Spatial Data Warehouses - Data Warehouse Systems: Design and Implementation - page 419

11.1 General Concepts of Spatial Databases
Spatial databases have been used for several decades for storing and
manipulating spatial data. These data are used to describe the spatial
properties of real-world phenomena. There are two complementary ways of
modeling spatial data. In the object-based approach, space is decomposed
into identifiable objects whose shapes are described. This allows us, for
example, to represent a road as a lineorastateasasurface.The field-based
approach is used to represent phenomena that vary on space, associating with
each point a value that characterizes a feature at that point. Typical examples
are temperature, altitude, and soil cover. In order to represent these two
alternative ways to model spatial features, we need appropriate data types.
In this section, we describe spatial data types for both the object-based and
the field-based approaches at a conceptual level.
11.1.1 Spatial Data Types
A spatial object corresponds to a real-world entity for which an application
needs to store spatial characteristics. Spatial objects consist of a descriptive
component and a spatial component. The descriptive component is
represented using traditional data types, such as integer, string, and date;
it contains general characteristics of the spatial object. For example, a state
object may be described by its name, population, and capital. The spatial
component defines the extent of the object in the space of interest.
Several spatial data types can be used to represent the spatial extent of
real-world objects. At the conceptual level, we use the spatial data types
defined by the MADS model, cited in the bibliographic notes at the end of
the chapter. These data types provide support for two-dimensional features.
They are organized in a hierarchy, shown in Fig. 11.1 .
Point represents zero-dimensional geometries denoting a single location in
space. A point can be used to represent, for instance, a village in a country.
Line represents one-dimensional geometries denoting a set of connected
points defined by a continuous curve in the plane. A line can be used to
represent, for instance, a road in a road network. A line is closed if it has no
identifiable extremities (i.e., its start point is equal to its end point).
OrientedLine represents lines whose extremities have the semantics of a
start point and an end point (the line has a given direction from the start
point to the end point). It is a specialization of Line and can be used to
represent, for instance, a river in a hydrographic network.
Surface represents two-dimensional geometries denoting a set of connected
points that lie inside a boundary formed by one or more disjoint closed lines.
If the boundary consists of more than one closed line, one of the closed lines