Geography Reference
In-Depth Information
Figure 10.7
Stereo view of earthquakes below Big Island, Hawaii, using luminescence/lightness
variations to enhance the depth effect (source: author; data from CNSS)
are rotated. In experiments using a head-mounted device to induce 3D perception, Ware
and Franck (1996) found not only that motion parallax alone is better than stereo alone for
tasks involving full 3D awareness, but also that motion parallax in combination with stereo
provides the best depth cues for the analyst. Stereo should not therefore be considered a
'must have' facility but, as Ware and Franck (1996) suggest, individual data visualizations
should use particular combinations of 3D-inducing effects. Few hard and fast rules are
available, because the relevant combinations need to be task-specific, and further research
is needed to evaluate which 3D effects are best suited to particular data visualization tasks.
10.3.5
z
-Axis contention
With the increasing availability of spatial tracking data, several attempts have been made to
automate the production of the space-time cube (Oculus, 2006; Kraak, 2007), first intro-
duced in the 1960s (Hagerstrand, 1970). In the resulting visualizations, the third dimension
is used to display time, with the
x
- and
y
-axes retaining their traditional role of displaying
ground surface features. Unfortunately, when spatial data are visualized in a space-time
cube, the
z
-axis is not really available as a spare dimension to be used exclusively to show
time, because the movement of people and objects also takes place within the vertical spatial
dimension. Although it is tempting to 'overload' the
z
-axis by using it to display both the
Search WWH ::
Custom Search