open TELEMAC-MASCARET The mathematically superior suite of solvers

I am running 2D simulations of the North Sea. The tidal signal is imposed at the shelf break using TPXO database and the subroutine TIDAL_MODEL_T2D from the fortran file of test case 123_tide. It works very well if no other liquid boundary condition is dealt with.

However, the Scheldt Estuary is included in the domain and I would like to impose the water elevation at the upstream part of it, using a liquid boundaries file (the water elevation is supposed spatially homogeneous there, but it still varies a lot in time).

The problem is that, apparently, calling the subroutine TIDAL_MODEL_T2D has the effect to override the information of the liquid boundaries file, and the water elevation remains 0 in the vicinity of the upstream boundary (because there is no TPXO data over there, I guess).

It is not surprising because, using my setup, there is no way for Telemac to understand on which boundary it must apply TPXO and on which it must apply the time series.

I would be pleased if someone could help me on this.

Update (my first post was not very clear about the way I impose the tide).

I am running 2D simulations of the North Sea. The tidal signal is imposed at the shelf break using TPXO database, with the keywords:

OPTION FOR TIDAL BOUNDARY CONDITIONS : 1
TIDAL DATA BASE : 2

It works very well if no other liquid boundary condition is dealt with.

However, the Scheldt Estuary is included in the domain and I would like to impose the water elevation at the upstream part of it, using a liquid boundaries file (the water elevation varies in time).

The problem is that, apparently, using the above keywords overrides the information of the liquid boundaries file, and the water elevation remains 0 in the vicinity of the upstream boundary (because there is no TPXO data over there, I guess).

It is not so surprising because, using my setup, there is no way for Telemac to understand on which boundary it must apply TPXO and on which it must apply the water elevation data.

I would be pleased if someone could help me on this.

Update

From the user manual:

The keyword OPTION FOR TIDAL BOUNDARY CONDITIONS activates the use of one of the available database when set to value 1 (default value is 0, meaning that this function is not activated). When this keyword is activated, every boundary is treated using the prescribed algorithms except the boundaries with prescribed flow rate.

I guess I just have to find appropriate data...

However, if someone could come up with a solution to impose TPXO at the shelf break and water elevation at the upstream river boundary, I would appreciate it.

Hello,

You are right, if you use the harmonic constants coming from Oregon State University such as the TPXO global solution by activating the keyword OPTION FOR TIDAL BOUNDARY CONDITIONS, every liquid boundary with boundary code like 5 4 4, 5 6 6 or 4 6 6, are treated as if you want to prescribe tidal boundary conditions. Your case, with a river boundary, was not taken into account with this automatic calculation of boundary conditions for tide.

Nevertheless, can you tell me if you only have to prescribe water elevation for your upstream boundary which global nodes are number 9, 1985, 1986, 1987, 1988, 10 (last column with number 1797 to 1802)?

If yes, you can try the FORTRAN FILE for your specific case: tidal_model_t2_for_OGourgue.f

Hope this helps,

Chi-Tuan

Thank you for the answer. I had to modify slightly your fortran file because node id cannot be < 1797 AND > 1802 (change AND by OR).

But it works now!

However, this hardcoding solution is not very elegant, and could be annoying when changing mesh or adding new rivers.

Would it be possible for the developers to offer another solution in a future release?

For example, the keyword OPTION FOR TIDAL BOUNDARY CONDITIONS could be defined like this:

Type: Integer
Dimension: Number of liquid boundaries
Default value: 0

This would offer the possibility to the user to choose on which liquid boundary to apply a tidal BC, and on which to apply what is defined in the .cli file.

Anyway, thank you very much for your help pham.

Hello,

Your are right for several points:
- sorry for the first point (.OR. rather than .AND.), I have not thought so much for your problem,
- the suggested implementation was very specific to your case because your case was easy, but it was the simplest way to implement to solve your specific case in my opinion, but I am aware that this solution is not generic, is mesh dependent and cannot be easy to implement for several rivers,

I have to think of the most efficient and elegant solution for this case. It may be another keyword as you suggest. I will discuss with other users of TELEMAC who may model river and sea flows at the same time (or even some specific waterworks) and the solution should be implemented for the next version.

Thanks for your remarks,

Chi-Tuan

Hello,

I am encountering same issue and wondering if the new version has added this option.

Thanks.

Qinghui

Dear Qinghui,

I think we can do it since version 7.1.

See the reference manual:

Hey,

That's cool! Thank you.

Qinghui