Geographic data model

The core data component of a GIS is often represented by a geographic data model, which is an industry
or discipline-specific template for geographic data. A geographic data model offers the user flexibility in
the design of the file management and database hierarchy. Geographic data models typically utilize a
grid-based structure (known as raster) or a coordinate point structure (known as vector).

Vector data
Many times, geographic data are modeled in what is known as vector space. A vector space is simply a
platform for geographic vector data, which use x-y coordinates with lines and shapes to depict Earth
features. Geographic vector data store non-topological coordinate geometry and attribute information
for spatial features.
Most standard GIS vector file formats consist of a feature file, an index file, and a linked attribute table.
A feature file contains geographic object feature information, such as representative point, line, and
polygon information. An index file contains unique identifiers that comprise more detailed information
and help speed spatial feature queries. A linked attribute table is a matrixed table that contains
explicative attributes for a group of spatial features.
The index file links the feature file data to the attribute table. Consequently, attributes and features
exist in a strict one-to-many relationship, whereby geographic features can have multiple attributes. The
vector model is readily recognized by those who use computer-aided design (CAD) environments, using
feature geometry in a GIS vector model to instantiate points, lines, and polygons as objects.
Sample vector data

In most cases geographic vector data are discrete data occurring in cases where there are well-defined
boundaries for physical representations or limited data values. The above figure involves tax lots that

have visibly evident and discrete boundaries. Vector data can be created or modified directly by
digitizing features using GIS development environments.

Raster data
Raster data are digital images represented by a grid of valued pixels, or cells. The image type and the
number of colors represented determine the properties and appearances of these pixels. In most cases,
raster data are more suited for representing continuous data than vector data. Continuous data are a
numeric form of data usually associated with the physical measurement of boundaries that are not well
defined. Additionally, the surfaces represented are, in many cases, estimated through statistical analysis.
Raster imagery is perfectly suited for GIS sensor webs, a current industry trend that harnesses real-time
observation and measurements over large, even global, regions.
Sample Raster data

Geodatabase
A geodatabase is a collection of geographic data sets, real-world object definitions, and relationships.
Comparable to a Microsoft Access file, a geodatabase is a collection of geographic data sets and
geometric features. A geodatabase furnishes the data organizational structure and workflow process
model for the creation and maintenance of the core data product. In essence, the geodatabase is the
heart of a GISs management capability.

Georeferencing
Many data sources lack formal spatial referencing. Some CAD and GIS data sets are developed in a
generic design space and have unique, often proprietary, types of referencing that simply need
reinterpretation to be spatially integrated into a GIS environment. However, many of these sources are
scanned raster data (digital imagery) that have only the coordinates of a raw pixel grid from an original
scan. While these raster sources are often times unique and critical to a GIS project, images also need to
be referenced from scratch, spatially transformed into a defined coordinate referencing system, then
integrated and overlaid in a GIS environment. This process is known as Georeferencing.
Georeferencing is the art of selecting common point locations in the real world using at least two data
sources: an unreferenced source (such as a raster map) and a referenced source of the same area
providing positional information. Basic georeferencing procedures involve point selection and
transformation. For example, when a hardcopy map is scanned to an electronic file, it has no
relationship to any real-world coordinate system. The georeferencing process establishes (or in some
cases reestablishes) the relationship between image pixel locations and real-world locations.
Georeferencing is accomplished by first selecting points on a source image (scanned raster map) with
known coordinates for the real-world surface location (benchmarks, grid ticks, road intersections, and so
on). These real-world coordinates are then linked to the corresponding pixel grid coordinates in the
raster source image. After the image is georeferenced, each pixel has a real-world coordinate value
assigned to it.

Queries: Locations and Attributes

Once data are adequately georeferenced and resident in a GIS, the user can then create a query
expression to find the relevant data for a specific application. A query enables the user to search
geographic data to collect location, feature, and attribute information from a relational database
management system or geodatabase. The most fundamental test of geographic coincidence between
data sources is a query by location in tandem with a targeting expression, such as Intersection, Within
distance of, Contained by, Share features with, Touch the boundary of, Are crossed by the outline of, and
Completely in.
Queries can also be executed on feature attribute tables and performed on specific fields. Records are
selected in this manner; each record (or row) in the table corresponds and is linked to a distinct
geographic feature data set. Features conforming to certain attribute criteria can be selected and
extracted using this method.
Quite simply, queries are the users refining tool for taking the massive available data and selecting only
those pieces that pertain to the application at hand.

Geoprocessing
Geoprocessing is the fundamental process of creating a derived set of geographic data from various
existing data sets using operations such as feature overlay and data conversion. In a typical

geoprocessing environment, the user applies GIS functions to a group of geographic (input) data to yield
a precise output data set suitable for a particular application.
Most professional GIS software environments include a mission-specific geoprocessing interface, or
workbench, of geospatial dialogs and tools. These software environments usually include extensible
scripting tools and compilers to automate, customize, and document geoprocessing workflows. The
most important contribution of geoprocessing to the GIS big picture is the automation of repetitive
tasks. Geoprocessing is an elaborate turnkey for efficient and clean geographic output.
Geoprocessors come in different forms. Many geoprocessing functions are embedded in a GIS
environment. A GIS environment is a package of integrated GIS components: a geographic map control,
a map layout designer, a data tree catalog, and so forth, of which geoprocessing is a member. However,
many powerful stand-alone software applications offer specific, related subsets of geoprocessing
functionality. Some include file format translators or spatial referencing transformation tools. Many
professional GIS efforts actually require licensing these stand-alone accessories in order to reap the
often-advantageous outputs of these stand-alone software applications.
There are eight categories of predominant geoprocessing operations, or
families of operations:
1. Conversion. Conversion is completely an issue of formatting. File format conversions
(translations) and coordinate system referencing conversions are the most common
geoprocessing conversion operations, and serve to characterize the conversion family.
2. Overlay (union, intersecting). Overlay involves superimposing two or more geographic data
layers to discover relationships. In fact, overlay is intimately associated with the discipline of set
topology, which defines the rules for valid spatial relationships between features in a geographic
data layer.
3. Intersect. Geoprocessing computes a geometric intersection of the input features. The resultant
features or portion of features common to all layers or assigned groups of same shape type
(called a feature class) will be written to the output.
4. Union. Like intersect, union computes a geometric intersection of the input features. All features
with the overlapping attributes from the input features will be written to the output feature
class.
5. Extraction (clip, query). Like overlay, extraction is also intimately associated with the discipline of
set topology. Queries help select the geographic data to be clipped or extracted, subject to a
specific group of topology rules.
6. Proximity (buffer). Proximity is initiated through a query that selects geographic features based
on their distance or proximity from other features. Geographic features include lines, points,
and polygons.
7. Management (copy, create). GIS data management software is generally designed to facilitate
the organization of a users unique personal catalog (or collection) of geographic data. The
intrinsic forms of all types of geographic data are accommodated by these applications.

Spatial Analysis
By their very nature, geographic data are intimately related to locations and feature attributes. Spatial
analysis harnesses this duplicity through the study of geographic feature locations and shapes. Spatial
analysis offers the user a range of procedures, tools, and interfaces varying in application and
complexity. For example, creating a simple map in a GIS environment is a basic form of spatial analysis.
Spatial analysis is not necessarily complex, but it is a process of reducing complex relationships to
something simpler, possibly bringing to attention things that otherwise would have remained hidden to
the user.