open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi There,

I am currently attempting to complete an uncoupled 2D sediment transportation model for dredging and reclamation activities using a point release source. This is implemented by updating the depth averaged concentration for a single node at each sub-iteration of the SISYPHE computation (modification of the condim_susp.f subroutine was previously made by another colleague).
I am experiencing large differences in the output of VELOCITY UV when comparing the time series from Sisyphe and Telemac-2D for the same node. When looking at the computation of Velocity UV for each sub iteration I can see that this seems to be related to some interpolation issue between the Telemac graphic output periods. More so the differences become extreme in areas of low water depth. The difference almost becomes non-existent away from the node of sediment release and in deeper water depths.
I have 30 minute graphic output periods for telemac with 30 min sisyphe sub-iterations (i.e every 1 minute of the simulation). For consecutive graphic output periods (30 mins) I can see that the initial sisyphe computational timestep has a non-zero value for UV MAG but for all sub-iterations there after the velocity is computed as 0.
Has anybody experienced this problem before as I am unable to work out what is going on. I was using the velocity outputs of sisyphe to assess the net erosion/deposition flux and noticed that this error was occurring.From my understanding the velocity results should not be modified between Telemac and Sisyphe in an uncoupled mode.

Any help would be greatly appreciated.

Best Regards,
Toby

Hi Toby,

In uncoupled mode, SISYPHE interpolates the veolcities between flow field times steps and then applies a correction based on the changes in the bathymetry due to sediment transport. If you add sediment locally at the release location, part of that material will drop out. This reduces the water depth, and through the conservation of mass momentum this leads to a local increase in flow velocities.
This effect is far stronger for the same amount of sediment in shallow water then it is in deep water. So your findings are in agreement with the way the uncoupled mode works.

Thanks for the quick repose. This makes sense and now that I have re-read the manual I can see that this is stated as well.
Continuing from the previous question I guess I want to ask how the velocities are updated. Are they only updated for nodes which experience evolution of the seabed? Which subroutine does this "flow updating" take place in? Sisyphe.f?
Also I'm interested in know exactly how the net deposition/erosion flux for cohesive sediments is calculated. From my understanding erosion will only occur when the bed shear stress exceeds the critical value and vice versa for deposition (however deposition will only occur when the bed shear stress is below the critical value for deposition). If this is the case how is the net deposition flux different from the actual deposition flux if for a given shear stress only erosion or deposition can occur.

Thanks,
Toby