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:

  1. Cartography : contributes with map generation techniques;
  2. 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;
  3. Database Management Systems (DBMS): data modeling, data structures, security, and processes for the maintenance of great data volumes;
  4. 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);
  5. 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;
  6. 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...;
  7. 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