error checking disabled for a particular processor and no file is created error checking enabled during initialisation phase only 2 error checking enabled throughout the whole program 534 CHAPTER 14 CONTROL PARAMETERS CHARACTER LEN leniofile errlog file Name of the errlog file Default is TRIM runtitle errlogA In parallel mode the name is appended with the process id number 14 2 4 Warning file LOGICAL warning Disables enables writing of a warning file TRUE CHARACTER LEN leniofile warlog file Name of the warning file Default is TRIM runtitle warlogA 14 2 5 Timer file INTEGER levtimer Determines the type of information in the timer report 0 0 No timer report is written 1 Writes the total execution time only 2 Writes time information in of total time for all timers In case of a parallel run the information is written as follows time on the master process mean minimum and maximum time over all processes 3 The same as previous but in case of a parallel run the in formation is additionally written for each individual process In the serial case behaviour is as for case 2 CHARACTER LEN leniofile timing file Name of the timing file Default is TRIM runtitle timingA INTEGER timer_format Format for total execution time in the tim
37. recipitation rate Eja Pre 2 precipitation rate P Selects type of input data for the barotropic mode i e sur face stress and pressure 0 0 no input 1 components of wind speed U10 V10 and unless iopt grid nodim 1 atmospheric pressure P 2 components of surface stress 77 77 and unless iopt grid nodim 1 atmospheric pressure P Type of wave input wave input 0 0 wave input disabled default 1 wave height period and wave direction 2 wave height period velocity excursion and direction e Note that all meteorological surface forcing is disabled if iopt meteo 0 This means that all surface fluxes are automatically set to zero and the input of any meteorological data is disabled 14 4 MODEL SWITCHES 549 e A wave current interaction module is currently not yet implemented Wave input is only used for the sediment module 14 4 17 Surface boundary conditions iopt_sflux_cds Formulation for the neutral surface drag coefficient Cas 0 0 constant value as given by the parameter cds cst see be low equation 4 286 from Large amp Pond 1981 equation 4 287 from Smith amp Bankel 1975 equation 4 288 from Geernaert et al 1986 ib 2 3 4 5 6 equation 4 289 from Kondo 1975 equation 4 290 from Wu 1980 equation 4 291 from 1955 iopt_sflux_cehs Formulation for the neutral surface heat exchange coeffi cients Ce Ch 0 0 constant valu
38. riolis term between 0 0 and 1 0 0 5 Parameter 0 in the Song amp Haidvogel 1994 vertical grid transformation 8 0 Implicity factor 0 for vertical advection between 0 0 and 1 0 0 501 Implicity factor 04 for vertical diffusion between 0 0 and 1 0 1 0 Constant coefficient for vertical diffusion of momentum used if iopt_vdif_coef 1 or as background value if iopt_turb_iwlim 0 m s 1075 Constant coefficient for vertical diffusion of scalars used if iopt_vdif_coef 1 or as background value if iopt_turb_iwlim 0 m s 1079 Time relaxation coefficient at U node weirs barriers 1 0 Time relaxation coefficient at V node weirs barriers 1 0 Value of the limiting ratio Emin for z5 zo H 2 0 Constant bottom roughness length zo when iopt_bstres_drag 3 m 0 0 14 5 5 Turbulence model parameters Parameters marked with a have been calibrated from experimental data or obtained from turbulence theory Their values should not be changed unless the user has sufficient experience in turbulence modelling 14 5 MODEL PARAMETERS alpha Black alpha ma alpha pp beta ma beta Xing cnu ad cl_eps c2_eps c31_eps c32_eps c_sk deltal_ad delta2_ad dissipmin expmom_ma expmom pp expscal_ma el_my e2_my e3_my 559 constant a in the Blackadar 1962 mixing length formulation 4 217 0 2 parameter am in the Munk amp Anderson 1948 scheme 4 136 4 139 10 0 12 oj
39. s or disables the inclusion of astronomical arguments and nodal corrections in the harmonic expansions and 4 354 0 0 astronomical argument set to zero nodal factors set to 1 nodal phases set to zero 1 evaluate astronomical phases at a given time and reference longitude nodal factors are set to 1 nodal phases set to Zero 2 evaluate astronomical phases and nodal corrections at a given time and reference longitude iopt astro tide Disables enables 0 1 the inclusion of the astronomical tidal force in the momentum equations 0 This requires that the model uses a spherical grid iopt grid sph 1 14 4 15 1 D applications iopt sur 1D Disables enables surface forcing surface slopes and elevations in case 1 D iopt grid nodim 1 water column applications 0 548 CHAPTER 14 CONTROL PARAMETERS 14 4 16 Surface forcing iopt meteo lopt meteo heat lopt meteo salflx iopt meteo stres lopt waves Remarks Disables enables 0 1 meteorological input and evaluation of all surface fluxes 0 Selects type of input data for the heat fluxes 0 0 no input 1 air temperature T relative humidity RH cloud cover fe 2 total downward non solar surface heat flux cloud cover fe 3 total downward non solar surface heat flux surface so lar radiance Qrad 4 cloud cover fe 5 surface solar radiance Qrad Selects type of input data for the salinity flux 0 0 no input 1 evaporation minus p
40. sharp gradients but con sumes more CPU time compared to the upwind scheme e TVD is the recommended scheme for 3 D scalars The advantage of using a T VD scheme is less evident for the 2 D mode which uses a much smaller time step than the 3 D mode The faster upwind scheme can be recommended for the 2 D mode in most cases TVD is recommended for resolving highly sheared 3 D currents as occurring in e g frontal zones e Advection of turbulence is considered of less importance than the pro duction and dissipation terms in the k k amp and kl transport equations It is recommended not to change the zero default value of iopt adv turb e The same limiting function applies for all transport equations solved with the TVD scheme 14 4 8 Diffusion coefficients iopt hdif coef Type of scheme for horizontal diffusion coefficients 0 0 not used 1 spatially uniform 2 Smagorinsky formulation 4 80 for momentum and 4 81 for scalars iopt_hdif_scal Disables enables 0 1 horizontal diffusion in the scalar trans port equations 0 iopt hdif turb Disables enables 0 1 horizontal diffusion in the turbulence transport equations 0 iopt hdif 2D Disables enables 0 1 horizontal diffusion in the 2 D trans port equations 0 542 CHAPTER 14 CONTROL PARAMETERS iopt hdif 3D Disables enables 0 1 horizontal diffusion in the 3 D current transport equations 0 iopt_kinvisc Formulation for kinematic viscosity 0 0 user
41. t surface exchange coefficient C when iopt_sflux_cehs 0 Constant surface exchange coefficient C when iopt_sflux_cehs 0 14 5 MODEL PARAMETERS dthd fld du BB dzetaresid conv gacc ref hcrit SH hdifmom cst hdifscal_cst kinvisc_cst optattcoefl cst optattcoef2 cst opt frac petsc_tol Rearth rho air sal ref sigstar DJ sig0 DJ 557 Threshold water depth d used in the mask criteria for drying and flooding see Section 5 4 2 m 0 1 Parameter d in the Burchard amp Bolding 2002 vertical grid transformation 4 26 1 5 Threshold value m used in the convergence criterium for the outer loop in the implicit scheme free surface corrector method 10 4 If different from real fill the acceleration of gravity taken as horizontally uniform Otherwise g is evaluated as function of latitude using 4 58 m s real fill Parameter her in the Song amp Haidvogel 1994 vertical grid transformation 0 1 Constant coefficient for horizontal momentum diffusion vg when iopt_hdif_coef 1 m s 0 0 Constant coefficient for horizontal scalar diffusion Ag when iopt hdif coef 1 m s 0 0 Constant value for the kinematic viscosity m s 1076 Inverse optical attenuation depth Aj for the absorption of long wave solar radiation as used in 4 59 m 10 0 Inverse optical attenuation depth A5 for the absorption of short wave solar radiation as used in 4 59
42. tion for the dissipation rate 2 1 Neumann condition 4 353 2 Dirichlet condition 4 351 Type of surface boundary condition for the dissipation rate e 2 1 Neumann condition 4 284 2 Dirichlet condition 4 281 Type of background mixing scheme as described in Section 4 4 3 6 0 0 using uniform background coefficients 1 using limiting conditions for turbulence parameters 2 the Large et al 1994 scheme given by 4 22 4 228 Selects type of background mixing mixing 0 0 user defined constant value vdifmom cst 1 kinematic viscosity as selected by iopt kinvisc Mixing length formulation as described in Section 4 4 3 5 4 1 parabolic law 4 213 2 modified parabolic law 4 214 3 Xing formulation 4 215 4 Blackadar asymptotic formulation 4 216 Number of transport equations as described in Section 4 4 3 4 1 0 zero equation model equilibrium or Mellor Yamada level 2 method with a mixing length selected by iopt turb Imix 1 turbulence energy equation with a mixing length se lected by iopt_turb_Imix 2 k amp of k kl equation depending on the value of iopt turb param Selects type of second turbulent variable 2 1 mixing length k l scheme 544 CHAPTER 14 CONTROL PARAMETERS 2 dissipation rate k e scheme iopt turb stab form Selects type of stability function 3 lopt turb stab lev iopt turb stab mod lopt turb stab tke
Download Pdf Manuals
Related Search