Geography Reference
In-Depth Information
Dimension. The goal just here is to describe the data model that lets the computer tell you about the
surface that the TIN represents.
The idea of a TIN is to approximate a surface with a set of planes—triangles, to be specific. A triangle has
some nice characteristics: If the computer knows the x-, y-, and z-coordinates of each of its three vertices,
it can calculate
The
z
value (e.g., elevation) of any point on the surface of the triangle
❏
The triangle's slope (the maximum quotient of rise over run)
❏
The triangle's aspect (the direction of the projection on the x-y plane of a line perpendicular to the
surface of the triangle-loosely, the direction in which the triangle's face is pointed. )
❏
One could also calculate the area and perimeter of the triangle, but it turns out these aren't usually
important. Since the triangle is (usually) at an angle to the x-y plane, the area of the triangle will be
greater than its projection onto the plane.
The trick to making a TIN yield useful information is to connect a number of irregularly spaced x-y-z
points, of known value, with straight lines so that they form triangles that are not too skinny. The
software, based on clever algorithms,
15
takes care of this for you.
TIN-Based Geographic Data Sets—Layout in the Computer
If you look at a TIN in terms of folders and files, you find a single folder, residing within a work space,
with a bunch of files within it. As with all spatial data sets, if you want to copy it, move it around, or
rename it, you must do so with ArcCatalog, not the operating system.
The primary apparent difference between a TIN data set and other GIS data sets is that there is no
attribute table. You can, however, use the Identify tool with TINs. It provides the elevation, slope, and
aspect, as well as some additional information tags that the user may add. How does it do this? By
calculating information on the fly when you click on a point in the TIN.
For particular problems such as surface analysis, surface display, and hydrological analysis, TINs can be
quite useful. You will see more of them in future chapters.
Terrains
An Esri Terrain dataset is a TIN-based surface that may be viewed at several levels of resolution derived
from measurements stored in one or more feature classes in a geodatabase. An Esri Terrain is not actually
stored in its entirety but is calculated as needed, because of the great size of the datasets that underlie it
and the fact that only smaller portions of those datasets are needed at any one time. The principle is sort
of similar to a handheld calculator approach to providing functions such as square root or sine. Rather
than storing the values in a table they are calculated from a formula as needed.
You met a Terrain in Chapter 2. In Chapter 9 you will build one. The somewhat unusual way in which
Esri Terrains are “stored” will be discussed then.
15
Algorithms involving Delaunay triangulation and Thiessen polygons (also known as Voronoi cells and Dirichlet
regions).
Search WWH ::
Custom Search