Cartographic Visualization

1
Cartographic Visualization
Joel Morrison Ohio State University Barbara
Buttenfield U. Colorado Boulder GIS-T
2003 Colorado Springs
2
Basic Premise
Using GIS The Cost of Well Designed
Visualizations EQUALS The Cost of Poorly Designed
Visualizations
3
Electronic Technology

• PRIOR to Availability of Electronic Technology
• MAP was both
• the database, and
• the visualization
• TODAY we use GIS to perform logical analyses for
  specific answers to queries and also to visualize
  the results - two separate operations

4
Major Question

• If visualization must convey accurately the
  results of GIS analysis, how can this be done?
• When the display does both, compromise is
  essential.
• Today, the visualization must reinforce the
  results of the GIS analysis.

5
Brief Review

• Nature of raster and vector data
• Geometric and topologic measurement
• Point, line, and area data types
• Nominal, ordinal, interval/ratio scales of
  measurement
• Volumetric Data
• Elements of spatial data accuracy
• GIS functionalities

6
Brief Review

• Raster Data
• Vector Data

7
Brief Review

• Geometric Measurement
• Topologic Measurement
• 2405 Elm Street
• 2401 Elm Street
• 1335 16th Avenue
• 1331 16th Avenue
• 1327 16th Avenue

8
Brief Review

• Data types
• Measurement scales

9
Brief Review

• Volumetric data
• Choropleth, Dasymetric, Isarithmic maps

10
Brief Review
Volumetric Symbols
11
Brief Review

• Spatial Data Accuracy
• Lineage
• Positional accuracy
• Attribute accuracy
• Completeness
• Logical consistency
• Temporal information
• Semantic accuracy

12
Brief Review

• GIS combines various data types, data
  measurements, and data accuracies, to produce an
  explicit answer to a spatial database query.
• How do we create a visualization that supports
  the answer?

13
Outline of Topics

• Principles of Design
• Controls on Design
• Visual Variables
• Typography
• Putting it all Together

14
GIS Visualization Design

• Principles of Design
• Clarity and Legibility
• Visual Contrast, Visual Grouping
• Balance and Stability
• Figure/Ground
• Hierarchical Organization

15
Principles of Design

• Clarity and Legibility
• Three levels of knowing
• Existence
• Associated existence
• Spatially associated existence
• Size is most important. Human eye subtends an
  angle of one minute of arc (so use at least two)
• Color adds Clarity
• Avoid ambiguity
• Recognize familiarity and use it to advantage

16
Clarity and Legibility

• Color adds clarity
• Size adds legibility

17
Principles of Design

• Visual Contrast
• The most important element
• Without recognizable contrast, existence will be
  overlooked
• Avoid guesswork, make different things different
• Contrast affects mood of display

18
Visual Contrast

• Poor contrast
• Good contrast

19
Principles of Design

• Balance and stability
• The optical center of a display does not coincide
  with its physical center
• Each major feature that is part of the display
  has a weight
• To achieve stability in a display requires the
  balancing of those weights

20
Balance and Stability

• Balanced features
• Visual Center B/S

21
Principles of Design

• Figure/Ground
• Eye and Mind create a Figure and separate it from
  a Ground in any display
• The Figure is perceived as a coherent shape or
  form with clear outlines in front of or above its
  surrounding or ground
• In creating a display, know what you want as
  Figure

22
Principles of Design

• Figure/Ground What we know
• To create Figure
• Add differentiation/detail to create homogeneous
  figure
• Make figure a closed form
• Use familiarity with shapes to make figure
• Lightness creates figure
• Good contour creates figure
• Small size tends to create figure

23
Figure/Ground

• Creating Figure
• Using good contour
• Using closed form
• Using small size

24
Figure/Ground

• Creating Figure
• Using familiarity of shapes
• Using lightness

25
Figure/Ground

• Creating Figure
• Using differentiation
• Or articulation

26
Principles of Design

• Hierarchical Organization
• Displays make use of visual layering
• Create a hierarchy of levels of importance to
  direct readers attention to your point
• Three types
• Stereogrammic
• Extensional
• Subdivisional

27
Hierarchical Organization

• Stereogrammic

28
Hierarchical Organization

• Extensional

29
Hierarchical Organization

• Subdivisional

30
GIS Visualization Design

• Controls on Design
• Purpose
• Reality
• Available Data
• Audience
• Technical Limits
• Conditions of Use (fitness for use?)

31
Controls on Design

• Purpose
• Substantive Objective
• What information must be included to make out
  point or to back-up our analysis
• Affective Objective
• What mood are we trying to create urgency to
  act, direness of a situation, happy solution

32
Controls on Design

• Reality
• Geography of area features have a shape and
  size
• Chile and Norway are long and narrow
• Denver is at the edge of mountains and plains
• Minnesota and Wisconsin have many small lakes

33
Controls on Design

• Available Data
• Date
• The earth changes with time. The transportation
  data may be of a different date from the
  hydrographic data
• Impression of Accuracy
• Visual displays often present a view of
  unwarranted accuracy. How to prevent this.

34
Controls on Design

• Audience
• Age
• Generally assumed that young readers need
  simplicity and like bright colors
• Mature adults prefer more subtle coloration
• Education level usually correlates with
  complexity of design
• Cultural constraints

35
Controls on Design

• Technical Limits
• How will visualization be distributed
• On-line
• CD
• Size of file and ability of user to receive it
• Will it also be printed?

36
Controls on Design

• Conditions of Use
• On screen
• In office
• In automobile or other mobile device
• In classroom
• Night light conditions
• Eye glasses of various sorts

37
GIS Visualization Design

• Visual Variables
• Size
• Value
• Texture
• Color
• Orientation
• Shape
• Typography
• Form
• Style
• Size

38
Visual Variables

• Size
• Value darkness
• Texture
• Color (Hue)
• Shape
• Orientation
• Position

39
Visual Variables
How do they operate?
(Bertin 1983 modified by Buttenfield, 1996)
40
Size
How Many Groups?
41
Size displays metric data progressions
42
Size displays metric data progressions
43
What We Know

• Human reactions to Size
• physiological reactions
• psychological reactions
• conventional connotations

44
Value
How Many Groups?
45
What We Know

• Human reactions to Value
• physiological reactions
• psychological reactions
• conventional connotations

46
Value
Value and Contrast
Herring grid Benussi ring Kanisza triangle
47
Value
Which square is darker?
Land and McCann 1971 (as shown in Marr Vision)
48
Value
Focus on areas
Focus on boundary between areas
49
Value
Create impression of transparency
50
Value
Create Metric Progressions
Scanning a figure from Slocum
51
Texture
Texture and Value Interfere
52
Texture
Preserving Texture in Data Reduction
53
Color
54
What We Know

• Human reactions to Color
• physiological reactions
• psychological reactions
• conventional connotations

55
Color

• Color adds clarity and supports the visual
  hierarchy

56
Color

• Color lends depth and supports a sense of spatial
  realism

57
Color

• How many colors do you really see?
• Color creation is not straightforward

58
Color

• Expectations about color meanings

59
Color

• Professor Cindy Brewer, Penn State
• http//www.personal.psu.edu/cab38/ColorSch/SchHome
  .html
• httpwww.personal.psu.edu/cab38/ColorSch/Schemes.h
  tml
• httpcolorbrewer.org

60
Color Brewer
61
Color Brewer
62
Color Brewer
63
Color Brewer
64
Color Brewer
65
Color Brewer
66
Color Brewer
67
Color Brewer
68
Color Brewer
69
Color Brewer
70
Color Brewer
71
Orientation
How Many Groups?
72
What We Know

• Human reactions to Orientation
• physiological reactions
• psychological reactions
• conventional connotations

73
Orientation
Direction reversal
74
Orientation
Swiss topographic map, 1945
75
Orientation
A different approach to terrain shading
76
Orientation
We have no trouble resolving two viewpoints
77
Orientation
Childs drawing (Arnheim, 1964)
78
Shape
How Many Groups?
79
Shape
How Many Groups?
80
Shape
How Many Groups?
81
Shape
Preserve shape during data generalization
82
Shape
When possible, legend placement should conform
with map shape
83
Shape
Transverse Mercator - conformal
Azimuthal Equidistant neither conformal nor
equivalent
Shape shows projection type and properties
84
Shape
Shape can serve as an analysis tool in its own
right
85
Shape
Shape can serve as an analysis tool in its own
right
Time travel transformations of Seattle (Tobler)
86
What We Know

• Human reactions to Shape
• physiological reactions
• psychological reactions
• conventional connotations

87
Shape
Shape perception affects what appears to be most
prominent (figure-ground)
88
Shape
Resolve shape ambiguities with other visual
variables (value, texture) and type
89
Typography

• Style
• Form
• Size
• Rules of Thumb
• New developments

90
Typography

• Styles
• Classic or Old Style Garamond

91
Typography

• Style
• Modern Bodoni

92
Typography

• Style
• Sans Serif Optima

93
Typography

• Style
• Computer designed Verdana

94
Typography
Comparison of Styles
95
Typography

• Form
• UPPER and lower case
• Upright versus slant
• Condensed versus expanded
• Light versus Bold
• italic

96
Typography

• 8 point
• 9 point
• 10 point
• 12 point
• 14 point
• 16 point
• 18 point
• 20 point
• 24 point
• 28 point
• 32 point
• 36 point
• 40 point
• 44 point

• 44 point
• 48 point
• 54 point
• 60 point
• 66 point
• 72 point

Type Sizes
97
Typography

• What we know
• Rules of Thumb - Style
• Never mix Classic and Modern styles
• Use sans serif with each of above
• Use as few different styles as possible

98
Typography

• What we know
• Rules of Thumb Form
• Caps and lower case are easier to read than all
  Caps
• Slant and/or italic are generally used for
  hydrographic features, upright are used for
  natural or man-made features

99
Typography

• What we know
• Rules of Thumb Size
• Is a more significant typographic variable than
  style or form
• Larger size connotes more importance
• 3-point type is the smallest capable of being
  read, for graphics 5-point is considered the
  absolute minimum
• Differences of less than 15 in point size are
  not recognized

100
Typography

• New Developments
• Use of audio instead of type to define features
  in a spatial display
• Dynamic text that moves across the display
• Sliding text that moves with zoom in or out,
  versus text fixed for one scale

101
Putting it all Together

• How many visual variables are at work?

102
Visual Variables in Practice

• Footnotes
• Crispness
• Resolution
• Representing Uncertainty
• Transparency (fog)
• Value
• Saturation

103
Brief Review
104
What We Know

• General Guidelines
• Legibility
• Harmony
• Ready association between label and feature
• Match visual hierarchy
• Suitability for reproduction
• Ease of execution

105
Typography
A few principles for good type placement (what
else could be improved?)

106
General References

• Elements of Cartography, John Wiley and Sons, 6th
  edition, 1995. Robinson, Morrison, Muehrcke,
  Kimerling and Guptill
• Thematic Cartography and Visualization, Prentice
  Hall, 1999. Slocum
• Cartographic Design Using ArcView GIS, OnWord
  Press, 2001. Madej

107
General References

• Cartography Thematic Map Design, McGraw-Hill,
  5th edition, 1999. Dent (includes CD of ArcView
  exercises w/ data)
• Some Truth with Maps, Assn. American Geographers,
  1994. MacEachren
• Cartography Visualization of Spatial Data,
  Longman, 1996. Kraak and Ormeling

108
Cartographic Visualization
Thank-you!
Joel Morrison Barbara Buttenfield morrison_at_cfm
.ohio-state.edu babs_at_colorado.edu
109
What We Know

• Visual Continua
• numerousness
• visual length
• visual area
• apparent volume
• color - value scale

110
Numerousness
111
Visual Length
112
Visual Area
113
Apparent Volume
114
Color Value Scale