open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear Friends,

The issue of non-erodable bed has been discussed several times in the forum but still, I am confused.

I am trying to run a T2D-SISYPHE coupled simulation for a small laboratory channel.

The simulation is with clear water (no-sediment supply from inlet)

The inlet and outlet channel consist of non-erodable bed. I modified the noerod.f file to define the location of my non-erodable bed.

According to the experimental results, there should be some deposition of sediments at the outlet end of the non-erodable bed. But the results shows no any deposition. And the final bed level is the same as the initial bed level of non-erodable bed. This is causing the problem to the overall simulation results.

So, in SISYPHE, non-erodable bed will not have any deposition? or should we make more adjustment in the noerod.f file?

How can I set in the noerod.f file so that I will not have any erosion on the non-erodable bed but the deposition could occur? or should we modify any other input condition?

The simulation files are attached.

Thank you.

Regards,
Saroj

Have you checked the flow rates and sediment transport rates?

With the dimensions, I estimate your velocities are about 0.16 m/s. That is unlikely to move sand with a grain size of 0.72 mm.

Dear Michael,

Thank you for your opinion.

Yes, the flow rate is relatively small. But under the given dimensions of the channel and the channel slope (1:550), the applied shear stress is quite higher than the critical shear stress required for the initiation of the sediment transport. Thus the movement of sediment is expected.

The only region I am unsure is at the non-erodable bed at the outlet. The boundary between the erodable and non-erodable bed is causing the problem since a sudden transition is imposed at this boundary.

I am not sure how to resolve this issue?

Regards,
Saroj

Hello Saroj,
Please find below some tips :

* Please run a test by adding a flag = 5 (instead of = 4) for your inflow sediment boundary in the .cli file.
4 5 5 0.000 0.000 0.000 0.000 5 0.000 0.000 0.000 5355 1 # Inflow (5155 - 5355)

* Indicate explicitly the sediment transport formula you're using (bedload transport problem with MPM formula) -> you can also check your results with different formulas

* Is your hotstart file 100% steady state ?

Let us know how's going.

Best wishes,
Pablo

Dear Pablo,

Thank you for the suggestion. I will have a try with it.

Yes, the hotstart file is a 100% steady state.

I just want to confirm that, Is it the inflow boundary that needs to modified? as you have pointed out or is the outflow boundary?
Because, I was having problem near the outflow boundary.

Thank you.
Saroj

Dear Pablo,

I used the boundary configurations as suggested by you. The results are much improved.

The use of the flag 5 like below you suggested,

4 5 5 0.000 0.000 0.000 0.000 5 0.000 0.000 0.000 5355 1 # Inflow (5155 - 5355)

According to the Sisyphe Manual in section (1.2.2) & (2.7.2), Does It mean that we are imposing equilibrium/imposed sediment concentration or imposed bottom evolution?
with EBOR/CBOR=0.0 means zero bottom evolution and zero concentration at the inflow?

In the above case, is the zero sediment supply from the inflow ensured? because my case is without sediment supply i.e. clear water.

And if we want to have a equilibrium bedload discharge for modelling real river case, how should we set the flag?

I am trying to carry out two cases, one is the experimental case under (clear water) no sediment supply from upstream and the other case is of the real river with equilibrium bed load discharge. What could be the possible inflow boundary condition for these two cases?
I tried with the same configuration for both cases but the bed evolution result of the real river case is not very satisfactory

And the other thing I wanted to confirm is about the Morphological factor.
How should I choose this value?

If I use a morphological factor = 10, and simulate for one hour, then does it mean my actual simulation is of 10 hours? I am really confused about this matter.

Finally, Does the coupling period for sisyphe has any effect on the results of bed evolution? If I use coupling period=10 or 60? Does it matter anything during the computation of bed evolution?

Hope I can get some clarification on this confusion.

Regards,
Saroj

Hi Saroj,

I had another look at your model domain. I noticed that the bed levels is the same everywhere in bottom_2018-10-17.slf. That implies that the transport capacity of the flow over the non erodible bed at the outlet is the same as that of the flow over the erodible bed. As soon as the erodible bed starts eroding, the transport capacity over the non erodible bed will higher than that of the eroded part of the channel. Therefore any sediment to reach the hard layer will be moved on in the next time step. So in the numerical model, there will be no sediment depositing, not even temporarily. Only if more sediment is arriving at a point than can be transported away from that point, deposition will occur.

Dear Michael,

Once again thanks for having a look at the model. I fully acknowledge your idea.

Actually, I impose a bottom topography file which is attached to this message. And another thing is that since the prescribed outlet elevation is about 3cm (i.e. mean flow depth), at least some sediment would be expected to be deposited upto the prescribed depth.
But that is just my understanding, I might be wrong.

Thank you.

Regards,
Saroj

I have run your case with the flat bed geometry in the original post now. I do get sedimentation on top of the non-erdible bed. However, the amounts are very small
(<1e-5 m)

Thank you.

I will have a try and share the results.

Sorry, the topography file mentioned couldn't be uploaded in the previous message. Here is the dropbox link in case you want to try.

www.dropbox.com/s/8867wowiacnekfc/topo.xyz?dl=0

Regards,
Saroj