open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi!!

I realize a study of flooding on the city of Bayonne (telemac 2d v6p2). The area is located on the maritime influence, so we want to use only the heights of the water in the boundary conditions.

I know it can cause problems on the model, but I still realized ​​the tests, obtaining the good results in relation to heights of water.

I would like to know your opinion of the use of the water levels in the all boundary conditions, and if telemac is able to solve these cases.

I add the model domain and the cas file. I wait your comments and advices! Thank u!

Manuel MARTIN




/

---

/ TELEMAC-2D Version 6.2
/ Juin 2013
/
/ MODELISATION BAYONNE 2D
/ SCHAPI - SPC GAD BORDEAUX
/ Manuel MARTIN
/
/
/

---

/
/ INFO
/

---

/
/MODELE 2D : EVENEMENT 2009 (PAS MESURE)
/CONDITIONS AUX FRONTIERES : HAUTEUR
/
/
//// CONSIDERATIONS POUR LES BANCS DECOUVRANTS (MANUEL D'UTILISATEUR)
//// HAUTEUR IMPOSE (3 FRONTIERES)
/
/LOI DE FROTTEMENT SUR LE FOND : 3 (STRICKLER)
/
/MODELE DE TURBULENCE : 1 (VISCOSITE CONSTANTE)
/
/NON METHODE DE C. THOMPSON
/

---

/
/ ENVIRONNEMENT INFORMATIQUE
/

---

/
/
TITRE = 'TELEMAC 2D : MODELISATION BAYONNE 2D'
/FICHIER FORTRAN : 'fortran3d.f'
FICHIER DES CONDITIONS AUX LIMITES : 'Bayonne2d_H_24_06mod.cli'
FICHIER DE GEOMETRIE : 'Bayonne2d_24_06mod.slf'
FICHIER DES FRONTIERES LIQUIDES : 'bayonne2d_liquide.txt'
FICHIER DES RESULTATS : 'Bayonne2d_1s_Res.slf'
/PROCESSEURS PARALLELES : 0
/
/
/

---

/ OPTIONS GENERALES
/

---

/
/SUITE 2D = NON
/
VARIABLES POUR LES SORTIES GRAPHIQUES : 'U,V,S,H,W'
PAS DE TEMPS : 4.
DUREE DU CALCUL : 604800 / heures
PERIODE POUR LES SORTIES GRAPHIQUES : 150
PERIODE DE SORTIE LISTING : 150
/
/
/

---

/ CONDITIONS INITIALES
/

---

/
CONDITIONS INITIALES : 'COTE CONSTANTE'
COTE INITIALE : 0.4
/DATE DE L'ORIGINE DES TEMPS : 2013;04;09
/
/

---

/ CONDITIONS AUX LIMITES
/

---

/
LOI DE FROTTEMENT SUR LE FOND : 3
COEFFICIENT DE FROTTEMENT : 30.
COTES IMPOSEES : 0.4;0.4;0.4
/DEBITS IMPOSES : 0.;0. / voir aussi liquide.txt
PROFILS DE VITESSE : 4 ; 4 ; 4
/
/
/
/
/

---

/ OPTIONS NUMERIQUES
/

---

/
OPTION POUR LES FRONTIERES LIQUIDES : 1;1;1
/
/
BANCS DECOUVRANTS : OUI
OPTION DE TRAITEMENT DES BANCS DECOUVRANTS : 1
TRAITEMENT DES HAUTEURS NEGATIVES : 1
COMPATIBILITE DU GRADIENT DE SURFACE LIBRE : 0.9
CORRECTION DE CONTINUITE : OUI
OPTION DE SUPG : 0;0
BILAN DE MASSE : OUI
PRECONDITIONNEMENT : 2
/
/
MODELE DE TURBULENCE : 1
COEFFICIENT DE DIFFUSION DES VITESSES : 1.
/
/
STOCKAGE DES MATRICES : 3
PRODUIT MATRICE-VECTEUR : 1
IMPLICITATION POUR LA HAUTEUR : 1.
IMPLICITATION POUR LA VITESSE : 0.55
/
/
MASS-LUMPING SUR H : 1
CLIPPING DE H : OUI
VALEUR MINIMUM DE H : -0.005
DISCRETISATIONS EN ESPACE : 11;11
/
/
FORME DE LA CONVECTION : 1;5
SOLVEUR : 1
PRECISION DU SOLVEUR : 1.E-5
/MAXIMUM D'ITERATIONS POUR LE SOLVEUR : 200
TRAITEMENT DU SYSTEME LINEAIRE : 2
/
/
&ETA
&FIN

Hello Manuel

You mean that you are using prescribed elevation (and not prescribed water depth)?
I can't give an objective opinion on your model because I don't have all the informations about it. However, to consider sea influence, you can use the boundary condition PRESCRIBED ELEVATION and in this case you don't consider the tide effect. If you estimate that the tide effects are not negligeable you have to improve your model (have a look on validation test case 123_tide).

With my best regards

Riadh ATA

hello!

Yes, sorry I had to express myself better .

The objective of this model is the modeling of floods with a coupling Mascaret1D-Telemac2D.

I use prescribed elevation boundary conditions, but the prescribed elevations are taken directly from the model Mascaret1D (currently used for flood prediction (SCHAPI)). So the prescribed elevation curves for every boundary consider the effects of the tide on the water level. Therefore I do not impose the tide with Telemac.

Consequently I impose boundary conditions PRESCRIBED ELEVATION on the three liquid boundaries, I do not impose any flow (Q) in the model. So I'd like to know your opinion and if it can cause problems, or if a model with all PRESCRIBED ELEVATION boundary conditions can work well.

Best regards

Manuel MARTIN

Hello Manuel,

I'm not sure that boundary conditions given by Mascaret include tide effects.
with tree prescribed elevations prescribed at the boundaries, Telemac will work well, i.e. it will satisfy these boundary conditions however, nothing will garantee that you will reproduce the real flow as if it was produced by a full 2D code.
The 1D/2D coupling has already been achieved by EDF team and if you are interested by its implementation you can contact directly the mascaret team (make your request on the Mascaret section of the forum).

With my best reagrds,

Riadh ATA

Hello!!

Thank you for your help.

The tide effects are taken by Mascaret and then the prescribed elevation boundary conditions consider these effects in the model. I add a elevation graph at a point of the model for a simulation (4 days).

I compared this simulation (all prescribed elevation boundary conditions) with other simulation : prescribed elevation boundary condition in the downstream section and prescribed flow in the upstream section (flows given by MASCARET). There are the importants differences for flood periods. I add a elevation graph of prescribed flood simulation at the same point of the model.
However, comparing with observations, the results with prescribed elevation boundary conditions are better that results with prescribed flows.

Thank you!!
Best regards

Manuel MARTIN

Hello,

Finally, we use the prescribed flowrates boundary conditions because it allows us to calibrate our model.

However I have some problems on the boundaries with negative flows and the software stops the simulation. For improving these results, I use the method of Thompson for this boundary. Telemac is capable of finishing the modelling and the results are good. Except at the boundary, we have significant speed oscillations.

I add a gif image of speed oscillations on the boundary.

Can we improve these boundary instabilities ?

Are there others methods to improve the prescribed flowrates boundary conditions with negatives values?

Best regards

Manuel MARTIN

Hello,

I understand that you have positive and negative prescribed discharges at an entrance, due to a tide influence. What should be the exit of the river becomes an entrance. I wonder if you should not change the boundary conditions when the exit becomes an entrance, changing for this boundary all the LIHBOR LIUBOR LIVBIR from 5 4 4 to 4 5 5. This could be done in bord.f (but no Thompson in this case).

Otherwise in thomps.f there is a hardcoded parameter TETA that you can use to stabilise. However Thompson is rather used in offshore conditions with generally large depths and small velocities, so it could be that it is not convenient in your case.

With best regards,

Jean-Michel Hervouet

Hello,

I have positive and negative flows, but it isn't an exit. It is an entrance of model (upstream section), all the domaine is under the maritime influence. So I should use prescribed flowrate boundary conditions (5 4 4) for calibrating the friction coefficient.

Changing the parameter TETA, Can I improve the results at the boundary ?

With best regards

Manuel MARTIN