Jump to content
GMS, SMS, and WMS User Forum
dahonegger

Depth-dependent mesh resolution

Recommended Posts

Hello,

I am looking into the unstructured (triangular) grid generation/editing capabilities of SMS, and I have limited time on the trial license for exploration. Can anyone point me to the manual page or a tutorial where I can enforce element size restrictions based upon the material properties? Example material properties might include depth or bathymetric gradient. Or, if SMS grid generation/editing does not have this capability, I would greatly appreciate a comment to that effect.

Many thanks!

Share this post


Link to post
Share on other sites

Check out the tutorial on CGWAVE. It uses the depth to compute wavelengths which are used to control the size of the generated elements. The specific formulations may not be of use to you, but this should give you an introduction to this feature in SMS.

Hello,

I am looking into the unstructured (triangular) grid generation/editing capabilities of SMS, and I have limited time on the trial license for exploration. Can anyone point me to the manual page or a tutorial where I can enforce element size restrictions based upon the material properties? Example material properties might include depth or bathymetric gradient. Or, if SMS grid generation/editing does not have this capability, I would greatly appreciate a comment to that effect.

Many thanks!

Share this post


Link to post
Share on other sites

Hi,

I am following the CGWAVE tutorial, applied to my domain, since I also want the triangles size in function of the bathymetry. However, the mesh generated has triangles which are all approximately the same size (500 m) (I asked in the range of 500-50000 m). Because of that, my mesh has more than 2 billions elements!!

Why is that happening?

When creating the coastline, what does Spacing mean (the tutorial suggests 10)?

Thanks,

Amy

Share this post


Link to post
Share on other sites

I believe the coastline spacing item indicates the spacing of your vertices along the arc that is used to form the coastline.

I would be surprised if SMS could handle 2 billion elements even in the 64-bit version.

I'm not sure why you aren't getting triangles with different sizes. Make sure you have selected to use scalar density paving and have selected the appropriate dataset in the options by this choice for size function dataset. If you continue to have problems, seeing your files would be helpful. Please contact support@aquaveo.com for upload instructions. Tell them to talk to Rusty about the files.

Hi,

I am following the CGWAVE tutorial, applied to my domain, since I also want the triangles size in function of the bathymetry. However, the mesh generated has triangles which are all approximately the same size (500 m) (I asked in the range of 500-50000 m). Because of that, my mesh has more than 2 billions elements!!

Why is that happening?

When creating the coastline, what does Spacing mean (the tutorial suggests 10)?

Thanks,

Amy

Share this post


Link to post
Share on other sites

Hi Rusty,

on fact I found that the mesh generated didn't consider the bathymetry in most of the domain because I was using a period of 20s, as suggested in the tutorial. Then I tried with a period of 44700s (the one of the M2

Share this post


Link to post
Share on other sites

Hi Rusty,

on fact I found that the mesh generated didn't consider the bathymetry in most of the domain (deep waters) because I was using a period of 20s, as suggested in the tutorial. Then I tried with a period of 44700s (the one of the M2 tidal constituent) and it was working! Maybe the 20s period is only applied to shallow waters. What do you think about that?

Amy

Share this post


Link to post
Share on other sites

I guess it depends upon the numeric engine you are using. CGWAVE recommends having 5-20 elements per wave length. We have some utilities in SMS to help estimate wave lengths based upon the parameters of the input 1D wave.

On the other hand, people use scalar density paving with ADCIRC which is a circulation model. The requirements for ADCIRC are quite a bit different as there isn't a well defined function you can compute from the bathymetry. We have a tool in SMS that uses an technique from a professor at the University of Central Florida to estimate what the element sizes should be.

For other numeric models, the requirements will be different. If you don't have well defined criteria for your numeric engine, you can estimate sizes that you think will be acceptable in different areas and build a size scatterset by hand. I would recommend smoothing such a dataset to make sure transitions between sizes are smooth. Looking at the generated mesh, you can get some idea if your mesh is fine enough to get adequate resolution. You can evaluate the quality of your mesh by comparing with calibration data. You can also generate one mesh and then a refined mesh and see how different the answers are in the area you are interested. This will give you some idea if the resolution is adequate.

Rusty

Hi Rusty,

on fact I found that the mesh generated didn't consider the bathymetry in most of the domain (deep waters) because I was using a period of 20s, as suggested in the tutorial. Then I tried with a period of 44700s (the one of the M2 tidal constituent) and it was working! Maybe the 20s period is only applied to shallow waters. What do you think about that?

Amy

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...