open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everyone.

I am currently discovering GAIA, and I struggle a bit on the solid discharge.
I am calculating a discharge implemented through the Soulsby-Van Rijn formula, without waves, coupling with Telemac2D in which I am only giving tidals conditions through the TPXO database. By doing so, I want to visualize and understand the behaviour of the sand movements.

However, I observed that whatever I am doing, it seems only the Bed Load discharge is taken into effect. Inded, here are the facts leading me to this conclusion
I wrote at first in my Gaia .cas file
BED LOAD FOR ALL SANDS : YES
SUSPENSION FOR ALL SANDS=YES
BED-LOAD TRANSPORT FORMULA FOR ALL SANDS : 5

The Soulsby-Van Rijn formula is supposed to calculate both supsension and bed load, but the results were that my solid discharge was exactly the same than my bed load discharge. Moreover the concentration for my sand was equal to 0

I thought that maybe I had to force the formula for the suspension only? so I did this:

BED LOAD FOR ALL SANDS : YES
SUSPENSION FOR ALL SANDS=YES
BED-LOAD TRANSPORT FORMULA FOR ALL SANDS : 5
SUSPENSION TRANSPORT FORMULA FOR ALL SANDS :4 (which is soulsby Van rijn also, calculationg only the suspension term)

The results were that my disharge didn't change at all, but my bed evolution were giving huge wrong values (like + 9000 m )

I tried then to calculate only suspension by writting
BED LOAD FOR ALL SANDS : NO

and the results were still diverging if I used the SUSPENSION TRANSPORT FORMULA FOR ALL SANDS, and the discharge were equals to 0 if I used only the BEDLOAD TRANSPORT FORMULA FOR ALL SANDS, but the bed evolution was still moving??

I precise that by doing the calculations on some cells thanks to my resuslts from the Telemac2d .slf file, the velocity were superiors to the critical one from the Soulsby Van Rijn formula.

Does anyone know what's happening? Also why is their a visual Graphic printout for Bed Load (QSBL) and the QSSUSP isn't their anymore?

Thank you

Hello Youenn,

I am not a sediment transport specialist nor a GAIA specialist, but which release do you use? If it not the latest, you should use this one as many improvements and bug fixes have been done since the first release of GAIA in 2019 (v8p1).

Chi-Tuan

Hello Pham

Thank you for your reply.

Indeed, I started at first on the V8P1 version, on which I had some bugs. That's why at the time I wrote, I was already using the v8p4 release. In fact, the bugs corrected between these two configurations seams to be more about the graphic printout than the results themself, because the weird results on discharges were exactly the same.

Jan Youenn

Hello everyone
I have an update.
By looking at the process by which gaia is doing the calculations, it seems that:

Telemac2d_init.f is calling gaia_init.f
-> gaia_init.F is calling init_transport_gaia.f
-> init_transport_gaia.f is calling bedload_formula_gaia.f
-> depending on the case, bedload_formula_gaia.f is calling the proper file for calculation, in my case bedload_soulsby_gaia.f This file is then calculating properly the suspended and bed load transport rate.

BUT: at the end of the init_transport_gaia.f (so after all the calculations at this time step) we have the line

CALL OS('X=Y ',X=QS,Y=QS_C)

which is copying y in x, so the bedload transport rate in the total sediment discharge??? not taking into account the QS_S (suspension transport rate)???

Can someone tell me if I'm wrong? because if not, the QS printed out cannot take the suspension into account i think.

I atached the file init_transport_gaia.f and the other so you can see.

That is correct. We are working on a solution to add suspended fluxes, but atm QS is the bedload transport only.

Thank you very much mafknaapen I was struggling a lot. I made a mistake indeed, It was the line X=QSCL%ADR(I)%P,Y=QSCL_C%ADR(I)%P) that I meant when I was talking about the transport rate as the unique bed load rate.

Assuming that this replacement is made at the end, what can I deduce about my bed evolution (in m)? Is it the real result of suspension+bed load or only the result of bed load as the final discharge doesn't take into account the result from the suspension part?

Jan Youenn

The erosion deposition calculations do take the suspended load into account. There is just no output variable for the suspended transport rates.

Thank you very much, I think this post can be closed now, I exactly have the answer to my answer

Goodby all
Jan Youenn

Hi everyone.

After a long standby, I choosed to work again on my case. By taking into account the suspension were calculated but not shown, I make some few arrangement in the fortran code in order to visualize my supension+bedload transport. It seems to work perfectly, as I have some values that match perfectly.


But, are you really sure the bed evolution is not only shown for the bedload effect? In my case, 94% of the transport are done through suspension, yet my bed evolution is not at all as it should be, the deposit are not corresponding. I attached a picture showing the bed evolution (blue is sand gained, red is sand lost)


It really seems like the bed evolution, as the transport, are showing only bedload effect

Picture showing bed evolution