- The need for breaklines has been debated for a number of years without a clear standard being developed.
- Producing contours from a LiDAR survey causes additional issues in the area of breaklines.
- Should we still be using contours in a 3D world?
A quick review of the literature reveals that the issue of LiDAR and breaklines has been debated for a number of years. This is an important issue because of the cost of creating them. Mark Romano from Earth Eye wrote a very important article on the topic in 2004. Dr. Qassim Abdullah, from Fugro also provided important insights in his PE&RS Mapping Matters column on the subject and how it relates to LiDAR point density. (The above image is from a presentation that he recently made.)
Unfortunately, like many of the other key technical issues involving LiDAR technology, there is no single correct answer, or accepted standard industry practice. I guess this is an indication of the maturity of the industry.
As both experts point out breaklines are a carry over from manual survey data collection workflows that produced a relatively sparse number of actual 3D data points. Without breaklines, you simply could not generate contours, unless you were in a field.
Enter LiDAR. Now the issue is too many points, and the fact that when combined with most digital terrain modeling software if the desired output is contours then breaklines and TINs may not be the best solution. The LiDAR data will certainly have to be thinned in the area of the breaklines if the end product is to be cartographically appealing.
My question is, “Why do we need contours in a 3D software environment? Weren’t contours needed to represent the third dimension on paper maps and for interpolation? Are there not better ways of describing and visualizing surfaces in 3D? And if that is the case, should we be using TIN to create the surfaces? Isn’t a grid approach better suited to LiDAR data collection methods?
This post was written by Gene V. Roe
I actually remember working on this project in the screenshot. I worked for Fugro EarthData at the time. It is (roughly) from about 2003-2004 where we processed LiDAR and aerial photography over Mecklenburg County, NC. We ran several tests and found that the only way to get the contours to “behave” properly while crossing the highway was to introduce breaklines. No matter how dense the lidar is, it is still point data which never completely defines an “edge”.
To answer your question; NO, we should no longer be delivering contours from LiDAR given the methods available now to visualize terrain. Some of us have been arguing this point with customers for years now. The problem is that the culture has been slow to change. Maybe the solution is to start winning over minds in academia. Or maybe this problem will correct itself as the generation raised on contours begins to retire. Either way, it has been a matter of frustration since we know that contours from LiDAR seem to be a waste of time and resources when the LiDAR can yield such an accurate surface. Breaklines may still be needed in some cases to “correct” a surface such as in places where a retaining wall (for instance) affects the terrain. Gridding – Agree here as well – When you are forced to deliver contours from LiDAR, gridding the source data is practically a necessity in order to reduce the number of isolations and to “smooth” the appearance of the contours.