Conclusions from Chapters 1 to 3: Where is Research Needed?

This is a short but nevertheless important part. I shall recapitulate and integrate the findings in the previous chapters to point out in which directions the research that follows should be targeted.

The following conclusions can be drawn:

1. A system similar to the (spatial) coordinate transformation between different coordinate systems needs to be developed for the temporal domain (section 2.2.1).
2. No concepts for the representation of moving objects are available in (commercial) GIS.
3. Major efforts in GIS concerning temporal data are focussed on representational issues. The analysis part has not received any attention yet.
4. There are no effective ways to visualize, handle and analyze moving objects within GIS.
5. Apart from cartographic symbology and video-like animations no true analytical methodology exists for spatio-temporal data with the exception of the cluster search engines developed by Openshaw and Perrée (1996).
6. Temporal aspects are reduced to a minimum in today's analysis (at best to time slices).
7. There is a major disparity between wildlife researcher needs, GIS, and the analytical methods available.
8. Time slicing of wildlife data has mostly been done using some artificial human (or civilization) centered calendars (months, weeks, hours etc.). New methods need to be developed to identify and define biologically meaningful and homogeneous phases within data.
9. Up to now time has only been considered as a linear component. Other temporal aspects such as cyclic phenomena or rhythms have not been recognized as important aspects in temporal GIS research.
10. No or only rudimentary concepts are available on how to handle, integrate and analyze the surrounding environment of an observation in a spatially explicit way, neither in biology nor in GIScience.
11. The temporal sampling scheme should be changed from 'avoidance of temporal autocorrelation' to a scheme which describes movements as accurately as possible (section 2.1).
12. Since an analysis is much more complex when conducted in a spatially explicit manner, the analysis might need to be performed step by step for one aspect after another. As it is well known from cartography the accumulation of large amounts of information can produce maps that are hiding more than they are revealing.

With the exception of the first two items mentioned above, the next three chapters are going to provide new approaches to these important issues, equally important for both Biology and GIScience. Thus the main question is the following:

How can we include temporal aspects in the analysis of the spatial behavior of an animal within a GIS environment?