Overview

Welcome to Lesson 8! In this lesson, we discuss another key topic in cartography: map generalization. Generalization, or transforming a map’s features (traditionally from large-scale to small-scale) to fit a map’s given scale and purpose, has been increasingly salient given the proliferation of interactive multi-scale web maps. Such maps are among a new generation of ”fast maps”, which include interactive, animated, and viral maps and mapping products. These maps, as well as other products of technological advancements in mapping, including 3D maps and extended reality applications, present new challenges and opportunities for geographers across many fields of interest.

In Lab 8, we break from our focus on mapping in ArcGIS and Tableau to design an interactive multiscale basemap using Esri's Vector Tile Style Editor online map design platform. To highlight the creative designs possible with such tools, we design this map by taking inspiration from a favorite piece of media/art/computer game/etc. Lab 8 thus ties together our discussion of generalization and interactivity with previous discussions of maps for advertising, map symbology, and basemap design.

Learning Outcomes

By the end of this lesson, you should be able to:

• describe the differences between content, geometry, and symbology generalization operators and their roles in multi-scale map design;
• evaluate a map designer’s use of selection and generalization on a map and the logical and ethical implications of these decisions;
• articulate the merits and challenges associated with designing and using animated and/or interactive maps (vs. static maps);
• discuss applications of cartography within the domain of Extended Reality (XR) – such as Virtual Reality (VR) and Augmented Reality (AR), while recognizing their respective disadvantages;
• design a custom multi-scale Mapbox basemap in the style of a movie, TV show, or other favorite piece of media and/or art.

Lesson Roadmap

Questions?

If you have questions, please feel free to post them to Lesson 8 Discussion Forum. While you are there, feel free to post your own responses if you, too, are able to help a classmate.

Map Generalization

In Lesson 7, we discussed a common theme in cartography: uncertainty visualization. In Lesson 8, we focus on how the complexities of our world are reduced into a visualization via a map. Specifically, when we reduce the complexities of Earth’s geography into a form more appropriate for a map’s given scale and purpose, this is called cartographic generalization. Thorough understanding of generalization and the related concept of scale is—and has always been—essential for creating high quality maps. The increased prevalence of web-maps, which simultaneous and seamless zooming and panning across multiple extents and scales, has encouraged increased research in the process of generalization. In Lesson 8, we discuss generalization, both in general, and in the context of multi-scale and interactive web maps.

Maps of Earth or other terrestrial bodies are set to some scale which reduces the complexities of Earth's features and their true size. As a result, all maps contain some level of generalization—maps would be unusable otherwise.  A map at a scale of 1:1 is rather pointless. Representing every element of the real world on a map is not feasible, nor would such a map be interpretable by readers. Generalization permits cartographers to construct maps with an appropriate level of detail while preserving spatial relationships, feature density, and complexity. In Lesson 4, we discussed the necessity of using the correct resolution of (raster) digital elevation data to create terrain visualizations of a large scale map. In Lesson 8, we focus primarily on the generalization of vector data, such as road networks, hydrologic features, and political boundaries.

When considering what level of detail is appropriate, it is important to consider your map's location, scale, and geographic extent. A map of seaside hotel locations in Massachusetts would, for example, show a much more detailed coastline of Cape Cod than would a map of the entire United States.

It’s also worth quickly reviewing the difference between small-scale and large-scale maps: small-scale maps represent a considerable extent of Earth's surface, while large-scale maps represent a reduced extent of Earth's surface. In terms of representative fractions, a 1:2,000,000 map scale shows a greater extent of Earth's surface than a 1:24,000 map scale. Whereas the former scale would be appropriate to map a state the latter scale would be more appropriate to map a limited portion of a city. Even professional cartographers mix these two up on occasion, so just do your best commiting this to memory.

It should also be noted that there is no hard-and-fast definition of what contitutes “small” and “large” scale, i.e. there is no quantitative agreement on the scale at which, for example, a map goes from small- to large-scale. Added to this confusion is the use of "medium" to indicate a scale. General, maps referred to as “small-scale” tend to depict large areas, such as regions, states, or continents. Large-scale maps tend to depict cities, neighborhoods, streets, and so on. But again, these are not immutable declarations. Instead, when we compare small- and large-scale maps, we are doing so in relative terms. However, small-scale maps nearly always benefit from increased generalization, and large-scale maps benefit from greater resolution or detail. Here is a brief overview [1] of map scale by the USGS. 

Natural Earth [2] is a source of boundary data that we have used extensively in this course. Figure 8.1.1 below demonstrates the differences in level of detail between different boundary datasets that Natural Earth offers. Natural Earth offers various categories of data and for each category at different scales. For example, the purple boundaries (left) show the most detail. Such data are appropriate for large-scale maps (scale = 1:10,000,000). Here, the level of detail shown in this large-scale vector linework is high. At this map scale, the intent is to present a dataset with a greater amount of detail compared to the other smaller scales. The pink (center) boundaries would be better suited for smaller-scale maps of countries or continents (scale = 1:50,000,000). The blue (right) boundaries are highly generalized, and would be best suited for very small-scale maps such as the globe or advertizing maps that use heavily stylized data (scale = 1:110,000,000).

Figure 8.1.1: Natural Earth data through scale.

Credit: Cary Anderson© Penn State is licensed under CC BY-NC-SA 4.0 [3].
Data Source: Natural Earth, Esri (basemap from ArcGIS).

Figure 8.1.2 shows each of the above boundary files at an appropriate scale given their level of detail. The extent of the largest-scale map (black rectangle of the frameline - left) is shown by the black rectangle extent indicator in the center and right maps. Note that as the scale becomes smaller (moving from the maps left to right in Figure 8.1.2, we see a diminished level of complexity with the administrative boundaries. Try to think of mapping purposes for which each level of boundary detail would be appropriate. 

Figure 8.1.2: Natural Earth data through scale: 1:10m (left), 1:50m, (center), 1:110m (right).

Credit: Cary Anderson© Penn State is licensed under CC BY-NC-SA 4.0 [3].
Data Source: Natural Earth, Esri (basemap from ArcGIS).

So far, we have talked about the overall idea of generalization – using data that is the correct level of detail for your map’s scale. A general-purpose map of a small town, for example, would likely show lakes, ponds, and reservoirs, while a small-scale map of a large region would show only the largest waterbodies (e.g., rivers, large lakes, and oceans). Often, rules decided upon by the cartographer are used to determine what elements are displayed on a map (e.g., “only show lakes that cover more than five square miles”). However, due to the uneven distribution of features across the landscape, cartographers also have to make some generalization decisions that are complex, subjective, and specific.

An example of this is demonstrated by Figure 8.1.3. Some cities are labeled, and some are not. At first, it may appear that the largest cities are labeled, and to some extent this is true. New York, NY is labeled, as well as Washington, DC. However, you may notice some cities that are absent—most notably Philadelphia, PA. A city with 1.5 million people is left off the map, while Reading, PA—a city of about 94,000—is included. Why?

Figure 8.1.3: Where’s Philadelphia? Scaling of cities in OpenStreetMap.

Credit: © OpenStreetMap contributors [6] CC BY-SA [7]

Philadelphia is located in a densely-populated region, with many nearby cities, such as Trenton, Baltimore, and Washington, D.C. By contrast, Reading, PA is surrounded only by smaller towns. Web-maps are designed to display—or not display—city labels based on a number of factors. These include population and general importance, but also design-relevant factors, such as the density of labels on the map. In the case of tools like Google Maps and Apple Maps, the features included on the map are also influenced by the user’s search history and/or other digital activity. Look at several web maps of the same location at the same scale (or zoom level) to see what similarities and differences in labels, road network, and hydrology features are apparent. 

Generalization Operators

As suggested by the previous OpenStreetMap example, generalization is also process for dealing with conflict and congestion among map symbols. Generalization can also be used as a strategy for creating a more readable and useful map by selectively including a limited set of labels, for example. Though this is a complex and context-dependent problem, some resources are available to help you determine the appropriate level of detail for your maps. The now-defunct website ScaleMaster (Brewer et al. 2007), for example, offered advice to mapmakers on which features ought to be included at different scales, and for different mapping purposes.

Figure 8.2.1: An example ScaleMaster diagram (scalemaster.org), as explained by (Roth, Brewer, and Stryker 2011).

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

We will not go into the details of ScaleMaster in this lesson, but you are encouraged to read more about ScaleMaster through the linked Cartographic Perspectives article if you are interested. The most important takeaway is that different scales require differing levels of detail, and that the appropriate level of detail is mediated by the map’s context (e.g., topographic vs. zoning maps).

Generalization can be broadly categorized as either selection or symbolization. In the context of scale, selection is simple—it refers to the decision of whether to include (or not) a feature at a certain scale, while symbolization refers to alteration of the way a feature is designed in order to make its design more appropriate for the scale at hand. For example, when designing a small-scale map, you might choose to not include cities unless they are high population (selection), and to symbolize these cites as labeled points rather than as areas (symbolization). Generalization traditionally refers to reducing detail in a map as much as is necessary to maintain legibility and usefulness at a specified scale. Generalizing multi-scale web maps (which exist at many rather than one scale) is more challenging, but not fundamentally different—we can think of every possible scale step (or zoom level) of a multi-scale web map as its own map for which an appropriate level of detail must be determined.

As generalization is a fundamental topic in cartography, many cartographers have proposed theoretical frameworks for discussing generalization. For simplicity, in this lesson, we will focus on the set of generalization operators proposed by Roth et al. (2011), as they were developed based on a comprehensive review of previous literature. As we discuss generalization operators, an important distinction should be made between generalization operators and generalization algorithms. Operator refers to a cartographer’s conceptualization of an intended change (e.g., I want to remove some roads to reduce the visual clutter of this road network), while an algorithm is a system followed for implementing this idea (e.g., I will remove all roads with speed limits below 25mph) (Roth et al. 2011). Like Roth et al., we focus on operators rather than algorithms in this lesson as they are more widely applicable to map-making tasks, and not dependent on the use of specific datasets or GIS software tools.

Roth et al. (2011) classify feature generalization operators into three groups: content, geometry, symbol. Content operators directly alter the content of the map, typically by adding or removing features at particular scales. An example would be deciding not to include local roads or trails in a small-scale map as these features would not be visible at small scales. These operators include: add, eliminate, reorder, and reclassify.

Figure 8.2.2: Content Operator: Add. Here, shaded relief is being added to a vector linework.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.3: Content Operator: Eliminate. Here selected local roads are being eliminated to simpllfy the visual appearance.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.4: Content Operator: Reorder. Here, linework and an area feature as being visually reorded to present the relationships in an appearance that is more consistent with reality (in most instances, roads do not travel on top of water features).

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.5: Content Operator: Reclassify. Here, the number of feature classes is being reduced according to the changes in scale. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Geometry operators describe the ways in which different features' geometry can be altered to create a map that is more legible and aesthetically pleasing. Examples include smoothing a line feature and representing a city as a point rather than an area. Geometry operators include: simplify, aggregate, collapse, merge, displace, exaggerate, and smooth.

Figure 8.2.6: Geometry Operator: Simplify. Here, the number of points along individual vector coastlines is being reduced.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.7: Geometry Operator: Aggregate. Here, the number of features are too dense give the map scale. Their symbology and dimension is changed from points to areas. Areas are viewed as having a higher dimension than points. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.8: Geometry Operator: Collapse. Here, a complex arrangement of symbols composing an area is changed to a new point symbol. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.9: Geometry Operator: Merge. Here, a series of small islands are merged into the larger coastline. In both cases, the islands and larger coastline is of the same dimension: area.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.10: Geometry Operator: Displace. Here, in order for both features to be visually detectable in their totality, each symbol is physically moved. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.11: Geometry Operator: Exaggerate. Here, the coastline feature collapsed upon itself during the scale change. That feature's areal extent can be exaggerated to enhance the appearance of that feature. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.12: Geometry Operator: Smooth. Here, a detailed linework segment has its complexity (or angularity) smoothed to improve its aesthetic appearance. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Symbol operators alter feature symbology to improve legibility, but do not change the features’ underlying geometry. An example would be simplifying the pattern in an area fill so it still looks good at a smaller scale. Symbol operators include adjust color, enhance, adjust iconicity, adjust pattern, rotate, adjust shape, adjust size, adjust transparency, and typify.

Figure 8.2.13: Symbol Operator: Adjust Color. Here, color contrast is increased to improve readability and emphasize the visual hierarchy. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.14: Symbol Operator: Enhance. Here, a symbolic enhancement is applied to the line feature to visually emphasize the presence of a bridge and the area fill to visually emphsize that the blue represents water.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.15: Symbol Operator: Adjust Iconicity. Here, geometric symbols replace symbols that are icons.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.16: Symbol Operator: Adjust Pattern. Here, the complex area fill pattern is replaced with a single hue fill to improve readabilty and reduce visual complexity of the map.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.17: Symbol Operator: Rotate. Here, symbols are rotated to better align to their associated feature.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.18: Symbol Operator: Adjust Shape. Here, symbol shapes are adjusted reducing the visual complexity.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.19: Symbol Operator: Adjust Size. Here, symbol sizes are reduced to improve their appearance. 

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.20: Symbol Operator: Adjust Transparency. Here, the transparency that was present in the larger scale is reduced to improve visual contrast.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

Figure 8.2.21: Symbol Operator: Typify. Here, the number of point symbols is reduced to show the relative number across a given space. In the center map, eight and two towers appear on the left and right side of the river, respectively. Their relative number has been reduced on the right map.

Credit: Adapted from (Roth et al. 2011) Cartographic Perspectives [8], (CC BY-NC-ND 3.0 [9])

It is not necessary to memorize the above operators, but you should aim to understand the difference between the three groups of operators (i.e., content, geometry, symbol) and think critically about situations in which each might be useful.

Dynamic Maps

The advent of the world wide web initiated many changes in the world of map-making. Though centuries-old cartographic principles are still relevant in a web-mapping world, digital map-making has presented new unique opportunities and challenges for cartographers.

The increasing ubiquity of the Internet has influenced cartography in many ways, from changing the nature of maps themselves (e.g., with new interactive and animated maps), to facilitating a system wherein map-making tools are widely accessible—a world in which almost anyone can make and share it widely.

Figure 8.3.1 demonstrates the evolution of the popular GIS software ArcGIS, from ArcMap/ArcView to ArcGIS Pro, designed with modern processing capabilities, searchable toolboxes, and a ribbon-based interface. Perhaps even more indicative of the times is the widespread availability of web-based mapping tools and libraries, including Felt [10], Leaflet [11], Mapbox [12], Social Explorer [13], and many more.

Figure 8.3.1: The evolution of Esri's ArcMap.

Credit: An early version of ArcMap (left) weather.gov [14], and ArcGIS Pro (right) fws.gov [15].

Geographer Mark Monmonier (2018) uses the term “fast maps” as an umbrella term to describe many new forms of maps and mapping products that have come about in the internet age. These include interactive maps, animated maps, and viral maps—maps that may be static or otherwise but are nevertheless a product of new technologies and widely spread due to the Internet and social media. New interests in virtual and augmented reality have also added to the variety of maps available in this widely-connected world.

Interactive Maps

We briefly discussed interactive maps in the previous lesson on multivariate mapping—interactivity is often used to solve problems related to multivariate mapping, such as the challenge of fitting all the necessary data into one map frame. New technologies (most notably, mobile smart phones) have both increased the challenge of designing maps and contributed their own solutions. Creating a map that can be viewed on a 4.7-inch screen, for example, can be quite a difficult design problem. Yet, accessibility to mobile cell data and location-aware devices have enabled the creation of zoomable, pan-able, user specific maps—thus reducing the amount of map content required in-view at any one time.

Figure 8.4.1: OpenStreetMap, an interactive web map.

Credit: OpenStreetMap [16] © OpenStreetMap contributors.
The data is available under the Open Database License (CC BY-SA [17]).

Web maps such as the one in Figure 8.4.1, which allow the user to zoom and pan around the extent of the map, are commonly called slippy maps. Such maps let you zoom and pan around (think of the map "slipping" around as you drag the mouse). While these slippy maps often serve as general purpose maps, they are most often used as basemaps that provide location context for a variety of thematic or functional overlays, such as the traffic volume data or navigational functionality of Google Maps.

We can categorize interactive maps by the level of user interaction (low to high) they permit. Some maps allow only simple interactions such as panning or zooming, or perhaps show additional information about features on mouse hover or click. Others may be developed for expert users, and include the ability to search, filter, and analyze data, as well as the option to upload the user’s own data for exploration and analysis. Many interactive maps, such the one in Figure 8.4.2, fall somewhere in the middle of this continuum allowing a multitude of user interactions/experiences.

Figure 8.4.2: An Online Interactive Map.

Credit: Cornell Lab of Ornithology [18].

While the term interactive map is most often used to describe maps such as the one in Figure 8.4.2, other maps are better characterized as containing passive interactivity. These are maps that respond to actions of the user, though not in the traditional sense of a user interacting with tools via a map interface (Monmonier 2018). An example of this is the navigation function in Google Maps, usable in some many car information/navigation screens, or an automotive personal navigation device (PND), such as those made by Garmin or TomTom.

Figure 8.4.3: A screenshot of Google maps being used via Apple CarPlay.

Credit: Harrison Cole © Penn State is licensed under CC BY-NC-SA 4.0 [3].

Though these devices also contain traditionally interactive components, they are primarily designed to respond to one particular user behavior—movement through space and time. Such mapping tools provide navigation, real-time traffic, and safety-zone warnings; some even provide advanced notifications such as lane departure warnings via unit-mounted cameras and other sensors.

Interactive maps have the potential to be useful in any geographic decision-making context wherein the computer or other device can provide an appropriate interface between the human and the map (Monmonier 2018). Due to the complexity of many of these products, however, the effectiveness of an interactive map is often dependent not only on the design of the map itself, but on its interface and related functions. This has made interactive map-making a particularly interdisciplinary subset of cartography, as successful approaches borrow increasingly from research in data visualization, human-computer interaction (HCI), and computer science. We will discuss the interface between maps and their users in more detail in Lesson 9.

Animated Maps

Though animated maps may also have interactive components, they are uniquely defined by their use of animation to display spatial data. A type of animated map you have likely seen and used is a weather radar map, such as the one shown in Figure 8.5.1. These ubiquitous maps typically contain little user-interaction capabilities—they are watched by the user as if watching a movie—though they may contain zooming or panning functionality, or the option to pause the animation at a point in time. Note the buttons at the bottom left of the image for controlling playback.

Figure 8.5.1: A Weather Radar animated map.

Credit: radar.weather.gov [19]

Animated maps are used to visualize a wide range of data topics, from weather to health data, demographic statistics to travel routes. Most common among these maps is the inclusion of time as the variable that is changed as the animation is performed. Though, theoretically, any quantitative variable could be depicted via animation, the use of animation to depict data through time is supported by the congruence principle which states that the external graphic representation of data should match its intrinsic characters (e.g., in the case of animation, the animation plays across time, and represents temporal data) (Tversky, Morrison, and Betrancourt 2002). Figure 8.5.2 shows a map animation where dots move across the map representing the day in the life of Chicago's extensive taxi network.

Despite the popularity of animated maps for data visualization, little research has yet been conducted that supports its use as a complete replacement for static graphics such as small multiple maps (Tversky, Morrison, and Betrancourt 2002; Griffin et al. 2006). Animated maps present unique challenges for users, who are often hindered by perceptive constraints, such as change blindness – the inability to detect changes in maps across animated frames, often combined with user overconfidence in map comprehension (Fish, Goldsberry, and Battersby 2011).

But there are some advantages: Griffin et al. (2006) conducted a map-cluster detection study with animated maps and small multiples and found that users did tend to be more successful with animated rather than static maps for this task. However, they note an important challenge in animated map design: the pace or speed of the animation is influential on user success, and that different paces are more useful for different maps. There is no ideal animation pace for maps, though cartographers ought to consider what pace might be most useful for their map’s intended audience and purpose. One way to address this issue is to add simple interaction features to animated maps, such as the ability to pause or step through time, so the user has control over the animation speed that works best for them (Tversky, Morrison, and Betrancourt 2002). Though many visual variables are used in animated mapping, pace is among the visual variables used specifically for encoding data via animation. Other animation-relevant variables include rate of change – how much the map changes between each animated frame, and order (DiBiase et al. 1992), which is the order in which individual frames are presented (often chronologically, but not always).

Alan MacEachren (1995) extended the above-mentioned visual animation variables to include display date – the starting time of a temporal sequence, frequency – the number of unique animation frames within each unit of time (e.g., animated frames per year vs. animated frames per month), and synchronization – the coincidence (or otherwise) of time series when two or more are displayed at once (e.g., snowfall and school attendance might be displayed out of sync).

Figure 8.5.3 An animated map of past and projected October temperatures in the US.

Credit: NOAA

Viral Cartography

Though the term “dynamic maps” implies movement within maps (i.e., animation and interaction), we discuss here a similar category of maps, as suggested by Monmonier (2018) in his categorization of “fast maps” – viral maps. Though there is no widely-accepted definition of a viral map, the term applies broadly to a map that is shared widely, and through non-traditional processes (i.e., through users sharing content with each other via social networks, rather than from a singular, popular provider) (Robinson 2018).

Maps that spread in this way tend to inspire emotion and be persuasive in nature (Monmonier 2018; Muehlenhaus 2014; Robinson 2018). Despite the heightened study of such emotive and persuasive maps due to their dispersion on social media, persuasive maps themselves are not new. Figure 8.6.1 shows a map from the Civil War, which illustrates General Winfield Scott’s plan to conquer the south. The snake illustrates a dark, emotional message.

Figure 8.6.1: Scott’s Great Snake, a persuasive map from the Civil War [21].

Credit: Library of Congress: Places in American Civil War History: Preparation for War

Social networking sites such as Facebook or WhatsApp have facilitated the spread of maps to a global audience with incredible speed—it is difficult to overstate the contrast between this new environment of online map distribution and cartography’s history of maps being made primarily by professional cartographers or those in other positions of power. In many ways, we find ourselves in an exciting, dynamic, more democratic era of map-making. It is important to note, however, the challenges that have arisen in this new era. The increasing ubiquity of maps and map-making has blurred the lines between mapmakers who make mistakes and those who deliberately mislead; between personal perspectives and dangerous propaganda.

Related to the increased availability of map-making tools and online map distribution channels, web technologies have facilitated increased access to wide amount of data within the public domain. Where debate tends to ensue, however, is when such data are made more visible and accessible to everyone, such as with the creation of an engaging map. Maps printed along with an article in a local newspaper titled “The Gun Owner Next Door: What You Don’t Know About the Weapons in Your Neighborhood” (linked below) provide a useful case study of such a debate. The article and accompanying maps identified gun-owners in the local area by their names and addresses. The map itself ‘went viral’ both due to people's interest in the data mapped, and the outrage that the discussions surrounding it incurred.

Maps are omnipresent in political media—consider the interactive maps used extensively on news channels while reporting election results. About a month before the 2016 US Presidential election, Nate Silver (Silver 2016) posted a map with the heading “Here’s what the election map would look like if only women voted [24]:”

In addition to reaching viral status itself, the map inspired many others to create similar maps, such as what the election map would look like if only millennials/white women/people of color voted. Robinson (2018) uses Silver’s map as an instrumental example of a viral map in his recent paper, Viral Elements of Cartography. He notes that it is characteristic of viral maps to inspire the creation of others.

Though viral and persuasive maps are often discussed in tandem (e.g., Muehlenhaus 2014), viral maps need not always be persuasive or political. The map in Figure 8.6.2 below was designed by Joshua Stevens, a cartographer at NASA who despite being well-known in the data visualization community, has only a fraction of the online following of journalist Nate Silver (Silver 2018). It was the creativity and entertainment value generated by Stevens’s map which was responsible for generating its viral status.

Figure 8.6.2: Joshua Stevens's viral eclipse map, Sunsquatch.

Credit: Sightings map by Joshua Stevens [25].

Like Silver’s map of women voters, Stevens’s Sunsquatch map inspired the design of many others, some of which went viral themselves, such as Jerry Shannon’s Smothered and Covered map (Figure 8.6.3) which illustrated where one could watch the eclipse while eating at Waffle House.

Figure 8.6.3: Professor Jerry Shannon's viral eclipse map, The Eclipse: Smothered and Covered.

Credit: Jerry Shannon [26], map presented at the 2018 NACIS conference in Norfolk, VA.
Watch his (entertaining) discussion of this map here: Viral Cartography: Or, how to make an affective map [27].

These maps by Silver, Stevens, and Shannon highlight the usefulness of Monmonier’s classification of new-era maps facilitated by web technologies as fast rather than dynamic or interactive maps (Monmonier 2018). The speed at which these maps were shared to thousands of users certainly qualifies them as fast, though they are simple, static maps. And though these static maps do not include animation or permit user interaction, they did instigate discussion and inspire further map-making, making them interactive in their own right. Certainly, interactive and/or animated maps can also ‘go viral.’ The above examples illustrate, however, the power in pairing a simple illustrative graphic with a creative idea.

3D Maps

Similar to how new web-based technologies have made it easier to design interactive and animated maps, technological advancements have altered mapmaking in another, related way: enabling more realistic depictions of the real world through more accessible 3D-mapping tools and virtual/augmented reality.

Physical, three-dimensional models have long been used to visualize geospatial information, such as cities (e.g., Figure 8.7.1), the Earth’s terrain, buildings, and so on.. These models are useful in that they provide a realistic view of the environment, but their realism and complexity often come at a cost. For example, the oblique view inherently obstructs some of the scene (e.g., locations behind tall buildings), and physical models are typically not built to scale.

Figure 8.7.1 A Model of the Golden Gateway Project from the San Francisco Redevelopment Agency (1960).

Credit: Eric Fischer [28], CC BY 2.0 [29]

In the past, creating complex 3D digital visualizations and physical models came at a near-prohibitory cost. Yet recent increases in the computational power of mainstream computers and new software tools have reduced the time, capital, and expertise required to create three-dimensional maps. Naturally, this has encouraged cartographers to make more of them. The inclusion of 3D visuals in mapping tools has become increasingly widespread; realistic modeling of buildings can now be seen, for example, in popular mapping applications such as Google Maps, even if their inclusion is questionably useful.

Increasing interest in and availability of 3D mapping tools has also resulted in an increased use of extruded or perspective height as a visual variable. Unlike the 3D map examples above, perspective height uses a third dimension to encode a variable distinct from the actual physical height of a feature.

An example is shown below (Figure 8.7.2). This is a choropleth map that uses a multi-hue sequential color scheme to encode the population density of the United Kingdom by postal code. In addition to color, however, another visual variable is used: height. Areas with higher densities are extruded from the map, giving them increased visual emphasis. The result is a map that echoes the look of varied terrain, only instead of actual physical terrain, it visualizes the terrain of population density across the landscape.

Figure 8.7.2: A map using extruded/perspective height to visually-encode population density in the UK.

Credit: plumplot [30]

Similar to the visual depiction of uncertainty, an important question surrounds the use of 3D visualization: is it useful? The answer, as suggested by recent cartographic research, is similar: it depends. Generally, studies have found that people enjoy using 3D maps more than their 2D counterparts. These studies also typically find, however, that people perform tasks less efficiently with 3D graphics than with simpler 2D visualizations (Smallman and John 2005).

There is no scientific consensus on whether 3D visualization tends to be helpful for users, and due to the context-dependence of such a question, it is unlikely that there will ever be a categorical answer. What the current state of research suggests is that 3D visualizations should be used with caution. In contexts where seconds count (e.g., emergency management; disaster response) for example, 3D visualization tools might be a risky option. In contexts where user enjoyment is of greater priority (e.g., in a university’s campus map), it might instead be an excellent choice.

Lesson 8 Lab

Multiscale Map Design in the ArcGIS Vector Tile Style Editor (VTSE)

In Lesson 6 and 7, we explored Tableau and the interactive maps that were possible. In Lesson 8 we will continue with the interactive map environment. You learned that the recent proliferation of such interactive maps has brought the challenges of map generalization back into focus. To create an effective interactive map, cartographers must consider not only how a map looks at one scale and extent (i.e., as in a typical static map) but at all locations and every scale.

As suggested above, creating an interactive web basemap can be a challenging task. Fortunately, tools exist to make this process easier and more efficient. In Lab 8, rather than using ArcGIS Pro or Tableau, we will be working in the ArcGIS Online ecosystem [32]. ArcGIS Online (AGOL) is Esri’s online mapping platform. It’s not exactly like ArcGIS Pro, but it does offer some of the same functionality—organizing and styling geospatial data, basic analysis—while providing the benefit of cloud-based data synchronization and the ability to more easily publish interactive maps online. AGOL also integrates with ArcGIS StoryMaps, which we’ll get to work with in Lesson 9, as well as data dashboards, which you may have used yourself at some point.

When one creates a thematic map, one often starts by creating or choosing the basemap. AGOL offers a handful of basemap style options, but they might not always be appropriate for the project you’re working on. For example, your client might want you to integrate their corporate style guidelines into the basemap design, or you might want to limit the amount of reference data being displayed in the basemap to reduce visual complexity. Fortunately, Esri has a tool for creating your own basemap style, called the Vector Tile Style Editor (VTSE), which we will be using for this lab.

Before getting into the details of working in AGOL, please be aware that the workflow you need to use is very different compared to ArcGIS Pro, even though they’re made by the same organization. AGOL is meant to be more accessible to a wider non-GIS specialist audience than ArcGIS Pro, at least in the sense that there are fewer buttons and tools to keep track of. You will probably expect some of a learning curve with this lab.

Lab Objectives

• Design an interactive basemap from the ground up using the ArcGIS Vector Tile Style Editor [33].
• Build a creative basemap design inspired by a favorite piece of media and/or art.
• Use your knowledge of map generalization to build a basemap that functions well at multiple scales.
• Reflect on the experience and challenges involved in designing an interactive web map.

Overall Lab Requirements

• Submit one PDF document: this should include a link to your basemap design as well as a short reflection statement.
• Example of media inspired basemaps:
  •  
    

    Figure 8.8.1: Nobuhiro Watsuki’s penultimate action-adventure story: Rurouni Kenshin.

    Credit: Map style by Patrick O’Shea.
  •  
    

    Figure 8.8.2: J. R. R. Tolkien’s Middle Earth themed map of Northwest Oregon.

    Source: Map style by Duncan Freeland.

Specific Requirements

Basemap

• Use at least 12 different layers (total) in your map. Each layer should be individually styled: do not use default settings.
• At least one layer should make use of a pattern fill.
• At least one point layer or label layer should include an icon, either the icon provided, or another one that you made yourself or is in the public domain.
• Data should transition appropriately across scales. Your map will be checked at large (local), medium (regional), and small (world) scales – you will need to set zoom-level controls for some of your layers so that they either appear/disappear or change their styling as the user zooms in and out.
• Draw inspiration from a favorite piece of media/art, such as a famous painting or a favorite TV series. Almost any media will work as an inspiration. However, do not use an existing basemap design from an organization such as Esri or National Geographic. Be creative and have fun with this lesson!

Reflection requirements (250+ words)

• Include the name of the basemap design.
• Include a screen capture of the basemap design that you created in the Vector Tile Style Editor environment.
• Explain the inspiration source (e.g., media, movie, TV show, art, etc.) behind your basemap design.
• Include at least one (1) image or illustration example that served as inspiration for your basemap design.
• Explain the key challenges you faced when working inside the VTSE environment designing your basemap, and how you overcame them.

Lab Instructions

1. You should have direct access to AGOL through Penn State.

   • Visit this site [34]

   • Under the ARCGIS ONLINE (AGOL): Open to all persons with PSU Access Accounts) heading, click on the Visit the Penn State ArcGIC Online Organization link.

   • On the page that appears, click on the Penn State WebAccess link.

   • You should automatically be directed to the ArcGIS Online environment.

   • All map design will take place within the Vector Tile Style Editor web interface which is an application found within the AGOL environment.

2. Download the Lab 8 zipped file (Lab8_Files.zip [35]) (approx. 12K MB). This zipped file contains an example json file and images that will be used to demonstrate certain design processes in this lab. Using these examples, you can create new images that would apply to your design work. This zipped file contains the following files:

   • blank_style.json

   • tree_icon.png

   • treen_icon2.png

   • tree_icon3.png

Grading Criteria

A rubric is posted for your review.

Ready to Begin?

Further instructions are available in Lesson 8 Lab Visual Guide.

Summary and Final Tasks

You've reached the end of Lesson 8! This lesson, we discussed the related topics of cartographic generalization and multiscale maps, as well as how these concepts are integral to creating effective interactive web maps. While introducing new mapping techniques, we discussed both the opportunities and challenges that new technologies in map-making provide. Using Mark Monmonier's conceptualization of fast maps, we discussed how even static maps have taken new forms in recent years due to the ability of social media to spread such maps fast, far, and wide.

In Lab 8, we used a new cartographic tool—the ArcGIS Online Vector Tile Style Editor—to create an interactive basemap inspired by a favorite piece of art. In Lesson 9, we continue along this trajectory of focus on interactivity and web-based map dissemination. We will move into Lesson 9 from creating an interactive basemap in Lesson 8 to an interactive, multimedia map story, pulling together everything that we’ve learned about design and its rhetorical impact in the world.

Reminder - Complete all of the Lesson 8 tasks!

You have reached the end of Lesson 8! Double-check the to-do list on the Lesson 8 Overview page [36] to make sure you have completed all of the activities listed there before you begin Lesson 9.