Accessible Color: The Work of Cynthia Brewer

A cartographer currently on the faculty of the Pennsylvania State University, Cynthia Brewer has spent her time researching color use on maps.  Her interest in this topic has led her to focus on two aspects of color: the accessibility of colors on maps for the user; and the accessibility of good color schemes for cartographers and other people who use color in design.  Brewer’s research in this area has helped create a cartographic field that produces maps that are easy for geographers to produce and also pleasant for the audience to view.  This has great implications for the future of map use for communication and visualization; perhaps even more importantly, Brewer has made geography noticed by people in other fields by virtue of her innovative color selection application, ColorBrewer.

After receiving her Bachelor of Arts degree in fine arts with a geography minor from the University of Guelph in 1983, Brewer moved to Michigan.  She studied geography with an emphasis in cartography at Michigan State University.  Brewer worked closely with Dr. Judy Olson, who served as her advisor for both her Masters Degree and Ph.D (Brewer 2004).

Even when she was an undergraduate student, Brewer had an interest in cartography.  Her undergraduate thesis was on the use of cubes as a map symbol.  After that paper, however, she chose to focus on color use in maps.  She discussed the development of Munsell charts for printing maps in her Master’s thesis and the topic of simultaneous contrast for her Ph.D (Brewer 2004).

Brewer went on to join the faculty of San Diego State University after she earned her Ph.D in 1991.  She spent three years there, and then moved to the Pennsylvania State University in 1994, where she still teaches and performs most of her research.  In addition to her work at the university, she also has done consulting work.  In this capacity, she has worked for ESRI, Inc., (helping them choose the color schemes they include for the display of cartographic data) in 2004 and the United States Census Bureau (producing an atlas of the ethnic makeup of the United States and consulting on the creation of a comprehensive atlas) from 2000 to 2004 (Brewer 2004).

As was previously stated, Brewer has a strong interest in color and its use in cartography.  Most of her important work has been focused in this field.  Her interests in color have been varied, but the overriding theme of her work is ultimately access – how easy is it for the audience to “get” the information the cartographer is trying to convey through color? how simple is it for cartographers to find and utilize appropriate color schemes?  There are myriad specific issues associated with these two questions.  Brewer has done a lot of research to explore the topic and a lot of work to provide solutions that work well for not only mapmakers but also other designers.  She is well-respected in the field, being one of the most-cited people in a leading cartographic design textbook (Slocum, McMaster, Kessler and Howard 2005).

Brewer’s research on the first topic – the accessibility of mapped information through color – has been both affirming and surprising.  Her review of simultaneous contrast research upheld the usefulness of testing colors in “a perceptual structure relevant to the map reading context” and found that “the opponent-process approach is the most common and promising avenue for modeling [simultaneous contrast],” meaning that the form of research has been sound, but that the testing environment needs to be constructed to be as similar as possible to those conditions that maps would typically be viewed under (Brewer 1992: 28; Brewer 1992: 28).

This was just the first step in Brewer’s work with color.  She went on to present guidelines for color with varying schemes of mapping (for example, for a balanced scheme, or a sequential/sequential scheme) for easy interpretation by the audience (Brewer 1994).  From there, she went on to address special cases of color use.

One of Brewer’s special cases is the selection of color to accommodate those with color-vision impairments.  Olson and Brewer (1997) performed an experiment to gauge the ability to read maps of people with normal vision and impaired vision.  Not surprisingly, people with color-impaired vision took more time to answer questions and performed worse on maps that were designed to be potentially confusing.  What was surprising was the fact that the color-impaired people performed just as well as normal-vision people when using maps with accommodating color schemes, but still took significantly longer to answer the questions.  They also found that people with normal vision have no problems with reading accommodating maps, indicating that maps should be made to aid those with color impairments as often as possible (Olson and Brewer 1997).  To see an example of a color-impaired accessible map, please see Figure 1.

Figure 1. An illustration of a map designed with colorblind people in mind (Pickle, Mungiole, Jones and White 1996:136).

Brewer has also been a champion of spectral schemes for use with quantitative data.  Spectral schemes are color schemes that go through the hues in order of their dominant wavelengths – red, orange, yellow, green, blue and purple.  Many noted cartographers have rejected such a scheme as “illogical and inappropriate” (Brewer 1997:203).  Brewer performed research to show that a well-designed spectral scheme can work as well as any other diverging scheme to show data that diverges from the mean (Brewer 1997).  As shown in Figure 2, a spectral scheme can convey the relevant information as well as other schemes.  This frees cartographers to use this scheme if they wish to, increasing the number of options available.

Some of Brewer’s most valuable experience that led to her most recent work has been working on mapping projects.  She worked on the Census diversity atlas, which helped her develop some ideas on creating map series that work together well – please view Figures 4 and 5 for an illustration (Brewer 2001).  It also led to considerable improvement in the maps produced (see Figures 3 and 4).  She also worked on the creation of epidemiological rate maps, which led to the revelation that people prefer cool colors for disease maps (Brewer and Pickle 2002).  Both experiences exposed her to people whose ability to create color schemes were not as strong as those of academic cartographers.

Figure 3

Figure 3. A pre-Brewer census map depicting the density of Hispanic people. Note the nonsensical color scheme – why include the pink at all? Or, if you want to include it, why not make it a diverging scheme, as opposed to this confusing one? (U.S. Census 1990).

Figure 4

Figure 4. A map Brewer produced for the Census Bureau. The color scheme makes it much easier to obtain information about the distribution of Hispanic individuals in the United States (Guzmán 2001).

Figure 5

Figure 5. Note how this map and the map in Figure 4 look like they go together, despite different themes (Brewer and Suchan 2001:7).

Brewer’s more recent work has focused on another issue of access: the accessibility of good color schemes for cartographers, professional and novice alike.  The most important of her developments in this regard is ColorBrewer (see Figure 6).  A lot of her current work revolves around this innovative web application.  One of the most important, and unexpected, consequences of putting ColorBrewer on the Internet was that not just cartographers had access to it.  Brewer mentions that she has heard from “GIS analysts and developers, remote sensing consultants, computer programmers, statisticians, medical instrument designers, geologists, and graphic artists,” all of whom used ColorBrewer’s schemes to produce maps that they can feel confident in disseminating (Brewer 2003:161-2).

While ColorBrewer provides well-designed color schemes for maps, it relies on the display of CRT monitors for the colors to look good.  Brewer’s goal is to provide good color schemes all-around, for either display on screens and in print.  Thus, she also published a work that “provided a catalog of printed choropleth map color schemes, so that their quality and suitability may be evaluated by mapmakers who are seeking assistance in choosing colors for print projects” (Brewer, Hatchard and Harrower 2003:7).  The main goal of this is to provide cartographers with printed out examples of the colors used in ColorBrewer, thus saving them time and money when producing maps.  Judy Olson suggested that it is one of Brewer’s more important contributions, indicating that having a printed version of the colors helps immensely with color choice (2005).

ColorBrewer has other benefits as well.  Its sample map allows the user to try out the scheme of their choice.  The application provides five different color specification types in order to make it easy to use with a variety of mapping programs.  It also indicates whether a specific scheme is appropriate for those with color-impaired vision, is good for printed maps and various types of electronic representation.  It is a very versatile application (Harrower and Brewer 2003).

So, why is this work important?  Academic work does not exist in a vacuum, and thus must be considered in a broader context.  Jay R. Harman has put forth a hypothesis that the scientific fields that garner the most public support and interest are those that provide valuable or interesting results.  He states that research may either be justified absolutely, which means “its products materially improve the lives of others or have useful policy implications”, or relatively if it is “then useful to others working on related research questions” (2003:418).  He also states that relative justification is undesirable, because it results in eventually having to rely on the work of others for one’s work to have relevance (Harman 2003).

Brewer’s work, especially her later work with ColorBrewer, can be absolutely justified.  Improving communication with the public certainly has policy implications, with governments and businesses being able to get across important information more quickly and efficiently.  Please take a look at Figures 3, 4 and 5 to see an illustration of improved communication made possible by implementing Brewer’s ideas.  Her maps for the 2000 census are clearly superior to the 1990 map in conveying the patterns occurring in the United States.  By helping to create a society in which maps are clear and easy to read, Brewer has created work that can stand on its own.  Her work can also be relatively justified, by aiding specific important information to get to the people it needs to, but the fact that she has improved access to information for many makes her work very valuable indeed.

Cynthia Brewer has performed research that has made access its central focus: access to sensible, well-designed color schemes for cartographers and access to geographic information for map audiences.  Her work will continue to influence the creation of maps, both for print and electronic display, for a long time to come.  ColorBrewer is an innovative program that solves design problems for many people.  Brewer has been a driving force in color use for the past decade, and will undoubtedly make many other lasting contributions to the field of geography.

Works Cited

Brewer, C. A.  (1992)  “Review of colour terms and simultaneous contrast research for cartography.”  Cartographica 29, no. 3 & 4:20-30.

Brewer, C. A. (1994)  “Color use guidelines for mapping and visualization.”  In Visualization in Modern Cartography, ed. by A. M. MacEachran and D. R. F. Taylor, pp. 123-147.  Oxford, England: Pergamon.

Brewer, C. A. (1997)  “Spectral schemes: Controversial color use on maps.”  Cartography and Geographic Information Systems, 24, no. 4:203-220.

Brewer, C. A.  (2001)  “Reflections on mapping Census 2000.”  Cartography and Geographic Information Science, 28, no. 4:213-235.

Brewer, C. A.  (2003)  “A transition in improving maps: The ColorBrewer example.”  Cartography and Geographic Information Science, 30, no. 2: 159-162.

Brewer, C. A.  (2004)  Curriculum Vitae.  Unpublished curriculum vitae, retrieved from http://www.personal.psu.edu/faculty/c/a/cab38/Brewer_CV.pdf on 25 February 2005.

Brewer, C. A. and Harrower, M. A.  (2002)  ColorBrewer.  Retrieved from http://www.personal.psu.edu/faculty/c/a/cab38/ColorBrewerBeta2.html on 15 March 2005.

Brewer, C. A., Hatchard, G. W., and Harrower, M. A.  (2003)  “ColorBrewer in print: A catalog of color schemes in maps.”  Cartography and Geographic Informaion Science 30, no. 1:5-32.

Brewer, C. A., MacEachran, A. M., Pickle, L. W., and Herrmann, D.  (1997)  “Mapping mortality: evaluating color schemes for choropleth maps.”  Annals of the Association of American Geographers, 87, no. 3:411-438.

Brewer, C. A. and Pickle, L.  (2002)  “Evaluation of methods for classifying epidemiological data on choropleth maps in series.”  Annals of the Association of American Geographers, 92, no. 4:662-681.

Brewer, C. A. and Suchan, T. A. (2001)  Mapping Census 2000: The Mapping of U.S. Diversity.  Washington, D.C.: U.S. Government Printing Office.

Guzmán, B.  (2001)  The Hispanic Population: 2000 Census Briefs.  Washington, D.C.: U.S. Government Printing Office.

Harman, J. R.  (2003)  “Whither geography?”  The Professional Geographer, 55, no. 4:415-421.

Harrower, M. A. and Brewer, C. A.  (2003)  “ColorBrewer.org: An online tool for selecting colour schemes for maps.”  The Cartographic Journal, 40, no. 1:27-37.

Olson, J. M.  (2005)  Personal conversation on 15 February 2005.  East Lansing, Michigan.

Olson, J. M. and Brewer, C. A.  “An evaluation of color selections to accommodate map users with color-vision impairments.”  Annals of the Association of American Geographers, 87, no. 1:103-134.

Pickle, L. W., Mungiole, M., Jones, G. K., and White, A. A.  (1996)  Atlas of United States Mortality.  Washington, D.C.: U.S. Government Printing Office.

Slocum, T. A., McMaster R. B., Kessler, F. C., and Howard, H. H.  (2005)  Thematic Cartography and Geographic Visualization.  Upper Saddle River, New Jersey: Pearson.

U.S. Census Bureau.  (1990)  “Hispanic Origin Persons.”  In Race and Hispanic Origin Population Density of the United States.  Retrieved from http://www.census.gov/geo/www/mapGallery/images/hispanic.jpg on 15 March 2005.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s