The Third Spatial Dimension - Introducing Geographic Information Systems with ArcGIS - page 563

Terrains in ArcGIS, although they are feature classes, don't really exist in the way that other
feature classes exist. Terrains resemble TINS but because of the massive number of data
points ArcGIS creates terrains on the fly, depending on the spatial extent that is being viewed.
A terrain sort of half-exists; the underlying points are stored, of course, but the location and
degree of triangulation that takes place in order to use or display the resulting TIN is variable,
depending on what is needed. When you are using larger scales (zoomed in), only a limited
portion of the underlying points need be used to create the image—which saves drawing time.
At smaller scales (zoomed out, so you would expect to see less detail), only a subset of the
points in the area is used to make the triangulation. What you see when you view a terrain is
an image that is calculated “on the fly.” The person who builds the terrain specifies parameters
that determine levels of detail used. Since the terrain is a dynamic entity the feature classes
that are used in producing the terrain (in our case, the points and shapefiles) have to be
accessible to ArcGIS. The software requires that these data must reside, in a new form, within a
single feature dataset.
78. Import the needed shapefiles into the feature dataset: In ArcCatalog, right-click Zone, pick
Import and then Feature class (multiple). In the resulting window browse to Terrain_Data and, in
the Input Features window, using the Ctrl key and clicking, pick both water_body.shp and break_
lines.shp. Click Add. Make sure the Output Geodatabase reads
IGIS-Arc_ YourInitials \Elevation_Data\Terrain_Data\Example.gdb\Zone
if it doesn't already. Click OK. When processing is completed (watch the Status bar), expand
Zone to check that the features of the two shapefiles have been duplicated inside the feature
dataset as feature classes. They aren't shapefiles anymore, but they contain the same
information as the originals.
As you determined earlier, LIDAR_Points.txt consists of hundreds of thousands of individual
geographic locations. While we could represent each of these as a point feature, it is more
sensible to group them into multipoints. This preserves the geographic location of each but
allows the software to produce many single records, each of which subsume many positions.
79. Make a multipoint feature class based on LIDAR data: For this you need ArcToolbox. Since
ArcCatalog is already open use the toolboxes there. In ArcToolbox expand 3D Analyst Tools.
Expand Conversion, followed by From File. The main tool you will use in a moment will ask you
for the average point spacing, so we will get that information first. There is a tool you can use
to obtain this number. Double-click on Point File Information. Browse for the input file
IGIS-Arc_ YourInitials \Elevation_Data\Terrain_Data\LIDAR_Points.xyz
Click Open. Accept the default Output Feature Class name. The File Format must be XYZ. Click
OK. In the Catalog Tree highlight
LIDAR_Points_PointFileInform.shp.
The Geography of that shapefile is completely unimpressive. But the Table is more interesting.
It consists of a single record. What is the number of points indicated ___________. What is the
average spacing (to the nearest tenth of a foot) between the points ______________. What is
the elevation of the lowest point? ______. The highest. ______.