Flow Maps

Flow maps depict the movement of phenomena between geographic locations, typically using lines of varying widths to depict quantitative data, although different color hues, in the case of qualitative data, may also be used. In the older literature, flow maps are also referred to as dynamic maps. In directed-flow maps, the flow travels along the line only in the direction of the arrowhead. In contrast, in undirected-flow maps, arrows are missing, and movement is possible in both directions.

Flow maps can be categorized into the following five types: distributive, network, radial, continuous, and telecommunications. A distributive flow map depicts the movement of commodities, people, or ideas between geographic regions. Flows within a network, such as the transportation network, are depicted in network flow maps. Radial flow maps possess a distinctive radial or spokelike pattern. Good examples are a street network based on traffic circles and an airline route network based on the airport hub structure. The movement of continuous phenomena, such as wind or ocean currents, is best shown with continuous flow maps, while the flows of information technology, including the Internet, are depicted in telecommunications flow maps.

The earliest flow map dates to 1837 and was created by Henry Drury Harness. Minard of France not only popularized the flow map among statisticians (Figure 1A) but also became known for the flow chart showing the demise of Napoleon's army in Russia. In the early decades of the 20th century, flow mapping became common in [Page 1139]geography textbooks with the impetus supplied by the study of economic geography. Among the first to develop an automatic solution for displaying migration flows was Waldo Tobler, with his Flow Mapper software (Figure 1B). Details about the software, including a free download option, can be found at http://www.csiss.org/clearinghouse/FlowMapper. A more recent example of a flow-mapping software was developed by Doantam Phan and his colleagues from Stanford University (Figure 1C). Details about their software, including a free download option, can be found at http://graphics.stanford.edu/papers/flow_map_layout.

Figure 1 (A) Minard's 1864 flow map of wine exports from France. (B) Tobler's computer-generated flow map of migration from California from 1995 to 2000. (C) A flow map produced by Doantam Phan and his colleagues from Stanford University shows the same migration.

Source: Phan, D., Xiao, L., Yeh, R., Hanrahan, P., & Winograd, T. (2005). Flow map layout. In INFOVIS ‘05: Proceedings of the 2005 IEEE Symposium on Information Visualization. Washington, DC: IEEE Computer Society. Used with permission.

Recently, Ningchuan Xiao and Yongwan Chun developed the kriskogram, which is a new method to visualize migration flows. In a kriskogram, geographical units are projected as a set of points on a straight line segment called a location line. The migration flow between two points on the location line is represented using a half-circle drawn from the origin to the destination in a clockwise direction (Figure 2, next page).

Figure 2 Kriskogram of interstate migration in the conterminous United States

Sources: U.S. Census state-to-state data; Xiao, N., & Chun, Y. (2009). Visualizing migration flows using kriskograms. Cartography and Geographic Information Science, 36(2), 183–191. Reprinted with permission from Cartography and Geographic Information Science.

Note: The states are arranged according to the longitude of their centroids. Migration flows with magnitude smaller than 77,059 are treated as background and are not shown.

...