Space-time cube analysis of animal behaviour

Organisator Laboratory of Geo-information Science and Remote Sensing

do 11 juli 2013 08:30 tot 09:00

Locatie Gaia, building number 101
Droevendaalsesteeg 3
6708 PB Wageningen
+31 317 48 16 00
Zaal/kamer 1

by Maarten Baas (The Netherlands)


In 1970 Hägerstrand developed the three-dimensional space-time cube. With this method spatiotemporal patterns can be visualized and analysed. Due to this ability the space-time cube is mentioned as potential method for exploratory analyses of animal time-tracking data in ecology. However, this has never been tested. Therefore a space-time cube was created visualizing a GPS-dataset of African buffalo (Syncerus caffer). No similar space-time cube has been found in literature. The possibility of adding additional attributes in the form of Orellana’s (2012) interactions was investigated. Eventually, one additional attribute was added, namely “distance to water source”.

A usability test was created in which the visualization operators described by Koua et al. (2006) were tested. Next to the visualization operators the ease-of-use of the visualization was tested among resource ecologists.

Preferably, a little practice time is given before the use of the space-time cube, for after a while its use becomes easier. For some visualization operators statistical alternatives were deemed favourable over the space-time cube. The space-time cube’s unique ability to visualize space-time paths resulted in an advantage. Spatiotemporal patterns are easily recognized within a space-time cube. Half of the respondents regarded the space-time cube as a useful tool for explorative data analyses. However, respondents had difficulties to keep their orientation, but this problem might be overcome with dynamic axis, labels and basemaps. There is no easy method to create a space-time cube yet and with the tools currently available it is a struggle to create one. As a result the space-time cube probably remains a nice concept for analysing animal tracking data instead of a used method for explorative data analyses.