open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,
I have been reviewing the micropol module of WAQTEL
and have found what appears to be errors in this module associated with the bed sediment.

No. 1 the bed sediment given in the output as mg/l should in reality based on the parameters used be a deposition rate of g per sec per m2
and
No. 2 the bed sediment is advected and diffused by the ambient flow velocities similar to a solute when in reality it should remain deposited on the bed when not re-suspended.

I demonstrate this by setting zero erosion rate and a high settling velocity in the attached example which should have produced a local deposition only at the outfall but the simulation shows bed sediment load away from the outfall advected by the ambient velocity field.

Regards
Tony C

Further to the earlier observations

there appears to be an error/bug in determining the erosion of deposited sediment in the MICROPOL application.

The sediment erosion subroutine erosion_fx.f
uses an ANTI_DIRAC function from the BIEF Library
The use of this function can promote erosion of negative sediment mass (which appears to occur in wetting and drying applications) and this unfortunately promotes the rapid growth of error in the solution affecting both deposited sediment and suspended sediment

I think it would be better to replace the ANTI_DIRAC() function by MAX() in erosion_fx.f

I attach modified subroutine

regards
Tony C

Hello Tony

Sorry for this late reply I was in an overloaded period with lots of travels and jetlag.

I will check your case as soon as possible. This module has just been added in waqtel and thus scarcely used. The origin of this module comes from an adaptation from a 1D code with lots of broad assumptions such as no advection neither diffusion of some tracers. This was implemented tentatively like you see which could be not in an that efficent way.
So again for your effort to improve this module and look forward to read again from you

with my kind regards

Riadh

Thanks Riadh,
Its great to have these modules for telemac.

I have got around the problem of the false advection and diffusion of the deposited sediment tracer(2) in Micropol option, with a temporary fix by using a private variable in its place which avoids it being advected and diffused and this appears to be working correctly.

Regards
Tony C

Dear Riadh and Tony,

I want to simulate a dispersion of dredge sediment plumes. I am in intership, and the first aim of this study is to use two method for the modelisation of dispersion dredge sediment plumes.

1- Fisrt with the use of TRACER. So with this method we ignor the deposition of sediment, we just look for the dispersion.

2-Second, i will try to use the NESTOR module.

So my question is: Do you think that i am right to use the micropol module to modelise a sediment source as tracers with constant discharge? And the error wich TONY mentionned, is-it in the exemple given in V7P3R1? because i use this exemple as a reference.

Thank you in advance for your help.
BEST REGARDS.
YASMINA.

Hello Yasmina
Sorry for this late reply.
Both of your solutions are OK. However, there is no need fot the use of Micropol. You are not looking for Water Quality but for sediment and matter plume.
To improve the use of Nestor, you can use Sisyphe in order to consider the sediment aspects.

with my best regards

Riadh

Hi,

Thank you for your reply.

Yes you right, with more reflection i had realise that i had no need to Micropol.

For the NESTOR yes i will use Sisyphe.
Thank you for your advices.

BEST REGARDS.
YASMINA.