[Luis Partida]

This seems very straightforward in that it is a time step vs cell size issue. I will say this, 15 ft cell sizes are typically never needed. To adhere to the courant conditions formula and just overall modeling guidance that scenario would require a very low computation interval and now your model takes 2 days to run (exaggeration). I would complete a cell size sensitivity analysis to see what you are gaining by using such small cell sizes then begin selecting an appropriate time step.

[Luis Partida]

Hi Chris, is this as steady state or unsteady state simulation? 1D or 2D or coupld?

The first thing I would do is remove the piers and run the model again to see if the problem still occurs. Also, I would check your internal XS’s.

[Luis Partida]

SOLUTION:

Thanks to brainstorming with my colleague Myles….The issue was that the problem bridge(s) had lengths that were shorter than the cells it was connected to.

For example, I had a bridge that’s CL length was 50 ft based on the drawn connector line. The overall spacing was 200x200ft with refined spacing of min 100 and max 150ft. Well RAS does not like this, so rule of thumb is that RAS reports ghost errors in saying “2D Bridge…” and will not identify the actual structure with the issue when the CL length of the structure is less than the cells it is connected to.

[Luis Partida]

@Chris, this is actually a much more complex issue than you have presented.

Geometric Preprocessor HEC-RAS 6.3.1 October 2022
An error occurred while reading 2D bridge

When RAS says “2D Bridge” as the error message than the HDF has become corrupt I believe as the error occurs in preprocessing from geometry to HDF. You can delete bridges one by one until the errors go away but then you still cannot rebuild the bridges. You can even try saving the geometry as which then assigns it a new .gx extension which also will populate the temp .gx HDF then ultimately transfers to the final geometry HDF. But this still does not fix the issue. I am still not sure how to fix this issue and am working with a colleague on a solution. Here are processes I have tried with no success:

-creating a new RAS model and importing the geometry into a new domain
-naming the “problem” bridge a new name (if you have found it…also note that it is random and nothing is actually wrong with the found bridge)
-deleting bridges one by one, then manually rebuild each (you will still hit a wall once RAS creates the problem)
-trying alternate locations of storing the model, diff path names (it was built and ran on my local drive)

This is also specific to bridges. If you add a culvert no error occurs, such that if you find the problem bridge and just convert it to a dummy culvert the model will run.

If anyone has any other ideas please let me know.

[Luis Partida]

Since this is 1D2D, my suspicion is that the results are not reflective of the underlying terrain. What that means is that I would not expect a 1D XS output in the table to match the RAS Mapper results if the 1D XS contains bathymetry not implemented in the terrain. If this is the case trust and document your Standard Table -1 output table results rather than anything from RAS Mapper

[Luis Partida]

I have had many meetings with the Civil Geo guys so I know what it is and never saw any benefits, I also had multiple colleagues tell me how continually buggy it was and typically only worked with example projects. If you have extensive years of experience in RAS, I suggest to do without it as sometimes HEC-RAS can have its own ridiculous bugs. I find no use in their product other than their auto mesh generating based on the underlying terrain, but as I mentioned, it was too buggy to be beneficial.

[Luis Partida]

Do you mean HEC- GeoRAS which was the old method to map floodplains? Or do you mean the alternative software GeoHECRAS. If you mean the first one, the process is different as it is mapping 1D results, if you mean the later, I recommend from an agency review standpoint, that you continue to use Corps products

[Luis Partida]

ST,

That is the point. When using stage as a downstream boundary condition, runoff will only exit the 2D domain if the calculated WSEL is greater than the stage boundary condition applied.

[Luis Partida]

Definitely use Scott’s method, that is WAY easier, and I will be using that from now on too so, Thanks Scott!

[Luis Partida]

When I have had to do this, it always ends up being a tedious Excel and RAS exercise. So here is what i do, i take the Original Cross-section data and paste it into Excel. In the same geometry within RAS select edit >move Points> and extend the cross section to where you want it at. Then within the cross-section editor select the button that cuts the cross section from the underlying terrain. Take the new station elevation data and paste that next to the original cross-section data in excel. From here you need to perform a routine of matching station data and cutting out data from the terrain where there is survey and merge the 2. From here you can re-paste this into the cross-section data editor and have a geo referenced survey/LiDAR cut cross section.

[Luis Partida]

Hi, I think I just responded to your question on Linkedin as well. In this instance that level of detail is rarely needed and would not impact results much by having a pier with odd shapes. Select a worst case scenario and model that. If you truly want that odd shape the use cross sections with Lids and create the entire shape.

[Luis Partida]

No, currently there is no way to obtain a 1D profile style plot for structures within a 2D domain. If you are curious about that information, plot the stage and flow hydrograph of the SA/2D connection and compare stage values over time with the elevation of the structure.

[Luis Partida]

Robert, when you download the HEC-RAS software you have the option to also download example models and data sets. You also can just download the data sets in the location below.

https://www.hec.usace.army.mil/software/hec-ras/download.aspx

[Luis Partida]

Hi Simon, I wouldnt say Oscillating rating curves is an instability as long as the output hydrographs AND stage does not have oscillations. If the hydrograph and stage oscillate, attention should be brought to the computational time step (either decrease or increase)

[Luis Partida]

You have 2 options here, the first is the old school way and that is to add a weir within the 2D domain that is elevated 0.01 (units) above the terrain using a very low weir coefficient of 0.2. The second option is using the new RAS feature called “reference lines” you will see the button at the top of the geometry data editor. This will also now be a part of your output HDF which if you are doing any type of automation with your output; then this is the best method.

[Author]

Posts