||Integrated computer tools for handling, processing and analysing geographic data, that is, data explicitly referenced to the surface of the Earth. They include self-contained software packages for personal computers and workstations, as well as tools for handling and processing geographic information over high-speed networks such as the Internet. Increasingly, the term \'GIS\' is used to imply any activity related to geographic information in digital form; hence \'GIS community\', \'doing GIS\', \'the GIS industry\' and \'GIS data\'.
Although the first mention of GIS occurred in the literature in the mid-1960s (Foresman, 1998), massive growth began only in 1980 with the introduction of super-minicomputers by manufacturers such as Digital Equipment Corporation and Prime Computer. Growth in the software industry followed, led by Intergraph Corporation and Environmental Systems Research Institute (ESRI), who remain the market leaders today. The technology of GIS found practical applications in resource management, particularly forestry, local government, the utility industries and geodemographics. More specifically, its uses include the automated measurement and analysis of geographically distributed resources, and the management of distributed facilities. Scientific applications also developed in the 1980s, as GIS was applied to the wide range of sciences and social sciences that deal with geographically distributed data and find value in a spatial perspective. These include epidemiology, archaeology, geology, ecology, geophysics, oceanography, regional science, and of course geography. The methods and concepts of GIS overlap strongly with the concerns of many more established disciplines, including cartography (particularly computer-assisted cartography), remote sensing, photogrammetry, geodesy and surveying.
Much of the impetus for the initial development of GIS came from the difficulties of analysing data shown on paper maps, including such simple tasks as measuring area. To measure areas from a paper map (for example, to measure the amount of Class 1 agricultural land, or to estimate the amount of timber in a forest stand based on the stand\'s area and the density and size of trees), it is necessary either to use a mechanical device known as a planimeter, or to overlay a transparent sheet of dots and then count them. Both methods are laborious and inaccurate. But if the map can be represented in a computer the calculation is accurate and fast. Today we make such calculations routinely, without realizing how difficult they were before computerization.
Roger Tomlinson was the moving force behind the development in the 1960s of the Canada Geographic Information System, a computer application for the analysis of the data collected by the Canada Land Inventory in the interests of improved land-use policy. Research groups at Harvard (the Laboratory for Computer Graphics and Spatial Analysis, led by William Warntz) and elsewhere conducted basic research into methods for handling geographic data in digital form during the 1960s and 1970s. A significant expansion of research occurred in the late 1980s, after it had become clear that GIS had great potential as a tool to support research and decision-making in a wide range of fields. In the US, the National Center for Geographic Information and Analysis was initiated in 1988 with major funding from the National Science Foundation, based largely on the efforts of Ron Abler, then Director of the NSF Geography and Regional Science Program. The Center exists to conduct basic research in GIS and its applications, particularly in science; funds were awarded to a consortium of the University of California, Santa Barbara; the State University of New York at Buffalo; and the University of Maine. In the UK, the Economic and Social Research Council funded the development of a network of Regional Research Labs, largely based in university Geography departments, to promote GIS applications in a mix of practical and scientific applications, and to build a stronger UK research base. The European Science Foundation launched its GISDATA programme in 1993, under the leadership of Ian Masser, aimed at fostering research collaboration between European countries. Programs similar to these exist in many other countries.
Two major traditions have developed in GIS for representing geographic distributions. The raster approach divides the study area into an array of rectangular cells, and describes the content of each cell; the vector approach describes a geographical distribution as a collection of discrete objects (points, lines or areas), and describes the location of each. In essence, the raster approach \'tells what is at every place\' and the vector approach \'tells where everything is\'. In addition, a GIS database contains information on the attributes of each cell or object, and on various kinds of relationships between objects. Broadly, the continuous view of space embedded in the raster approach is most commonly associated with environmental and physical science applications of GIS, while the view of space as a collection of discrete objects that is implicit in the vector approach has found more applications in the social and policy sciences, in the mapping industry, and in the management of geographically distributed facilities. Most currently available GIS software products can be identified with one approach, although most also provide limited capabilities for handling the other.
The ability to couple the input and output functions of a GIS with its more exploratory functions of browsing and simple statistical analysis, and with more sophisticated confirmatory techniques, has led to many GIS applications in human geography and related disciplines. GIS can be used to \'zoom in\' on parts of an area, displaying them at higher resolution, or to \'pan\' rapidly across large mapped areas. It can be used to explore statistical relationships between geographical variables, such as the relationship between rainfall and agriculture in arid regions, or between ethnic groups and voting patterns. GIS has been used to implement models of regional economies, transportation systems and urban growth; to develop archaeological hypotheses from complex catalogues of spatially distributed artifacts; to develop understanding from patterns of social deprivation and disease; to analyse voting behaviour; and to understand the sacred meaning many cultures give to space.
For example, GIS has been used to develop reconstructions of Iron Age landscapes in southern England. Three-dimensional models were built and used to gain insights into the criteria for the selection of sites for long barrows, based on intervisibility (Lock and Harris, 1996). GIS have also been used to construct models of pre-Columbian agricultural systems in the US Southwest. Agricultural production models were built using detailed maps of soils, drainage, and rainfall; and used in simulations of production under various climate scenarios. These were linked to data from other sources, such as tree-ring growth, to provide insights into the effects of drought and possible explanations for the catastrophic collapse of early cultures in the centuries before European contact (Van West and Kohler, 1996). Examples of applications can be found in the popular GIS magazines (GIS World, Geo Info Systems), in the proceedings of GIS conferences, and in introductory texts. In short, GIS has become a powerful tool for automating the geographer\'s processes of analysis and synthesis, and has been adopted in many other related disciplines.
Currently, GIS remains firmly bound to its cartographic roots; maps continue to be the primary means of input and output. It provides tools for recording and processing the positions of features in space, but has yet to develop much sophistication in its handling of time-dependence (see time-geography), or interaction. In this sense GIS preserves a container-like view of space, and cannot yet deal effectively with the temporal changes and interactions that drive or result from many social processes. However, basic research efforts over the past decade have yielded significant advances in these areas. Moreover, the growth of GIS has led to renewed interest in many of the more fundamental issues of geography and cartography: the accuracy of abstracted views of geographic distributions; the effects of scale and resolution; languages for describing spatial relations; methods for exploring the spatial perspective; the role of geographic information in empowerment and domination (cf. power; surveillance); and the importance of geographic information to the decision-making process.
A flurry of recent literature has drawn attention to the importance of the social context of GIS, the need to support alternative ways of knowing about the human environment, and the limitations of early GIS technology in that regard (Pickles, 1995; Openshaw, 1997). Although most developers of GIS technology would argue that their contributions are inherently neutral with respect to their social implications, Smith (1992) makes the case that much of GIS development has been driven by military and intelligence applications, which are far from neutral. Others have argued that GIS use in marketing has the potential to invade personal privacy, and that GIS use in environmental decision-making reinforces the power of its users (notably governments and corporations) over those who lack access to it (notably marginalized groups). Other arguments in this social critique of GIS focus on the limited representations that are possible in the digital environment, and how these favour centralized, uniform views over individual, more idiosyncratic ones.
GIS continues to be of great importance to the discipline of Geography, and its popularity has done much to mould the contemporary image of Geography as a discipline in the minds of others. The GIS Specialty Group has been the largest in the Association of American Geographers for many years. The growth of GIS has created a demand for students with GIS skills, and most Geography departments have responded by instituting sequences of courses. Recent surveys suggest that over half of all of University GIS courses worldwide are taught in Geography departments, and \'GIS/remote sensing\' was listed most often in a recent survey of US Geography department chairs as the occupation for which students were being prepared (Gober et al., 1995). But debate continues in many departments over the appropriateness of the massive and continuing investments in hardware and software that are needed to support a GIS program; about the intellectual importance of GIS (Wright et al., 1997); and about the importance of the specialty relative to other skills that compete for the attention of students and staff. Geographers have been among the leaders in identifying geographic information science as an important new research field addressing the fundamental issues arising from the use of digital computers to represent, process, and analyse geographic information. In the US, the University Consortium for Geographic Information Science was formed in 1996 in order to act as a national focus for promoting the new field, and now has close to 50 institutional members, including all of the key centres of basic research.Â (MG)
References Foresman, T.W., ed., 1998: The history of geographic information systems: perspectives from the pioneers. Upper Saddle River, NJ: Prentice Hall.Â Gober, P., Glasmeier A.K., Goodman, J.M., Plane, D.A., Stafford, H.A., and Wood, J.S. 1995: Employment trends in Geography, Part 2: Current demand conditions. Professional Geographer 47(3): 329-36.Â Lock, G.R. and Harris, T.M. 1996: Danebury revisited: An English Iron Age hillfort in a digital landscape. In M. Aldenderfer, and H.D.G. Maschner, eds, Anthropology, space, and geographic information systems. New York: Oxford University Press, 214-40.Â Openshaw, S. 1997: The truth about Ground Truth. Transactions in GIS 2 (1): 7-24.Â Pickles, J. 1995: Ground truth: the social implications of geographic information systems. New York: Guilford Press.Â Smith, N. 1992: History and philosophy of geography: real wars, theory wars. Progress in Human Geography 16 (2): 257-71.Â Van West, C, and Kohler, T.A. 1996: A time to rend, a time to sew: New perspectives on Northern Anasazi sociopolitical development in later prehistory. In M. Aldenderfer, and H.D.G. Maschner, eds, Anthropology, space, and geographic information systems. New York: Oxford University Press, 107-31.Â Wright, D.J., Goodchild, M.F., and Proctor, J.D. 1997: Demystifying the persistent ambiguity of GIS as \'tool\' versus \'science\'. Annals of the Association of American Geographers 87 (2): 346-62.
Suggested Reading Aldenderfer, M., and Maschner, H.D.G. 1996: Anthropology, space, and geographic information systems. New York: Oxford University Press.Â Chrisman, N.R. 1997: Exploring geographic information systems. New York: John Wiley & Sons.Â DeMers, M.N. 1997: Fundamentals of geographic information systems. New York: John Wiley & Sons.Â Longley, P.A., Goodchild, M.F., Maguire, D.J. and Rhind, D.W., eds, 1998: Geographical information systems: principles, techniques, management and applications, 2 vols. Cambridge: GeoInformation International.