Geography Reference
In-Depth Information
6.3
History, Related Work and New Directions
The history of spatio-temporal network modeling arguably begin in 1988 with the
landmark paper 'Framework for Temporal Geographic Information' (Langran and
Chrisman 1988 ). This paper illustrated the challenges associated with visualizing
temporal information and introduced the idea of world time and database time. More
importantly, it introduced a system to visualize spatio-temporal changes of an image
(e.g., a map). Each change would be layered on top of an old change, so that the
layers could be tracked through to detect where changes occurred (Shekhar and
Xiong 2007 ). This work established the field of temporal geography and laid the
foundation for new research in temporal and spatial representations.
Though much work has been done in conceptualization and modeling
(Christakos et al. 2001 ;Frank 2003 ), there is a fundamental struggle between spatial
and temporal databases, the former representing things with graphical snapshots
and the latter doing so with time-stamps. It is difficult to merge the two in an
efficient way and current research is moving slowly toward a unified theory for
these representations (Langran 1992 ; Egenhofer and Golledge 1998 ; O'Sullivan
2005 ; Peuquet and Duan 1995 ; Goodchild et al. 2007 ; Frank et al. 1999 ).
Conceptually, processes and events can be used to help create unified spatio-
temporal representations (Galton and Worboys 2005 ). This may eventually lead
to logical models useful for implementation in spatio-temporal databases. In the
meantime, these ideas led to an important new area of study in temporal geography
known as geographic dynamics. Geographic dynamics can be described as a series
of processes and events that allow for descriptive capabilities to illustrate spatio-
temporal changes. This idea of geographic dynamics is not new. It can be thought
of in the same light as describing the dynamics of weather; only here we are talking
about geographic changes such as land use change over time. These dynamics have
been used and represented in various ways, starting with animated maps and moving
towards more interactive models (DiBiase et al. 1992 ; Monmonier 1990 ; Harrower
2004 ). Current techniques combine visualization with the modeling itself. For
example, cellular automata models are made up of a large number of cells that must
change based upon a simple set of rules (Batty 2005 ). These rules allow for easy
creation of models which can then provide interesting insights into the real world,
and demonstrates the changes in action as the cells move and change. Visualization
has proven historically to be a powerful tool for understanding complex spatio-
temporal data (Kwan 2004 ; Torrens 2006 ; McIntosh and Yuan 2005 ).
Extensive work has also been done analyzing the human use of time in relation to
geography, the primary components being space-time paths and space-time prisms
(Burns 1979 ; Hägerstrand 1970 ; Janelle 2004 ; Lenntorp 1977 ; Miller 2005 ;Pred
1977 ; Timmermans et al. 2002 ). The focus here is on individual spatio-temporal
paths, allowing for interesting decisions and analyses to be made. Both approaches
represent a person's position over time by plotting geographic space on the X and Z
axis, and time on the Y axis. The path based graph shows where a person traveled
over time, while a prism shows their potential travels based upon maximum travel
Search WWH ::




Custom Search