Geography Reference
In-Depth Information
were call “arcs.” The geodatabase is a more recent development. You will spend a fair amount of time
and effort understanding and working with geodatabases. The coverage data structure is, frankly, rapidly
becoming obsolete. My recommendation is that GIS professionals of today convert their coverages (and,
admittedly, there are a lot of them around) to the File Geodatabase structure. Another data structure,
the middle-aged shapefile, is relatively simple, but because multitudinous data sets exist in this form,
nationally and internationally, it is also important. Understanding geographic data structure is vital to
being able to do some forms of analysis with GIS. Please note that you can convert any of the three of
these dataset forms into any of the others.
The primary elements of vector-based datasets are points (zero-dimensional entities), lines (one-dimensional
entities), and polygons (two-dimensional entities). Terms used with all Esri data models are described in
the sections that follow, and then when we look at particular data models. Figures that graphically show the
entities follow in the detailed discussions of geodatabases and shapefiles.
Zero-Dimensional Entities in a Two-Dimensional Field: Points
While vital to the functioning of a GIS, a zero-dimensional thing (generically a point) is pretty dull from a
geometric view. It is basically a pair of numbers (x- and y-coordinates, or perhaps, latitude and longitude
coordinates) stored as single- or double-precision numbers.
6
The concept of a point is used in a variety of ways in GIS to represent features, as end points of lines, as
vertices in sequences of line segments, vertices of triangles, as reference points tying to the feature set to
the real world, as locations to hang labels on, as centroids of areas, as junctions and nodes in geometric
networks, as centers or corners of raster cells, and others.
We work mainly with points in a two-dimensional arena, but of course they exist truly in three-
dimensional space. ArcGIS will let us add information about this third dimension, sometimes as what
amounts to an attribute (a “z” value) and sometimes (for example, in a TIN) as a measure in the true third
dimension. Even when a true 3-D point is represented, the units of measurement of the vertical may not
be the same as those used in the horizontal plane. Some of the uses of points are described below.
Points representing features
—In these cases, a point has associated with it a row in a relational database
table that identifies the point and allows the user to add other (attribute) information about the feature
the point represents. You became acquainted with points representing features in the fire hydrant
example of Chapter 1. A point may be used to represent a feature that is too small to have a meaningful
area. How does the concept of a point fit into a vector system? A point is simply a vector with zero
magnitude and an unimportant direction.
Multipoints
—A multipoint is a collection of points that share the same attribute values (e.g., several gas
wells which have the same characteristics and the same owner). The collection of points is represented by
a single row in a table.
Vertices
—Sets of coordinates where two line segments are joined or where a line segment ends. Also
considered vertices are the corners of a triangle in a TIN. Usually, no database table row is associated
with a vertex.
6
To really locate a point in real (3-D) space, one would need a trio of numbers—but vector GIS either assumes that
the third dimension is the elevation of the surface of the Earth, or is not relevant, or is stored as attribute data.
An exception to this is a TIN, which stores coordinate triples.
Search WWH ::
Custom Search