intro story D-Flow FM

 

D-Flow Flexible Mesh

D-Flow Flexible Mesh (D-Flow FM) is the new software engine for hydrodynamical simulations on unstructured grids in 1D-2D-3D. Together with the familiar curvilinear meshes from Delft3D 4, the unstructured grid can consist of triangles, pentagons (etc.) and 1D channel networks, all in one single mesh. It combines proven technology from the hydrodynamic engines of Delft3D 4 and SOBEK 2 and adds flexible administration, resulting in:

  • Easier 1D-2D-3D model coupling, intuitive setup of boundary conditions and meteorological forcings (amongst others).
  • More flexible 2D gridding in delta regions, river junctions, harbours, intertidal flats and more.
  • High performance by smart use of multicore architectures, and grid computing clusters.
An overview of the current developments can be found here.
 
The D-Flow FM - team would be delighted if you would participate in discussions on the generation of meshes, the specification of boundary conditions, the running of computations, and all kinds of other relevant topics. Feel free to share your smart questions and/or brilliant solutions! 

 

=======================================================
We have launched a new website (still under construction so expect continuous improvements) and a new forum dedicated to Delft3D Flexible Mesh.

Please follow this link to the new forum: 
/web/delft3dfm/forum

Post your questions, issues, suggestions, difficulties related to our Delft3D Flexible Mesh Suite on the new forum.

=======================================================

** PLEASE TAG YOUR POST! **

 

 

Sub groups
D-Flow Flexible Mesh
DELWAQ
Cohesive sediments & muddy systems

 


Message Boards

Loss of wave height.

EA
Ed Atkin, modified 8 Years ago.

Loss of wave height.

Padawan Posts: 27 Join Date: 3/27/12 Recent Posts
Dear delft3d'ites,

I have a Cartesian grid. there are technically 4 open boundaries of varying size. However land covers the majority of the North and East boundaries so we will assume I have 2 open boundaries, west and south. 2 problems, possibly related.

Wave heights and directions are as per time series along the western boundary, but wave heights are not would they should be in the nearshore. A look at a plot of dissipation shows what you would expect, high values in the nearshore and zero everywhere else. So why are my wave heights so small in the nearshore?

I run the wave module with my western boundary forced by a uniform time series of conditions. I have no conditions for the southern boundary. A look at the vector plot shows vectors heading toward the southern boundary, this starts one cell away from the western boundary, when they should really be heading east like the first vector on the western side. It is an open boundary, it makes sense.
I have tried putting a time series for the first ten cells along the southern boundary to counteract the exodus toward the southern boundary. No joy.
Can anyone suggest how the southern boundary can be structured??

Regards

Ed
SH
Shawn Harrison, modified 8 Years ago.

RE: Loss of wave height.

Padawan Posts: 70 Join Date: 6/25/12 Recent Posts
Hi Ed,

It's hard to tell without seeing a sample of your wave boundary file. Will you please attach the file or just cut&paste the first few entries of the time series here? Also, it would help to know how you are prescribing the wave boundary.. e.g. defined by orientation, or by grid coordinates, etc.

It sounds like you might have the orientation confused, since the waves are going the wrong way beginning at the boundary.
The wave direction asks to be given in "nautical" orientation in the boundary input window... but check in the physical parameters | constants window to make sure that you are not rotating North.. I think if you set North with respect to the x-axis to 90deg, then choose the nautical convention that your wave direction prescriptions should be as you expect.

Good luck! Let us know how you get on.

Shawn
EA
Ed Atkin, modified 8 Years ago.

RE: Loss of wave height.

Padawan Posts: 27 Join Date: 3/27/12 Recent Posts
Hi Shawn,

I have attached two of my test cases (inputs and outputs) in a zip file. I have been thourhg and checked everything I can think of to make a difference, including directional formats (ran a bunch of simplified test cases to ensure I have it right).
The max wave height in the time series in 1.3m, the max wave height I see anywhere on the shore is around 0.8m
If anyone gets a chance to look that would be great.

Attachments:

JH
Jeff Hansen, modified 8 Years ago.

RE: Loss of wave height.

Padawan Posts: 29 Join Date: 7/12/11 Recent Posts
Hi Ed,

I took a quick look at you wave model, I am pretty sure part of your problem is that you essentially have an open boundary to the south but no forcing on it. While swan doesn't really do diffraction, it will try and fill this empty area with energy (in your plot you see the wave vectors heading toward that boundary), so I suspect you need to have something on this boundary. There are two ways to do this, first and the easiest is to just force it the same way as the west boundary. If your area of interest is sufficiently far from the south boundary this will not impact your results, this is what many people do but this often involves moving the lateral boundaries far from the area of interest so the boundary effects do not contaminate the area of interest. A second slightly more complicated approach is to use a simple 1-D model to calculate what the waves should realistically be on the south boundary and then use that as the boundary forcing. Most people use the prior approach and just make their wave domain big.

Also given the variability in depth along your west boundary and that it is all <10 m, forcing that boundary with spatially constant wave conditions will likely not result in realistic conditions at the shoreline. I would suggest moving that boundary into much deeper water.

hope this helps

jeff
SH
Shawn Harrison, modified 8 Years ago.

RE: Loss of wave height. (Answer)

Padawan Posts: 70 Join Date: 6/25/12 Recent Posts
Hi Ed,

Your results look as if the wave energy is dissipating to quickly without actually breaking.
It could be that there is too much friction and the energy is dissipating on the bottom.. did you try decreasing friction?

Also, looking at your wave boundary file, I noticed something (below).. You have no directional spreading.
This could also be a problem.. If I recall I think the scheme doesn't like 0. Try some value here, even if it's small.
My friend Cyprien used to have a magic formula for determining the most appropriate directional spread.

TPAR
20120101.000000 0.300000 8.000000 5.000000 0.000000
20120101.004200 0.500000 8.000000 3.000000 0.000000
20120101.012400 0.300000 8.000000 3.000000 0.000000
20120101.020600 0.700000 9.000000 -1.000000 0.000000
20120101.024800 0.900000 10.000000 -3.000000 0.000000
...


We're excited to hear how you get on..
Cheers,
Shawn
EA
Ed Atkin, modified 8 Years ago.

RE: Loss of wave height.

Padawan Posts: 27 Join Date: 3/27/12 Recent Posts
Shawn,

The lack of Dspd was the problem. As I was lacking an actual value, so I decided to leave as 0. Not good.
I did a couple of tests. First I ran with Dspd as 1 for all conditions. The difference in transmission across the domain is great.
By making the Dspd value 4 (which is the default in parametric/single condition entry in the front end) it’s even better, and closer to what I would expect.
Could do with a little improving but see comparison attached.
I will probably go with Jeffs 1D boundary creation for the southern bnd.


Thanks
Ed
EA
Ed Atkin, modified 8 Years ago.

RE: Loss of wave height.

Padawan Posts: 27 Join Date: 3/27/12 Recent Posts
Hi Jeff,

Thanks for your reply.
Problem solved(ish), see below.
The idea of creating a boundary file with a 1D is most excellent and I did entertain the thought a while ago but some reason discarded it. Hearing suggested from another user is encouraging and I will have a go.
EA
Ed Atkin, modified 8 Years ago.

RE: Loss of wave height.

Padawan Posts: 27 Join Date: 3/27/12 Recent Posts
Just for reference,

I extracted south boundary condtiiosn using a 1D swan run, as you suggested.
This again improves the result. See attached ans compare to Dspd4_station.tif in previous post
Thanks for you comments