Geographic Information Systems

views updated

GEOGRAPHIC INFORMATION SYSTEMS

Information systems may be broadly divided into nonspatial and spatial categories. Most information systems, including management information systems, do not refer their data to a spatial coordinate system. For example, payroll records are usually linked to a person rather than a specific location. Spatial information systems refer data to some coordinate system. For example, architectural software records the spatial relationship of beams to the foundation of a building but not necessarily to the location of the beams or the building on Earth's surface. Geographic information systems (GISs) are a subset of spatial information systems that do refer information to location.

Locations in a GIS are usually referred either directly or indirectly to coordinates denoted by latitude, longitude, and elevation or depth—or some projection of these onto a flat surface. The sciences of geodesy (concerned with the size and shape of Earth and the determination of exact positions on its surface) and cartography (concerned with the creation of maps) are therefore integral parts of geographic information science in general and of geographic information systems in particular. The combination of science and technology required for a GIS is reflected in the range of available GIS software. Some GIS software has been developed from specialized architectural drawing packages, some from computerized mapping systems, some from geographic extensions of relational databases, and others from mathematical graphing and visualization software. GISs vary in sophistication from systems that simply show the location of items ("what-where systems") or that allow the user to search for information with a location-based query ("where-what systems") to relational databases that are searchable by feature, attribute, relative location, and absolute location.

Conceptual Elements

Conceptually, a GIS consists of a data input subsystem, a data storage and retrieval subsystem, a data manipulation and analysis subsystem, and a reporting subsystem. The data input subsystem may include arrays of values (raster data or imagery), points, lines or areas constrained by coordinates (vector data), or tables of attribute data. Common data input methods include raster scans of existing maps or images, satellite and aircraft imaging systems, digitizing the vertices of map points, lines, or areas to generate vector files (electronic line drawings), uploads of coordinate data from global positioning system (GPS) receivers, or uploads of flat files of attribute data.

The data storage and retrieval subsystem is usually a relational database similar to those used in other branches of information science with extensions for manipulating location-based or georeferenced metadata and data. Georeferenced data are usually stored in one of three forms: raster, vector, or flat files. Noncoordinate-based locations, such as street addresses, may be referred to coordinate-based locations through a process known as geocoding, which assigns coordinates to each feature.

The data manipulation and analysis subsystem translates the data into a common format or formats, transforms the data into a common map projection and datum for compilation, performs calculations, and facilitates analysis of the data with regard to shape, contiguity, orientation, size, scale, neighborhood, pattern, and distribution with location-based queries. Therefore, this manipulation and analysis component of a GIS enables geographic knowledge management. Such systems are used to estimate landslide current risk, calculate property values and tax assessments, and manage wildlife habitat.

Some sophisticated GISs also allow the user to take geographic knowledge management to the next level by incorporating a modeling capability. The modeling capability allows the user to adjust variables and mathematical relationships between variables (algorithms) in the situation represented by the GIS to help determine the probable cause of the situation and to simulate what might occur if the situation were to change. Some GISs automate this last capability via built-in or modular computer programs such that the GIS becomes a geographic decision support system (GDSS). GDSSs are used to model the effect of various land-use policies on urban growth, the effect of timber cutting patterns on soil erosion, and the effect of toxic waste dumps on groundwater quality. The accuracy of a GDSS is commonly evaluated with time-series historical and real-time data.

The reporting subsystem of the GIS displays the processed information in tabular, graphic (e.g., histograms, pie charts, and surfaces), or map form, depending on the information need of the user and the capability of the GIS. The first two types of report are similar to those of most other information reporting systems. The map form links GIS to the sciences of geodesy and cartography as well as computer mapping, image processing, and visualization technologies.

Physical Elements

Physically, a GIS usually consists of input hardware and software, magnetic or optical storage, a central processing unit with GIS software, and output hardware and software. The advent of networks and the Internet has resulted in a phenomenal increase in the variety and amount of input hardware for GISs. Traditional sources of geographic information (input) include scanners, digitizing tables, keyboards, pointers, and mice. Software for GIS input includes raster-to-vector conversion programs, image processing and classification programs, geocoding packages, and various digitizing packages for translating analog maps and drawings into a vector-based GIS.

The heart of a GIS is computer hardware and software. Growth in the number and capability of computers was essential to the development and widespread use of GISs. Rapid increases in the performance/price ratio of computer hardware and decreases in overall cost for entry-level computer systems between 1985 and 2000 facilitated the rapid growth of GISs. Input hardware such as color scanners and GPS receivers decreased by a factor of ten during that period. A usable workstation including a central processing unit, disk storage, CD-ROM drive, monitor, random-access memory (RAM), keyboard, and mouse decreased in cost from several thousand dollars in 1985 to less than $1,000 in 2000.

Affordable input and processing hardware expanded the market for GIS software from tens of thousands of users to hundreds of thousands of users between 1985 and 2000. This resulted in the evolution of easy-to-use commercial GIS software with graphic user interfaces from the previous, difficult-to-use UNIX-based command-line systems of the mid-1980s. Similarly, the cost of sophisticated graphic user interface (GUI) GIS software decreased to less than $3,000 and simple systems to less than $100. The core technology necessary for GIS is now well developed, and the price of GIS technology is no longer a barrier to its widespread use.

The Information

While technology and its cost were key factors in the explosive growth of geographic information science after 1985, the primary fuel for the adoption of GIS and the development of geographic information science was the availability of low-cost, entry-level geographic information from federal, state, provincial, and tribal governments. Efforts by the U.S. Geological Survey, the Bureau of the Census, the National Oceanic and Atmospheric Administration (NOAA), the Environmental Protection Agency (EPA), the military, and academia led to the development of a limited number of data formats and metadata exchange standards based on those of the Library of Congress and the Federal Geographic Data Committee. These agencies made their data and standards available to the public and the value-added industry at little or no cost. These early efforts led to increased governmental and public awareness of the utility of GIS and, at least, the beginnings of interoperability with regard to GIS software, data, and metadata.

The Future

The growth rate of GISs seems likely to increase. The advent of the Internet has resulted in new possibilities for geographic information capture, exchange, storage, analysis, and dissemination. Internet map and geographic information servers, distributed storage, federated electronic clearinghouses, online data access, enhanced visualization tools viewable through simple and GPS-aware Internet browsers and other thin-clients decrease the cost barriers for the GIS and flatten its learning curve still further. The economies of scale and tremendous value added by networking geographic information systems allow users to share expensive information resources such as satellite data for collaborative education, research, and planning purposes, all of which used to be prohibitively expensive.

Low-cost, high-accuracy GPS receivers in combination with wireless telecommunications are rapidly increasing the amount of georeferenced information. GISs are now used to help manage the construction and maintenance of both the wired and wireless parts of the Internet as well as to synchronize its servers. As bandwidth expands and compression techniques improve, large amounts of real-time, raster-based information, including satellite imagery and three-dimensional visualizations, will be added to Internet-based GISs. Wireless networks, bandwidth, GPS, satellite imagery, and improved visualization make geographic information more plentiful and easier for users to interpret. This will drive the demand for ancillary georeferenced information. For example, if users know where they are via GPS, they may wish to query an Internet-based GIS to find the nearest restaurant. Eventually, georeferenced information will increase to the point where the terms "information system" and "geographic information system" are synonymous in the minds of most users and information scientists.