open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I am doing a theoretical model 3d of an estuary to study the development of plume river in the ocean and in the estuary.

From the calculation of several different models, I have obtained some results of inconsistent speeds. About 7-8 days after modeling, the speeds begin to oscilate in the coastal zone of the plume. This phenomenon alters the salinity and the elevation of the water surface.
Presented below is the image corresponding to the problem. I hope that you can give me an answer to this problem.

Thanks in advance

Manuel MARTIN

Hello,

I don't really see if it is a problem at the open boundary or due to dry zones. If it is at the open boundary, it is certainly yet another time the fact that the problem is not well posed at open boundaries (it is the case of all studies of estuaries). If you prescribe only the free surface at the sea boundary, with free velocities, it is not a well posed problem especially when the velocities comme from the sea and enter the domain. The solutions are :

1) have your open boundaries in the deepest waters possible (so that the velocities ramain small). A head loss on such boundaries is also useful sometimes.

2) use Thompson boundary conditions (now possible in 3D just like in 3D, see feature logs on the bottom of the home page)

If it is near the coast and dry zones, we have to check how the treatment of tidal flats is done and the advection schemes you use.

With best regards,

Jean-Michel Hervouet

Hello,

Thanks for your reply.

I already had a domain large enough to avoid these problems in the open boundaries.Therefore I did not think that the problem was not well posed in the sea boundaries. Despite this, I've made ​​the changes proposed and am still getting the same velocity oscillations.
I think the problem may come from dry zones and the advection schemes, thus I will I add the cas file. This model has a 'TIME STEP' = 30s (but I've also done the same test with 10s) and the tide is not yet included.
I will also add a graph of velocity fluctuations at a point near the mouth in the coastal zone.

Best regards.

Manuel MARTIN

Hello,

You just state :

BANCS DECOUVRANTS : NON

are you sure you don't have some ?

JMH

Hello,

I am almost certain. Normally I should not have BANCS DECOUVRANTS, because the entire model is submerged during the modeling. The minimum height of the water level is at least 1 m.

Anyway I'm going to run the test. About the CAS file, are the rest of parameters consistent?

Thanks in advance

Manuel MARTIN

Hello,

In relation to the problems of oscillations in speed results, I tried to make the model with BANC DECOUVRANT but I obtained the same results. So I confirmed that effectively my model is always submerged.

So I decided to make changes in the mesh and the salinity variation. First changing the mesh, we noticed that the previous mesh did not allow modelling well above the speed change.

Second, the previous variation of salinity (4-32 kg/m3) was too strong and that caused the problem of oscillations in the velocity. After making several tests changing salinity variation, we obtained to avoid fluctuations in speed results. I add a graph of velocity oscillations depending on the salinity variation at a point near the mouth in the coastal zone.

Red : 4-32 Kg/m3
Green : 12-32 Kg/m3
Blue : 20-32 Kg/m3
Yellow : 28-32 Kg/m3

However I do not know if I could have another problem types of the advection schemes



Best regards.

Manuel MARTIN