Have you ever found driving directions and maps online, used a smartphone to ‘check in’ to your favorite restaurant, or entered a town name or zip code to retrieve the local weather forecast? Every time you and millions of other users perform these tasks, you use Geographic Information Science (GIScience) and related spatial technologies. Many of these technologies, such as Global Positioning Systems (GPS) and in-vehicle navigation units, are very well-known, and you can recall the last time you used them.

Other applications and services that are the products of GIScience are a little less obvious, but they are every bit as standard. For example, you use geospatial technologies if you are connected to the Internet. A geographic lookup occurs whenever your browser requests a web page from a Content Delivery Network (CDN). The server you are connected to contacts other servers closest to it and retrieves the information. This happens so that the delay between your request to view the data and the data sent to you is as short as possible.

GIScience and related technologies are everywhere, and we use them every day. However, when it comes to information, “spatial is special.” Reliance on spatial attributes separates geographic information from other types of information. There are several distinguishing properties of geographic information. Understanding them and their implications for the practice of geographic information science is critical to utilizing geographic data.

• Geographic data represent spatial locations and nonspatial attributes measured at certain times.
• Geographic space is continuous.
• Geographic space is nearly spherical.
• Geographic data tend to be spatially dependent.

Spatial attributes tell us where things are or where things were when the data were collected. Geographic data allows us to ask many geographic questions by including spatial attributes. For example, we might ask, “are gas prices in Puyallup high?” The interactive map from GasBuddy.com can help us with such a question while enabling us to generate many other spatial inquiries related to the geographic variation in fuel prices.

Another essential characteristic of geographic space is that it is “continuous.” Although the Earth has valleys, canyons, caves, oceans, and more, there are no places on Earth without a location, and connections exist from one place to another. Outside of science fiction, there are no tears in the fabric of space-time. Modern technology can measure location precisely, making it possible to generate incredibly detailed depictions of geographic feature locations (e.g., of the coastline of the eastern U.S.). Unfortunately, it is often possible to measure so precisely that we collect more location data than we can store and much more than is helpful for practical applications. How much information is useful to store or display on a map will depend on the map scale (how much of the world we represent within a fixed display, such as the size of your computer screen) and the map’s purpose.

In addition to being continuous, geographic data tend to be spatially dependent. More simply, “everything is related to everything else, but near things are more related than distant things” (which implies that things near one another tend to be more alike than things far apart). A statistical calculation known as spatial autocorrelation can measure how alike things are concerning their proximity to other things. Without this fundamental property, geographic information science as we know it today would not be possible.

Data Collection

Geographic data comes in many types from diverse sources and is captured using many techniques; they are collected, sold, and distributed by many public and private entities. In general, we can divide geographic data collection into two main types.

Directly collected data are generated at the source of the phenomena being measured. Examples of directly collected data include temperature readings at specific weather stations, elevations recorded by visiting the location of interest, or the position of a grizzly bear equipped with a GPS-enabled collar. Also included here are data derived from surveys (e.g., the census) or observations (e.g., Audubon Christmas bird count).

Remotely sensed data are measured remotely without direct contact with the phenomena or needing to visit the locations of interest. Satellite images, sonar readings, and radar are all remotely sensed data.

Maps are both raw material and geographic information systems (GIS) products. All maps represent features and characteristics of locations, and that representation depends upon data relevant at a particular time. All maps are also selective; they do not show us everything about the place depicted; they only show the features and characteristics their maker decided to include. Maps are often categorized into reference or thematic maps based on the producer’s decision about what to have and the expectations about how the map will be used. The prototypical reference map depicts the location of “things” usually visible worldwide; examples include road maps and topographic maps showing terrain.

Thematic maps, in contrast, typically depict “themes.” They are more abstract, involve more processing and interpretation of data, and often describe not directly visible concepts; examples include maps of income, health, climate, or ecological diversity. There is no clear-cut line between reference and thematic maps. However, the categories are helpful to recognize because they relate directly to how the maps are intended to be used and their cartographers’ decisions in shrinking and abstracting aspects of the world to generate the map. Several types of thematic maps include:

• Choropleth – a thematic map that uses tones or colors to represent spatial data as average values per unit area.
• Proportional symbol – uses symbols of assorted sizes to represent data associated with different areas or locations within the map.
• Isopleth–contour or isopleth maps depict smooth, continuous phenomena such as precipitation or elevation.
• Dot – uses a dot symbol to show the presence of a feature or phenomenon – dot maps rely on a visual scatter to offer a spatial pattern.
• Dasymetric – an alternative to a choropleth map, but instead of mapping the data so that the region appears uniform, ancillary information is used to model the internal distribution of the data.

Click the “Previous” button on the lower left or the ‘Next” button on the lower right to navigate throughout the textbook.