open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear All,

I am simulating transport and first order decay of a tracer in a stretch of a river. My decay rate is 0.15/day and constant inflow discharge of 8 m3/s and inflow tracer conc of 3.8 mg/l

I am using T2D with wetting and drying activated. I conducted two day simulation with above boundary condition on 24 processors [computational time consumed just 1 hr, what a joy to use the parallel version :)], my problem is that although you can see the flow of water in the main channel. I don't see tracer getting transported. I have done the same with HEC-RAS and I can see tracer making it all the way to the downstream end.

Attached is my case file and user code and a pic displaying the problem. All help is immensely appreciated.

kind regards,
Sumit

Hello,

Beware that the method of characteristics is not mass-conservative (but it is less diffusive).

You might try also :

TYPE OF ADVECTION : 1;5;13
TREATMENT OF NEGATIVE DEPTHS : 2

though it is more expensive...

However the explanation when a tracer stops in the middle of a river is that your :

ACCURACY FOR DIFFUSION OF TRACERS is too high

the default is 1.E-6, if you see that the solver does only 1 or 0 iteration, try e.g. 1.E-10 and see what happens.

Have a nice week-end,

Jean-Michel Hervouet

Hi JMH,

Thanks a ton for your suggestion but the problem still persist.

I increased the coeff. of diffusion to 1.E-11 and still all the tracer decays after 150 m from the upstream end.

Kindly let me know if you have any other suggestion.

Have a great weekend.

Kind regards
Sumit

Hello,

Did you really change : ACCURACY FOR DIFFUSION OF TRACERS

or the coefficient of diffusion as you say ?

Anyway the tracer mass balance perhaps tells us what really happens ?

Regards,

JMH

Hi JMH,

Yeah my bad, downside of trying stuff on Friday.

I increased the ACCURACY FOR DIFFUSION OF TRACERS = 1.E-11 and that did not change the result.

I also tried to simulate the transport without any decay and still the tracer doesn't get transported however you can still see the flow progressing.

I wonder what's going wrong. By the way I am still sticking to
TYPE OF ADVECTION = 1;5;1 this has given me good result with flat bottom channel.

Thanks,
Sumit

By the way for shutting down the reaction I just make the reaction rate = 0 and still have DIFSOU subroutine in the user program.

Thanks,
Sumit

Hello,

So with no reaction rate the listing will give interesting data: quantity of tracer released, quantity of tracer in the domain, quantity lost. The latter may be important with the method of characteristics, this is why scheme 13 would help understand. As we have no proof of mass conservation with characteristics it is more difficult to analyse the problem, and this may be the problem itself...

Regards,

JMH

Hello JMH,

TYPE OF ADVECTION = 1;5;13 did the trick for me. Now I can see the tracer getting transported as well as decayed with the flow.

Thanks a ton for all your help.

Kind regards,
Sumit