Geography Reference
In-Depth Information
as from LIDAR data, can be mapped to colour or height. Two-dimensional maps are often
used as the main reference visualization, with other statistical information being layered on
top. Furthermore, different types of data, such as agriculture, transportation, boundaries or
population density, can be represented as different layers of the map. Other forms of maps
include choropleths, where the land areas are shaded with values proportional to statistical
measures of that land, while cartograms distort the map dependent on another statistical
measure, e.g. the time it takes to travel from one point to another.
Networks
describe relational and associational information, e.g. that a connects to b and
then c. Networks include trees, hierarchies, routing diagrams and graph visualizations. A
well-known hand-crafted network is that of Harry Beck's London underground map; it
is really a network presentation because it represents relational information of the under-
ground rail lines. In fact, aspects of Beck's map have inspired network layout algorithms
(Stott and Rodgers, 2004). Network visualizations have been used in geovisualization to
represent various types of associated data and have been realized through different layout
strategies. Rodgers (2004) provides a good overview of graph techniques for geovisualiza-
tion. Depictions range from node-edge graphs and treemaps to trails.
Charts
display statistical or mathematical information. In this category we include line
graphs, histograms, circular histograms, pie charts, surface plots, scatter plots and parallel
coordinate plots (Edsall, 2003). Each of these visualizations addresses a specific need; line
graphs and histograms visualize continuous data on a two-dimensional plot, surface plots
display continuous data in three dimensions, bar charts and pie charts display quantitative
information and parallel coordinates display multidimensional data, while scatter plots
allow users to see clusters.
Tabular and matrix
layouts are popular for displaying statistical quantities and numerical
information contained within geographical databases. They are familiar forms and as such
can be extremely useful. Spreadsheets enable large amounts of numbers to be shown and
manipulated; with table lens views (which utilizes distortion techniques) even more infor-
mation can be displayed (Rao and Card, 1994), while reorderable matrix techniques map
the values into the size attributes of symbols to allow trends and similarities to be viewed
(Siirtola and Makinen, 2005).
Symbols
may be used in two ways. Either they are used to identify individual aspects of
the information, such that objects or buildings can be located on a map, or they are used to
notify trends. In the latter case, the trend is understood through perceiving the texture that
is formed from viewing multiple symbols close together. For instance, Nocke
et al.
(2005)
plotted multiple icons onto a map to visualize land usage.
Diagrams
realize some process, concept or phenomenon; most are hand-crafted to display
a particular phenomenon or result. They are often used in teaching to clearly explain a
process. There are many examples in the literature, demonstrating phenomena like how
volcanoes function, the migration of various animals or the campaigns of Napoleon's army.
Pictures
are often associated with geographical datasets. Aerial photographs or site
photographs can be easily associated with ground or land-use data. With the ubiquity
of GPS it is easy to take a picture with position information. Applications like Google
Maps (maps.google.com; maps.google.co.uk, accessed May 2007) montage multiple aerial
photographs to make a detailed view of the Earth, while projects such as 'MyLifeBits'
from Microsoft Research utilize cameras that automatically take pictures at regular in-
stances throughout the day and correlate them with positional information from GPS (Bell,
Gemmell and Lueder, 2007).
Search WWH ::
Custom Search