open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi all!

I just set up a T3D model with a suspended sediment as source inflow.
I specify 0.85 as inflow concentration but the concentration read from the nearest point from the source only shows 0.25 kg/m3.

I was expecting to have a value close from the specified one since I chose the x,y,z coordinates of my source so that it is located very close from a mesh point.
Is the volume around my inflow node have any role on the concentration value?

A last question: SEDI-3D does not allow to define several sediment classes, right?

Here is my sediment set-up:

/

---

/
/ SEDIMENT
/

---

/
/
SEDIMENT = YES
COHESIVE SEDIMENT = YES
/
NUMBER OF TRACERS = 1
NAMES OF TRACERS = 'SEDIMENT KG/M3 '
STANDARD VALUES FOR TRACERS = 0.
INITIAL VALUES OF TRACERS = 0.
COEFFICIENT FOR HORIZONTAL DIFFUSION OF TRACERS = 1.E-6
COEFFICIENT FOR VERTICAL DIFFUSION OF TRACERS = 1.E-6
TRACERS VERTICAL PROFILES = 1
/
ABSCISSAE OF SOURCES = 1797.7407226562500
ORDINATES OF SOURCES = 604.31365966796875
ELEVATIONS OF SOURCES = 0.6 / plane 11, ca 1 m under vy
/WATER DISCHARGE OF SOURCES = 1. / cf. source file
/VALUE OF THE TRACERS AT THE SOURCES = 0.85 / cf. source file
/
DENSITY OF THE SEDIMENT = 2650.
CONSTANT SEDIMENT SETTLING VELOCITY = -0.00128
MEAN DIAMETER OF THE SEDIMENT = 0.00006
CRITICAL SHEAR STRESS FOR EROSION = 1000. /no erosion / 0.2
CRITICAL SHEAR STRESS FOR DEPOSITION = 0.2
EROSION COEFFICIENT = 0.000 /no erosion
INFLUENCE OF TURBULENCE ON SETTLING VELOCITY = NO
CONCENTRATION OF FRESH DEPOSITS = 160 /default 160
/
DENSITY LAW = 0
/
SOLVER FOR DIFFUSION OF THE SEDIMENT = 7
MAXIMUM NUMBER OF ITERATIONS FOR DIFFUSION OF SEDIMENT = 2000 /60
ACCURACY FOR DIFFUSION OF SEDIMENT = 1.E-6
/


Thank you in advance!

Best regards
PL

Hi PL
I not an expert at sediment modelling using telemac3d but in the above CAS file the tracer concentration is commented out as is the discharge perhaps you have done this while testing after you found the anomaly in the concentration.

In any case I would have thought that mixing and transport would immediately reduce the concentration as would settling given that you are inputting it close to the bed.

regards
tony c

Hi Tony,

Thank you for your reply!

As mentioned in comments in my cas file, source discharge amd concentration are provided through a SOURCE FILE.

Regarding the local concentration, I agree that advection-diffusion and settling should influence my input value, but I was (wrongly?) expecting to observe a concentration much closer from the input value when looking at the node where the source is applied (according to info provided in the log).

My run is not mass conservative. In this test i use MOC for advection of tracers. Maybe i should change to MURD 14?

Best regards
PL

Hi PL,
yes certainly the MOC method is not conservative and could explain the differences that you are observing. I have previously used the MOC for tracers and found large error and variation in the mass conservation.

regards
Tonyc

Hi Tony!

Another question: did you use a same advection-diffusion scheme (MURD 14?) for both currents and tracers?

I ran the hydrodynamics with MOC so far. It would be great if I could keep it for velocities and change to 14 only for tracers.

Best regards
PL

I used the default MOC for hydrodynamics and the conservative scheme 14 for tracers as recommended by JMH

regards
Tony

HI PL

possibly use the following
TYPE OF ADVECTION : 1;5;14
TREATMENT OF NEGATIVE DEPTHS : 2

regards
Tony

Hello! Could you give me your Source File to know how to make one.
Thanks!