Database Reference
In-Depth Information
In order to precisely describe the spatial extent of geographic objects or cells
in a raster image, it is important to have a
spatial reference system (SRS)
(or
coordinate reference system (CRS)
) underlying the space in which features
and phenomena are modeled. A reference system is a particular
map projec-
tion
that represents the two-dimensional curved surface of the Earth. There
are numerous such map projections used in practice, ranging from global pro-
jections such as latitude/longitude or Universal Transverse Mercator (UTM)
to parameterized local ones tailored to specific regions on the Earth's surface,
such as the State Plane Coordinate System
18
used in the United States. From
a spatial data management point of view and in particular for the integration
of diverse datasets, an important aspect is to be able to re-project geospatial
data from one reference system to another one.
19
,
20
10.2.2 Geospatial Data Management Systems and Formats
Compared with data management systems for relational data, which are all
based on the same model (the relational model) and make use of the same
language (SQL), there is a plethora of commercial and open-source systems
for managing geospatial data. In the following, we give a brief overview of
the different types of systems and focus on aspects that are relevant to data
integration approaches.
GISs are the predominant type of systems to manage, store, analyze, and
display geographic data and associated attributes that are spatially refer-
enced to the Earth.
21
A widely used type of GIS is ESRI's (Environmental
Systems Research Institute's) ArcGIS products, such as ArcView to view spa-
tial data, create maps, and perform basic spatial analysis operations. ArcInfo
is an advanced version of the ArcGIS product line that also includes functions
for manipulating, editing, and analyzing geospatial data
22
,
23
and services for
geoprocessing and geocoding.
24
ArcGIS also provides different types of Web
services to access geospatial data.
There are also traditional relational database management system vendors
that offer spatial extensions to their relational engines. For example, Oracle
Spatial provides several functions for storing, querying, and indexing spatial
data, including raster and gridded data.
25
The spatial extension models a
majority of the spatial types and operations described in the SQL/MM spatial
standard.
26
IBM's DB2 product line also offers spatial extensions to their
relational DB2 core system such as the DB2 Spatial Extender and the DB2
Geodetic Extender.
27
Also here, the spatial extensions implement types and
functions specified in the SQL/MM standard.
Prominent open-source GIS type systems are PostGIS,
28
the spatial exten-
sion of the object-relational database management system PostgreSQL, and
the Geographic Resources Analysis Support System (GRASS).
29
,
30
Like the
spatial extensions for Oracle and DB2, PostGIS follows the
Simple Features
for SQL specification
developed as an implementation specification by the
OGC.
31
This standard specifies the storage of different types of geographic
