open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everyone,

Could someone tell me what TELEMAC2D does when it has as input a discharge and water depth level?

Attached you can find a picture of the mesh of my simulation (BOTTOM_mesh).



In my simulation I have 6 open boundaries: 5 inputs and 1 output.
I would have liked to have as inputs 5 hydrographs, one for each input boundary condition, and one fixed water level for the output section. Because after a few iterations some of the boundary conditions became dry (in the forum it was told me that it happens probably because the flow is torrential), I decide to set as input, for these boundary conditions, both water depth levels and discharges. The results are in picture (BC).



I created a stage-discharge curve for the sections with given discharges and water depth levels. Each curve it has been created with the hypothesys of uniform flow, large section, etc.
The results are into file hydro.txt.




Then, I decided to check the trend of discharge in time by creating 5 control sections immediately downstream the boundary conditions. The results are in pictures (Input2, Input3, Input4, Input5, Input6).

Where:
SezInput"N" = the recorded discharge by the control section "N" located a few meters downstream the boundary condition "N".
Q("N") = is the input discharge of section "N".












As you can see, for the input sections 4 and 5, that have only a hydrograph as input, the discharges in input are very similar to the ones recorded a few meters downstream.
For sections 2,3 and 6, the ones with given discharge and water depth level, the discharges given in the file "hydro.txt" are different from the ones recorded in the downstream sections. This is what I don't understand. If someone can help me it would be really appreciated.

Thank you in advance,

Daniele

Hello,

It is not correct either to impose depth and discharge on a fluvial boundary, this may degrade a lot mass conservation since continuity is replaced by the imposed depth equation at the entrance. Conditions which do not respect the regime inside the domain will simply create a jump. What you need to know is whether these boundaries should be normally fluvial or not. If they are fluvial, in order to keep them fluvial a trick is to put a lower bottom at these entries so that the depth always ensures a fluvial flow, whatever the fluctuations.
If conditions may be both fluvial and supercritical at a boundary it would be better to change the type of boundary conditions in subroutine bord.

With best regards,

Jean-Michel Hervouet