These materials were developed by Kenneth E. Foote and Donald J. Huebner, Department of Geography, University of Texas at Austin, 1996.  These materials may be used for study, research, and education in not-for-profit applications.  If you link to or cite these materials, please credit the authors, Kenneth E. Foote and Donald J. Huebner, The Geographer's Craft Project, Department of Geography, The University of Colorado at Boulder.  These materials may not be copied to or issued from another Web server without the authors' express permission.  Copyright © 2000 All commercial rights are reserved.  If you have comments or suggestions, please contact the author or Kenneth E. Foote at k.foote@colorado.edu.

This page is also available in a framed version.  For convenience, a Full Table of Contents is provided.

 1. GIS as Representations of Reality

Perhaps we should use the acronym gIs, rather than GIS for geographic information systems. These are really geographic INFORMATION systems. It is the information they contain that makes them so valuable.

The database is also important because its creation will often account for up to three-quarters of the time and effort involved in developing a geographic information system. Once an organization compiles this information, the database may be maintained for ten to fifty years. For this reason, shortcuts are not recommended.

 It is important, however, to view these GIS databases as more than simple stores of information. The database is used to abstract very specific sorts of information about reality and organize it in a way that will prove useful. The database should be viewed as a representation or model of the world developed for a very specific application.

 One of the reasons that there are so many software and hardware systems employed for GIS is because each system allows users to represent and model certain types of phenomena.

 2. Basic Types of Representation: Raster and Vector Reality

One of the sharpest distinctions among GIS is the way that location is represented in a database, as either a raster or vector position.
 
 

2.1. The Raster View of the World

A raster based system displays, locates, and stores graphical data by using a matrix or grid of cells. A unique reference coordinate represents each pixel either at a corner or the centroid. In turn each cell or pixel has discrete attribute data assigned to it. Raster data resolution is dependent on the pixel or grid size and may vary from sub-meter to many kilometers. Because these data are two-dimensional, GISs store various information such as forest cover, soil type, land use, wetland habitat, or other data in different layers. Layers are functionally related map features. Generally, raster data requires less processing than vector data, but it consumes more computer storage space. Scanning remote sensors on satellites store data in raster format. Digital terrain models (DTM) and digital elevation models (DEM) are examples of raster data (Koeln et al 1994 and Huxhold 1991).
 
 

2.2. The Vector View of the World

A vector based system displays graphical data as points, lines or curves, or areas with attributes. Cartesian coordinates (i.e., x and y) and computational algorithms of the coordinates define points in a vector system. Lines or arcs are a series of ordered points. Areas or polygons are also stored as ordered lists of points, but by making the beginning and end points the same node the shape is closed and defined. Topological models define the connectivity of vector based systems. Vector systems are capable of very high resolution (less than or equal to .001 inch) and graphical output is similar to hand-drawn maps. This system works well with azimuths, distances, and points, but it requires complex data structures and is less compatible with remote sensing data. Vector data requires less computer storage space and maintaining topological relationships is easier in this system. Digital line graphs (DLG) and TIGER files are examples or vector data (Koeln et al 1994; and Huxhold 1991).
 
 

2.3 Graphical Comparison of Raster and Vector Systems

It is important to stress that any given real world situation can be represented in both raster and vector modes, the choice is up to the user.

 Each of these systems of representation has its advantages and disadvantages:
 
 
Method Advantages Disadvantages 
Raster
  • Simple data structure
  • Compatible with remotely sensed or scanned data
  • Simple spatial analysis procedures
  • Requires greater storage space on computer
  • Depending on pixel size, graphical output may be less pleasing
  • Projection transformations are more difficult
  • More difficult to represent topological relationships 
    		•
    Vector
  • Requires less disk storage space
  • Topological relationships are readily maintained
  • Graphical output more closely resembles hand-drawn maps
    		•
  • More complex data structure
  • Not as compatible with remotely sensed data
  • Software and hardware are often more expensive
  • Some spatial analysis procedures may be more difficult
  • Overlaying multiple vector maps is often time consuming 
    		•
    It should also be stressed that data modeled in one system can be converted into the other. That is, raster data can be vectorized and vice versa. Many systems even allow data modeled in raster form to be overlaid on vector data and vice versa. In this graphical example, an aerial photo (raster) is overlaid with with supplemental vector information.

     3. Organizing Attribute Data

    GIS use raster and vector representations to model location, but how they must also record information about the real-world phenomena positioned at each location and the attributes of these phenomena. That is, the GIS must provide a linkage between spatial and non-spatial data. These linkages make the GIS "intelligent" insofar as the user can store and examine information about where things are and what they are like.

     The relationship can be diagrammed as a linkage between:
     
     

    At the most abstract level, this is a relationship between: In a raster system, this symbol is a grid cell location in a matrix. In a vector system, the locational symbol may be a one-dimensional point; a two-dimensional line, curve, boundary, or vector; or a three- dimensional area, region, or polygon.

     The linkage between symbol and meaning is established by giving every geographic feature at least one unique means of identification, a name or number usually just called its ID. Non-spatial attributes of the feature are then stored, usually in one or more separate files, under this ID number. In other words, locational information is linked to specific information in a database

     It is important to realize that this non-spatial data can be filed away in several different forms depending on how it needs to be used and accessed. Perhaps the simplist method is the flat file or spreadsheet, where each geographic feature is matched to one row of data.
     
     

    3.1 Flat Files and Spreadsheets

    A flat file or spreadsheet is a simple method for storing data. All records in this data base have the same number of "fields". Individual records have different data in each field with one field serving as a key to locate a particular record. For example, your social security number may be the key field in a record of your name, address, phone number, sex, ethnicity, place of birth, date of birth, and so on. For an person, or a tract of land there could be hundreds of fields associated with the record. When the number of fields becomes lengthy a flat file is cumbersome to search. Also the key field is usually determined by the programmer and searching by other determinants may be difficult for the user. Although this type of database is simple in its structure, expanding the number of fields usually entails reprogramming. Additionally, adding new records is time consuming, particularly when there are numerous fields. Other methods offer more flexibility and responsiveness in GIS.

    3.2 Hierarchical Files

    Hierarchical files store data in more than one type of record. This method is usually described as a "parent-child, one-to-many" relationship. One field is key to all records, but data in one record does not have to be repeated in another. This system allows records with similar attributes to be associated together. The records are linked to each other by a key field in a hierarchy of files. Each record, except for the master record, has a higher level record file linked by a key field "pointer". In other words, one record may lead to another and so on in a relatively descending pattern. An advantage is that when the relationship is clearly defined, and queries follow a standard routine, a very efficient data structure results. The database is arranged according to its use and needs. Access to different records is readily available, or easy to deny to a user by not furnishing that particular file of the database. One of the disadvantages is one must access the master record, with the key field determinant, in order to link "downward" to other records.
     
     

    3.3 Relational Files

    Relational files connect different files or tables (relations) without using internal pointers or keys. Instead a common link of data is used to join or associate records. The link is not hierchical. A "matrices of tables" is used to store the information. As long as the tables have a common link they may be combined by the user to form new inquires and data output. This is the most flexible system and is particularly suited to SQL (structured query language). Queries are not limited by a hierarchy of files, but instead are based on relationships from one type of record to another that the user establishes. Because of its flexibility this system is the most popular database model for GIS.
     
     

    3.4 Flat, Hierarchical, and Relational Files Compared

    Structure Advantages Disadvantages 
    Flat Files
    • Fast data retrieval
    • Simple structure and easy to program
    • Difficult to process multiple values of a data item
    • Adding new data categories requires reprogramming
    • Slow data retrieval without the key
    Hierarchical Files
    • Adding and deleting records is easy
    • Fast data retrieval through higher level records
    • Multiple associations with like records in different files
    • Pointer path restricts access
    • Each association requires repetitive data in other records
    • Pointers require large amount of computer storage
    Relational Files
    • Easy access and minimal technical training for users
    • Flexibility for unforeseen inquiries
    • Easy modification and addition of new relationships, data, and records
    • Physical storage of data can change without affecting relationships between records
    • New relations can require considerable processing
    • Sequential access is slow
    • Method of storage an disks impacts processing time
    • Easy to make logical mistakes due to flexibility of relationships between records
    Now, let us consider a couple of examples of matching applications to database structures.

     Exploratory research--flat files are easy to organize, space is not particular problem

     Government agencies--hierarchical systems are particularly attractive

     Planning and development--relational might be justified for flexibility

     4. Representing Relationships

    GIS have the power to record more than location and simple attribute information. In some situations, we will want to examine spatial relationships based upon location, as well as functional and logical relationships among geographic features.

    Spatial Relationships
     
     

    Functional Relationships among Geographic Features and Their Attributes. This includes information about how features are connected and interact in real-life terms. A road network might be classified functionally from the largest superhighway down to the most isolated rural road or suburban cul-de-sac based upon their role in the overall transportation system. Minor roads and suburban streets "feed" major highways, but are not directly connected to them. As another example in assessing wildlife habitats, various environmental conditions function together to define the optimal living environments for certain species. Within cities, ownership is a functional classification of great importance as is landuse and zoning classification.

     Logical Relationships among Geographic Features and Their Attributes. Logical relationships involve "if-then" and "and-or" conditions that must exist among features stored in the dataset. For example, no land may be permitted to be zoned for residential use if it lies within a rivers five-year flood plain. Development may disallowed in the habitat of some endangered species.

     Databases can be designed to represent, model, and store information about these relationships as needed for particular applications.

     5. The Example of Topological Relationships

    Topology is one of the most useful relationships maintained in many spatial databases. It is defined as the mathematics of connectivity or adjacency of points or lines that determines spatial relationships in a GIS. The topological data structure logically determines exactly how and where points and lines connect on a map by means of nodes (topological junctions). The order of connectivity defines the shape of an arc or polygon. The computer stores this information in various tables of the database structure. By storing information in a logical and ordered relationship missing information, e.g., a line segment of a polygon, is readily apparent. A GIS manipulates, analyzes, and uses topological data in determining data relationships.

     Network analysis uses topological modeling for determining shortest paths and alternate routes. For example, a GIS for emergency service dispatch may use topological models to quickly ascertain optional routes for emergency vehicles. Automobile commuters perform a similar mental task by altering their route to avoid accidents and traffic congestion. Likewise an electrical utility GIS could rapidly determine different circuit paths to route electricity when service is interrupted by equipment damage. Similarly, political redistricting planners could use certain algorithms to determine logical relationships between population groups and areas for district boundaries.

     To see how topology is represented or modeled, it is useful to consider an example to see how connections are coded into a database. This involves recording more than use the absolute location of points, lines, and regions.

     The first step is to record the location of all "nodes," that is endpoints and intersections of lines and boundaries.
     
     

    Based upon these nodes, "arcs" are defined. These arcs have endpoints, but they are also assigned a direction indicated by the arrowheads. The starting point of the vector is referred to as the "from node" and the destination the "to node." The orientation of a given vector can be assigned in either direction, as long as this direction is recorded and stored in the database.
     
     

    By keeping track of the orientation of arcs, it is possible to use this information to establish routes from node to node or place to place. Thus, if one wants to move from node 3 to node 1, we can locate the necessary connections in the database.

     Now, "polygons" are defined by arcs. To define a given polygon, trace around its area in a clockwise direction recording the component arcs and their orientations. If an arc has to be followed in its reverse orientation to make the tracing, it is assigned a negative sign in the database.
     
     

    Finally, for each arc, one records which polygon lies to the left and right side of its direction of orientation. If an arc is on the edge of the study area, it is bounded by the "universe."
     
     

    Now that this information has been recorded in the database, it is possible to pose questions about connectivity and location. For example:
     
     

    Arc-node topology, as this is called, was developed several decades ago as a convenient way of store information of this sort. It is used to encode information used in the US Bureau of Census TIGER boundary files and is the basis of the spatial modeling system used by the Arc/Info software system.

     6. Object-oriented Databases

    The methods of file organization discussed above depend upon the careful description of real-world phenomena in terms of their attributes, such as height, weight, or age. It is these attributes that are stored in the database and together they provide a sort of abstracted depiction of the real-world feature. Much recent attention has focused on how to organize this information in ways that more readily represent the way users gather and use information about the world around them. That is, humans recognize "objects" immediately in terms of their totality or "wholeness." Houses and skyscrapers are recognized immediately by form and function. The differences can be described in terms of the underlying attributes, but people recognize these from experience.

     The idea of "object-oriented" database is to organize information (that is group attributes) into the sorts of "wholes" that people recognize. Instead of "decomposing" each feature a distinctive list of attributes, emphasis is placed on "grouping" the attributes of a given object into a unit or template that can be stored or retrieved by its natural name.

     Consider the following situation involving two ways of organizing information about buildings zoned for different uses.

     This information can be broken down into attributes, as follows:
     
     
    Parcel Use Height Minimum Lot Size Maximum Number of Dwelling Units
    01-4567 Residential 35 ft 10,000 SF 1
    01-5632 Residential 35 ft 7,000 SF 2
    04-6781 Residential 40 ft 43,560 SF 23
    05-3759 Residential 60 ft 43,560 SF 54
    06-3962 Office 40 ft 5,750 SF 0
    06-9977 Office 60 ft 5,750 SF 0
    To organize this information differently, let us first define some "templates" that reflect the different "objects" we wish to include in the database.
     
     
    SF Single Family  Token 1=Large Lot  Token 3=Duplex 
    MF Multi-family  Token 1=Low Density  Token 5=High Density
    LO Limited Office  Must Specify Predominate Use  Maximum Height=40 ft Minimum Lot Size=5,750 SF
    GO General Office  Must Specify Predominate Use  Maximum Height=60 feet Minimum Lot Size=5,750 SF 
    Once these are created, information can be added to our database by referring to the template. The template maintains in one place all attributes held in common by a certain class of object. It may be the case that slight differences exist between objects of a given category. These differences can be stored as "tokens" or additional attributes.
     
     
    Feature Number Token Description
    SF-1 1
    • Single Family
    •  Height=35 ft
    • Large Lot
    SF-1 3
    • Family Residence
    • Height=35 ft
    • Duplex
    MF-2 2
    • Multi-family
    •  Height=40ft
    • Low Density
    MF-5 5
    • Multi-family
    • Height=60 ft
    • High Density
    LO 40
    • Limited Office
    • Neighborhood Needs
    GO 50
    • Offices
    • City-wide Needs
    Although templates and tokens may be stored in two different files, it is easy to see how this method of organization changes the database. It is not merely a process of simplication. By using templates, users can enter and retrieve data in terms of "real" items. A query might ask for all "Single Family Houses."

     Object-oriented databases thus have the advantage of organizing information in ways that users often find easier to use. The database has as an intuitive feel because it employs that categories that users employ naturally in day-to-day life. For this reason, object-oriented databases are gaining increased attention in GIS.

     7. The Idea of the Expert Systems

    If a database has been designed to store information about spatial, functional, and logical relationships, the user can pose more complex questions of the data. That is, the user can program the system to consider a variety of spatial, functional, and logical conditions during query or analysis. Such efforts result in what are termed expert systems or, if carried further, artificially intelligent systems. At there simplest, expert systems allow the user to set "rules" that must be followed as data is analyzed. These rules are written to mirror the way an experienced user would compare or judge data. As more and more rules are written, the system becomes more adept or "expert" at finding solutions with less directed guidance by users.

     The point of expert systems is to build sets of rules that reflect the sorts of comparisons and judgments that experienced users would make. By programming these rules into the system, more and more of the work of decision making can be passed on to the computer system--including complex comparisons that may be difficult or time consuming for even experienced users to undertake.

    Such systems are of interest to GIS practioners in many fields including urban planning and resource analysis. Complex issues involving zoning and land use can often be written in terms of rules that need to be followed.

    At the same time, following rules in only a step toward "intelligence." The difference between expert systems and artificial intelligence is much in debate. But to be truly "intelligent" a system must be able to "learn," "think," or "reason," perhaps really to write its own rules from experience. The definition of artificial intelligence is, in fact, still a contentious issue. So far, it has been very difficult to program computer systems to provide a semblance of human thought processes. Yet, the potential of such systems makes the effort irresistible. The idea that computer systems might one day be able to reason about real- world environmental and geographical problems and issues is a reason why GIS theorists maintain an interest in developments in the area of artificial intelligence.

     8. Required Reading

     9. References and Supplemental Reading

     Antennucci, J.C., Brown, K., Croswell, P.L., Kevany, M.J. 1991. Geographic Information Systems. New York and London: Chapman & Hall.

     Burrough, P.A. 1986. Principles of Geographical Information Systems for Land Resource Assessment. New York: Oxford University Press.

     Huxhold, W.E. 1991. An Introduction of Urban Geographic Information Systems. New York and Oxford: Oxford University Press.

     Koeln, G.T., Cowardin, L.M., Strong, L.L. 1994. Geographic information systems. in T.A. Bookhout, ed. Research and Management Techniques for Wildlife and Habitats. Fifth Edition. Bethesda: The Wildlife Society Pages. pp. 540-566.

     Walker, J.D., Black, R.A., Linn, J.K., Thomas, A.J., Wiseman, R., and D'Attilio, M.G. 1996. Development of Geographic Information Systems-Oriented Database for Integrated Geological and Geophysical Applications. GSA Today: A Publication of the Geological Society of America 6(3):2-7.

    10. Examination Study Questions

    Last revised on 2000.3.27. LNC