open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello Guys,

I am trying to model the sediment settling and bed evolution on a settling basin.
The domain is a rigid bed everywhere with no sediment at the beginning at the simulations (ZR=ZF).

I already simulate with success using T2D coupled with Sisyphe and now I am trying to use the T3D, but I am not happy about how it is going.


Im my first attempt I coupled T3D+Sisyphe using rigid bed and get the bed evolution that is in agreement with the 2D results. Both suspension and bed-load were made in Sisyphe.
This way I get results of the suspended sediments as a 2D data on the Sisyphe result file, and not as a 3D data at the horizontals levels of the T3D result file.

My second attempt was to simulate the suspension in the T3D using SEDI3D (not coupling with Sisyphe). In this case I get the result as as a 3D data, but in this case I wasn't able to set rigid bed. The NOEROD file (the same that worked without T3D calculating sediments) did not work.

An Finally I try to simulate suspension using T3D and the bed load on Sisyphe. Again the NOEROD file did not work. It seems that I need a new setting or a different way to set rigid bed on T3D SEDI3D.

Additionally, with T3d calculating suspension and Sisyphe the bed-load the bed evolution computed by Sisyphe and computed by T3D were different!!! I coupled as INTER-SISYPHE and at each time-step.

I also was a little confused about how the bed erosion was calculated by T3D and why it was different than the Sisyphe.


So the questions are:

>> Is it possible to measure suspension as a 3D data using Sisyphe ou make the T3D read and plot the sisyphe suspension results as a 3D?

>> How to set rigid bed when use the SEDI3D from T3D?



Thanks in advance.


Cheers!

Hello,

For what I know, the implementation of the rigid bed ZR is in subroutine condis.f in library telemac3d. I see that there is an implementation based on a call to NOEROD that has been commented out, and ZR is here computed with the bottom ZF and the sediment layer given above in condis.f. Maybe this can help you.

Actually this is a temporary situation, in the next release we plan to force the coupling to Sisyphe as soon as sediment layers are involved, and only Sisyphe will deal with layers, which will avoid all duplications and differences of behaviour.

With best regards,

Jean-Michel Hervouet

Thanks for your prompt response.

Yes, I was using the NOEROD routine to set the rigid bed.
I edited it to reckon the 3D mesh and it worked for Sisyphe if I do not computes sediments in Telemac3D.
When turn on the SEDI3D it starts to provides different bed evolution for Telemac3D and Sisyphe (rigid bed in Sisyphe and erodable in Telemac results).

I used the CONDIS routine (and setting the initial sediment layer thickness as zero for my case) without coupling with Sosyphe and it works just for simulationg sediments in Telemac3D.

If I try to couple with Sisyphe the bet becomes no erodable for both models. I tryed to use CONDIS and NOEROD together, but also did not work.


Some questions appears.


1) Using the Telemac3D SEDI-3D, how is the bed-load computed?
It is affected by the key-words CRITICAL SHEAR STRESS FOR DEPOSITION and EROSION COEFFICIENT, but the manual did not provide much information about it.

2) Using the Telemac3D SEDI-3D - I used a 3 sediment classes, imposed concentrations at the boundaries and a TRACERS VERTICAL PROFILE = 3 (Rouse normalized). Despite the Rouse profile the 2 finer classes are been computed as a constant profile but the concentration match to the imposed values. For the coarser, from the free surface to the bottom the concentration increases until the layer 2, and the layer 1 (the deepest one) presents the lowest concentration, additionally the concentration did not agree with the imposed value.
I think by the Rouse condition that some of the sediment behavioured as bed-load, which physically makes sense. As the T3D do not provide as an output the bed-load, could it a possible result? How could I check it?

3)Coupling Telemac3D and Sisyphe: The Sisyphe manual did not provide informations about how the suspension is computed when coupled with Telemac3D, it is just described the 2D sediment transport. How suspended transport is handled by coupling T3D + Sisyphe?

4)Coupling Telemac3D and Sisyphe: The Sisyphe manual describe that the velocity on the deepest layer is use to compute bed shear stress. I understand that it means the bed-load is transport equations are handled in the same way. Am I right?




Thanks again for your help.