open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello Chi-Tuan,

Thank you very much for your reply. I managed to translate all manual on Tidal Model Simulation from french to english and have better understanding how it works. I used Google translator so some phrases do not make sense. But if Telemac users are interested I could provide you with it and after some professional touch it could become useful.

In terms of JMJ data, I am afraid you are right as after including the mentioned files, it is still asking for Harmonic file. It does not generate the file itself, unless I am doing something wrong there. I am battling with it for few days now and cannot understand why it is not working, maybe because the data does not cover that area, but error message does not indicate so. I also have TPXO data, but really wanted to try both to see if I get any close results. I thought that with JMJ data it would be much better as I can see from Harmonic file that tidal constituencies are applied on each node on open boundary, but with TPXO file it is not so visible, so it is mystery to me how TPXO works.

Thank you for coming back to me on Initial Conditions, I will try to make changes there. Thanks.

Kind Regards!

Violeta

Hello again,

Which version of TELEMAC-3D are you using? Version 6.3 release 2? With Python scripts or Perl scripts?

First, you have to create the harmonic constant file, even empty, otherwise the computation cannot start (I suppose you did right).
A few weeks ago, a colleague of mine (yugi) and I discovered a bug with files that must be read-write, like the harmonic constant file, with the Python environment. It is created in the temporary directory (with the name T3DHAR) but not copied one level upper at the end of the computation. A temporary solution is to run telemac3d with the option -t, which does not delete the temporary directory. Then you can copy the T3DHAR file in the current directory with the expected name, and that is it.

I do not know if yugi has changed the trunk or not.

Hope this helps,

Chi-Tuan

PS: do not forget to play with the XSHIFT or YSHIFT variables in tidal_model_t3d subroutine as explained in the manual, in particular if you have some warning messages when interpolating the constants.

Brilliant, thank you very much Chi-Tuan. I will do so. I am using in parallel version 6.3 with Python. I will run the case with -t option and will check the temporary directory. Actually, to generate the harmonic file, I wanted to practice on validation case tide from example/telemac3d directory. In that file I did not change much as XYshifts already there and BETA0 as well. But I will try -t option when I run and see if it is working. Thanks again.

Kind Regards!

Violeta

Hello Chi-Tuan,

OK, well received,

from somewhere near Guangzhou (Canton), monsoon starting here, all OK.

JMH

Hello,

I see it is spoken about TPXO and JMJ tidal modelling here. I am precisely trying to compare both on the "tide" example case in Telemac2d default files.
Please tell me if this was already addressed in another past topic in the forum (and sorry then).
Otherwise, here are my questions:

1/ For applying TPXO conditions, should the XSHIFT, YSHIFT and BETA be set to 0 in the Fortran file? (I don't think they are read actually but just to make sure)

2/ Are there similar problems of shifting coordinates in the TPXO database? or is it fine anytime?

3/ In "tide", should the open boundary conditions be changed for TPXO (i.e., NOT 566 but 544 or 466) in order to get consistent results?

4/ When one wants to use JMJ for any area covered by the JMJ database, is there any nice way of determining XSHIFT and YSHIFT prior to running T2D? (for instance, using Blue Kenue or other)

If there are other important points that need to be stressed for such a comparison JMJ/TPXO, please let me know.

Thanks a lot in advance,
JBS.

Hello JBS,

I am using TPXO files to run telemac3d. I probably can answer only one or two of your questions as my area is outside of JMJ database, so I could not get good results there.

1) For using TPXO I do not need XSHIFT, YSHIFT and BETA0 options, they are all 0. I had some problems at the beginning, but since I changed my coordinates domain to UTM and picked correct UTM zone in the steering file, it all worked very well. I am not aware if there are similar options for TPXO.

2) Also in terms of Initial Conditions you do need to stick to TPXO only, you still can use other options, Constant Elevation in my case, even with TPXO file. In boundary conditions I have 566 and 5 (I use tracers as well in my case).

I hope somebody can advise you on JMJ data. I wanted to run them both and compare but I think that JMJ database does not cover whole North Sea f.e. as I wanted to use it for that area as well.

Good luck!

Kind Regards!

Violeta

Hello 716469,

I am trying to run a tidal model, but no luck so far. Can I ask your advice on the following points?

1. I am using TPXO for my tide database. The files needed for binary database 1 and 2 for tide are h_tpxo7.2 and u_tpxo7.2, am I right?

2. Does it really matter how many open boundaries I have in my model? I have set by BC to both 544 (H) and 566 (UV and H), and it is still not working.

3. After I put in the command in DOS terminal, the message stopped after " END OF TPXO INITIALISATION" . I have no clue why it wont continue running.

Thank you

Hello,

I am not sure what problem might be, but I will try to answer your questions:

1) yes, you are right by using binary data file as u_tpxo7.2 and h_tpxo7.2

2) I am using two open boundaries in my domain (North Sea) and it is working fine. The setting at open boundaries I have as 566 and 4, as your binary tpxo data files contain tidal velocity and amplitude data.

3) "End of tpxo initialisation" is not an error and normal process of run with tpxo files,but I do not think that your run has started to use the files.

Few things to check:
1) Check your coordinate system, as far as I remember TPXO uses UTM system, so make sure it is set up as UTM in cas file and your bathymetry has correct UTM coordinates. I had this problem too, as my bathymetry was in British coordinate system and I had to convert it into UTM and then it worked.

2) Check that fortran file has 1 on NODALCORR field.

Also as I have seen on some posts that we do not need to use Harmonic Constants file if we are using TPXO. It works fine for me without harmonic file.

I have attached my steering and fortran files as examples. I hope it will help.

Kind Regards!

Violeta

Thank you so much Violeta for your help!

I hope you dont mind me asking a few more questions regarding the model

1. regarding the binary data file (u_tpxo7.2 and h_tpxo7.2), do I need to extract the data to fit the domain that I am working on, or can I just take the whole file (without data extraction) and let Telemac extrapolate the points at the boundaries?

2. As for coordinate system, I am a bit confused as well. I used Geodas coastline extractor (from NOAA) for my coastline points, then convert them into lines. The coordinate used was in geographic coordinate system. Thus I had to change the projection to UTM in Blue Kenue. My bathymetry was from GEBCO 08.

everything works fine in BK after I change the coordinate projection to UTM. However, when I imported the geometry and BC file into Fudaa, I got the following warning "Cartesian grid is not supported".

Could this be the problem for my model? But, from my understanding, that was a warning, and not an error.

3. And, do I need fortran file for my model? I have seen that some users did not include fortran file in their steering file, so I am not really sure the need of the file.

Again, I really appreciate your guidance!

No problems at all. I went through various problems myself until I made it working.

I do not think that BK actually convert the geographical coordinates into UTM, I have tried it too. You need to do it manually with help of coordinate converter. There are many on google you can find. However you probably will need to use batch converter (well not point by point manually, but if you could upload whole file with coordinates and software will convert them all). I had many points to convert but I had problem to find batch converter, so I used below, but you might be lucky and get some free software to do it. Delft Dashboard can do it as well.

www.uwgb.edu/dutchs/usefuldata/utmconversions1.xls


I think GEBCO gives you bathymetry in lat and long so I suspect your problem is because tpxo files are using UTM then try to convert all coordinates into UTM and see if it is working.

In terms of TPXO file, no you do not need to extract your area, use whole files. TPXO files have data for extended area for example North Atlantic or another is just for Europe. Telemac extracts the points for your domain that is why you need to have match between TPXO coordinates and your case, so both need to be in UTM. Also I think you need fortran file as there is an option to use TPXO in the code. You do not need to change much there, just one option I think, the one I mentioned on my previous post.

Good luck!

Violeta