Geographic Information Systems

Many difficult decision-making processes in geographic planning, environmental, and social studies arise because they require the manipulation and integration of large volumes of spatially-indexed data. Computerized geographical information systems (GIS) provide the key to the efficient use of such data sets because they can efficiently perform such functions on vast quantities of spatially addressed data. These powerful systems acquire data from many sources; change the data into a variety of useful formats; store the data; retrieve and manipulate the data for analysis; and then generate outputs required by a given user.

Underlying each GIS is a powerful spatial database management subsystem that provides such functions as efficient data storage, retrieval and updating, as well as security, integrity, and redundancy control. Within the database, a variety of mappable characteristics associated with any given geographical location are organized as a series of spatially-registered maps, or "overlays." Each of these overlays describes a single theme of geographical features (such as water features, geology, or vegetation), and is stored as a separate data layer. Each feature in a map layer is described by the linking of two components: geometry and attributes.

For example, a road is represented geometrically as a line linked with attributes describing road type, name, and conditions. A land parcel has an interior and a boundary, and is represented by a polygon associated with attributes describing ownership, value, and tax. Additional topological or metric information (such as neighborhood, connectivity, distance, and directions) may also be computed from geometry and stored for fast response to topology-related queries.

A GIS can integrate information from a variety of sources and formats. Common geographical feature maps that describe administrative boundaries, hydrologic features, land use, parcels, transportation, and population data, are widely available from government agencies and private data vendors. Additional data can be captured through field measurement, use of GPS (global positioning system), and direct interpretation of remotely sensed images. There are also common data standards for sharing digital geographic data among users and producers. In order to answer queries that require combinations of multiple data inputs, source data must be pre-processed—converted for consistent format and scale and errors corrected—to achieve consistency and compatibility in data types, scales, spatial reference systems, spatial units, and coverage.

A typical GIS is capable of performing a variety of data analysis procedures that involve access to both spatial and non-spatial information of one or more geographical feature classes. As a generic spatial analysis engine, a GIS provides a number of primitive operations that can be combined to implement a given analytical procedure.

Map reclassification operations merely repackage information from an existing map layer into a new one that has feature classes and attribute categories different from the original one. Map overlay operations involve the creation of a new map layer from two or more existing map overlays. It begins by finding point-by-point or region-based spatial coincidence among input maps, and then assigns a new value to each location based on the spatial coincident values of input overlays. An overlay operation requires that input map layers must be compatible in scale, coordinate systems, error characteristics, and data coding methods.

Other analytical functions of GIS include buffer zones, neighborhood characterization, and connectivity measurement. A particular feature of GIS is the ability to calculate more realistic distance measures among objects based on actual geometry, travel time, and cost, rather than straight-line distance. The trend in developing GIS analytical functions is to better integrate GIS with other software in statistical analysis, operations research, and artificial intelligence (AI) tools.

In the forty years since their first appearance in the 1960s, geographical information systems have experienced phenomenal growth in sophistication, size, and popularity. Today, GIS is a multi-billion dollar industry and has become part of a basic information infrastructure for private enterprises, government agencies, and academic institutions. The majority of the operational GIS are used for thematic mapping, handling spatial queries, and decision-making support.

Thematic mapping involves making special purpose maps that provide appropriate content, context, symbolization, and illustrations. Geographic information systems provide utilities to make all stages of map-making faster, more accurate, and easier to do. Spatial query processing selects objects based on the combination of their locational, topological, and descriptive attributes. Decision-support functions of GIS allow users to ask "what if"questions to evaluate alternative actions.

In terms of practical applications, GIS programs assist with resource inventory, administrative recordkeeping, communication or transportation infrastructure and facility management, emergency response, and environmental monitoring preservation. Examples of GIS applications include understanding the process of water and nutrient flows over the land surface; identifying potential markets for a product in the local area; and deciding on the fastest route over which to send a fire truck to a downtown restaurant. All these tasks utilize the functionalities of a GIS to deal with complex spatial data sets.

Geographical information systems represent a synergistic development of multiple technical fields, including computer cartography, computational geometry, spatial and relational databases, and information visualization. Fast retrieval of spatial data is made possible by using efficient indexing schemes such as quadtrees or R-trees. The processing of spatial queries (e.g, point-in-polygon) relies on computational geometry algorithms such as line sweeping, Voronoi diagrams, and convex hulls. The increasing trend to build spatial data types and spatial operators in database engines make it possible to model the complexities of spatial data with object-relational or object-oriented approaches. Computer graphics and visualization techniques have increased the effectiveness of communicating spatial information to users.

Looking into the future, geographical information systems will likely become more user-friendly. With the development of web-based GIS, there is an evolution from single-user systems to more open, multiple-user systems. There has been much cross-fertilization between the fields of artificial intelligence and spatial information systems due to increased needs to deduct new information from retrieved data. Knowledge-based geographic information systems that incorporate deductive reasoning and learning facilities will be more powerful in modeling human spatial intelligence in the environment. An object-oriented, intelligent, dynamic, multimedia geographical information system will revolutionize people's views of the spaces and worlds in which we live.