RAS 2D Terrain Accuracy

[JakeNCD]

Hello, I am working with a large, 1 foot detail lidar data set and have processed the data to display in two separate ways, one as contour bands (a 3D topo map) and the second as processed lidar data (lots of tiny pyramids) shown in images below. The small flow accuracy is important on this project as it is a restoration based project utilizing low flows. Is one of these styles of terrain data preferable to the other in the way 2D analyzes the surface and the geometry data? Do the thousands of tiny pyramids artificially increase roughness? Or do the flattened countour bands artificially smooth the surface too much? Does the manning’s ‘n’ value need to be adjusted to account for the artificial lengthening or shortening of the ground surface depending on the terrain type? Thanks for the help and sorry about the wordy question.

[Scott Miller]

Jake. Don’t worry about the texture you see. 2D HEC RAS reads cell volume and face cross-section from the terrain, not the roughness. You can use a land-use layer for roughness or digitize it in manually.

One thing to consider is whether the channel topography shown is the dry bed or the water surface. You may need survey to get the channel bottom.

My understanding is that the flattened contour bands would mess with the rendition of inundated areas in 2D. Consider that a sloped water surface will cross those steps.

Are you planning to use full 2D, or use 1D for the main channel?

[JakeNCD]

Thanks Scott!

That makes sense, I have added land use layers for the roughness. My only other concern with the “rougher” terrain is an increased surface area that the model would apply the manning’s values to. Is the extra area of all of the sides of the little pyramids a problem? Or is the increase so small that I’m over analyzing it?

The channel bottom was dry at the time of the flight, it is generally flat and sandy bottomed.

I agree with your point on the inundated areas, after running both terrains in the same model the flat banks fill with water more quickly than the “rougher” looking terrain.

I am planning on using a full 2D model only for the main channel and floodplain.

[cameron]

The second terrain is what you will want to use. Terrain is not flat like you have for the first terrain and will for mess up the mapping results. Each cell face is a cross-section and applies the n value the same way the 1D models do.

You are over thinking it.

[Lonnie A]

I agree with Cameron, it looks like the first surface was converted to integer which is why it is flattened between contours.

[Author]

Posts