An InRoads user asked for Mark's thoughts on this question :
Mark,
I've been told that the Exterior Boundary in an InRoads surface can be at a 'zero' elevation and to use the 'Exclude from Triangulation' option when it is. Is this okay?
- Thanks, Alf
Here is the Zen Dude's response:
Hi Alf!
There are a couple reasons why I need to publish everything I learn about this software; 1) Because I lose information (I've covered this topic in the past but can't find any record of it >: (
...and 2) Because these sort of technical things get thrown around so carelessly and without enough information to assess whether or not it should be done. And if it is done, what are the implications?
Ahhhhhhh . . . (deep exhale)
First, the software will allow you to add an exterior boundary with a zero elevation. If it is added to the surface and incorporated into the DTM as part of the triangulation, I hope everyone within the sound of my clicking keyboard will understand the implications. Bottom line is . . . that's really bad and should not be done because your triangles will now go to 'zero' around the outer edges.
So the next thought is, "Hey! What if I just set it to 'Exclude from Triangulation'!!"
Someone who has been involved with InRoads from the start once posted on a public InRoads forum that this was okay to do. I was in shock when I read what he wrote. I contacted him and asked, "Why did you say that!!?" He said, "Because you can".
You can do a LOT of things in InRoads, but that doesn't mean that you should.
But I digress.
Here is the scoop.
Take this area with breaklines and spot elevations.
The triangulation looks like this:
Let's say that an Exterior Boundary is added shown by this green shape.
It is added to the DTM at elevation zero, and set to Exclude from Triangulation.
If the Perimeter is viewed, something very deceiving happens. The perimeter by definition is a shape defined by the connection of the outer leg of all of the outer triangles around the surface model. This is not what happens in this instance, and I don't know why. It appears that the View Perimeter command is viewing some sort of 'limit' as defined by the shape added as the 'Exterior Boundary'.
You should be aware of the importance and value of the triangles within an InRoads surface. Dozens upon dozens of important commands are 'triangle-based' and woven into the location of the surface triangles such as, View Contours, View Elevations, Color-coding, the Create Profile and Create Cross Sections commands, End Condition intercepts during modeling, earthwork quantities and on and on. It is very important to know where you do and where you don't have triangles. The triangles define the useful extents of a surface model. And the Perimeter is always used to gauge that usable limit.
Viewing the triangles reveals my root problem with this 'technique'.
This is the actual Perimeter: (as shown by the triangles, as compared to the viewed perimeter)
Depending on how the triangles formed (which you know are irregular), the results of using the 'zero' elevation Exterior Boundary set to Exclude from Triangulation creates a very unpredictable, unexpected resulting surface limit.
As I mentioned earlier, the surface triangles drive many of the InRoads commands. Viewing the Contours results in this:
Areas of the surface model are more or less unusable.
Every surface, composed of different data, will react in a different way.
At one point in time working with InRoads, I believed that using this 'technique' would cause any triangles coming into contact with the zero elevation Exterior Boundary to be eliminated. This is not the case now (as it may have been in the past). From what I can observe, any triangle that contains the majority of surface area within the shape is retained, and triangles with the majority of its surface area outside the boundary shape are deleted.
The blue shaded triangles are some that are removed, whereas the yellow ones are retained in the surface.
My recommendation? . . . Caution.
I personally do not use this 'technique'.
Move forward understanding the implications of your actions.
I'm glad this is finally documented . . . I hope this helps.
Civilly yours,
- zen
Mark was great. Very Very knowledgeable on the subjects and really wanted the students to learn. He has a great personality but never got away from the subject at hand. Great person and instructor. Excellent ! 
