open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,
I would like to use the Smart & Jaeggi formula for sediment transport in a simulation with Telemac2D and Sisyphe.
Did anybody already use this formula? I would like to write it in the subroutine QSFORM.f, but I'm a beginner with Fortan. Do you have any advice?
Thank you.

Marina

Hello,

the Meyer-Peter and Müller bed load formula is similar to Smart & Jäggi in terms of the required variables, so you can start by adapting the subroutine bedload_meyer.f and you don't have to build the subroutine from scratch.

You should consider also to read the Telemac programming guide available in the download section of the website.

Hope this helps,

Clemens

Hello Marina,

are you dealing with steep slopes? Then you may also have to think about the different flow resistance in steep channels and the corresponding formulas for Smart and Jaeggi, because they are influencing the sediment transport.

If you are german speaking, you can find a very good overview in
Rickenmann, D., 2014: Methoden zur quantitativen Beurteilung von Gerinneprozessen in Wildbächen. WSL Ber. 9: 105 S.

Best regards,
Gabi

Hello,
Thanks for your quick answers.

@ Clemens:
I've been following your advice and starting from the bedload_meyer.f, but I'm facing troubles. The required variables are not exactly the same (or I'm missing something ?)

For Meyer-Peter, I just need the following variables: GRAV, DENS, DM, MPM_ARAY, TETAP and ACP

If I'm right, in the Smart Formula, I also need the friction slope and the liquid specific discharge.
For the friction slope, I found these variables in the subroutine bedload_effpnt.f :
DZFDX |<->| BOTTOM SLOPE IN THE X-DIRECTION
DZFDY |<->| BOTTOM SLOPE IN THE Y-DIRECTION
Is it correct to use them here (I used DZFDY) ?
And for the liquid specific discharge, I used Q, but is it the specific discharge ?

I obtain :
DO I=1,QSC%DIM1
QSC%R(I)=4.2/DENS*DZFDY**1.6*Q*MAX(1-(ACP%R(I)*DENS*DM/(HN*DZFDY)),0.D0)
ENDDO

Is this correct ?

@ Gabi : I downloaded the paper and started to read it. In this case, I start my simulation with a flat channel and an erodible bed, and I impose a dam break upstream. At the end of the simulation, I should get local steep slopes due to erosion. Maybe in this case, the impact of different flow resistance is negligible due to the small extent of areas concerned by steep slopes ?

Thanks again for your help.
Best regards.
Marina

Hello,

good start.
Yes, to access other variables you can add the following line in the subroutine.
USE DECLARATIONS_SISYPHE, ONLY : name of variables
For example: USE DECLARATIONS_SISYPHE, ONLY : H, U, V

In the declarations_sisyphe.f subroutine you can check the available variables.
The friction slope is equal to the bed slope only in uniform flow conditions which especially in sediment transport calculations hardly occurs. You have some possibilities to calculate the friction slope e.g. via the Strickler law and quadratic friction coefficient. The specific discharge in one node is velocity times water depth.

Hope this helps,

Clemens