Tortuous River and Overbank/Channel Downstream Distances
Thank you in advance for any insights offered in response to this post.
I have a couple questions and first I will provide the background. I am modelling a very tortuous river (meandering) over a broad floodplain stretching for about 30 km. I would like to think that I'm an intermediate HEC-RAS user. That said, I question my understanding of the definition of Overbank and how the downstream distances are measured. In this very tortuous, low slope river my cross-sections are spaced fairly widely apart (700 m) based on Samuel's equation. In some places the meanders are so "intertwined" that it's difficult to have a cross-section through some areas as the floodplain portions would have to "bend" significantly through the flood plain.
1. Are the Downstream Distances for LOB, ROB and Channel meant to be the straight line distance to the next downstream cross-section or should they follow the lines of the river (channel edges and thalweg)? Straight line (or crow flight) distances from point to point would capture how the channel should look in schematic but wouldn't capture the potentially significant distance if there are one or two horseshoe bends in the river in between. Distances along the banks and thalweg would capture the longitudinal distance but will the schematic still appear correctly? Have not had to model something this large before.
2. Should "horseshoes" and other meander patterns that block the proper alignment of cross-sections through the flood plain be "skipped" or is it reasonable to bend a cross-section over the floodplain (for what feels like a significant distance)? I know in other posts the term "center (or centroid?) of flow" has been discussed by I feel like I'm really stretching the term "cross-section" in this case.
Re: Tortuous River and Overbank/Channel Downstream Distances
The key thing to remember for cross sections is that it is a 1-dimensional model.
Ideally, you want the channel length to represent the distance the water flows in the channel and the overbank length to be the distance that that water flows in the overbank.
If most of the overbank flow is going straight "as the crow flies" then this is the length you would want to use. If the overbank flow curves around generally following the channel, then that is the length you would want.
It is possible that you might have considerably different flow length at low flows versus high flows. Then it is just a matter of engineering judgment deciding what length to use. And if this is a really serious problem, you might want to consider going to 2D flow.
Having weird, "doglegged" cross sections is not a problem in and of itself. The thing to keep in mind is that each cross section has a single energy grade and that you generally want to draw the cross sections such that the flow is (to the extent reasonable possible) perpendicular along the cross section.