Introduction to Geoprocessing
Geoprocessing can be defined as a set of technologies aimed at the
gathering and processing of spatial information with a specific
objective. The activities that make part of geoprocessing are performed
by systems that are specific for each application. These systems are
more usually known as Geographic
Information Systems (GIS).
A geoprocessing system is designed for the processing of
geographically referenced data (or georeferenced), from their
collection to the generation of outputs in the form of conventional
maps, reports, digital files, etc..., providing features for their
archiving, management, handling and analysis.
With the evolution of geoprocessing technology and of graphic
software many new terms were created for the many disciplines. The term
Geographic Information System
(GIS) is widely used and many times misinterpreted as geoprocessing.
Geoprocessing has a wider meaning representing any type of processing
of georeferenced data, while a GIS processes both graphical and
non-graphical (alphanumeric) data with emphasis on spatial analysis and
surface modeling.
The SPRING is also a GIS, although here it will be treated as a Geographic Database System since it
was conceived with all the tools of a GIS within a relational database
structure.
As a geoprocessing system SPRING is not only a computational system
for the production of maps, although it can create maps in different
scales, in different projections, and with different colors. It is
mainly an analysis tool that helps in the decision-making process.
See also:
Characteristics of a GIS
The term geographic information
system (GIS) is applied to systems that perform the
computational treatment of geographic data. Due to its wide application
spectrum, ranging from agriculture, forestry, cartography, urban
cadastre, and urban networks (water, energy, telephone) there are at
least three main ways of using a GIS:
- as a map production tool;
- as a tool for the spatial analysis of phenomena;
- as a geographic database, with spatial data archiving and query
functions.
These three visions of a GIS are convergent rather than conflicting,
and reflect the relative importance of the geographic data treatment
inside an organization. To make the subject even clearer we present
below some definitions of a GIS:
- "A manual or computational set of procedures used for
georeferenced data archiving and manipulation "
(Aronoff, 1989).
- "A powerful set of tools for the collection, archiving,
retrieval, transformation, and visualization of data about the real
world" (Burroughs, 1986).
- "A decision-making support system that integrates spatially
referenced data into a problem-solving environment" (Cowen, 1988).
- "A spatially-indexed database where a set of procedures to query
spatial entities operate" (Smith et al., 1987).
These definitions of GIS reflect,
on its own way, the multiplicity of possible uses and visions of this
technology while pointing towards a multidisciplinary perspective of
its uses. From these concepts it is possible to indicate the main
characteristics of SPRING:
- Integrate, in a single database, spatial information from
cartographic data, censitary, rural and urban cadastre data, satellite
imagery, network and digital terrain models.
- Provide ways for the combination of different information,
through data manipulation and analysis algorithms, and the query,
retrieval, visualization, and plotting of the geocoded database.
Introduction to
Geoprocessing
The concept of space and of spatial relations
What is special about spatial data? From this question we can try to
unveil the particularity of geographic data.
The most fundamental aspect of the data processed by a GIS is the
dual nature of the information: a geographic data has a geographic localization (expressed
as coordinates in a map) and descriptive
attributes (that can be represented in a conventional database
system). Another very important aspect is that geographic data do not
exist alone in space: as important as localizing them is
discovering and representing the relationship among them. Some examples
of the spatial analysis process typical of a GIS are presented in
the table below (adapted from Maguire, 1991).
EXAMPLES OF
SPATIAL ANALYSIS
Analysis |
General Question
|
Example |
Condition |
What is...” |
What is the population of this
city ?” |
Localization |
Where is...? |
What are the areas with
declivity greater than 20% ? |
Trend |
What changed...? |
Was this land productive 5
years ago ? |
Routing |
Which way to go.. ? |
What is the best way to the
subway ? |
Patterns |
What pattern....? |
What is the distribution of the
"dengue" in Fortaleza ? |
Models |
What if...? |
What is the impact on climate
if we deforest the Amazon ? |
Let's take a concrete example to make clear the concepts about
spatial analysis (in this case manually performed). In 1854 London was
suffering a massive cholera epidemics, a disease whose contagion
mechanism was unknown at the time. More than 500 deaths had occurred by
then. Dr.John Snow had in insight: plotting on a map the location of
the cholera sick persons and of the water wells (at that time the main
source of water for the Londoners). The map is shown in the figure
below.

Map of London showing cholera cases
(points) and water wells (crosses) (adapted from E. Tufte, 1983).
With the spatialization of the data Dr.Snow noticed that most
of the cases was distributed
around the Broad Street well, and thus ordered its sealing, that
contributed a lot to the control of the epidemics. Such case provided
empirical evidence for the hypothesis (later confirmed) that the
cholera is transmitted via the drinking of contaminated water. This is
a typical situation where the spatial relationship of the data would
not be inferred by the simple listing of the cholera cases and of the
wells.
Dr.Snow's map entered history as one of the first examples
illustrating the explanatory power of spatial analysis.
Introduction to
Geoprocessing
Related Technologies
Geoprocessing can be considered a multidisciplinary science that
benefit from the contribution of many areas. In general the
technologies that contribute the most are:
- Cartography : contributes with map generation techniques;
- CAD (Computer Aided Design) and Computer Graphics:
contributes with software, hardware, data input techniques,
presentation, visual representation in 2D and 3D, manipulation and
presentation of graphic objects;
- Database Management Systems (DBMS): data modeling, data
structures, security, and processes for the maintenance of great data
volumes;
- Remote Sensing: techniques for image acquisition and
processing with facilities for data gathering from any place on earth,
either through orbital sensors (satellites) or photographic sensors
(airborne);
- Artificial Intelligence: technology that employs
computers to emulate human intelligence. The computer acts as a
specialist in the functions of drawing, mapping, classification,
generalization of map characteristics;
- Statistics: provides models and data analysis methods,
graphical and non-graphical. Statistical techniques are used for the
verification of the quality during preprocessing, to summarize a file
in a data management report, to create derived data during analysis,
etc...;
- Computer Science: besides covering some of the
items above, computer science also contributes with system development
techniques, hardware development for the support of great data
processing power and the computer network technology that allow the
data exchange among local or remote computers.
Introduction to
Geoprocessing
General Structure of a GIS
In a comprehensive point of view we can say that a GIS has the
following components:
- User interface;
- Data input and integration;
- Image and graphic processing functions;
- Visualization and plotting;
- Data archiving and retrieval (organized in the form of a
geographic database).
These components relate in a hierarchical way. At the level closer
to the user, the man-machine
interface defines how the system is operated and controlled. At
the intermediate level a GIS should have tools for the processing of
spatial data (input, editing, analysis, visualization and output). At the most internal level
of the system, a geographic database
management system provides facilities for the archiving and
retrieval of spatial data and attributes.
In general, the processing functions of a GIS operate on the data at
a workspace in the main
memory. The link among the geographic data and the GIS processing
functions is done via data query and selection tools that define
restrictions on the dataset. Illustrative examples on the data
selection are:
- "Retrieve the data related to the Guajara-Mirim map" (restriction by the definition of the region
of interest).
- Retrieve the cities in the state of Sao Paulo with a population
between 100.000 and 500.000" (query by
non-spatial attributes).
- "Show the health-care units within a 1 km radius from the City
Hospital in Sao Jose dos Campos" (query
with spatial restrictions).
The figure below shows the relationship among the main components.
Each system will implement these components in a distinctive way,
depending on its objectives and needs, but every GIS features these
subsystems.

Geographic Information Systems Architecture.
Introduction to
Geoprocessing
Data types in Geoprocessing
A Geoprocessing system stores the geometry and attributes of the
georeferenced data, that is data which is localized on a cartographic
projection of the surface of the earth. The data manipulated in
geoprocessing have the main characteristics of coming from many
different sources and in many different formats.
Differently from Computer Aided Design (CAD) systems, one important
characteristics of a GIS is its ability to treat the spatial
relationships among the spatial objects. We call topology the structure
of spatial relationships (neighborhood, proximity, pertinence) that can
be established among geographic objects.
The requirement of archiving the geometry and the attributes of the
geographic objects represent a basic duality for a GIS. For every
geographic object the GIS needs to archive their attributes and their
various associated graphical representations.
Introduction to
Geoprocessing
Classes of projects in Geoprocessing
The spatial database is the most expensive component in a
Geoprocessing system. The comprehensiveness of the work allows
the distinction between two types of environment.
- spatial analysis projects over small or medium-sized regions;
- spatial inventory over great regions.
As an example of the first case we
can take the case of the RIMA's (Environmental Impact Assessment
Report), as in the case of a railway or hydroelectric plant, for
example. In the second case we can point the systematic surveys like
the ones carried out by INPE to map the deforestation of the Amazon
region.
Spatial Analysis Projects
Spatial analysis projects usually
require great flexibility and comprehensiveness of the GIS functions,
for a limited quantity of data but with great variety. To illustrate,
we present some applications of INPE's GIS:
- PETROBRAS/CENPES (study of the Recôncavo basin): 50
"information layers" for an area of about 50.000 km2, with different
data types (geophysical, magnetometry, geologic map, aerial photos),
and a great quantity of derived data (conversion of DTM into images,
IHS-RGB transformation, among others...).
- EMBRAPA/NMA (zoning of the Fernando de Noronha island): 80
"information layers" for an area of 3.000 km2,
with maps of declivity, land use, geomorphology, wind direction,
localization of species, touristic places, among others.
We consider that the present Geographic Database technology is
already capable of supporting these types of application.
Spatial Inventory Projects
In the case of inventories the greatest emphasis lays on the
treatment of great databases, where the same procedures are repeated on
all the data. The regions covered are very big, as can be seen from the
examples:
- INPE (Amazon Deforestation Atlas): 228 "projects",
for a total area of about 5 million km2. Each project has 5
"information layers" (forest physiognomy, hydrography, road map, urban
base and deforestation).
- SOS Mata Atlântica (mapping of the remaining atlantic
forest): 200
projects, for a total area of 3 million km2. Each project has 5
"information layers".
In the case of inventories the work is reduced, in many cases, to
the phases of data interpretation, integration to the spatial database
and plotting of the results.
Introduction to
Geoprocessing
|