I’m teaching CV118-4 Modeling Rivers, Channels, and Streams With AutoCAD® Civil 3D® 2010 this year at Autodesk University. I spent some time this weekend working on some new datasets and trying out new techniques.
(I’ve taught a similar class in the past: Modeling Floodplains and Designing Stream Corridors with Civil 3D® at AU 2006)
The stream I was working on had two types of a sections. A riffle section with a v-shaped bottom:
And a pool section with a flat bottom and one side that “kicked out” on the inside of the curve.
The engineer who designed the restoration provided a meandering centerline and profiles for both stream bottom and bankfull elevation. I used alignment labels that referenced PVI locations to mark things out in plan. You can snap to lines that you build into these labels, which is handy.
Using the typical sections and my alignment labels as a guide, I created alignments that trace the path of the bankfull location. I used this as a target a bit later.
Something I had always struggled with in my previous stream models was that these targets become longer than the stream centerline. The bankfull profile was drawn on the profile view for the stream centerlines. If you have ever tried to hook an alignment to a profile that was too short, you know that it drops down to zero once it runs out of stations. Plus, it may not match up station numbers properly.
I borrowed a solution from a post I did about road widenings. I made a flat assembly.
Then, I made a quick corridor that hooked the flat assembly along the centerline alignment and the bankfull profile. I made a surface from this assembly and sampled my bankfull alignments against that surface. Bang. I have long-enough profiles. I used these profiles as-is, but if you wanted to you could retrace them with layout profiles or use them as a guide for other aspects of the design.
Now that all of my alignments have corresponding profiles, I simply make a corridor where the bankfull points on a simple assembly target the left and right bankfull elevations.
The result is almost right. If both the riffle and pool sections had v-shaped bottoms with the same side slopes, we’d be done. You can see in the image below that the corridor has a steady bottom section. We didn’t give the lower links anything to target so that they stretched, flattened, etc. This particular combination is difficult because not only does it move from flat to v-shaped, but it also goes from being less than a foot deep to over three feet deep in the pools.
There are a few ways to approach this. I tried about 700 of them yesterday and today. Some of them are more elegant than others. One solution was to make another set of alignments and profiles that follow the path and target them.
I’d rather not have to do this- its a lot of working making those transitions. Yes, once the model is built it gets easy and edits are a breeze, but I know there is some way to build some more intelligence into my assembly so that it “reads” the situation. I’ve been trying things with regions, link-to-marked-point, and more. If you’ve used something that works well for a road or channel project, let me know- I’d love to hear it.
Other posts about streams:
Overlay Analysis in Map 3D 2010 for Floodplain Study
Subscribe
Comments