open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi everyone,

I am interested in road design and I just stumbeled upon this software, so I am interested if you could help me and answer wether you think TELEMAC 2D can be used to simulate surface runoff on pavements.
This should be a smaller scale model, maybe a few houndreds meter long, but with mesh sizes around 0,2m. It should include only pavement surface and nothing else. And wather depths as small as few milimeters.
Do you think this software could be used for something like that?
Sorry for my english

Thank you!

Hi,

I think the main difficulty would be to model friction with so small water depths. The main friction laws are generally valid for common hydraulic cases in which the water depth are much larger than millimeters. You can check the proceedings from the last TUC where an article presents the implementation of a variable friction coefficient depending on the submergence ratio (water depth / irregularities).

I have no experience of similar cases, maybe some other users?

Runoff can be modelled with the RAIN OR EVAPORATION function which assumes direct runoff (ni infiltration), with:

RAIN OR EVAPORATION = YES
RAIN OR EVAPORATION IN MM PER DAY = your value

It is probably not relevant for your case but I can inform that a rainfall-runoff model will be included in the next version (7.2) to take infiltration into account.

Best regards
PL

Hello,

Several such studies have already been done and seemingly gave good results. Of course, PL is right, friction is an issue, and to make sense it should at least correspond to a roughness smaller than the water depth.

A good point is that the specific algorithms developed for tidal flats in Telemac require no minimum depth (otherwise in the case on rain nothing would move...), however you should try to obey the following criterion:

mesh size < depth/slope

For example a depth of 1 cm and a slope of 1% would give a maximum mesh size of 1 m. This is due to the fact that within an element we consider that we have dry zones when the highest bottom is higher than the lowest free surface, and this criterion may become wrong with very shallow waters on steep slopes.

With best regards,

Jean-Michel Hervouet

Hi Jean-Michel,

Can you describe what is happening numerically if the mesh size does not fulfill the criterion you mention?

It is indeed very difficult to ensure that this criterion is fulfilled for common rainfall applications which normally include steep slopes locally and for which the mesh size cannot be reduced too much.

This hints that regions with steep slopes should be modelled with a mesh size that is a function of (assumed water depth / slope).

Another limitation is that the 2D theory is commonly restricted to slopes not steeper than approx. 10%...

Best regards
PL

Hello Pierre-Louis,

When the criterion :

"highest bottom higher than lowest free surface" within a triangle is reached, we start specific treatments of tidal flats, which may slow down the flow (part of the free surface gradient may be canceled because on dry zones this gradient is the gradient of the bottom and should not be a driving term in the momentum equation). Generally rainfall runoff studies are done in cities or parking lots where the slope is rather small.

With best regards,

Jean-Michel Hervouet

Thank you for the explanation!

There is then no big differences compared with a classic model with tidal flats on the side slopes of a canal for example (I mean, there is no clear limitation for rainfall applications - apart maybe due to the fact that there are more partially wet elements in such models but the influence should be zero on entirely wet elements).

Regarding rainfall-runoff simulations, I actually believe that it is rather commom to have steep slopes since one normally defines the model boundary according to the catchment delimitation and it should be rather commom to have somewhat steep slopes in the higher part of the catchment.

Best regards,
PL