open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I'm wondering about the erosion generated by Gaia/Sisiphye when the NCO diameter is small (~2mm). The Telemac2d is force by only wave.
Attached the bed evolution after 24 hours as well as the current field and bellow the main options used by Gaia:
=====================================================
BED MODEL = 1
/--
NUMBER OF LAYERS FOR INITIAL STRATIFICATION = 1
LAYERS INITIAL THICKNESS = 2.
CLASSES TYPE OF SEDIMENT = NCO
CLASSES SEDIMENT DIAMETERS = 0.00020
CLASSES SEDIMENT DENSITY = 2650
LAYERS NON COHESIVE BED POROSITY = 0.4
/

---

COMPUTE BED ROUGHNESS AT SEDIMENT SCALE = YES
BED ROUGHNESS PREDICTOR OPTION = 1
/======================================================================
BED LOAD FOR ALL SANDS = YES
BED-LOAD TRANSPORT FORMULA FOR ALL SANDS = 4
/

---

/
SLOPE EFFECT = YES
ZERO = 1.E-12
MASS-BALANCE = YES
/======================================================================
/SUSPENSION FOR ALL SANDS = YES
/

---

SUSPENSION TRANSPORT FORMULA FOR ALL SANDS = 4

/

---

MAXIMUM NUMBER OF ITERATIONS FOR SOLVER FOR SUSPENSION = 500
MAXIMUM NUMBER OF ITERATIONS FOR POSITIVE THICKNESS = 500
/

---

SCHEME FOR ADVECTION OF SUSPENDED SEDIMENTS = 13
MORPHOLOGICAL FACTOR ON BED EVOLUTION = 1
/

---

CLASSES SHIELDS PARAMETERS = 0.047
=====================================================================

regards,

Given the grain size in the fine sand range (0.0002 equals 0.2mm !) and the velocity reaching 1.25m/s or more with strong gradients in the currents, I don't think the erosion (and deposition) are ridiculously large.

Dear Michiel,
thank you for your reply. In fact, I am currently working an other site for which I unfortunately do not have the permission to provide nor the configuration neither the simulation results. So I configured a fictitious site to illustrate the problem. For the real site, I forced the model with a very frequent sea state and the erosion exceeds 2m in several areas. If I decrease the density, or change the morphological factor to 2 which is more suitable for my case study according to the manual, the erosion is accentuated.
regards,
Hafedh

Dear Hafedh,

2m in a day, does seem high. But without the details I won't be able to help you much. The only think I can see is that you use
BED-LOAD TRANSPORT FORMULA FOR ALL SANDS = 4
which is the Bijker Formula, but also
SUSPENSION TRANSPORT FORMULA FOR ALL SANDS = 4
which is Soulsby-van Rijn.
It would make sense to make these choices consistent.
If you use Soulsby van Rijn, using the settling lag option might reduce the erosion/deposition
The only other thing I can suggest is using a stronger (rippled based) skin friction correction.

Michiel,

which seems logical to me, the erosion decreases when:
- the diameter is increased (CLASSES SEDIMENT DIAMETERS)
- the porosity is reduced (LAYERS NON COHESIVE BED POROSITY)
- we increase the height of the roughness (BED ROUGHNESS PREDICTOR OPTION)
What I found is the higher the morphological factor (MORPHOLOGICAL FACTOR ON BED EVOLUTION), the greater the erosion

for your information, the figures I attached in my previous post were extracted from a simulations using the attached steering file:
Hence, the configuration uses the stronger friction correction:
BED ROUGHNESS PREDICTOR OPTION = 2
and not:
BED ROUGHNESS PREDICTOR OPTION = 1
I apologize for that.

Keep in mind that with
MORPHOLOGICAL FACTOR ON BED EVOLUTION = 2
The model assumes that twice as much time has passed as you are setting in the steering file. So if you run 1 day of hydrodynamics, the bed changes are those occurring over 2 days.

Hello,

Apologies for hijacking a thread, but I had a related question. Regarding the keyword MORPHOLOGICAL FACTOR ON BED EVOLUTION, I was wondering if there is any difference with MORPHOLOGICAL FACTOR ON TIME SCALE. It seems to me that setting either keyword equal to 2. for instance would double the speed of the sediment processes for the same hydrodynamic time.

I did find this in the manual:
I assume (i) corresponds to TIME SCALE and (ii) corresponds to BED EVOLUTION? In this case, I'm still not sure what the practical difference between these two settings would be.

Thanks,
André Renault

Hello André,

the description was quite confusing and the 2 morphological factors too. For these reasons you will find only one morphological factor in GAIA's next release. I copy here its description (this will be in the user manual):
"The morphological factor is automatically activated setting the coefficient to apply to the bed evolution equation through the keyword MORPHOLOGICAL FACTOR (real type variable, = 1 by default). This factor allows to accelerate a morphological simulation by modifying the values
of sediment flux for bedload transport and the erosion and deposition fluxes for suspended transport. This accelerator is based on a constant scale factor applied to the divergence operator of the Exner equation for the case of bedload transport [36, 45]. When suspended transport processes are considered, the constant scale factor is applied to the net exchange flux term at the bed-water interface. It is thus available for all the transport modes included in GAIA (suspended, bedload and both simultaneously). "

This factor corresponds to the actual keyword MORPHOLOGICAL FACTOR ON BED EVOLUTION.

Hope this helps.
Best regards,

Sara

I had forgot this. Thank's