open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I have set up a hydrodynamic model which looks quite reasonable, but as soon as I like to calibrate to the measured values I experience a time shift of around 1h.
As input I used the same data as and the same times and dates as the one present during measurement. First I thought it could be caused due to different time zones, but I already checked on this and this is not the case. Moreover summer and wintertime are matching.
What could be the reason, that the model is always behind the actual measurements?

Thanks so much in advance.

Hi Mustermann,

The model is only as good as it input!

If I have this kind of problem I go and check all the input i give to the model. It usually means I screwed up somewhere.
Check again all your input and also check the measurement data you are comparing with.


Hope you find a solution soon

Sven

Hi Sven,

that makes a lot of sense, but I already double checked my input and can not find any kind of mistake. Apart from the time difference the model is matching the measured data quite nice. I will check on the measured data again and maybe I need to make some kind of adjustments.

Regards

Hi Sven,

I checked the input and tried to change some physical parameters but I still have a time lag. What is quite strange the the tide compared to the water level change has not time lag but the flux along my measured cross sections has a time lag. I do not really know what could be the problem for that. Do you have maybe some ideas.

Thanks
- Clemens

Hi Clemens,

It sounds like it is only a date issue. Your model is fine nd matches measured data; only this time difference is a problem. So maybe the problem is in the postprocessing? somewhere there must be something that shifts a date or a date has to be set wrong.

If you want I can take a look at your case. You can attach files here or send me a private message. A different set of eyes see different things

Sven

Hi Sven,

attached you can find the model.
First I thought about the a Daylight Saving Time issue, but since the model is located in Iceland that can be the problem because Iceland does not follow the DST since 1967.
I use the flux along the cross section to calibrate the model at the moment I assume that something is wrong with that. Not totally sure how TELEMAC calculates the flux along a cross section. I assume it just integrates over the area and the flux in x and y direction.

Thanks a lot for taking a look at it.

Hi Clemens,

I'm running your case. The input files look normal.
You have output every 12 minutes.
Can you also give me the measurements that you compare with? So I can see the time lag myself.

Sven

Hi Sven,

I thought I had uploaded it, but it seems like it got rejected. Attached is the measurement file for control section 3.

Regards,
- Clemens

Hi Clemens,

for control section 3 you reach a minimum discharge in the model around the same time as the measurement. The shape of the curve of measurement and model are different. Have you tried defining your cross section slightly different? A small difference in angle might give different results. Test the sensitivity on that.
I also saw that your free surface results oscillate a little. Do you have free surface measurements to calibrate your model? Make sure the hydrodynamics are reproduced good.
It could still be a time shift, but according to me it could also be that the hydrodynamics need to be improved and I would check the sensitivity of your definition of the cross section for the results.
Did you see the same result for the other two cross sections?

Sven

Hi Seven,

thanks first for taking the time to look at it.
I have not tested to change the angle, but I will check on that tomorrow, do not have the time today.
Regarding the free surface I saw the same issue, but could not figure out how to get rid of it. I guess maybe continued computing could help a bit. Unfortunately I do not have any kind of free surface measurements to calibrate. The only measurement I have is the flow.
The other cross sections look quite similar so I will check on that regarding the angle as well.

Could you maybe upload a picture of the graphs you got, just that I am sure that we got the same results, since my discharge is a little shifted as well.

- Clemens