Geovisualisation

1
Geovisualisation
Phil Bartie (GRCNZ)
2
(No Transcript)
3
The BIG Picture
The BIG Picture
Geovisualisation - assist in spatial data
exploration - assist in decision-making

• Background on mapping and map design
• Examples
• More background info
• More examples

http//www.datenform.de/map14_1000.jpg
4
Whats Special about Spatial?

• Toblers first law of geography
• everything is related to everything else, but
  near things are more related than distant things
  
• Tobler, W. R. 1970. A computer movie simulating
  urban growth in the Detroit region. Economic
  Geography 46 23440.

5
Geographic Understanding
task
Real World Data Model Data Structure File Format
conceptual data model of space around us
reality
6
What is a map?

• A graphic depiction of all or part of a
  geographic realm in which the real-world features
  have been replaced by symbols in their correct
  spatial location at a reduced scale.

power line
7
Limitations of Paper Maps

• Fixed scale
• Fixed extent
• Static view
• Flat and hence limited for 3D visualization
• Only presents complete world view
• Map producer-centric

8
Data Layers
9
SELECT FROM buildings WHERE "NAME" LIKE
Lecture AND AREA gt 1400
Query uses more than one field in where clause
(name and area)
10
Data Type 1
11
Vector Data Model
represents geography via coordinates
(arcs)
(first and last point is the same closed shape)
12
Bertin's 10 Graphic Primitives
LOCATION (x,y)
13
Vector Data Model
(see Heywood et al 2002 pp 46-58)
advanced concepts
14
Path of Least Cost
(quickest route, shortest route)
Topological dataset of Roads
15
(No Transcript)
16
NORWICH - UK
http//www.spatialmetro.org/
17
Data Type 2
18
Raster Data Model
represents geography via grid cells
Digital Camera captures a Raster Image
consisting of many pixels (megapixels)
19
Raster vs Vectorraster is faster but vector is
corrector - Joseph Berry

• Vector data model
• location referenced by x,y coordinates which can
  be linked into lines, and polygons
• attributes referenced through unique ID number to
  tables
• much data comes in this form
• GPS
• On Screen digitization
• LINZ
• best for features with discrete boundaries
• property lines
• political boundaries
• transportation

• Raster data model
• location is referenced by a grid cell in a
  rectangular array (matrix)
• attribute is represented as a single value for
  that cell
• much data comes in this form
• images from remote sensing (LANDSAT, SPOT)
• scanned maps
• elevation data
• best for continuous features
• elevation
• temperature
• soil type
• land use

20
Raster and Vector Data Models
21
Discrete and continuous data
Discrete data
Continuous data
22
(No Transcript)
23
Changing resolution..
24
Map Algebra
25
Data Type 3
26
Attribute Data
Bird Type KiwiDate 20 Feb 2002 Weight 1.2
kg Height 27 cm
Bird Survey
27
Attribute Data
Numerical(known difference between values)
Categorical (name)
Interval no natural zero cant say twice as
much (eg temperature in Celsius) Ratio natural
zero ratios make sense (e.g. twice as
much) (eg income, age, rainfall, temp. in Kelvin)
Nominal no inherent ordering(eg land use types,
city names) Ordinal inherent order(eg road
class stream class city/town/village
high/medium/low)

• may be expressed as integer whole number
  floating point decimal point

• often coded (eg GZ green zone)
• cant do arithmetic

Nominal Le Race establishing identity Ordinal
1st, 2nd, 3rd place establish an
order Interval difference between things gt 1st
place 9am , 2nd place 915am (15 min
later) Ratio 1st place took 2hr, 2nd place 2hr
15mins 50th place took 4 hours twice as long
as the first person to finish
28
(No Transcript)
29
Scale
30
Scale Generalisation

• Large scale maps tend to have more detail on them
  
• Small scale tend to have less local detail on
  them but cover a large geographical area

Large scale
Small scale
31
Examples
32
(No Transcript)
33
(No Transcript)
34
Canterbury Campus 3D
Exaggerated vertical scale
35
The colours show the strength of the signal
reflection where the darker browns are the
strongest reflections (typically flat grassy
areas).
36
Projection Systems
Challenge in making a world map is representing
the Earth (spheroid) as a flat surface
Peters Equal Area Projection
37
Cartograms
Shape distorted by a variable (here population)
http//www.worldmapper.org/
38
Entity Relationship Diagram
Custom Functions in the Database Server (MS SQL)
T-SQL Function
Soil Map
39
(No Transcript)
40
Circle Map of Indian Population in Leicestershire
Same Map with Moved Circles
41
(No Transcript)
42
(No Transcript)
43
More Background Theory
44
(No Transcript)
45
(No Transcript)
46
Colour schemes

• Colours have connotations which can make or
  break a map
• Blue water, cool, positive numbers
• Green vegetation, lowlands, forests
• Yellow/Tan dryness, lack of vegetation
• Brown Landforms, contours
• Red warm, important items, negative numerical
  numbers

47
(No Transcript)
48
Choropleth Class Schemes
49
Spatial Boundaries

• MAUP data aggregation

50
More Examples
51
John Snow 1854
Cholera Soho, London Thought cholera spread by
water not air as thought at the time Mapped
deaths and water pumps
52
A reconstruction of John Snows map of cholera
cases in London and three choropleth maps
produced by different areal aggregations
53
Geographical Analysis Machine - GAM by Openshaw
Draw circle at random location, of random
radius Count points (incidents) in circle If
greater than expected (based on global values)
for circle area leave circle on map Repeat
54

• Christchurch Crash data from the LTNZCrash
  Analysis System (CAS) Accident Investigation
  System (AIS)
• 25 years (1980 2004) of data
• 28,645 geocoded crash points with attributes
• 600 different crash cause codes
• Only Minor Injury, Serious Injury, Fatal are
  recorded in dataset
• Modelled traffic flow for major roads
• Actual Traffic Flow counts from about 800points
  around Christchurch
• Road network (number of lanes, road class,
  surface)

55
(No Transcript)
56
Observed gt Expected
Top 2.5
Monte Carlo Simulation Output
(traffic flow input)
57
Aquiferim for Canterbury, NZ
GIS (setup)
C (.dll)
GDAL (.dll)
GIS (visualisation)
58
Flowpath pattern
Slice (side view through selected path showing
age, nitrates etc)
59
(No Transcript)
60
Digital Elevation Models
Hydrological modelling
61
Example surface types
62
(No Transcript)
63
3D Topographic Map - Switzerland
64
3D Topographic Map Milford Sound
65
http//www.mapmyrun.com
66
LiDAR
67
(No Transcript)
68
LiDAR 3D points of some trees
Surface model of trees as 2.5D representation
69
View created from LIDAR data for Edinburgh City,
Scotland
70
(No Transcript)
71
Cumulative Visibility Map of Christchurch
This shows how many times a part of the city
centre could be seen from all other pedestrian
accessible areasin the city. The more red areas
are the most visible parts.
72
(No Transcript)
73
Mapping Visibility Metric FLOW - Direction and
Magnitude of Change of a Metric
74
Map Display Considerations
In Car Navigation Systems
75
TomTom Rider
Used with Gloves
76
http//www.apple.com/iphone/
77
(No Transcript)
78
Thermal Imagery on Christchurch City DSM
79
(No Transcript)
80
http//strangemaps.wordpress.com/
81
Thanks for listening! Any questions?