open TELEMAC-MASCARET The mathematically superior suite of solvers

Good morning

I am modelling suspended sediment transport using telemac2d couple with GAIA. I have a time-varying input concentration, which I indicate in g/l (I attach the liquid boundary file)

I am printing the mass balance, and when I check the mass in the domain (both what it is in suspension and what have already been deposited) for my last time step, it makes sense when compared with the total mass input. It is less (because some sediment leaves the model) but in a similar order of magnitude.

Nevertheless, this mass is not in accordance with the volume deposition my model is showing (the deposition volume is much much more), so I was wondering how does GAIA make this mass-volume transformation. To show some numbers, I am modelling sedimentation in a reservoir, and over a 3 year period

a) I have an input mass of close to 4.5 million tons.
Input_mass=4.5e9 [kg]
b) Using a density of 2650kg/m3 this means I have an
Input volume= 4.5e9/2650= 1.7 million m3
c) Using a porosity of 0.4, this should take a volume inside my reservoir of:
Volume_Reservoir=1.7 million/(1-0.4)=2.8 million m3

But if I check the bottom elevation change at the end of the model (postprocessing the results), I have a deposition close to 56 million m3.

I have the simplest bed model, without consolidation. So I was wondering what could be my error to obtained so much volume with that mass? Is Gaia considering another sediment density? (but in my case I need a sediment density of rho=4.5e9/(56e6*0.6)=134 kg/m3 to get results similar to what I am getting, which is much much less than what I am assigning (2650 kg/m3).

I highly appreciate any input.

Kind regards,
Eduardo A

As additional information:

1. All my sediment inputs are from cohesive sediments.

2. I have a small layer of 10 cm of non-cohesive sediment as an initial condition all over my domain, just to increase the stability of the bed in the first time steps. But I do not have any input of non-cohesive sediment.

My presumption is that your depositions of cohesive sediment are not consolidated at all (bed model 1) which is why the dry bulk density might be actually as small as you calculated.

Perhaps, a GAIA developer can confirm that cohesive sediments in bed model 1 have such a low initial dry bulk density because they are not consolidated? Otherwise there might be another explanation.


I think it's a interesting question how the deposited mass of suspended cohesive sediments is converted to the volume or thickness of mud deposition.

Best,

Kilian

Hi Eduardo

For cohesive sediment, the volume of the bed is calculated using the user defined dry density of the bed. It does not use the porosity (that is for non-cohesive sediment).

So the volume of the bed is calculated as: VOLUME = MASS / DRY DENSITY.

The density (also called concentration) of the bed is set using the keyword:
LAYERS MUD CONCENTRATION

The default value for dry density for layer 1 is 50 kg/m3. So using you value of sediment input to the model of 4.5e9 kg...

Volume = 4.5e9/50 = 75 million m3

This is getting close to your value of 56 million m3. Presumably some material is still in suspension and some may have left the model through a boundary.

Hope that helps
Tom Benson (HR Wallingford)

Good afternoon Tom

Thank you very much for your explanation! I am really grateful for the help.

And yes you are completely right about the material in suspension or leaving the boundary. Looking at the mass balance in detail, from the 4.5e9 kg that entered the domain, just 2.82e9 kg is sedimented (the rest leaves the domain or is in suspension) which give us the 56 million m3 of sedimentation volume.

I think there is also a part of the cohesive sediment that goes into the interstitial volume between the non-cohesive sediment classes (mentioned in section 3.2 bottom stratigraphy of the gaia manual) but in my case the initial layer of non-cohesive sediment is just 10 cm thick, so this volume is small.

Again, thank you very much for your help!

OK Eduardo, glad to be of help.

Just one thing to point out, the default value of 50 kg/m3 for bed density is rather low when considering the bed to be made up of a single layer. A value of approximately 500 kg/m3 would probably be more appropriate.

Good luck with your study.

Tom

Perfect! Thank you for the advice, I will definitely take it into consideration.