«

»

Jul 14 2011

Hydro Survey: The Process to a Map (Part 2)

In my last post (The Process to a Map (Part 1), I explained how the process to building a bathymetry map begins before you even get in the field. I walked through the spacing of your single beam cross sections and the importance of your cross sections being perpendicular to flow. After the extracting and thinning to just the perpendicular data, it left us with this:

Data thinned to just perpendicular section points

Correctly Modeling the Curvature of the River

So how do we then get to a full bathymetry digital terrain model (DTM) correctly representing the curvature of the channel? You cannot simply manually add break lines to the surface and then triangulate to a DTM. It won’t model the thalweg correctly nor tie in well to the water’s edge. There are a few ways to do this. First, you could use a method utilizing b-splines. You see this in “A Hybrid Meshing Scheme,” and “Hydromesh.” This route is also well explained in peer-reviewed article “Hydraulic Splines.” We’ve looked into going this route, particularly because of how you could better automate the mapping.

However, our method of processing single beam bathymetric data to a DTM entails taking the data from a curved alignment to a straight alignment and then back. We do this using Microstation InRoads. We take the cross section data from the curved alignment shown above and map it to same stationing along a straight alignment of the same length. This is where we then triangulate it and breakline the data, weeding out bad points and correctly modeling the thalweg and other features. When this process is completed, the straight alignment DTM is used to generate the data in 50 foot stationed cross-sections. These 50 foot cross-sections are translated back to the curved alignment to get the result you see below.

50 foot cross-section data translated back to the curved alignment.

The Final Steps to the Map

From the 50 foot cross-section data, you can then combine it with your overbank topography for the final product. The overbank mapping for this area is shown below:

06_Overbank_thumb[1]

Here is the bathymetry processed data points shown combined with the topography:

07_Combined_thumb[1]

Finally, we have a full DTM with which we can use for a 2D hydraulic model, cut cross sections for use in a 1D hydraulic model, among other uses.

08_Color Contours_thumb[1]

The full process of mapping the channel from the use of single beam channel width cross-section data is not incredibly complex but it does entail a working knowledge of the river surveyed to correctly breakline the thalweg and accurately model the existing bathymetry. This is where multi-beam depth sounder systems are obviously more advanced: a dense array of points representing the surface of the river bottom is collected via a type of sonar scan leaving less post-processing.

However, there are various weaknesses to multi-beam hydro surveying that we’ll discuss in future posts. The main relevant weakness: multi-beam can only go so shallow. That means, for now, in typical shallow western streams, you need to have some method of processing cross-section type data collected from a single beam depth sounder.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>