open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I have a trouble with Sisyphe and the reading of the hydrodynamic file.

I am using a simple flat bottom with initial water depth of 1 m, one boundary condition with prescribed water level at 2 m and velocity 0.2m/s.

Either with the uncoupled or coupled mode, I have the following message when Sisyphe starts:
"variable velocity V is in the file but will not be read" and same for velocity U, water depth and bottom.
Then I get "Negative depth prescribed on the boundary, check your specific subroutine bord3d" and the program stops.

First, is there an error in the message that refers to "bord3d" whereas I am using telemac2d?
Then, I don't understand this negative depth, I have tried any possible boundary condition and initial conditions and it does not seem to be related to that.

If somebody has already had a similar error aor/and has some ideas about where it comes from I would be happy to get your help.

Thank you,
Caroline

Hello,

I would first advice that you use the coupled mode, which is always better.
Then the error message comes from the subroutine positive_depths in library bief. I thought this problem could happen only with Telemac-3D, so I'll have to change the message when we have found what is the problem.

It happens that the subroutine positive_depths stops because a prescribed depth (in HBOR) is negative. It may happen e.g. if you prescribe a constant elevation at 0. and have a bottom at an elevation greater than 0.

As the subroutine positive_depths is also called by Sisyphe to solve the Exner equation, I would hint that HBOR is in fact EBOR in Sisyphe, for some reason (this should normally checked or clipped). So you should check why the sediment height is negative.

If you find: tell us and we'll try to create an error message that is more informative.

If you do not find, we'll try to have a look at your case and see what happens.

With best regards,

Jean-Michel Hervouet

Hi,

I have eventually found out where the problem was. I was using the definition of NON COHESIVE BED POROSITY given in the user manual v6p0, with a value 1.6, instead of the new definition of v6p1 which corresponds to a value of 0.6. This gave me a negative value of CFS_SABLE then a negative depth for the bed layer.
The message error was called by sisyphe > bedload_main > bedload_evol > bedload_solvs_fe > positive_depths.

While checking thism I have come across a few definitions that were unclear for me, either in the manual or in the fortran files. I enclose a file with my comments just to let you know in case it can help you improving it.

Thanks for your help,
Caroline