open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I'm trying to model cohesive sediment transport in Scheldt River. I have a Telemac-3D model internally coupled with SISYPHE. I got logical results of QS Suspension from the model but the mass concentration seems not correct. Does anyone know what is wrong with my model? Is there anything wrong in my steering file?

Thanks in advance!

Hello,

I adjusted the scale of the values of concentration. Now I have the more realistic result except for those spots in shallow areas. I looked at the velocities in those spots and it always has big velocities there. I think this caused the high erosion in those spots.

Hello,

This is to keep you posted on the hydrodynamics part. I worked on your case and found a number of improvements that will be in version 6.2. One you can try in version 6.1 is the attached subroutine diff3d.f. On tidal flats the salinity will stay at 0.5, not 0.
I also noticed that the loss of tracer mass is due to the accuracy of the solver which is at 1.E-6, with a better accuracy or with a direct system it decreases a lot.

I hope this helps, but the sub iterations problem will only be fully solved in version 6.2.

with best regards,

Jean-Michel Hervouet

Hello,

Well, attaching files seems difficult, in subroutine diff3d.f (library telemac3d), you have 4 lines :

DO I=0,IPBOT(I2D)-1

just remove the -1 of the first two lines. It will have the effect that points at the free surface, when they have no volume (i.e. on tidal flats with all neighbours without water) the points will keep their original temperature instead of getting 0). If it is OK for you I'll put it as a bug correction.

JMH

Thank you very much! I'll try this in my model.
By the way, when will the version 6.2 be released?

Hello,

We plan to release version 6.2 in the end of july 2012.

JMH

That will be great. I look forward to the new release.

Another question is that is there any subroutine in Telemac-3d could be programmed to re-calculate friction coefficient vary in space at each time step just like SUBROUTINE CORSTR in Telemac-2d? Could those modified friction coefficients be used by SISYPHE while coupling with Telemac-3d?

Hello,

CORSTR is also called in Telemac-3D (with slightly different user implementation as it will be based on declarations_telemac3d instead of declarations_telemac2d), and arrays CF, RUGOF and UETCAR (U* squared) are transmitted to Sisyphe.

JMH

Hello Catherine,

I'm sorry I didn't post the frame at the same time step so it might look strange.
This plot shows the concentration and QS suspension at the same time step. At this time, water flows from sea to the river.

Hello : it is rather strange, that the concentration 2D plot dosnt't show a similar pattern than the QS plot. I recommend that you try to adjust the scale. Let me know,
Catherine