Welcome, Guest
Username: Password: Remember me

TOPIC: horizontal/vertical diffusion of the tracer has no effect?why?

horizontal/vertical diffusion of the tracer has no effect?why? 6 years 2 months ago #31527

Hello!
So I am trying to play with the horizontal/vertical diffusion of the tracer to see how could that impact the salinity over the domain. The thing is the results seemed not to be affected even by using some high values of diffusions. Any ideas why that could happen?

Thanks
The administrator has disabled public write access.

horizontal/vertical diffusion of the tracer has no effect?why? 6 years 2 months ago #31531

  • c.coulet
  • c.coulet's Avatar
  • OFFLINE
  • Moderator
  • Posts: 3722
  • Thank you received: 1031
Could we see the global parametrization of the computation?
Christophe
The administrator has disabled public write access.

horizontal/vertical diffusion of the tracer has no effect?why? 6 years 2 months ago #31547

Yes sure.
Here is the steering file content.
PARALLEL PROCESSORS : 100
/DEBUGGER=1
/----------------------------------------------------------------------
/   TELEMAC 3D : The Gironde Estuary  
/                ***  20 plans  ***
/ Runs the model for the period Oct1 to Nov15
/----------------------------------------------------------------------
/
TITLE = 'The Gironde Estuary Model'

/--------------------------------------------------
/     			Files
/--------------------------------------------------

FORTRAN FILE           : Wind_and_Friction2.f
BOUNDARY CONDITIONS FILE : BC_50m.cli
GEOMETRY FILE           :  Geo_50m.slf
BINARY DATABASE 1 FOR TIDE='h_tpxo7.2'
BINARY DATABASE 2 FOR TIDE ='u_tpxo7.2'
ASCII ATMOSPHERIC DATA FILE     = wind_data
LIQUID BOUNDARIES FILE    : River_Discharge.txt
INITIAL TIME SET TO ZERO=YES
2D RESULT FILE FORMAT            ='SERAFIN'
3D RESULT FILE FORMAT            ='SERAFIN'
3D RESULT FILE  : The_Girond_dens_dif2_3D.slf
2D RESULT FILE  : The_Girond_dens_dif2_2D.slf

/--------------------------------------------------
/     			Initializing
/--------------------------------------------------
/		---2d file---
/2D CONTINUATION = YES
/FILE FOR 2D CONTINUATION FORMAT = 'SERAFIND'
/FILE FOR 2D CONTINUATION = 'INITIAL_DENSITY_2D.slf'
/		---3d file---
PREVIOUS COMPUTATION FILE FORMAT = 'SERAFIN'
COMPUTATION CONTINUED = YES
PREVIOUS COMPUTATION FILE = 'T3DRES'
/RECORD NUMBER FOR RESTART = 3276

/--------------------------------------------------
/     			Coordinate System 
/--------------------------------------------------
GEOGRAPHIC SYSTEM=4
ZONE NUMBER IN GEOGRAPHIC SYSTEM=3

/--------------------------------------------------
/     			Time Steps And Variables
/--------------------------------------------------

NUMBER OF TIME STEPS = 64800
TIME STEP   = 60
VARIABLES FOR 2D GRAPHIC PRINTOUTS : U,V,S,H,W,B,X,Y,TA1
VARIABLES FOR 3D GRAPHIC PRINTOUTS : Z,U,V,W,TA1
LISTING PRINTOUT PERIOD    : 1
GRAPHIC PRINTOUT PERIOD : 30
MASS-BALANCE    = YES
INFORMATION ABOUT MASS-BALANCE FOR EACH LISTING PRINTOUT = YES

/--------------------------------------------------
/     			Initial Condition and B.C
/--------------------------------------------------

INITIAL CONDITIONS ='CONSTANT ELEVATION'
INITIAL ELEVATION       =3.5
PRESCRIBED FLOWRATES     =0.;0.0;0.0
PRESCRIBED VELOCITIES=0;0;0
VELOCITY PROFILES=1;4;4  /-----it was 4;4;;
VELOCITY VERTICAL PROFILES =2;2;2
NUMBER OF HORIZONTAL LEVELS= 20
MESH TRANSFORMATION : 1
OPTION FOR TIDAL BOUNDARY CONDITIONS=1;0;0
OPTION FOR LIQUID BOUNDARIES=2;1;1
ORIGINAL DATE OF TIME=2015;09;30
ORIGINAL HOUR OF TIME=21;51;21
TIDAL DATA BASE=2
CORIOLIS	=	YES
CORIOLIS COEFFICIENT : 0.000137
LAW OF BOTTOM FRICTION = 3
FRICTION COEFFICIENT FOR THE BOTTOM = 45.44
TREATMENT OF FLUXES AT THE BOUNDARIES : 2;2;2
INITIAL GUESS FOR DEPTH  =1

/--------------------------------------------------
/     					Tidal Flat
/--------------------------------------------------

TIDAL FLATS: YES
OPTION FOR THE TREATMENT OF TIDAL FLATS=1
TREATMENT OF NEGATIVE DEPTHS = 2
MINIMAL VALUE FOR DEPTH = 0.0
TREATMENT ON TIDAL FLATS FOR VELOCITIES = 0
TREATMENT ON TIDAL FLATS FOR TRACERS = 0
BYPASS VOID VOLUMES: YES

/--------------------------------------------------
/     					Tracers
/--------------------------------------------------

NUMBER OF TRACERS=1
TYPE OF SOURCES=2
PRESCRIBED TRACERS VALUES = 35;0;0
NAMES OF TRACERS : 'SALINITY'
INITIAL VALUES OF TRACERS : 0.0
DENSITY LAW : 2
COEFFICIENT FOR VERTICAL DIFFUSION OF TRACERS = 0.01
COEFFICIENT FOR HORIZONTAL DIFFUSION OF TRACERS = 0.01
/--------------------------------------------------
/    Steps and Schemes
/--------------------------------------------------

ADVECTION STEP = YES
SCHEME FOR ADVECTION OF VELOCITIES = 1 / this could cause a problem if the CFL condition is not met
SCHEME FOR ADVECTION OF DEPTH = 5
/SUPG OPTION = 1;1;1;1
SCHEME FOR ADVECTION OF TRACERS = 14
/NUMBER OF SUB ITERATIONS FOR NON LINEARITIES=4 /for better mass conservation
DIFFUSION STEP = YES
SCHEME FOR DIFFUSION OF VELOCITIES = 1
SCHEME FOR DIFFUSION OF TRACERS = 1
PROPAGATION STEP = YES
NON-HYDROSTATIC VERSION = YES
DYNAMIC PRESSURE IN WAVE EQUATION =YES

/--------------------------------------------------
/    Solvers and Accuracies
/--------------------------------------------------

SOLVER FOR DIFFUSION OF VELOCITIES = 1
SOLVER FOR PROPAGATION = 1
SOLVER FOR PPE = 1
SOLVER FOR VERTICAL VELOCITY = 1
SOLVER FOR DIFFUSION OF TRACERS = 1

ACCURACY FOR DIFFUSION OF VELOCITIES = 1.E-6
ACCURACY FOR PROPAGATION = 1.E-6
ACCURACY FOR PPE = 1.E-6
ACCURACY FOR VERTICAL VELOCITY =1.E-6
ACCURACY FOR DIFFUSION OF TRACERS =  1.E-15

MAXIMUM NUMBER OF ITERATIONS FOR DIFFUSION OF VELOCITIES = 1000000
MAXIMUM NUMBER OF ITERATIONS FOR PROPAGATION = 1000000
MAXIMUM NUMBER OF ITERATIONS FOR PPE = 1000000
MAXIMUM NUMBER OF ITERATIONS FOR VERTICAL VELOCITY = 1000000
MAXIMUM NUMBER OF ITERATIONS FOR DIFFUSION OF TRACERS = 1000000
MAXIMUM NUMBER OF ITERATIONS FOR ADVECTION SCHEMES = 1000000

IMPLICITATION FOR DEPTH            : 1.
IMPLICITATION FOR VELOCITIES         : 1.
IMPLICITATION FOR DIFFUSION         : 1.
MASS-LUMPING FOR DEPTH   : 1
MASS-LUMPING FOR VELOCITIES : 1
MASS-LUMPING FOR DIFFUSION : 1
MATRIX STORAGE : 3

/--------------------------------------------------
/     TURBULENCE MODEL
/--------------------------------------------------

TURBULENCE REGIME FOR THE BOTTOM = 2
/ in case of 1 the friction will be computed based on Richardson number
LAW OF FRICTION ON LATERAL BOUNDARIES = 5 /  value 1 means no friction
FRICTION COEFFICIENT FOR LATERAL SOLID BOUNDARIES = 60
FRICTION COEFFICIENT FOR LATERAL SOLID BOUNDARIES = 2
VERTICAL TURBULENCE MODEL = 2
MIXING LENGTH MODEL = 6
HORIZONTAL TURBULENCE MODEL = 4

/--------------------------------------------------
/     WIND
/--------------------------------------------------

WIND                                         = YES
OPTION FOR WIND                              = 3
COEFFICIENT OF WIND INFLUENCE                =  6.78E-7
WIND VELOCITY ALONG X                        = 0
WIND VELOCITY ALONG Y                        = 0
THRESHOLD DEPTH FOR WIND = 0.5

/--------------------------------------------------
/     PRECONDITIONING
/--------------------------------------------------
 /PRECONDITIONING FOR DIFFUSION OF VELOCITIES: 34
 /PRECONDITIONING FOR PROPAGATION : 34
 /PRECONDITIONING FOR PPE : 34
 /PRECONDITIONING FOR VERTICAL VELOCITY : 34
/ PRECONDITIONING FOR DIFFUSION OF TRACERS : 34 
 /PRECONDITIONING FOR DIFFUSION OF K-EPSILON : 34

/--------------------------------------------------
/     OTHER OPTIONS
/--------------------------------------------------
COEFFICIENT TO CALIBRATE SEA LEVEL=0.4758
COEFFICIENT TO CALIBRATE TIDAL RANGE = 0.8905882
AVERAGE WATER DENSITY =1000
HYDROSTATIC INCONSISTENCY FILTER : NO
/CHECKING THE MESH: YES
NUMBER OF BOTTOM SMOOTHINGS = 1

&ETA
&FIN
The administrator has disabled public write access.

horizontal/vertical diffusion of the tracer has no effect?why? 6 years 1 month ago #31587

  • Leballeur
  • Leballeur's Avatar
  • OFFLINE
  • openTELEMAC Guru
  • Posts: 434
  • Thank you received: 163
Hello,

The only thing I see in your steering file is the high values for the maximum number of iterations of your different solvers. Try to decrease it to about 200-300 iterations (reduce the general time step if needed).
Depending on your mesh element size (horizontal and vertical), the numerical diffusion can be higher than the physical one but by trying different values for the diffusion coefficient of tracers (ie 1e-6, 1e-2, 10 for instance) you should see some differences.

I hope it helps.

Regards,
Laurent
The administrator has disabled public write access.

horizontal/vertical diffusion of the tracer has no effect?why? 6 years 1 month ago #31589

Thank you. I will try that.
The administrator has disabled public write access.
Moderators: pham

The open TELEMAC-MASCARET template for Joomla!2.5, the HTML 4 version.