15 5 2018 Geospatial 221124 064741

International Journal of Cartography
ISSN: 2372-9333 (Print) 2372-9341 (Online) Journal homepage: http://www.tandfonline.com/loi/tica20
Geospatial big data and cartography: research

challenges and opportunities for making maps
that matter
Anthony C. Robinson, Urška Demšar, Antoni B. Moore, Aileen Buckley, Bin

Jiang, Kenneth Field, Menno-Jan Kraak, Silvana P. Camboim & Claudia R.
Sluter
To cite this article: Anthony C. Robinson, Urška Demšar, Antoni B. Moore, Aileen Buckley,
Bin Jiang, Kenneth Field, Menno-Jan Kraak, Silvana P. Camboim & Claudia R. Sluter (2017)
Geospatial big data and cartography: research challenges and opportunities for making maps that
matter, International Journal of Cartography, 3:sup1, 32-60, DOI: 10.1080/23729333.2016.1278151
To link to this article: https://doi.org/10.1080/23729333.2016.1278151
© 2017 The Author(s). Published by Informa Published online: 13 Mar 2017.

UK Limited, trading as Taylor & Francis
Group
Submit your article to this journal Article views: 1747
View related articles View Crossmark data
Citing articles: 5 View citing articles
Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tica20
INTERNATIONAL JOURNAL OF CARTOGRAPHY, 2017
VOL. 3, NO. S1, 32–60
https://doi.org/10.1080/23729333.2016.1278151
Geospatial big data and cartography: research challenges

and opportunities for making maps that matter
Anthony C. Robinson a, Urška Demšar b, Antoni B. Moorec, Aileen Buckleyd,
Bin Jiange, Kenneth Field d, Menno-Jan Kraak f, Silvana P. Camboimg and
Claudia R. Sluterg
a
GeoVISTA Center, Department of Geography, The Pennsylvania State University, University Park, PA, USA;
b
School of Geography and Sustainable Development, University of St Andrews, St Andrews, Scotland; cSchool
of Surveying, University of Otago, Dunedin, New Zealand; dEsri, Inc, Redlands, USA; eDivision of GIScience,
University of Gävle, Gävle, Sweden; fDepartment of Geoinformation Processing, Faculty of Geoinformation
Science and Earth Observation, University of Twente, Enschede, The Netherlands; gDepartment of Geomatics,
Universidade Federal do Paraná, Curitiba, Brazil
ABSTRACT ARTICLE HISTORY

Geospatial big data present a new set of challenges and Received 26 May 2016
opportunities for cartographic researchers in technical, Accepted 29 December 2016
methodological and artistic realms. New computational and
KEYWORDS
technical paradigms for cartography are accompanying the rise of Big data; cartography;
geospatial big data. Additionally, the art and science of geovisualization; visual
cartography needs to focus its contemporary efforts on work that analytics; research agenda
connects to outside disciplines and is grounded in problems that
are important to humankind and its sustainability. Following the
development of position papers and a collaborative workshop to
craft consensus around key topics, this article presents a new
cartographic research agenda focused on making maps that
matter using geospatial big data. This agenda provides both long-
term challenges that require significant attention and short-term
opportunities that we believe could be addressed in more
concentrated studies.
RÉSUMÉ
Les Big Data géospatialisées représentent un nouvel ensemble de
défis et d’opportunités pour les chercheurs en cartographie dans
les domaines technique, méthodologique et artistique. Les
nouveaux paradigmes informatiques et techniques en
cartographie suivent la progression des big data géospatiales. De
plus, les composantes artistiques et scientifiques de la
cartographie doivent actuellement concentrer leurs efforts sur les
travaux de mise en relation avec d’autres disciplines et se fondent
sur des problèmes qui sont importants pour l’humanité et sa
durabilité. A partir de papiers de position et d’un atelier
collaboratif pour élaborer un consensus sur des sujets clés, ce
papier présente un nouvel agenda de recherche en cartographie
centré sur la fabrication de cartes qui comptent pour la société,
conçues à partir de Big Data géospatiales. Cet agenda propose
CONTACT Anthony C. Robinson arobinson@psu.edu

© 2017 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License
(http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any
medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
INTERNATIONAL JOURNAL OF CARTOGRAPHY 33
non seulement des défis à long terme qui requièrent une attention
particulière mais aussi des opportunités à court terme qui, à notre
avis, pourraient être traitées par des recherches spécifiques.
Introduction
The emergence of big data presents a call to action for cartographers. The process of
making a map is, at its core, an act of generalization to make sense out of an infinitely
complex world. As data sources creep closer towards the ability to describe every detail,
all the time, for every place, the ways in which we make maps to make decisions must
adapt to handle this data windfall. New sources of information, including streaming
imagery from satellites and millions of conversations via location-enabled social media,
are examples which stretch the limits of what and how we map. These new data
sources are of limited utility if we cannot find meaning in them; therefore, an overarching
goal for cartographers is to find a way to use these data to make maps that matter to
people.
In this article, we present a new research agenda for making maps that matter from big
data. Maps that matter are those that pique interest, are tacitly understandable and are
relevant to our society. Achieving any one of these goals with a given map is non-
trivial, and achieving them together is even more difficult. And yet cartography and geo-
graphic visualization are uniquely placed to develop tailored representations and methods
of interaction that can help humans to visually discover the hidden content of big spatio-
temporal data (MacEachren & Kraak, 2001, p. 3). Many of the challenges outlined in pre-
vious cartographic research agendas remain relevant (Andrienko et al., 2007; Fairbairn,
Andrienko, Andrienko, Buziek, & Dykes, 2001; MacEachren, 1994; MacEachren & Kraak,
1997; MacEachren & Kraak, 2001; Virrantaus, Fairbairn, & Kraak, 2009), but we focus here
on new challenges posed by the emergence of big data. To develop this focus for a
research agenda and associated key challenges and opportunities, we followed a multi-
phase collaborative ideation process. Members of four International Cartographic Associ-
ation (ICA) Commissions collaborated using Google Moderator to propose, comment on
and vote to promote key research issues in cartography. Twenty-nine research needs
emerged from this activity, which were then categorized into three major areas: research
methods in cartography, designing for human ability and map use contexts, and lever-
aging geospatial big data. Each research aim was further expanded upon through the
development of 15 position papers for an ICA workshop titled Envisioning the future of car-
tographic research held in 2015. At this workshop, the challenges and opportunities
described in this agenda were iteratively refined and elaborated. A complete description
of the agenda development process is described in Griffin, Robinson, and Roth
(submitted).
Since the emergence of the concept in recent years, big data have been discussed in
many disciplines and a number of research agendas for big data have appeared, generat-
ing ample new scholarship. The definition of big data is nebulous, however. It was not so
long ago that one megabyte of data was seen as ‘big’ and which required vast compu-
tational power to explore. While there is no widely accepted common definition for
what constitutes big data, the most commonly cited definition suggests that big data
34 A. C. ROBINSON ET AL.
are characterized by large volumes, high velocity and a high degree of variety (Laney,
2001). Additional dimensions have been subsequently proposed to include veracity (Li
et al., in press; Thatcher, 2014; Tsou, 2015), high resolution, a high degree of flexibility, a
relational nature and an exhaustive scope. For the purposes of this agenda, we focus on
big data volume, velocity, variety and veracity. Recent research agendas on big data in
GIScience have focused on methods and algorithmical complexities (Jin, Wah, Cheng, &
Wang, 2015), on social media (Li et al., in press; Tsou, 2015) and on defining geospatial
big data (Lee & Kang, 2015; Li et al., in press). Researchers are also pondering what big
data really mean (Jagadish, 2015) or how to contextualize them within the traditional dis-
ciplines, such as geography (Graham & Shelton, 2013; Kitchin, 2013).
Our agenda contributes to this discussion to identify what cartography and geovisua-
lization can do to efficiently and effectively visualize complex spatial data to make
decisions and support reasoning. Cartography and geovisualization can support explora-
tory as well as confirmatory analysis of big data, serving a role to help users identify pat-
terns worthy of analysis as well as to interrogate previously known problems. We also
emphasize the critical need for interdisciplinary knowledge transfer between various
data sciences, given the complexity of geospatial big data in its current and anticipated
future forms. Each of the research aims in this article deserves collaborative attention
between cartographers and experts from allied domains such as computer science,
human–computer interaction (HCI), game design, virtual reality, information visualization,
data mining and the visual arts.
We begin by highlighting core concepts in cartography that provide a basis for carto-
graphic research with big data, many sources of which are location based (or have location
as an attribute), which makes mapping them essential. Then we present our research
agenda in two major sections. First, we present the long-term and large-scale research
challenges that face cartography and geovisualization in relation to big data and its
four primary dimensions. To go further, we look at potential approaches informed by
art, an essential component of cartography. In the second part of this agenda, we intro-
duce short-term research opportunities that we believe can be achieved in more concen-
trated investigations. We structure these around the development of new visual,
computational and artistic methods for geospatial big data. Finally, we explain how carto-
graphy and geovisualization can tackle research challenges and opportunities to make
maps that matter; maps that generate insights from complex, large, unstructured,
varied data on problems that have broad impacts to society and our environment.
Core concepts
Geospatial big data, a special type of big data, can be categorized into two classes. The first
is geolocalized big data in which location is an additional, accessory attribute. These data
are often points, such as GPS locations from smartphones or customer addresses from
business intelligence systems. The other category of geospatial big data is spatially
grounded, in which location, shape, size, orientation and spatial relationships are integral
to the data. These data come from sources such as sensor networks, collections of text
reports with spatial references, high-resolution imagery from drones and satellites, and
3D laser scans. In the context of geospatial big data, space and time are inherently
linked as many big data sources include temporal information. In the following sections,
we outline five crosscutting themes that provide key perspectives on cartographic

research with geospatial big data.
Place, space and time

A core theme concerns an understanding of space and place so that we can appreciate the
cartographic complexity of representing geospatial big data. Space might be regarded as
abstract, while place is something more tangible that people can identify with, model and
map as part of a space (Seamon & Sowers, 2008; Tuan, 1977). A place gives a space
meaning. It is those features of a place that can be captured, modelled and mapped. In this
sense, geospatial big data allow us to model a place. Geospatial big data can go beyond
mapping of individual places and features and begin to allow us to define combinations of
places. Perhaps this will lead to a mechanism to better characterize the linkages of places
that form a space. We, therefore, envision the potential that mapping big data may lead to
big cartography which conceptualizes and visualizes complex representations of place and
space. Coupled with the important but poorly researched temporal dimension (Kraak, 2014;
Rosenberg & Grafton, 2010), this has the promise of making it possible for cartographers to
tackle pressing and non-trivial issues in visualization concerned with references to place.
Representation
In the context of cartography, representations are the constructions we develop to signify
features and concepts in reality in a simplified form for interpretation via maps (Dodge,
Kitchin, & Perkins, 2011; Fairbairn et al., 2001; MacEachren, 1995). In the context of big
data, advances in representation are necessary to produce graphics, including maps,
that help people see patterns and outliers, as well as derive meaning from massive,
complex data sets (and computational results derived from them). It is also essential to
craft new means for representation in computational structures, as complex data and
their interrelations will need to be carefully modelled so that we can effectively and effi-
ciently compute on massive data collections.
Interaction
Maps as overviews to big data can provide a powerful visual gateway for analysis, but only
if their interactive affordances are carefully designed and evaluated. Interaction is the
essential mechanism by which users can navigate, search, filter and compare (among
other actions) using geographic data sources. The wide range of form factors for digital
maps today presents a new set of challenges to support user interaction, including
touch, voice and gesture-based interfaces. We have much to learn yet to determine
best practices for manipulating maps via digital interaction (Andrienko & Andrienko,
1999; Roth, 2013). Interaction also plays a key role now in determining what data are cap-
tured or emphasized in the future, as evidenced by new predictive computational
approaches by industrial giants such as Google and Amazon. It is possible that interaction
with a digital map today could influence what is collected and/or emphasized tomorrow –
highlighting a new age for map use in which map interaction itself can influence which
data we can map later.
Scale
Map scale is the relationship between distances on the map and their corresponding ground
distances (Kimerling, Buckley, Muehrcke, & Muehrcke, 2011). Geographic or spatial scale is
the operational level at which the analyses are conducted, for example, from cities to conti-
nents. In other sciences such as physics, biology and mathematics (Mandelbrot, 1982), scale is
primarily defined in a manner in which a series of scales or sizes ranging from the smallest to
the largest form a scaling hierarchy. This meaning of scale has not received its deserved
attention in the literature of geography and cartography (Jiang & Brandt, 2016). These con-
cepts of scale overlap in interesting ways with big data. The degree of resolution (map scale)
associated with big data can influence types of operational scales for analysis. Both present
challenges associated with showing detail on maps. Requirements for cartographic abstrac-
tion (generalization, simplification, classification and symbolization) are exacerbated for big
data, and while there are well-known issues associated with abstracting to smaller scale rep-
resentations (Fotheringham & Wong, 1991), new approaches are providing promising
avenues for exploration, such as those that support visual analysis across multiple scales
of social statistics (Dykes & Brunsdon, 2007; Goodwin, Dykes, & Slingsby, 2014).
Users and usability

Understanding users and prioritizing their needs are key to helping people make sense of
geospatial big data. A major goal is to support a high degree of usability of the tools and
techniques for mapping big data, as these are increasingly complicated in terms of their
form and function (Brown et al., 2013). This challenge is not a new one (Andrienko et al.,
2002; Robinson, Chen, Lengerich, Meyer, & MacEachren, 2005; Slocum et al., 2001), but it
retains importance today as users expect real-time interaction and the ability to manip-
ulate model parameters to explore outcomes. This need challenges existing compu-
tational and rendering capabilities, and requires that complex algorithms and their
parameters are readily understandable to non-experts. The emergence of crowdsourcing
as a mechanism for conducting user studies (Liu, Bias, Lease, & Kuipers, 2012) and cogni-
tive testing (Klippel et al., 2013) also suggests the potential for evaluating usability at scale
to help refine cartographic approaches to geospatial big data.
Context
Every map is situated within a specific context of use, which is frequently defined as the
information that can be used to describe the situation of an entity (Brézillon, 1999; Dey,
2001; Tomaszewski & MacEachren, 2012). Contextual factors impact the interface
between cartography and big data because the scientific challenges associated with
mapping big data are inextricable from the relevance of real-world problems. Context
may also determine whether or not we consider a specific data set to be big. Advances
in knowledge and technology must be linkable to contexts that matter for society, for
example, via connection to global goals for sustainable development (United Nations
General Assembly, 2015). The type of visualization a user might need will depend on
the context in which it is used, and little is known regarding which visualization types
are best suited to which contexts.
Figure 1. Research challenges in cartography for geospatial big data.
Research challenges
Here we outline key long-term research challenges for cartography and geospatial big
data, organized into six broad areas: making sense of geospatial big data, volume,
variety, velocity, veracity, and art and geospatial big data (Figure 1).
Making sense of geospatial big data

Big data have become a buzzword that carries different meanings in a variety of contexts
(Jagadish, 2015). Consequently, there is a need to define what we actually mean when we
consider big data in the context of cartography. Further, a known challenge lies in how to
find the hidden knowledge inherent in big data, to make sense of big data. This challenge
is not unique to big data per se, but big data’s attributes make revealing knowledge more
difficult to achieve. Progress made towards the development of better map overview
methods, pattern analysis techniques and dynamic spatial visualizations will set the stage
for research to improve methods and interfaces for synthesizing results and for analytical
reasoning about space and geography. Map users need the ability to add meaning to indi-
vidual findings in mapping interfaces, to ensure that results provenance is maintained and
to develop sharable stories to support decisions (Robinson, 2011). In this section, we
present challenges that refer to these issues in the context of cartography, geovisualiza-
tion and visual analytics.
Challenge: develop visual analytical reasoning systems that help users add
meaning to and organize what they discover from geospatial big data
To support visual analytics with geospatial big data, we have to move beyond naïve
exploration and focus attention on tools that help people reason about what they are
seeing. It is not enough to build systems that help users find patterns. Those same
users need to be able to save, annotate and compare their findings as they work on
complex problems. This action of information synthesis was proposed early on by
DiBiase (1990) and expanded upon by MacEachren (1995), but has received relatively
little attention since that time. One thing we do know is that once analysts have made dis-
coveries using analytical systems, many turn to generic office productivity software to
collect, organize and add meaning to their results (Robinson, 2009). Supporting infor-
mation synthesis will require significant advances in how we help users capture insights
and maintain provenance (Gotz & Zhou, 2009; Morisette et al., 2013; Ragan, Endert,
Sanyal, & Chen, 2016) while telling stories about data using geovisual analytics tools
(Eccles, Kapler, Harper, & Wright, 2008).
Challenge: design effective map-based interfaces to support long-term analytical

engagement with complex spatiotemporal problems and geospatial big data
Support for analytical reasoning with geospatial big data must enable long-term engage-
ment on complex societal and environmental problems. Many research efforts on evalu-
ating geovisual support for analytical reasoning focus on short-term, tactical problems
that are easier to observe and compare (Robinson, 2011). However, many problems that
have broad impacts on society and the environment will require months, years or even
decades of concerted effort to solve (Kang & Stasko, 2011). We do not yet know how
best to support long-term, collaborative analytical engagement (Jeong, Ji, Suma, Yu, &
Chang, 2015), effective recall of past spatiotemporal insights or the use of structured ana-
lytic techniques with geospatial big data.
Challenge: develop new approaches for supporting collaborative decision-making

using the map as a mutual interface for analysing geospatial big data
Decision-making is almost always a collaborative task, yet guidelines for designing effec-
tive maps and map-based interfaces to support collaborative decision-making are missing
(Andrienko et al., 2007). Supporting collaborative analysis using map-based interfaces
(Brewer, MacEachren, Abdo, Gundrum, & Otto, 2000; Rinner, 2001) has been mentioned
in most of the previous research agendas in our field. Much of the published research
on this area was conducted in the early days of web-mapping, and does not account
for mobile devices, high-performance touch/gesture/voice interfaces, cloud computing,
augmented reality and other significant advancements.
Challenge: develop techniques that allow users to express and match a

spatiotemporal pattern they have in mind using the map as an interface for
analysing geospatial big data
Exploratory spatial data analysis typically seeks to reveal new patterns in complex data sets
to prompt engagement and (hopefully) new discoveries. Pattern detection methods are
abundant today, employing a wide range of quantitative techniques for uncovering clus-
ters and outliers. The next generation of techniques and tools must also support users who
wish to seek examples for what they already have in mind. This should build on work
initiated as early as 1990 by cartographers who proposed a cognitive framework for
how humans explore data on maps which relies on the recognition of patterns (MacEach-
ren & Ganter, 1990). We need to provide users with effective visualizations that show inter-
esting spatiotemporal patterns, and we also must give them the ability to provide their
own concept of a relevant pattern and have the system find examples of similar patterns
in their data. To support the latter, we must design and evaluate approaches that support
the flexible expression of a user-desired multivariate spatiotemporal pattern. An expressed
pattern could come in the form of a verbal or written description, or via a graphical depic-
tion, as demonstrated by Shao, Sacha, Neldner, Stein, and Schreck (2016).
Challenge: understand when, how and if maps can help us understand geospatial
big data
The final challenge we highlight in this section is that as cartographers we have a natural
thirst to find a way to map something. It is rare that anyone asks for a map of something,
yet we make them nevertheless. We value the cartographic perspective and the way in
which mapping something brings us insight. It remains entirely possible that for many
situations, maps of big data are not the ideal solution. Knowing when to make a map is
just as important as knowing how to make a map. We need to determine in which situ-
ations maps help people make decisions, just as much as we need to understand what
qualities aid (or hinder) effectiveness in terms of map design.
Volume
Volume refers to data size (Gandomi & Haider, 2015), and depending on the discipline,
these sizes can vary considerably, from several million data points in a movement data
set (Stienen et al., 2016), to petabytes in imagery sources. What these definitions have
in common is that the volume of the data exceeds the handling capabilities of current
computational systems. We focus here on how the challenges associated with big data
volume prompt new directions for cartographic research.
Challenge: identify effective methods for creating overviews of geospatial big data
Maps remain one of the best ways to reduce complexity and render actionable complex
spatial data sets. However, we routinely come up against the limits of traditional map-
based overviews of big data, for example when we attempt to show millions of social
media conversations (MacEachren et al., 2011) or movement trajectories captured from
mobile phone users (Andrienko & Andrienko, 2011). We need new approaches for gener-
ating overviews and we need to know which overview methods work better than others.
We also need to know more about user requirements for interaction with displays of big
data, for example controls for manipulating overview maps and linking them to other visu-
alizations. Coordinated view systems (Roberts, 2005) may provide solutions for visualizing
big data, and a significant amount of work has focused on their development in the field of
information visualization (Shneiderman, 1996), but we know relatively little about their
limits (Andrienko & Andrienko, 2007) in generating insights about geographic
phenomena.
Challenge: develop methods that embody the volume of geospatial big data
In terms of data volume, we need cartography that can intelligently process and display
big data at a size and in a format that users can realistically handle. We need methods
that can highlight the most salient aspects of the data to reveal something useful for a
user. A number of researchers are working on geospatial visual analytics, which focuses
on this area of cartographic inquiry (Andrienko et al., 2016). To embody large volumes,
we need mapping methods that handle volume at each phase in the analysis and visual-
ization pipeline: collation and categorization of vast collections of data into constituent
parts; processing and analysis to draw out essential characteristics; and graphical
display and manipulation of the results to reveal insights. This will require solutions that
support coupled analysis and visualization, as big data often need to be analysed
before they are visualized (this order is reversed in exploratory visualization).
Variety
Traditionally, variety refers to data heterogeneity (Gandomi & Haider, 2015); that is, data lie
in different formats and representations that can be structured, semi-structured or
unstructured. Most existing visual exploratory and analytical systems are only able to
deal with geospatial big data of particular types, such as trajectory data from animal track-
ing or phone records that include location information. Geospatial big data are diverse,
often less structured, include a temporal component and will feature qualitative and quan-
titative dimensions. While the integration of geospatial big data is a problem, location can
be used as a common denominator, and the linked data concept (Heath, Bizer, & Hendler,
2011) is also promising. Additionally, we recognize the significant analytical potential that
can come from diverse data representing different perspectives on a problem.
We also propose that it is not only the data that have this property, but that the variety
dimension should also be considered in the terms of users of mapping systems, and this
needs to be addressed through user-centred design. We are still quite far from the goal of
developing walk-up-and-use geovisual systems that are accessible for first-time users
(Thomas & Cook, 2005).
Challenge: design and develop cartographic interfaces that can handle the
complexity of geospatial big data
Data are becoming cheaper to acquire and can be sensed in automated ways to provide
live feeds that populate repositories with vast quantities of information with varying
degrees of structure. Traditional database frameworks are ill-equipped to do analytical
work with these sources, but distributed computing provides a framework to support
their use with geospatial big data. To expose the variety within geospatial big data, we
need front-end visualization methods that integrate and synchronize disparate displays
to provide multiple windows into data (Agrawal et al., 2012). Synchronized displays, often
referred to as Coordinated Multiple View systems (Roberts, 2005), include a variety of
forms (e.g. maps, graphs, tables and reports), in which an operation (e.g. zoom or filter)
on one display automatically applies to all other displays. Such approaches are not
novel, but we do not yet know how best to design coordinated geovisual environments
in the context of geospatial big data. Recent work to define new types of data structures
around spatial dimensions (Bédard, Proulx, Rivest, & Badard, 2006) and trajectories (Leo-
nardi et al., 2013) could help support rapid interaction with new cartographic interfaces
to big data.
Because big data are collected in a variety of formats and the data represent a variety of
phenomena, cartographic methods should be developed that directly link visualization
methods with the format of the data and the phenomena they represent. For example,
temporal data could be displayed dynamically to leverage intuitive connection between
dynamic processes and data that describe them (Buckley, 2013). Similarly, virtual reality
(VR) could be used to explore geospatial big data in an immersive environment that simu-
lates a physical presence in real (or imagined) places (Olshannikova, Ometov, Koucheryavy,
& Olsson, 2015). We need to evaluate the potential utility of VR in the context of geospatial
big data and cartographic representation.
Challenge: develop techniques for understanding change over time in geospatial

big data
In the geographic context, change in geospatial big data can not only refer to changes in
location or in character, but also include dynamic processes over time. From a temporal
perspective, events might occur at changing time intervals; they might occur only once,
or they might return in irregular overlapping and different cycles (Andrienko & Andrienko,
2012). The advantage of sensors continuously providing data means that there is less
chance we might miss events entirely due to intervals between collections.
So how do we detect changes over time? Finding spatial patterns at one moment in
time is not a problem. Finding temporal patterns and subsequent changes over time
can be done with existing algorithms. Examples include the analysis of changing tempera-
ture over the last 30 years at one particular weather station (Wu, Zurita-Milla, & Kraak,
2015). Another example could be the changing wind pattern over multiple years at a par-
ticular location observed at different heights (Yusof, Zurita-Milla, Kraak, & Retsios, 2016).
For both examples, the complexity will increase if we consider multiple locations with
observations. Data mining algorithms alone will not suffice. Visual interfaces to guide
and support their action as well as understanding their results will be necessary.
Challenge: craft new approaches to support predictive analytics of dynamic

phenomena with maps leveraging geospatial big data
Solving problems associated with recognizing change in maps is an essential step towards
tackling the larger issue of prompting change prediction using maps. Effective situational
awareness requires knowledge of the key spatiotemporal elements, understanding their
meaning and making predictions about their status in the future (Endsley, 1995). Most
dynamic mapping and visualization systems are only designed to support the first two pri-
orities of identifying key elements and understanding their meaning. Predictive analytics
can use past patterns and modelled outcomes to alert users to important changes in data
streams and to suggest potential future outcomes (Maciejewski et al., 2011).
Challenge: develop spatiotemporal visualization methods for geospatial big data

that support a variety of uses and users
Once big data have been analysed, they need to be presented in meaningful visualizations
that assist interpretation and build understanding. In doing so, one has to consider the
potential tasks that users might have to address in respect to the spatiotemporal data
(Roth, 2012). Geospatial big data visualization systems should include a rich palette of
display options that convey the significance of analytical results. These displays must be
presented in a way that is best understood by users within their particular domains, or,
in some cases, by the general public. Cartographers can draw upon Bertin’s concept of
the Map-to-See, which is ‘a clear graphic representation which can be comprehended in
a short moment’, and as such is ‘the most efficient means to communicate a message’
(Kraak, 1989). At the same time, the messy nature of big data may be more effectively
expressed in a ‘messy map’, which is an example of a more complex Map-to-Read
(Kraak, 1989).
Geospatial big data visualizations are used in a wide variety of applications and
domains: science and research; education; intelligent transportation; environmental pres-
ervation; business intelligence; personalized healthcare; urban planning; homeland secur-
ity and more. Our challenge is to provide cartographic visualization solutions that can be
applied across these domains for users at varying levels of expertise. Alternative
approaches such as spatialization may be intuitive for some users (Fabrikant, 2000). For
non-expert audiences, intelligent automated mapping may provide solutions through
dashboards (Few, 2013) or via interfaces that incorporate storyboarding techniques
from the cinematic arts to convey a broader understanding of big data (Roth, Hart,
Mead, & Quinn, in press), and could also support data-driven journalism (Lewis, 2015).
Velocity
Velocity is the speed at which geospatial big data are generated and at which they should
be analysed (Gandomi & Haider, 2015). Recent technological developments in data cre-
ation have led to fast, continuous and ubiquitous data streams that exceed capabilities
of contemporary computing systems to map and analyse in real time.
Challenge: develop methods that embody the velocity of geospatial big data
In many situations, the results of geospatial big data analysis and visualization are required
immediately – to detect the locations of fraudulent credit card transactions, for example. In
these cases, the ideal solution is to flag the activity before the event has ended. A full
analysis of all relevant data in this sort of situation is not feasible in real time – nor is it
necessary. Instead, we need to produce predictive partial results, presented in personal-
ized displays, so that incremental computation with new data can be used to make
quick decisions.
Challenge: create maps and map-oriented interfaces that prompt attention to

important changes in dynamic geospatial big data sources
The dynamic behaviour of people and their environments is now captured and streamed in
real time (Andrienko et al., 2010). Although there are a range of representation techniques
for showing change, maps suffer well-known issues associated with change blindness (Fish,
Goldsberry, & Battersby, 2011). Yet it is more important than ever for users to recognize
change when it is occurring in their data sets, and to be able to anticipate future changes
through the use of predictive techniques. Designing visualizations that are able to handle
data generated at high velocity is not necessarily a problem if the changes to trends and
patterns in those data are limited. However, if the data changes frequently and/or with
high magnitudes, traditional graphics will not suffice. Further development in cartographic
animation techniques could be one direction to take; however, its success will depend on
the nature of changes and the scale at which they occur. We will need to work with
global changes, local changes and combinations across scales. In addition, if we display
every possible change at once, then the graphical displays become cluttered. Creating sum-
maries of change may be the solution, but we do not yet know how to select important pat-
terns and generalize to something that a user can understand.
A dashboard with streaming visual summaries of geospatial big data’s vital statistics
could be part of the solution (Few, 2013). It could offer a selective (analytical) overview,
including the data’s spatial extent, the range and variability of the attributes, and time
interval, as well as information about its geographic context. Although dashboards are
supposed to be simple in design, at times they may become complex due to the nature
of the data at hand (Andrienko, Andrienko, Fuchs, Rinzivillo, & Betz, 2015). Embedded
within this challenge is the need for design guidelines for summary maps that can
readily characterize geospatial big data.
Veracity
In data analytics, veracity refers to the inherent unreliability of data (Gandomi & Haider,
2015) in terms of precision and other aspects of uncertainty. For example, social media
data, which contain human judgements, are subjective in their basic nature and may
refer only vaguely to places and times. This however is not novel to cartography, which
has a long tradition of dealing with uncertainty in spatial data (Zhang & Goodchild, 2002).
Challenge: characterize the quality and certainty of geospatial big data

Maps can claim to be authoritative, and the more fuzzy, dubious or nefarious the data, the
more the map’s message is brought into question. With increasingly large spatial data sets
that are distilled into a simplistic metric for reporting, the potential for oversimplification is
clear. What may be more problematic is quality in terms of the bias that the data may
contain. For instance, many have mapped dimensions of social media feeds (Field &
O’Brien, 2010; MacEachren et al., 2011), but the resulting maps, which at first glance
may appear to reveal insights, are rarely representative of what they purport to show.
Their ability to be representative of a societal view is fundamentally flawed, given what
we know about which people use (and do not use) social media. This requires map
readers to be cautioned and aware of the limits of what they are seeing.
The same could be said of data sets captured and published by non-traditional sources.
We are no longer limited to making maps with official data gathered by National Mapping
and Census agencies (Goodchild, 2007). Anyone can capture, store and publish data from
their mobile devices. The many situations that these sensor networks operate within can
also create problems. Error and uncertainty undoubtedly exist, sometimes in unknown
ways and quantities.
Challenge: develop new approaches for visualizing the quality and certainty of
geospatial big data
In Kinkeldey, MacEachren, and Schiewe (2014), the authors reviewed decades of efforts to
find a way to communicate data quality information on maps, and have concluded that
the task to retrieve quality values can still be achieved using traditional cartographic
methods. However, analytical and exploratory tasks, such as those involved in analysing
big data, need dynamic approaches for application in real time. MacEachren (2015) pre-
sents a new and complementary perspective, claiming that the role of uncertainty in
decision-making, reasoning and outcomes is often overlooked in the research of visual
methods to represent uncertainty in maps.
Use and interpretation caveats could become commonplace on maps based on geos-
patial big data to highlight aspects of uncertainty. Such a future approach might lead not
to a single map but a set of maps, or an interactive component that reveals the extent of
uncertainty. While mapping uncertainty is not new to academic cartography (MacEachren
et al., 2005), representing it on maps in the public realm would be novel to many map
readers who are used to seeing one map and considering one message. Engendering
an ethos of caution through appreciating error and uncertainty could become a major
goal for effective mapping of geospatial big data.
Art and geospatial big data

The potential of art (including aesthetics and beauty) in the context of exploring spatial
data has not been fully realized, despite calls for investigation into ‘non-conventional
graphics’ in earlier agendas (Fairbairn et al., 2001). Geospatial big data highlight the
increased need for this investigation to take place. Art has always been present in the
history of maps, assuming a major but decorative aspect early on (Casey, 2005), taking
a background role in the nineteenth and twentieth centuries before recent recognition
within cartography that art is a powerful representation of the world that is different in
kind from maps (Caquard & Fraser Taylor, 2005; Cartwright, Gartner, & Lehn, 2009). Further-
more, art is a key component of the ICA’s definition of cartography (Rystedt et al., 2003),
described as the ‘ … discipline dealing with the art, science and technology of making and
using maps’.
There are many reasons to explore artistic approaches towards understanding big data.
Artistic methods have the flexibility to represent multiple scales, dimensions, uncertainty
and variety (Laney, 2001; MacEachren & Kraak, 2001) in a single representation. Cartwright
(2004) uses the analogy of the soft pencil, for vague, impressionistic and artistic represen-
tations that foster exploration and make serendipitous discoveries more likely. Art is also a
natural representation for the narratives and stories implicit in big data, being able to
convey underlying meaning, as well as representing process and phenomenon, potentially
with emotional heft. The linkage between cartographic design principles and emotions
and ideas has been considered, as well as the role of the map relative to mood (Buckley
& Jenny, 2012). Here we focus on the potential of creative expression, artistic rendering
methods and their generation, and link artistic methods with conventional spatial and
aspatial displays for big data analysis.
Challenge: encourage and maximize creative contributions for expressing

geospatial big data
There is an acknowledged overlap of fine art in particular with cartography, sharing
aspects of form, composition, framing and perspective, as well as content selection,
emphasis, line, colour, medium and symbolization aspects (Ehrensvärd, 1987). There has
also been an increased number of artists using map content or mapping processes in
their work, possibly because artists recognize that maps are very flexible (Watson,
2009). An example is the ‘flowing city maps’ by Chaotic Atmospheres (2015) which
conveys their dynamic and chaotic essence by digitally extending flow lines from map fea-
tures to artistic effect. In art, complex information can be represented in a readily con-
sumed way, which makes it ideal for communicating the broad sense of geospatial big
data. This shared basis must be built upon further. But how do we engage artists to visually
address this new order of spatial data, quite unlike the geospatial subject matter that they
have tackled up to now? And what can we learn from art’s attempts at geospatial
expression and cartography so far? Examples include artworks with a realistic content
(Priestnall & Hampson, 2008) to abstracted works, even those that are maps themselves
(Patterson, 1992).
Challenge: build a library of artistic methods and techniques for representing

geospatial big data
Despite the observed commonalities in the respective technical methods and approaches
of fine art and cartography (Ehrensvärd, 1987), there is large scope to exploit geometric
and technical aspects that have not yet made it into map making, from common processes
(e.g. generalization) to what would be completely new aspects for cartography (e.g. ‘messy
maps’). To initiate this, we can learn from examples in recent visualization work which use
types of artistic expression as a starting point for the development of new techniques
(Etemad, Carpendale, & Samavati, 2014; McCurdy, Lein, Coles, & Meyer, 2016). Likewise,
there are aspects of cinematic and aural art processes that also have great potential in
the visualization of geospatial big data. We need to know the full scope of potential artistic
methods and techniques, drawing from fine, illustrative, cinematic and music/audio arts,
that can be co-opted to represent geospatial big data.
Challenge: generate artistic renderings of geospatial big data

The first challenge points to solutions that would have a heavy reliance on the output of
artists, which is hardly a real-time solution, so it is not a good approach for the velocity
aspect of big data (Laney, 2001). The second challenge at least would facilitate identification
of artistic techniques for depicting big data, effectively packaging them for easy reuse com-
putationally, which could enable real-time visualization. However, before this can happen, a
solution needs to be found for the digital generation of artworks that are both meaningful
wholes and are somehow a true representation of a big data set. This is aesthetics as ‘the
science of the beautiful in nature and art’ (Merriam-Webster, 2014). Recent neurological
research indicates that there is a systematic basis for our cognitive processing of art and
its attributes, despite its apparent subjectivity and variability (Zeki, 2001), paving the way
towards generation of artworks. For example, the aesthetic properties of beauty in art are
processed by the same area of the brain, regardless of genre (e.g. landscape or portrait)
(Kawabata & Zeki, 2004) or medium (e.g. painting or music) (Ishizu & Zeki, 2011). Given
this, how can we semi-automatically generate artistic renderings of big data?
Challenge: dynamically link artworks to conventional geospatial big data

representations
Artworks that have been created or generated from big data may be effective in isolation,
but are likely to realize a greater potential if put into a context that enables dynamic
exploration. Such a connection could be effective for dealing with the dynamism of big
data, adopting the standard overview, zoom and filter, details-on-demand visualization
approach (Shneiderman, 1996) and coupling this with new artistic methods for carto-
graphic design (Christophe & Hoarau, 2012). New artistic interfaces should be able to be
manipulated visually in real time, connecting standard tools (choropleth maps, scatter-
plots, etc.) and non-conventional tools (abstracted and virtual representations) via
Figure 2. Research opportunities in cartography for geospatial big data.

linking and brushing. There is a clear challenge in engineering the linkages needed to
build such an interface: how do we determine the meaningful linkages from created art-
works and artistically rendered maps to ‘conventional’ spatial and non-spatial represen-
tations in a visual analytics context?
Research opportunities
In the sections that follow, we highlight key research opportunities in the broad categories
of visual, computational and artistic methods (Figure 2). We distinguish opportunities from
challenges based on our estimation that they can be solved in the near term, rather than
long term.
Visual methods for geospatial big data

The common refrain of ‘ … a picture is worth a thousand words’ is only worthy when the
picture in question is understandable. Bertin (1983) states that ‘the most efficient construc-
tions are those in which any question, whatever its type or level, can be answered in a
single instant of perception, in a single image’. MacEachren (1995) proposed that ‘maps
that connote truth (or even reality) are likely to work better than those that do not’.
While we propose a research agenda for creating and developing new visual methods
for big data, we must not ignore lessons learned from these and other predecessors.
Opportunity: systematically evaluate the ability of existing visual methods in

thematic cartography to support the analysis of geospatial big data
An overarching goal for cartographers is to develop techniques for processing and manip-
ulating geospatial big data that make it possible to generalize and symbolize geographic
results through graphics and visualizations that match or exceed quality we already
achieve with smaller data sources and analytical methods. Rather than focusing solely
on the development of new visual methods, we propose to first evaluate current thematic
mapping solutions to characterize their strengths and limitations when it comes to their
application with geospatial big data. To accomplish this goal, we need to connect big
data characteristics and the visual affordances of existing cartographic approaches. In
addition to cartographic approaches, it is important to look at how allied fields have
approached similar problems. Solutions for visualizing large data sets have been proposed
in statistics, computer science, psychology and related domains (Wills & Wilkinson, 2010).
Opportunity: adapt cartographic generalization principles and techniques to

support visual analysis of geospatial big data
Key considerations for visual designs include the number of the spatial phenomenon
occurrences and the variation of their spatial density and accuracy; the variation of
related data set semantics and the spatiotemporal dynamics of data streams. The core
characteristics of geospatial big data can influence the level of generalization, both geo-
metric and semantic, that may be possible. In terms of Bertin’s (1983) framework for
graphics, the components that must be considered for map generalization are invariant
information and components; the organization of information components (qualitative,
ordered and quantitative) and the level of retinal variations (association, selection, order
and quantity). Additionally, decisions about generalizing geospatial big data must take
into consideration the variation of data representations from discrete elements to continu-
ous phenomena (MacEachren, 1992). This concern is also present for situations in which
temporal resolution must be considered.
Opportunity: couple computational methods and cartographic representation best

practices into an automated framework that suggests appropriate design decisions
on the fly for geospatial big data
When the information component is quantitative, there are two well-known situations that
have been studied in cartography. One of them is when the data can be represented as
choropleth maps or proportional symbol maps. The other is related to continuous
phenomena that can be represented as a smoothed surface. For choropleth or pro-
portional symbol representations, we propose the development of computational sol-
utions to automatically evaluate the characteristics of the phenomenon, in terms of
both spatial and semantic dimensions, in order to render decisions on the level of gener-
alization, and the appropriateness of representing the data with these kinds of maps. For
continuous surface representations, the level of generalization also depends on the spatial
density of samples, including the variation in quality within the study area. A compu-
tational solution could assess the variation in the spatial density of the data sample and
calculate various levels of accuracy. Such a computational solution must also include
the means to dynamically generate a surface as data streams change.
When the information component is qualitative, the phenomenon is discrete and
known at point locations, and the study region is large enough to be depicted on a
scale that requires semantic generalization, the decisions about attribute classification
can be based on the geographic relations between data attributes and key geographic fea-
tures, and the maximum level of generalization possible. The qualitative level always
involves two perceptual approaches: associative and dissociative. At a dissociative level,
it is not possible to combine information components because they are different from
each other. On the other hand, at the associative level, components can be grouped.
There are some solutions in statistics that have attempted to automate design decisions
for data graphics (Wills & Wilkinson, 2010). One potential way forward is to evaluate the
applicability of Wilkinson’s Grammar of graphics (2005) for displaying geospatial big data.
Opportunity: leverage what we know about map animation and interactive

cartography to construct visual solutions for dynamic sources of geospatial big
data
Conventional solutions for interactive mapping, animated mapping or geovisual analytics
can be used for representing big data. However, because of the high velocity character-
istic of big data, it is necessary to develop solutions that can automate map design
decisions to support interactive design solutions that respond (or potentially precede
based on modelled outcomes) as the data changes. New mapping solutions could
allow users to understand different aspects of phenomena by reviewing multiple alterna-
tive scenes from the same data set. Animated mapping solutions could be developed
specifically to target geospatial big data velocity, volume, variety and veracity aspects.
These approaches could be incorporated into dashboards and virtual environments to
support new avenues for user engagement. Efficient interaction in such systems will,
of course, depend on concomitant advances in computational methods to drive such

tools.
Computational methods for geospatial big data

Maps appear to be a perfect means for showing things that are too big and/or too complex
to readily perceive in raw form. As a result, maps and map metaphors have received sig-
nificant attention in information visualization and visual analytics (Aggarwal, 2011;
Andrienko & Andrienko, 2011; Chen, 2013; Dykes, Wood, & Slingsby, 2010; Guo, Chen,
MacEachren, & Liao, 2006). However, in most analytical settings, maps must be combined
with statistical methods or computational methods to explore or uncover underlying pat-
terns or structure. Computational methods constitute the third scientific paradigm (Hey,
Tansley, & Tolle, 2009), and in the field of geography, computational geography and geoin-
formatics have been well developed since the 1960s (Openshaw, 1998). Today, compu-
tational geography is data-driven (Miller & Goodchild, 2014) or data-intensive (Jiang,
2013b). In addition to volume concerns, there are other big data aspects to be addressed,
including veracity and variety. The former is closely related to uncertainty modelling
(Zhang & Goodchild, 2002), while the latter illustrates an underlying scaling pattern of
far more small things than large ones. Big data are likely to show diversity and heterogen-
eity – highlighting the problem of spatial heterogeneity (Jiang, 2015a). Here we describe
several key research opportunities focused on advances in computation to support carto-
graphy with geospatial big data.
Opportunity: leverage knowledge about patterns across scales in the development

of new computational methods for geospatial big data
Computational methods are particularly important for uncovering underlying patterns.
One such universal pattern is the scaling or fractal pattern (Mandelbrot, 1982), in
which a pattern repeats at every scale. In this regard, head/tail breaks (Jiang, 2013a)
provides an alternative approach to conventional clustering techniques such as k-
means for showing underlying scaling patterns, and could be employed for cartographic
visualization through recursively filtering out data in the tail (Jiang, 2015b). Conventional
clustering techniques, for example k-means or natural breaks, are based on the premise
that variance within classes should be minimized, and variance between classes should
be maximized. However, these assumptions do not necessarily make sense for big data,
which often demonstrate a long tail distribution because of their diversity and hetero-
geneity. New efforts should be made to integrate computational and visual methods
that can help develop new insights into geospatial big data (Endert, Chang, North, &
Zhou, 2015).
Opportunity: use what we know about human dynamics to find patterns in

geospatial big data
It has been claimed that 95% of big data are unstructured (Gandomi & Haider, 2015). As
more and more big data are becoming georeferenced and time stamped, geographic fea-
tures such as cities and temporal rhythms become an implicit means for structuring the
data. Because big data can be structured geographically and/or temporally, maps will
play an important role in visualizing the underlying patterns and towards developing
new insights. For example, the notion of natural cities has been applied to location-based
social media data to uncover how users aggregate spatially and temporally (Jiang, 2015b).
Opportunity: connect concepts from complexity science to new visual analytics

methods for geospatial big data
Conventional science or Newtonian science is simple science focused on the correlation
between two parameters, or causality, and represents a long-standing paradigm under
which geography and cartography have been significantly influenced. Complexity
science focuses on individual interactions from the bottom up, and is interested in emer-
gence rather than simple correlation or causality (Balcan et al., 2010). Complexity science
has developed a range of tools including discrete models such as cellular automata and
agent-based modelling; complex networks such as small-world and scale-free networks;
scaling hierarchies, for example, Zipf’s law; fractal geometry; self-organized criticality
and chaos theory. All of these approaches attempt to reveal underlying mechanisms,
linking complex surface forms to underlying mechanisms (or deep simplicity) through
agent-based simulations from the bottom up. To date, relatively little has been accom-
plished to design cartographic solutions for representing phenomena through complexity
science methods.
Adapting artistic methods for geospatial big data

Opportunity: facilitate engagement of artists with geospatial big data and the
creation of an artistic geospatial language
There is a need for new work which introduces artists to geospatial big data as well as
creating artworks that could be used as part of a geovisual analysis of a big data set.
One such investigation would be to simply ask artists to draw overviews for data sets
that are quite well known already. These artwork overviews could then be examined for
commonalities in how data characteristics are expressed; these common elements
would become the start of a language which we could use in future investigations of
big data. This language would be added to by mining geospatial concepts that are seman-
tically depicted in existing ‘map’ artworks. A similar process could be applied to sonic
works, realizing the potential of sound (Cartwright et al., 2001; Krygier, 1994). Art and
science collaborations can be engineered through joint art shows, such as those featured
at SIGGRAPH and IEEE VIS conferences. Digital creativity support tools (Shneiderman,
2007) or similar, for example, COSTART (Candy & Edmonds, 2000), could provide a platform
for collaboration between artists and scientists.
Opportunity: co-opt artistic methods and techniques to represent geospatial big

data
We can look to fine art, cinema and music and the techniques they employ to represent
spatiotemporal data, compiling these techniques into a library through which geospatial
big data can be expressed. A further step would be to organize this resource into a
‘grammar of artistic methods’, inspired by Wilkinson’s (2005) grammar of graphics. Promi-
nent in such a library would be the methods of caricaturing in illustrative art, being ana-
logous to cartographic generalization (Jones, 1997), an approach used by Döllner (2007) in
non-photorealistic rendering of 3D objects. Simplicity in drawing makes for effective
communication to a greater audience (McCloud, 1994), a powerful aspect for depicting big
data while addressing its volume aspect. Alternatively, certain styles of modern art such as
Jackson Pollock’s drip maps can be used to create ‘messy maps’ for revealing spatial and
attribute veracity (Field, 2015).
Another prime example is storyboarding, an essential visual planning tool in cinema
(Caquard, 2013) and one that has seen previous use in various geospatial and HCI contexts
(Cartwright, 1994; Dix, Finlay, Abowd, & Beale, 2004; Riedl, 2012). The goal here is to depict
the spatiotemporal narratives that may be implicit in big data. Narratives can also be spatially
vague (Caquard, 2013), speaking to the aspect of the veracity in big data. This is a continu-
ation of the observed convergence of maps and narratives, manifesting itself as ‘story maps’,
and enabled by technologies such as geoparsing (Caquard, 2013). The technique is also
similar to comic strips (Moore, 2009), given knowledge of the narrative. Synthesized discov-
eries from the geovisual analytics process will also lend themselves to storyboarding.
Opportunity: generate artistic renderings of geospatial big data

In fractal geometry, we have a way of generating a complex artefact of an aesthetic, or
even artistic value from simple global inputs. Jiang and Sui (2014) build upon this and
Alexander’s living geometry (Alexander, 2002), stressing that geographic features, if
taken a right perspective and scope, all demonstrate structural beauty, which is defined
mathematically, exists in space and matter physically, and reflects in our minds and cogni-
tion psychologically. The structural beauty with geospatial big data can be effectively
revealed by head/tail breaks (Jiang, 2013a). How we translate attributes of geospatial
big data to fractal parameters is a particular research challenge. There have been projects
to generate artworks from simple initial rules (Sundararajan, 2014) and artificial intelli-
gence (Koch, 2015). Another potential method of artistic rendering is to adapt highly
abstracted spatiotemporal representations (Bertin, 1981; Guo et al., 2006) with aesthetic
qualities that have attractive geometries and/or colour schemes, such as those found in
Adaptive Relative Motion (Moore & Rodda, 2015).
Opportunity: link artworks to conventional representations in a visual analytics

context to leverage geospatial big data
To identify meaningful linkages from artworks to typical spatial and aspatial represen-
tations as a path towards artistically enabled geovisual analytics, we have a starting
point with the map: anti-map (i.e. art) category of art and cartography manifestations
(Caquard & Fraser Taylor, 2005). This linkage between artwork and map (Moore, Marinescu,
& Tenzer, 2011) can be built upon by establishing links from art to other graphical rep-
resentations (e.g. scatterplots).
Finally, agents should be a part of future research too. For example, agents could guide
users through the bewildering amount of possible representations, setting their par-
ameters and managing possible combinations with other representations (Fairbairn
et al., 2001). AutoVis, an automatic visualization system (Wills & Wilkinson, 2010), builds
on Wilkinson’s grammar of graphics and has similar aims to what we propose here.
Agents could also have a role in using geocomputation to uncover structure in complex
representations (Gahegan, Wachowicz, Harrower, & Rhyne, 2001; MacGill & Openshaw,
1998). Cartwright (2004) proposes ‘engineered serendipity’, a combination of free explora-
tion and engineered guidance to information that may have been missed. If the artwork,
whether created or generated, and flexible links with other representations (Thudt, Hin-
richs, and Carpendale (2012) present an interface that encourages serendipity in discover-
ing books) can foster serendipity, then agents can help direct attention to overlooked
visual elements, representations and linkages between them.
Making maps that matter from geospatial big data

In this article, we have presented a series of key research challenges and opportunities
associated with mapping geospatial big data with the intention of spurring the next
wave of forward-thinking cartographic research. Cartographers in concert with experts
in allied domains such as computer science, visual arts, HCIs and data mining have the
ability to contribute new solutions to pressing problems that require novel approaches.
In particular, cartographers have the ability to provide effective visual products that lever-
age what we know about how people see and make sense out of geographic information.
In addition to the research imperatives we have outlined in this article, we call attention in
this concluding section to the critical need for cartographic research on big data to focus
on relevance to society, and we specifically suggest the development of close connections
to the global goals set forth by the United Nations through their 2030 Agenda for Sustain-
able Development (United Nations General Assembly, 2015). We also draw inspiration
from earlier work to theorize visual analytics and connect its aims to key application
areas (Keim, Mansmann, Schneidewind, Thomas, & Ziegler, 2008).
For example, the United Nations 2030 Agenda includes Goal 6, which aims to ‘Ensure avail-
ability and sustainable management of water and sanitation for all’. A way to support this
goal via future cartographic research would be to ground a project on geospatial big data
veracity, for example, in the context of water modelling results under a range of potential
climate change scenarios. Taken one step further, one could include an effort to explore artis-
tic representations and their affordances for communicating such model results via maps.
A concerted effort must take place to define potential linkages across research goals in
cartography and tackle grand societal challenges. As the recognized international scientific
body of cartography, we recommend that the ICA should coordinate and host workshops
and web-based collaborative efforts to pursue these goals. We also suggest that the ICA
should compile a repository of exemplary maps and visualizations that can be easily repro-
duced with other data, have demonstrated success in use and interpretation, and are well
understood in terms of their effectiveness for explaining key dimensions of big data. In
order to make progress in collecting, sharing and promoting such things, we have to
define what we mean by best practices for the cartography of big data as a community.
We believe that solutions to the challenges and opportunities we present here will
result in cartographic contributions to big data that will deliver maps that matter to
society and the environment. To enact those solutions, cartographers will need to
explore each key dimension of geospatial big data and develop clever solutions for over-
coming their complexity and dynamism.
Acknowledgements
The authors wish to thank Alan MacEachren and Gennady Andrienko for their feedback on an early
version of this manuscript.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes on contributors
Dr Anthony C. Robinson is Assistant Professor, Director for Online Geospatial Education programmes
and Assistant Director for the GeoVISTA research centre in the Department of Geography at Penn
State University. Dr Robinson’s research focuses on the science of interface and interaction design
for geographic visualization software tools. He currently serves as the Chair of the Commission on
Visual Analytics for the International Cartographic Association.
Dr Urška Demšar is Lecturer (Assistant Professor) in Geoinformatics in the School of Geography &
Sustainable Development, University of St Andrews, Scotland, UK. She has a Ph.D. in Geoinfor-
matics from the Royal Institute of Technology (KTH), Stockholm, Sweden, and a background in
Applied Mathematics from the University of Ljubljana, Slovenia. Her research interests are in
spatio-temporal visual analytics and in particular in visualization and analysis of movement – a
topic on which she is collaborating with movement researchers from other disciplines (movement
ecologists and human–computer interaction specialists). For more information, refer http://
udemsar.com, @udemsar.
Antoni B. Moore is an associate professor in Geographical Information Science at the National School
of Surveying, University of Otago, NZ. He has a BSc(Hons) in Geographical Science (Portsmouth), an MSc
in GIS (Leicester) and a PhD (on the application of spatial artificial intelligence to integrated coastal zone
management, Plymouth). He currently researches in the areas of geovisualization, cartography and
spatial analysis, with specific emphasis on art and maps, also representation of spatiotemporal data.
Dr Aileen Buckley has been a research cartographer with Esri since 2003 and a professional carto-
grapher for almost 30 years. Dr Buckley has published and lectured widely on topics relating to car-
tography and GIS. In addition, she is the author of the Atlas of Oregon (2001) and the last three
editions of Map use: Reading, analysis, interpretation (2009, 2012 and 2016).
Dr Bin Jiang is Professor in Computational Geography at the University of Gävle, Sweden. His
research interests centre on geospatial analysis of urban structure and dynamics, for example, topo-
logical analysis, scaling hierarchy and agent-based modelling applied to buildings, streets and cities,
or geospatial big data in general. He developed Axwoman for topological analysis, and head/tail
breaks for scaling analysis. Inspired by Christopher Alexander’s work, he developed a mathematical
model of beauty – beautimeter, which helps address not only why a design is beautiful, but also how
much beauty the design has.
Kenneth Field has over 25 years of experience in cartography as an academic and now at Esri Inc. He
holds a BSc in Cartography and a PhD in GIS and is widely published, makes maps and regularly pre-
sents on map design. A former Editor of The Cartographic Journal, he is current Chair of the ICA Com-
mission on Map Design. Often opinionated; always passionate researcher, teacher, writer, presenter,
blogger and tweeter.
Menno-Jan Kraak is a professor in Geovisual Analytics and Cartography at the University of Twente,
ITC. Currently he is head of ITC’s Geo-Information Processing Department. He has written more than
200 publications on cartography and GIS. He is a member of the editorial board of several inter-
national journals in the field of Cartography and GIScience, and currently serves as the President
of the International Cartographic Association.
Dr Silvana P. Camboim is a faculty member at the Department of Geomatics at the Federal Univer-
sity of Paraná – UFPR, Brazil. She serves as the Chair of the Commission of Open Source Geospatial
Technologies for the International Cartographic Association. She is the co-chair of ‘Geo for All’ in
South America. Dr Camboim has a focused on research in Geospatial Science, Spatial Data Infrastruc-
ture (SDI) and Geographic Information (GI) standards. Before joining UFPR, Dr Camboim held a pos-
ition at the Brazilian National Mapping Agency.
Claudia R. Sluter is Full Professor at Federal University of Paraná (UFPR), Brazil, where she teaches
and researches on the following subjects: geovisualization, thematic mapping, cartographic gener-
alization, topographic mapping, interactive map designs and GIS. She holds a bachelor degree in
Cartographic Engineering (1986) and a master’s in Geodetic Science (1993) from UFPR, and
earned a doctorate in Computer Science (2000) from National Institute for Space Research, Brazil.
During her doctorate, she has studied for one year at Geography Department of the University of
Kansas (1998) as a sandwich doctorate.
ORCID
Anthony C. Robinson http://orcid.org/0000-0002-5249-8010
Urška Demšar http://orcid.org/0000-0001-7791-2807
Kenneth Field http://orcid.org/0000-0003-0423-7892
Menno-Jan Kraak http://orcid.org/0000-0002-8605-0484
References
Aggarwal, C. C. (Ed.) (2011). Social network data analytics. New York, NY: Springer US.
Agrawal, D., Bernstein, P., Bertino, E., Davidson, S., Dayal, U., Franklin, M., … Widom, J. (2012).
Challenges and opportunities with big data: A community white paper developed by leading research-
ers across the United States. Retrieved from http://cra.org/ccc/wp-content/uploads/sites/2/2015/
05/bigdatawhitepaper.pdf.
Alexander, C. (2002). Book 1 – the phenomenon of life. Berkeley, CA: Center for Environmental
Structure.
Andrienko, G., & Andrienko, N. (1999). Interactive maps for visual data exploration. International
Journal of Geographical Information Science, 13(4), 355–374.
Andrienko, G., & Andrienko, N. (2007). Coordinated multiple views: A critical view. Zurich: Coordinated
& Multiple Views in Exploratory Visualization.
Andrienko, G., Andrienko, N., Demsar, U., Dransch, D., Dykes, J., Fabrikant, S. I., … Tominski, C. (2010).
Space, time and visual analytics. International Journal of Geographical Information Science, 24(10),
1577–1600.
Andrienko, G., Andrienko, N., Dykes, J., Kraak, M. J., Robinson, A., & Schumann, H. (2016). GeoVisual
analytics: Interactivity, dynamics, and scale. Cartography and Geographic Information Science, 43
(1), 1–2. doi:10.1080/15230406.2016.1095006
Andrienko, G., Andrienko, N., Jankowski, P., Keim, D., Kraak, M.-J., MacEachren, A. M., & Wrobel, S.
(2007). Geovisual analytics for spatial decision support: Setting the research agenda.
International Journal of Geographical Information Science, 21(8), 839–857.
Andrienko, G. L., Andrienko, N. V., Voss, H., Bernardo, F., Hipolito, J., & Kretchmer, U. (2002). Testing the
usability of interactive maps in CommonGIS. Cartography and Geographic Information Science, 29
(4), 325–342.
Andrienko, N., & Andrienko, G. (2011). Spatial generalization and aggregation of massive movement
data. IEEE Transactions on Visualization and Computer Graphics, 17(2), 205–219.
Andrienko, N., & Andrienko, G. (2012). A visual analytics framework for spatio-temporal analysis and
modelling. Data Mining and Knowledge Discovery, 27(1), 55–83. doi:10.1007/s10618-012-0285-7
Andrienko, N., Andrienko, G., Fuchs, G., Rinzivillo, S., & Betz, H. D. (2015). Detection, tracking, and visu-
alization of spatial event clusters for real time monitoring. 36678 2015 IEEE international conference
on data science and advanced analytics (DSAA), pp. 1–10.
Balcan, D., Gonçalves, B., Hu, H., Ramasco, J. J., Colizza, V., & Vespignani, A. (2010). Modeling the
spatial spread of infectious diseases: The global epidemic and mobility computational model.
Journal of Computational Science, 1(3), 132–145. doi:10.1016/j.jocs.2010.07.002
Bédard, Y., Proulx, M.-J., Rivest, S., & Badard, T. (2006). Merging hypermedia GIS with spatial on-line
analytical processing: Towards hypermedia SOLAP. In E. Stefanakis, M. P. Peterson, C. Armenakis, &
V. Delis (Eds.), Geographic hypermedia: Concepts and systems (pp. 167–187). Berlin: Springer.
Bertin, J. (1981). Graphics and graphic information processing. Berlin: De Gruyter.

Bertin, J. (1983). Semiology of graphics: Diagrams, networks, maps (W. Berg, Trans. (translation from
French 1967 edition) ed.). Madison: University of Wisconsin Press.
Brewer, I., MacEachren, A. M., Abdo, H., Gundrum, J., & Otto, G. (2000). Collaborative geographic visu-
alization: Enabling shared understanding of environmental processes Information Visualization, 2000.
IEEE Symposium on InfoVis 2000, pp. 137–141.
Brézillon, P. (1999). Context in artificial intelligence II. Key elements of contexts. Computers and
Artificial Intelligence, 18(5), 425–446.
Brown, M., Sharples, S., Harding, J., Parker, C. J., Bearman, N., Maguire, M., … Jackson, M. (2013).
Usability of geographic information: Current challenges and future directions. Applied
Ergonomics, 44(6), 855–865. doi:10.1016/j.apergo.2012.10.013
Buckley, A. (2013). Guidelines for the effective design of spatio-temporal maps 26th International
Cartographic Congress. Dresden: International Cartographic Association.
Buckley, A., & Jenny, B. (2012). Letter from the Guest Editors. Cartographic Perspectives, (73), 3–12.
doi:10.14714/CP73.786.
Candy, L., & Edmonds, E. (2000). Creativity enhancement with emerging technologies.
Communications of the ACM, 43(8), 63–65. doi:10.1145/345124.345144
Caquard, S. (2013). Cartography I: Mapping narrative cartography. Progress in Human Geography, 37
(1), 135–144. doi:10.1177/0309132511423796
Caquard, S., & Fraser Taylor, D. R. (2005). Chapter 12 Art, maps and cybercartography: Stimulating
reflexivity among map-users. In D. R. F. Taylor (Ed.), Modern cartography series (Vol. 4, pp. 285–
307). Amsterdam: Academic Press.
Cartwright, W. (1994). Chapter 5 – interactive multimedia for mapping. In M. Alan & T. D .R. Fraser
(Eds.), Modern cartography series (Vol. 2, pp. 63–89). Oxford: Academic Press.
Cartwright, W. (2004). Engineered serendipity: Thoughts on the design of conglomerate GIS and
geographical new media artifacts. Transactions in GIS, 8(1), 1–12. doi:10.1111/j.1467-9671.2004.
00164.x
Cartwright, W., Crampton, J., Gartner, G., Miller, S., Mitchell, K., Siekierska, E., & Wood, J. (2001).
Geospatial information visualization user interface issues. Cartography and Geographic
Information Science, 28(1), 45–60. doi:10.1559/152304001782173961
Cartwright, W., Gartner, G., & Lehn, A. (Eds.). (2009). Cartography and art. Berlin: Springer-Verlag.
Casey, E. S. (2005). Earth-mapping: Artists reshaping landscape. Minneapolis: University of Minnesota
Press.
Chaotic Atmospheres, I. (2015). Flowing city maps. Cartographic Perspectives, (81), 49–52. doi:10.
14714/CP81.1316
Chen, C. (2013). Mapping scientific frontiers: The quest for knowledge visualization. London: Springer-
Verlag.
Christophe, S., & Hoarau, C. (2012). Expressive map design based on pop art: Revisit of semiology of
graphics? Cartographic Perspectives, 73(2012), 61–74.
Dey, A. K. (2001). Understanding and using context. Personal and Ubiquitous Computing, 5(1), 4–7.
doi:10.1007/s007790170019
DiBiase, D. (1990). Visualization in the earth sciences. Earth and Mineral Sciences, Bulletin of the College
of Earth and Mineral Sciences, The Pennsylvania State University, 59(2), 13–18.
Dix, A., Finlay, J., Abowd, G. D., & Beale, R. (2004). Human-computer interaction (3rd ed.). Upper Saddle
River, NJ: Prentice Hall.
Dodge, M., Kitchin, R., & Perkins, C. (Eds.). (2011). The map reader: Theories of mapping practice and
cartographic representation. Chichester: John Wiley & Sons.
Döllner, J. (2007). Non-photorealistic 3D geovisualization. In W. Cartwright, M. P. Peterson, & G.
Gartner (Eds.), Multimedia cartography (pp. 229–240). Berlin: Springer.
Dykes, J., & Brunsdon, C. (2007). Geographically weighted visualization: Interactive graphics for scale-
varying exploratory analysis. IEEE Transactions on Visualization and Computer Graphics, 13(6), 1161–
1168. doi:10.1109/TVCG.2007.70558
Dykes, J., Wood, J., & Slingsby, A. (2010). Rethinking map legends with visualization. IEEE Transactions
on Visualization and Computer Graphics, 16(6), 890–899. doi:10.1109/TVCG.2010.191
Eccles, R., Kapler, T., Harper, R., & Wright, W. (2008). Stories in GeoTime. Information Visualization, 7(1),
3–17.
Ehrensvärd, U. (1987). Color in cartography. In D. Woodward (Ed.), Art and cartography: Six historical
essays (pp. 123–146). Chicago, IL: Chicago University Press.
Endert, A., Chang, R., North, C., & Zhou, M. (2015). Semantic interaction: Coupling cognition and
computation through usable interactive analytics. IEEE Computer Graphics and Applications, 35
(4), 94–99.
Endsley, M. R. (1995). Toward a theory of situation awareness in dynamic systems. Human Factors: The
Journal of the Human Factors and Ergonomics Society, 37(1), 32–64.
Etemad, K., Carpendale, S., & Samavati, F. (2014). Node-ring graph visualization clears edge congestion.
Proceedings of the IEEE VIS 2014 Arts Program, VISAP’14, Paris, pp. 67–74.
Fabrikant, S. I. (2000). Spatialized browsing in large data archives. Transactions in GIS, 4(1), 65–78.
Fairbairn, D., Andrienko, G., Andrienko, N., Buziek, G., & Dykes, J. (2001). Representation and its
relationship with cartographic visualization. Cartography and Geographic Information Science, 28
(1), 13–28.
Few, S. (2013). Information dashboard design: Displaying data for at-a-glance monitoring. Berkeley, CA:
Analytics Press.
Field, K. (Cartographer). (2015). Pollock-style election map. Retrieved from http://www.arcgis.com/
apps/webappviewer/index.html?id=661d96be9fe64b3db4124546b67b2f39
Field, K., & O’Brien, J. (2010). Cartoblography: Experiments in using and organising the spatial context
of micro-blogging. Transactions in GIS, 14(1), 5–23.
Fish, C., Goldsberry, K. P., & Battersby, S. (2011). Change blindness in animated choropleth maps: An
empirical study. Cartography and Geographic Information Science, 38(4), 350–362.
Fotheringham, A. S., & Wong, D. W. S. (1991). The modifiable areal unit problem in multivariate stat-
istical analysis. Environment and Planning A, 23(7), 1025–1044. doi:10.1068/a231025
Gahegan, M., Wachowicz, M., Harrower, M., & Rhyne, T.-M. (2001). The integration of geographic visu-
alization with knowledge discovery in databases and geocomputation. Cartography and
Geographic Information Science, 28(1), 29–44. doi:10.1559/152304001782173952
Gandomi, A., & Haider, M. (2015). Beyond the hype: Big data concepts, methods, and analytics.
International Journal of Information Management, 35(2), 137–144. doi:10.1016/j.ijinfomgt.2014.
10.007
Goodchild, M. F. (2007). Citizens as voluntary sensors: Spatial data infrastructure in the world of Web
2.0. International Journal of Spatial Data Infrastructures, 2(1), 24–32.
Goodwin, S., Dykes, J., & Slingsby, A. (2014). Visualizing the effects of scale and geography in multi-
variate comparison. 2014 IEEE Conference on Visual Analytics Science and Technology (VAST),
pp. 251–252.
Gotz, D., & Zhou, M. X. (2009). Characterizing users’ visual analytic activity for insight provenance.
Information Visualization, 8(1), 42–55. doi:10.1057/ivs.2008.31
Graham, M., & Shelton, T. (2013). Geography and the future of big data, big data and the future of
geography. Dialogues in Human Geography, 3(3), 255–261. doi:10.1177/2043820613513121
Griffin, A. L., Robinson, A. C., & Roth, R. E. (submitted). Envisioning the future of cartographic research.
International Journal of Cartography.
Guo, D., Chen, J., MacEachren, A. M., & Liao, K. (2006). A visualization system for space-time and multi-
variate patterns (VIS-STAMP). IEEE Transactions on Visualization and Computer Graphics, 12(6),
1461–1474.
Heath, T., Bizer, C., & Hendler, J. (2011). Linked data. Williston, VT: Morgan & Claypool.
Hey, T., Tansley, S., & Tolle, K. (Eds.). (2009). The fourth paradigm: Data intensive scientific discovery.
Redmond, WA: Microsoft Research.
Ishizu, T., & Zeki, S. (2011). Toward a brain-based theory of beauty. PLoS ONE, 6(7), e21852. doi:10.
1371/journal.pone.0021852
Jagadish, H. V. (2015). Big data and science: Myths and reality. Big Data Research, 2(2), 49–52. doi:10.
1016/j.bdr.2015.01.005
Jeong, D. H., Ji, S.-Y., Suma, E. A., Yu, B., & Chang, R. (2015). Designing a collaborative visual analytics
system to support users’ continuous analytical processes. Human-centric Computing and
Information Sciences, 5(1), 5–20. doi:10.1186/s13673-015-0023-4
Jiang, B. (2013a). Head/tail breaks: A new classification scheme for data with a heavy-tailed distri-
bution. The Professional Geographer, 65(3), 482–494. doi:10.1080/00330124.2012.700499
Jiang, B. (2013b). Volunteered geographic information and computational geography: New perspec-
tives. In D. Sui, S. Elwood, & M. Goodchild (Eds.), Crowdsourcing geographic knowledge (pp. 125–
138). Berlin: Springer.
Jiang, B. (2015a). Geospatial analysis requires a different way of thinking: The problem of spatial het-
erogeneity. GeoJournal, 80(1), 1–13. doi:10.1007/s10708-014-9537-y
Jiang, B. (2015b). Head/tail breaks for visualization of city structure and dynamics. Cities, 43, 69–77.
doi:10.1016/j.cities.2014.11.013
Jiang, B., & Brandt, A. (2016). A fractal perspective on scale in geography. ISPRS International Journal
of Geo-Information, 5(6), 95. doi:10.3390/ijgi5060095
Jiang, B., & Sui, D. Z. (2014). A new kind of beauty out of the underlying scaling of geographic Space.
The Professional Geographer, 66(4), 676–686. doi:10.1080/00330124.2013.852037
Jin, X., Wah, B. W., Cheng, X., & Wang, Y. (2015). Significance and challenges of big data research. Big
Data Research, 2(2), 59–64. doi:10.1016/j.bdr.2015.01.006
Jones, C. B. (1997). Geographical information systems and computer cartography. New York, NY:
Routledge.
Kang, Y., & Stasko, J. (2011). Characterizing the intelligence analysis process: Informing visual analytics
design through a longitudinal field study. 2011 IEEE conference on visual analytics science and tech-
nology (VAST), pp. 21–30.
Kawabata, H., & Zeki, S. (2004). Neural correlates of beauty. Journal of Neurophysiology, 91(4), 1699–
1705. doi:10.1152/jn.00696.2003
Keim, D. A., Mansmann, F., Schneidewind, J., Thomas, J., & Ziegler, H. (2008). Visual analytics: Scope
and challenges. In S. J. Simoff, M. H. Böhlen, & A. Mazeika (Eds.), Visual data mining: Theory, tech-
niques and tools for visual analytics (pp. 76–90). Berlin: Springer.
Kimerling, A. J., Buckley, A. R., Muehrcke, P. C., & Muehrcke, J. O. (2011). Map use: Reading, analysis,
interpretation (7th ed.). Redlands, CA: Esri Press.
Kinkeldey, C., MacEachren, A. M., & Schiewe, J. (2014). How to assess visual communication of uncer-
tainty? A systematic review of geospatial uncertainty visualisation user studies. The Cartographic
Journal, 51(4), 372–386. doi:10.1179/1743277414Y.0000000099
Kitchin, R. (2013). Big data and human geography: Opportunities, challenges and risks. Dialogues in
Human Geography, 3(3), 262–267. doi:10.1177/2043820613513388
Klippel, A., Wallgrün, J. O., Yang, J., Mason, J. S., Kim, E.-K., & Mark, D. M. (2013). Fundamental cognitive
concepts of space (and time): Using cross-linguistic, crowdsourced data to cognitively calibrate
modes of overlap. In T. Tenbrink, J. Stell, A. Galton, & Z. Wood (Eds.), COSIT 2013 proceedings of
the 11th International conference on spatial information theory – Scarborough, UK, September 2–
6, 2013 (pp. 377–396). Cham: Springer International.
Koch, C. (2015). Do Androids dream? Scientific American Mind, 26, 24–27.
Kraak, M. J. (1989). Computer-assisted cartographical 3D imaging & techniques. In J. Raper (Ed.),
Three-dimensional applications in geographical information systems (pp. 99–114). London: Taylor
& Francis.
Kraak, M. J. (2014). Mapping time. Redlands, CA: Esri Press.
Krygier, J. B. (1994). Chapter 8 – sound and geographic visualization. In M. Alan & T. D. R. Fraser (Eds.),
Modern cartography series (Vol. 2, pp. 149–166). Oxford: Academic Press.
Laney, D. (2001, February 6). 3D data management: Controlling data volume, velocity, and variety.
Retrieved from http://blogs.gartner.com/doug-laney/files/2012/01/ad949-3D-Data-Management-
Controlling-Data-Volume-Velocity-and-Variety.pdf
Lee, J.-G., & Kang, M. (2015). Geospatial big data: Challenges and opportunities. Big Data Research, 2
(2), 74–81. doi:10.1016/j.bdr.2015.01.003
Leonardi, L., Orlando, S., Raffaetà, A., Roncato, A., Silvestri, C., Andrienko, G., & Andrienko, N. (2013). A
general framework for trajectory data warehousing and visual OLAP. GeoInformatica, 18(2), 273–
312. doi:10.1007/s10707-013-0181-3
Lewis, S. C. (2015). Journalism in an era of big data. Digital Journalism, 3(3), 321–330. doi:10.1080/
21670811.2014.976399
Li, S., Dragicevic, S., Castro, F. A., Sester, M., Winter, S., Coltekin, A., … Cheng, T. (in press). Geospatial
big data handling theory and methods: A review and research challenges. ISPRS Journal of
Photogrammetry and Remote Sensing. doi:10.1016/j.isprsjprs.2015.10.012
Liu, D., Bias, R. G., Lease, M., & Kuipers, R. (2012). Crowdsourcing for usability testing. Proceedings of
the American Society for Information Science and Technology, 49(1), 1–10. doi:10.1002/meet.
14504901100
MacEachren, A. M. (1992). Visualizing uncertain information. Cartographic Perspectives, (13), 10–19.
doi:10.14714/cp13.1000
MacEachren, A. M. (1994). Visualization in modern cartography: Setting the agenda. In A. M. MacEachren
& D. R. F. Taylor (Eds.), Visualization in modern cartography (pp. 1–12). Oxford: Pergamon.
MacEachren, A. M. (1995). How maps work: Representation, visualization and design. New York, NY:
Guilford Press.
MacEachren, A. M. (2015). Visual analytics and uncertainty: It’s not about the data. In EuroVis workshop
on visual analytics (EuroVA) (pp. 55–60). Cagliari: Eurographics Association.
MacEachren, A. M., & Ganter, J. H. (1990). A pattern identification approach to cartographic visualiza-
tion. Cartographica: The International Journal for Geographic Information and Geovisualization, 27
(2), 64–81.
MacEachren, A. M., Jaiswal, A., Robinson, A. C., Pezanowski, S., Savelyev, A., Mitra, P., … Blanford, J.
(2011). SensePlace2: Geotwitter analytics support for situation awareness. IEEE Conference on
Visual Analytics Science and Technology, Providence, RI.
MacEachren, A. M., & Kraak, M. J. (1997). Exploratory cartographic visualization: Advancing the
agenda. Computers & Geosciences, 23(4), 335–343.
MacEachren, A. M., & Kraak, M.-J. (2001). Research challenges in geovisualization. Cartography and
Geographic Information Science, 28(1), 3–12.
MacEachren, A. M., Robinson, A. C., Hopper, S., Gardner, S., Murray, R., & Gahegan, M. (2005).
Visualizing geospatial information uncertainty: What we know and what we need to know.
Cartography and Geographic Information Science, 32(3), 139–160.
MacGill, J., & Openshaw, S. (1998). The use of flocks to drive a geographic analysis machine. Third
International Conference on Geocomputation, Bristol, UK.
Maciejewski, R., Hafen, R., Rudolph, S., Larew, S. G., Mitchell, M. A., Cleveland, W. S., & Ebert, D. S.
(2011). Forecasting hotspots – a predictive analytics approach. IEEE Transactions on Visualization
and Computer Graphics, 17(4), 440–453. doi:10.1109/TVCG.2010.82
Mandelbrot, B. (1982). The fractal geometry of nature. New York, NY: W.H. Freeman.
McCloud, S. (1994). Understanding comics. New York, NY: William Morrow Paperbacks.
McCurdy, N., Lein, J., Coles, K., & Meyer, M. (2016). Poemage: Visualizing the sonic topology of a poem.
IEEE Transactions on Visualization and Computer Graphics, 22(1), 439–448. doi:10.1109/TVCG.2015.
2467811
Merriam-Webster. (2014). Webster’s new world college dictionary (5th ed., pp. 1728). Boston, MA:
Webster’s New World.
Miller, H. J., & Goodchild, M. F. (2014). Data-driven geography. GeoJournal, 80(4), 449–461. doi:10.
1007/s10708-014-9602-6
Moore, A., Marinescu, D., & Tenzer, R. (2011). A visual art interface to an embedded sequence of maps.
Cartography and Geographic Information Science, 38(2), 184–191. doi:10.1559/15230406382184
Moore, A. B. (2009). Maps as comics, comics as maps. In International cartographic conference (pp. 1–
13). Santiago: International Cartographic Association.
Moore, A. B., & Rodda, J. (2015). Adaptive relative motion representation of space–time trajectories.
The Cartographic Journal, 52(2), 204–209. doi:10.1080/00087041.2015.1119463
Morisette, J. T., Jarnevich, C. S., Holcombe, T. R., Talbert, C. B., Ignizio, D., Talbert, M. K., … Young, N. E.
(2013). VisTrails SAHM: Visualization and workflow management for species habitat modeling.
Ecography, 36(2), 129–135. doi:10.1111/j.1600-0587.2012.07815.x
Olshannikova, E., Ometov, A., Koucheryavy, Y., & Olsson, T. (2015). Visualizing big data with augmen-
ted and virtual reality: Challenges and research agenda. Journal of Big Data, 2(1), 1–27. doi:10.
1186/s40537-015-0031-2
Openshaw, S. (1998). Towards a more computationally minded scientific human geography.
Environment and Planning A, 30(2), 317–332. doi:10.1068/a300317
Patterson, S. (1992). The great bear. Retrieved from http://www.tate.org.uk/art/artworks/patterson-
the-great-bear-p77880.
Priestnall, G., & Hampson, D. (2008). Landscape visualization: Science and art. In M. Dodge, M.
McDerby, & M. Turner (Eds.), Geographic visualization (pp. 241–258). Chichester: John Wiley & Sons.
Ragan, E. D., Endert, A., Sanyal, J., & Chen, J. (2016). Characterizing provenance in visualization and data
analysis: An organizational framework of provenance types and purposes. IEEE Transactions on
Visualization and Computer Graphics, 22(1), 31–40. doi:10.1109/TVCG.2015.2467551
Riedl, A. (2012). State-of-the-art of tactile hyperglobes. In M. Buchroithner (Ed.), True-3D in cartogra-
phy (pp. 215–226). Berlin: Springer.
Rinner, C. (2001). Argumentation maps: GIS-based discussion support for on-line planning.
Environment and Planning B: Planning and Design, 28(6), 847–863. doi:10.1068/b2748t
Roberts, J. (2005). Exploratory visualization with multiple linked views. In J. Dykes, A. M. MacEachren,
& M. J. Kraak (Eds.), Exploring geovisualization (pp. 159–180). Amsterdam: Elsevier.
Robinson, A. C. (2009). Needs assessment for the design of information synthesis visual analytics
tools. In IEEE international conference on information visualization (pp. 353–360). Barcelona: IEEE
Computer Society.
Robinson, A. C. (2011). Supporting synthesis in geovisualization. International Journal of Geographical
Information Science, 25(2), 211–227.
Robinson, A. C., Chen, J., Lengerich, G., Meyer, H., & MacEachren, A. M. (2005). Combining usability
techniques to design geovisualization tools for epidemiology. Cartography and Geographic
Information Science, 32(4), 243–255.
Rosenberg, D., & Grafton, A. (2010). Cartographies of time. New York, NY: Princeton Architectural
Press.
Roth, R. E. (2012). Cartographic interaction primitives: Framework and synthesis. The Cartographic
Journal, 49(4), 376–395. doi:10.1179/1743277412y.0000000019
Roth, R. E. (2013). Interactive maps: What we know and what we need to know. The Journal of Spatial
Information Science, 6, 59–115.
Roth, R. E., Hart, D., Mead, R., & Quinn, C. (in press). Wireframing for interactive & web-based mapping:
Designing the NOAA Lake Level Viewer. Cartography and Geographic Information Science
Retrieved from http://www.tandfonline.com/doi/abs/10.1080/15230406.2016.1171166.
Rystedt, B., Ormeling, F., Konecny, M., Li, L., Liebenberg, E., McMaster, R., … Wood, M. (2003). A stra-
tegic plan for the International Cartographic Association, 2003–2011. I. C. Association Retrieved from
http://icaci.org/files/documents/reference_docs/ICA_Strategic_Plan_2003-2011.pdf.
Seamon, D., & Sowers, J. (2008). Place and placelessness, Edward Relph. In P. Hubbard, R. Kitchin, & G.
Vallentine (Eds.), Key texts in human geography (pp. 43–51). London: Sage.
Shao, L., Sacha, D., Neldner, B., Stein, M., & Schreck, T. (2016). Visual-interactive search for soccer tra-
jectories to identify Interesting game situations. Electronic Imaging, 2016(1), 1–10.
Shneiderman, B. (1996). The eyes have it: A task by data type taxonomy for information visualizations
1996. In IEEE conference on visual languages (pp. 336–343). Boulder, CO: IEEE Computer Society.
Shneiderman, B. (2007). Creativity support tools: Accelerating discovery and innovation.
Communications of the ACM, 50(12), 20–32. doi:10.1145/1323688.1323689
Slocum, T., Blok, C., Jiang, B., Koussoulakou, A., Montello, D., Fuhrmann, S., & Hedley, N. (2001).
Cognitive and usability issues in geovisualization. Cartography and Geographic Information
Science, 28(1), 61–75.
Stienen, E. W. M., Desmet, P., Aelterman, B., Courtens, W., Feys, S., Vanermen, N., … Lens, L. (2016).
GPS tracking data of lesser black-backed gulls and herring gulls breeding at the southern North
Sea coast. ZooKeys, 555, 115–124. doi:10.3897/zookeys.555.6173
Sundararajan, L. (2014). Mind, machine, and creativity: An artist’s perspective. The Journal of Creative
Behavior, 48(2), 136–151. doi:10.1002/jocb.44
Thatcher, J. (2014). Living on fumes: Digital footprints, data fumes, and the limitations of spatial big
data. International Journal of Communication, 8(2014), 1765–1783.
Thomas, J. J., & Cook, K. A. (Eds.). (2005). Illuminating the path: The research and development agenda
for visual analytics. New York, NY: IEEE CS Press.
Thudt, A., Hinrichs, U., & Carpendale, S. (2012). The bohemian bookshelf: Supporting serendipitous book
discoveries through information visualization. Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems, Austin, TX, USA.
Tomaszewski, B., & MacEachren, A. M. (2012). Geovisual analytics to support crisis management:
Information foraging for geo-historical context. Information Visualization, 11(4), 339–359. doi:10.
1177/1473871612456122
Tsou, M.-H. (2015). Research challenges and opportunities in mapping social media and Big Data.
Cartography and Geographic Information Science, 42(supp1), 70–74. doi:10.1080/15230406.2015.
1059251
Tuan, Y.-F. (1977). Space and place: The perspective of experience. Minneapolis: University of Minnesota
Press.
United Nations General Assembly. (2015). Transforming our world: The 2030 agenda for sustainable
development (United Nations Ed., Vol. A/RES/70/1). Geneva: United Nations. Retrieved from
http://www.un.org/ga/search/view_doc.asp?symbol=A/RES/70/1&Lang=E
Virrantaus, K., Fairbairn, D., & Kraak, M. J. (2009). ICA research agenda on cartography and GI science.
The Cartographic Journal, 36(2), 209–222.
Watson, R. (2009). Mapping and contemporary art. The Cartographic Journal, 46(4), 293–307.
Wilkinson, L. (2005). The grammar of graphics (2nd ed.). New York, NY: Springer-Verlag.
Wills, G., & Wilkinson, L. (2010). AutoVis: Automatic visualization. Information Visualization, 9(1), 47–69.
doi:10.1057/ivs.2008.27
Wu, X., Zurita-Milla, R., & Kraak, M.-J. (2015). Co-clustering geo-referenced time series: Exploring
spatio-temporal patterns in Dutch temperature data. International Journal of Geographical
Information Science, 29(4), 624–642. doi:10.1080/13658816.2014.994520
Yusof, N., Zurita-Milla, R., Kraak, M.-J., & Retsios, B. (2016). Interactive discovery of sequential patterns
in time series of wind data. International Journal of Geographical Information Science, 1–21. doi:10.
1080/13658816.2015.1135928
Zeki, S. (2001). Essays on science and society: Artistic creativity and the Brain. Science, 293(5527), 51–
52. doi:10.1126/science.1062331
Zhang, J., & Goodchild, M. (2002). Uncertainty in geographical information. New York, NY: CRC Press.

15 5 2018 Geospatial 221124 064741

Uploaded by

Copyright:

Available Formats

15 5 2018 Geospatial 221124 064741

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

15 5 2018 Geospatial 221124 064741

Uploaded by

Copyright:

Available Formats

International Journal of Cartography

ISSN: 2372-9333 (Print) 2372-9341 (Online) Journal homepage: http://www.tandfonline.com/loi/tica20

Geospatial big data and cartography: research

Anthony C. Robinson, Urška Demšar, Antoni B. Moore, Aileen Buckley, Bin

To link to this article: https://doi.org/10.1080/23729333.2016.1278151

© 2017 The Author(s). Published by Informa Published online: 13 Mar 2017.

Submit your article to this journal Article views: 1747

View related articles View Crossmark data

Citing articles: 5 View citing articles

Full Terms & Conditions of access and use can be found at

Geospatial big data and cartography: research challenges

ABSTRACT ARTICLE HISTORY

CONTACT Anthony C. Robinson arobinson@psu.edu

we outline five crosscutting themes that provide key perspectives on cartographic

Place, space and time

Users and usability

Figure 1. Research challenges in cartography for geospatial big data.

Making sense of geospatial big data

Challenge: design effective map-based interfaces to support long-term analytical

Challenge: develop new approaches for supporting collaborative decision-making

Challenge: develop techniques that allow users to express and match a

Challenge: develop techniques for understanding change over time in geospatial

Challenge: craft new approaches to support predictive analytics of dynamic

Challenge: develop spatiotemporal visualization methods for geospatial big data

Challenge: create maps and map-oriented interfaces that prompt attention to

Challenge: characterize the quality and certainty of geospatial big data

Art and geospatial big data

Challenge: encourage and maximize creative contributions for expressing

Challenge: build a library of artistic methods and techniques for representing

Challenge: generate artistic renderings of geospatial big data

Challenge: dynamically link artworks to conventional geospatial big data

Figure 2. Research opportunities in cartography for geospatial big data.

Visual methods for geospatial big data

Opportunity: systematically evaluate the ability of existing visual methods in

Opportunity: adapt cartographic generalization principles and techniques to

Opportunity: couple computational methods and cartographic representation best

Opportunity: leverage what we know about map animation and interactive

of course, depend on concomitant advances in computational methods to drive such

Computational methods for geospatial big data

Opportunity: leverage knowledge about patterns across scales in the development

Opportunity: use what we know about human dynamics to find patterns in

Opportunity: connect concepts from complexity science to new visual analytics

Adapting artistic methods for geospatial big data

Opportunity: co-opt artistic methods and techniques to represent geospatial big

Opportunity: generate artistic renderings of geospatial big data

Opportunity: link artworks to conventional representations in a visual analytics

Making maps that matter from geospatial big data

Bertin, J. (1981). Graphics and graphic information processing. Berlin: De Gruyter.

You might also like