What is a GIS?
GIS stands for geographic information systems. A GIS refers to the hardware, software and people used to store, organize, display, and analyze geographically referenced data – i.e. data that is tied to specific locations.
A GIS may be used to manipulate a wide-range of geographic data and address any number of questions. For example, ecologists and conservationists may use a GIS to find suitable habitat for a particular species or to determine potential migration routes. A city planner may use a GIS to keep track of zoning areas, parcels of property, and utility lines. Companies often use a GIS to determine the best location for their new stores.
While a GIS cannot function without hardware or “brainware” (i.e. people), the term GIS is often used to describe just the software within which the data is manipulated. GIS software, like ESRI’s ArcGIS, combines cartography (the art of map-making), with database management and spatial statistics. A GIS allows users to create accurate and detailed maps of nearly anything, anywhere. Like hand-drawn maps, these simplified and scaled-down models of the real world allow us to see and understand spatial relationships that may be difficult (or impossible) to determine otherwise.
Just like on a hard-copy map, maps in a GIS use symbols or features to represent real objects, whether they are part of the built environment (e.g. buildings, roads, parking lots) or the natural environment (e.g. trees, rivers, lakes, soils). Each feature on a map is associated with a particular location on the Earth’s surface. Because maps are two-dimensional representations of a three-dimensional Earth, there is always some location-related error. This error, however, is minimized using map projections, discussed in forthcoming article.
The capabilities of a GIS map extend well beyond those of a standard hard-copy map. In a GIS, map features can be associated with any number of attributes that tell you something about that particular feature. For example, you might have points indicating the approximate location of archaeological dig sites. When you select one of the points within the GIS’s user interface (UI) you can bring up the associated database, which can can tell you not only where the site is located, but maybe whether the site is active, what has been found there, or who has been working there.
Spatial data within a GIS can be of two types: vector and raster. Vector data is most commonly used to represent features that have exact locations.Vector data is represented by three geometric shapes: points, lines and polygons. Points represent features that – at a particular scale – do not have (or need) dimensions. This may be as simple as a particular latitude-longitude coordinate used to represent the center of a city or the location of a well. At a finer (more detailed) scale, point features may actually become polygon features. Lines are one-dimensional shapes that represent linear features such as rivers, streets, and pipelines. Finally, polygons are two-dimensional features used to represent any object with area at a particular scale.
Scale is key here. Cities, for example, may be represented as dots on a map showing an entire country, but polygons on a map depicting a small region.
In the raster data format, information is contained within discrete cells known as pixels. A lot of digital data is stored as as a raster, including the color values displayed on your computer screen and in digital photographs. You can think of raster data as a grid of information with rows and columns, and each individual cell containing a particular value.
In a GIS, raster data sets are most often used to store and display continuous data – that is, data that varies continuously across space. One typical example of continuous data is rainfall. Rainfall over an area varies continuously from location to location. Raster data, like rainfall, may be obtained by first sampling data and plotting their location using points. We use sample points because we cannot know the value of a continuous variable at every single location. Using sample data points, however, we can estimate a surface of values, a raster data set, for any number of continuous variables such as rainfall, temperature, land cover, or elevation.
One of the most powerful functions of a GIS is the ability to overlay multiple data layers and perform spatial analyses. Users can investigate an endless variety of spatial relationships that may exist between multiple data layers or spatially-referenced variables. For instance, one might ask how topography (the “lay of the land”) affects local air quality in an region with high mountain ranges like California’s Central Valley. By overlaying a layer of a common air pollutant such as ozone atop an elevation layer (right) it is possible to see that the Sierra Nevada mountain range does seem to trap high levels of ozone within the valley. Combining layers like this clearly aids our ability to explain spatial patterns.