open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi

I have successfully set up the telemac2d to model storm surge with variable wind in time and space. But the effect of cyclone is not clearly visible in my results.

When I tried to improve the influence of wind by changing the coefficient its affecting the results in all other parts of entire domain.

When I tried to reduce influence of wind the results with and without wind are almost same.

Is there any option to clearly model the effect of cyclone which influences current speed and surface elevation as such in reality?
Thank you

Hello

You have also the atmospheric pressure which is a major parameter in driving the harricane. Did you try it ?

For your information, we are developping a harricane model using Holland model which will be introudced in the next release of Telemac.

with my kind regards

Riadh

Hi

My atmospheric binary file also have varying atmospheric pressure. So that major parameter is definitely included.

Is there any other option to try in telemac2d itself.

May I know when will be the next release?

If you use the latest version (v7p3) as you did not activate the keyword AIR PRESSURE, you didn't take in account the pressure event if it exist in your binary file...

Hope this helps

Hi

I am attaching the cas file.

Wind data I used ecmwf


Thank you

Hello,

As you're prescribing wind thanks to the binary atmospheric data file, be sure that your file is at the right format. You can find an example in the test case wind_txy (atm_wind_txy.slf).
It can also be useful to print out the wind variables during the simulation so as to check that the model well reads them (VARIABLES FOR GRAPHIC PRINTOUTS=X;Y). Let us know how are these variables during your simulation.

2 other points:
- you're mixing varying and constant time step options. you can't use all the following keywords together: duration, variable time step, time step, number of time step, desired courant number. You have to choice one option (maybe constant time step should be easier), and read the user documentation to set up the right parameters.
Desired courant number value should be lower than 1.
- you set the coefficient to calibrate tidal velocities to 0. It means that all your velocities will be set to 0 on the tidal boundaries. Is it really what you want ?

I hope it helps.

Regards,
Laurent

Hi

Yes. My file format is correct and the file is read while simulation.
For variable time step it is required to give the desired courant number and duration and also I am not using number of time step as per the user manual.

If time step is not specified it takes the minimum time step as 1s inorder to avoid this I am setting the minimum time step as 50s.

Yes I want to calibrate the the tidal velocities to zero.


Thank you

Ok. Just to be sure that your atmospheric data are correctly read, can you output the wind variables, and potentially air pressure if you use it (VARIABLES FOR GRAPHIC PRINTOUTS=X;Y;P), and check if you have the same values in your input file.

For the time step, you have 2 solutions:
- either VARIABLE TIME-STEP = yes, DURATION = duration of your simulation, DESIRED COURANT NUMBER = something lower than 1. There is no minimum time step value as you mentioned, Telemac2D computes the time step automatically.
- or constant time step with VARIABLE TIME-STEP = no (default setting), TIME STEP and NUMBER OF TIME STEP. and then you have to set the TIME STEP as lower to have a courant number lower than 1 (VARIABLES FOR GRAPHIC PRINTOUTS=L) according your mesh resolution and the velocities of the currents in your domain.
You can't mix both solutions has you did in your steering file.

I don't understand why you want to have 0 velocities on your tidal boundaries. Try to comment this keyword.

Regards,
Laurent

Just small moderation of the last post.

In case of variable time-step, the keyword TIME-STEP is used for the first computation step which allow the computation of the Courant number which is then compared to the DESIRED COURANT NUMBER.
Not fully sure but the value could also be used as the maximum time step value...

For constant time step, you could replace NUMBER OF TIME STEP per DURATION. This is, in my opinion, recommended as you don't have to take care about changing it when you change the time-step...

Another precision, if you have the keyword DESIRED COURANT NUMBER with the keyword VARIABLE TIME-STEP = NO, the keyword is not used... But it could exist in the steering file...

And finally, event it's not recommended, you could choose to have a COURANT NUMBER higher than 1 as the maximum is local and not necessarily located at the most interesting area if your mesh have large and small elements...

Regards

Hi

I am sure time step is not causing any problem. Anyway I need to give time step to write output at certain intervals. The velocity I am obtaining for the specific regions in my domain are quite satisfactory for without wind condition. With wind I have some changes in terms of velocity vectors and magnitude which also reasonable. My concern is difference between with wind and without wind condition is not as expected when compared to other models like ADCIRC etc.

I cannot find what is the problem.

Thank you