open TELEMAC-MASCARET The mathematically superior suite of solvers

HI everyone

I do not understand the difference between these 2 variables, the manual says:

M = "bed-load discharge (m2 / s)";
QSBL = "bed load transport rate (m2 / s)";

Another question, I already have my runs getting the QSBL, somehow I can get the QSBL in direction "x" and direction "y", to then be able to integrate it into a section?

Thanks

Joaquin

Hi Joaquin,

I am not an expert of Sisyphe, but it seems to me that the variables M and QSBL are the same in sisyphe. I hope someone will confirm that.

Concerning your second question about "how to compute bed-load discharge" from QSBL or (QSBLX and QSBLY), a tool we developed can perform that computation, see another thread dedicated to them: www.opentelemac.org/index.php/kunena/scr...ost-processing#27759 and github.com/CNR-Engineering/PyTelTools/wiki/Compute-flux.

Regards,
Luc

Hello Luc

I have checked what you tell me. I have reviewed the table of the second link and shows how to obtain the total flow, I would like to obtain the solid flow in x and y (QSBLX and QSBLY) from these parameters:

"U,V,H,S,B,E,QSBL,F,Q,TOB"

Thanks

Joaquin

Hi Joaquin,

After looking at sisyphe.dico in french, it seems that M corresponds to "total solid discharge", where QSBL stands for "bed load discharge". Therefore, if you do not consider suspension, they should be equal (and have same unit). It is part of some inconsistencies in variable names, description or units in sedimentary modules (and sometimes between modules and/or between english/french).

Computing accurately a solid discharge with "U,V,H,S,B,E,QSBL,F,Q,TOB" is not simple. If your cross-section is perpendicular to the flow, the simple integral of the scalar QSBL is probably a good approximation.
Nevertheless, if you want to compute the solid flux with a projection on the cross-section, you have to know the flux direction.
You can assume solid flux follows the mean flow velocity, you can compute the components of QSBL on X and Y directions (named QSBLX and QSBLY).
Quoting the sisyphe documentation, it is not true in the following cases :
If QSBLX and QSBLY were not asked by the user, you have to recompute them with this latter assumption.
Recomputing vector components is now possible in PyTelTools (since my last commit, see list of implemented equations in this document), within the Extract variables in classic interface (named Select variables for workflow).
I did not tried it extensively, but I hope it works properly on any data (the tool handles the zero divison case, and sets the component to 0). I would be interested in a comparison between the 2 methods, based on your case, to know how much the solid discharge computed with only QSBL (ie. without projection) is over-estimated.

PS : If you still have a problem with Froude number computation in sisyphe (see this thread), you can also compute this variable with this tool (from U, V, H existing variables).

Regards,
Luc

Hi,

Although computing transport vectors is possible from the available information (QSBLx=QSBL*U/(sqrt(U2+V2)), this will not account for any slope effects.

Michiel

Thanks for your comments Luc y Michiel.

Finally, I think the best approach is to use projection with respect to velocity. With respect to the approach of integrating the scalar flow with the perpendicular cross-section is quite bad (I have done with the flow of liquid some time ago).

regards

Joaquin