Data Time
Resume:
Development of a geovisual analytics environment for
investigating archaeological events based upon the Space-time
Cube.
Extended abstract for GIScience Geovisual Analytics Workshop
Otto Huisman, Irvin Feliciano Santiago, Menno-Jan Kraak, Bas Retsios
International Institute for Geo-Information Science and Earth Observation (ITC),
Hengelosestraat 99, P.O. Box 6, 7500 AA Enschede, The Netherlands.
*******@***.**
Archaeology is an application domain that deals with data having inherent temporal and
spatial components. The purpose of this paper is to present research results on the
development of functions for Archaeological investigation within a prototype Geovisual
Analytics environment based upon the Space-time Cube (STC). Generally speaking, the
discipline of Archaeology focuses on recovering and analyzing remains and artefacts of
past societies in order to reconstruct architectural structures, social organization, and gain
insight into human behavior of past societies through consideration of the immediate
geographical context and relationships to other finds or artefacts recovered in the area.
Over the past two decades, archaeologists have been developing and improving methods
for collection, analysis, visualization and modelling of site information. GIS has been
deployed in a range of decision-support applications (van Leusen, 1995; White, 2002), as
well as analytical and modelling applications (Kohler et. al. 2000; Dean, et. al. 2000) in
archaeological research. Specifically, applications such as viewshed analysis (Ruggles,
et.al. 1993; Lobera et. al. 2003), spatial analytical applications for inverstigating or
simulating the potential extent of a habitat/environment (Madry and Rakos, 1996;
Williams, 2004), as well as a range of micro-level analyses documenting individual sites.
This paper argues that, despite the increasing capabilities of GIScience tools in recent
years, archaeology is an application domain where these capabilities have not been fully
utilized, and there is significant scope for the development of new or extended tools and
methods. The paper explores the potential of the STC environment to contribute to
insights on archaeological phenomena, and in so doing, reports on extended
geovisualisation functionality to support archaeological investigation. These are
implemented and tested on a small subset of a significant archaeological database of
approximately 900 archaeological sites from Puerto Rico.
The field of Geovisualization concerns itself with providing the necessary tools and the
adaptability to handle complex geospatial datasets. It is important for geovisualizers to
reassess the ways to explore, understand and satisfy the user s need in order to develop
new applications that provide more meaningful results (Dykes et al, 2005). The Space-
time Cube (STC) is a GIS-based implementation of H gerstrand s (1970) original Space-
time Aquarium for the visualization and analysis of space-time data. Two horizontal axes
are used to represent the x-y geographic coordinates and the vertical axis the time (t)
dimension. Together with attribute information, it is possible to explicitly represent the
three main components of spatio-temporal data, namely when, where, and what the key
components of Pequet s Triad framework (Peuquet, 1994). In order to address questions
at different scales, it is possible to employ different time granularities along the temporal
(z) axis. Within the STC, the location of a find can be represented by a vertical line that
has a color/thickness at the relevant time period. To indicate the findings of a particular
type of artifact (i.e., bones, shells, pottery, etc.) the assignment of a color classification
schema can provide the necessarily visual output to identify patterns or clustered artifact
locations, and help the archaeologist to differentiate, associate and identify relationships
in a more meaningful manner (Kraak and Koussoulakou, 2004). Alternatively, qualitative
icons could be used to distinguish these.
Figure 1: The extended Space-time Cube (STC) environment, drawing on relationships
that exist between data from various sites in the database. These are connected according
to selected criteria and can be visualized using multiple linked views.
A key issue in archaeological research concerns the relationships between sites and
artefacts discovered at various sites in order to understand the interaction between
cultures. To extend the existing capabilities of the STC, this research employs concepts
from graph theory to examine the relationships between objects and events in space and
time by abstracting them to a mathematical structure, i.e., a graph. Mathematically, a
graph is defined as a pair of G = (V, E) of sets satisfying E [V]2; thus, the elements of
E are 2-elements subsets of V. Therefore, elements of V represent vertices (nodes or
points) of a graph G, and elements of E its edges (Diestel, 2000). In the field of
visualization such graphs (or layouts) are referred to as networks composed of nodes and
edges, in which each node in the graph can represent a particular entity such as classes,
variables, or items of interest; and the edges (lines) between them represent various
relationship characteristics (Ware, 2004).
The criteria used in this research for determining the tools and options to be added was
based on an extensive literature review of graph theory, search levels and operational task
topology (Andrienko, 2003), an evaluation of similar geovisualization environments, and
an assessment of the current status of the STC software under development at ITC (Kraak
and Huisman, in press). The realization of the prototype involved the development of
automated functions for identifying/establishing relationships between attribute data in
the database (data-mining), and facilitating direct import into the STC. One advantage of
the extended functionality offered by the prototype is that relationships between attributes
(database tuples) can be explicitly visualized and manipulated with various grouping and
sorting functions, providing the user with a range of options for knowledge discovery.
Figure 1 above illustrates the stages and elements involved in this procedure.
A range of visual variables can be employed for interactive geovisualisation to help
answer the question such as: which cultures were present in the area, and how are they
related? To demonstrate, Figure 2 above illustrates the notions of grouping nodes
(archaeological sites with a color classification representing the culture variable) into
temporal intervals. This relates specific known cultures associated with finds at an
archaeological site with either (Figure 2A) the first known associated period of a
particular artifact found at a site, or the last known period for the same, and the result
illustrates the number of sites participating in the relationships for the time periods. These
can be combined into a single cube to give an insight into the periods with which specific
finds are associated, or alternatively, viewed as stations (Figure 2B) to illustrate
potential associations between cultures (in the form of overlap).
The remainder of the paper discusses the issues above in greater depth, and demonstrates
various geovisual analytical approaches to reasoning with the space-time cube. A range
of artefacts can be derived from this information. For instance, an animated series of
time-slices could be used to investigate the dominance of one culture over another
(possibly revealed as the emergence of one culture and the displacement and eventual
disappearance of another) as revealed by the discovery of artefacts related to those
respective cultures. Multiple views can be used to support this investigation, aided by
tools such as interactive filtering and sorting functions which can be applied to clarify
patterns and relationships hidden in the data. The paper concludes with a critical
discussion of the effectiveness and potential of the prototype and its functionality as well
as reliability of observations and issues relating to data quality.
A
B
Cultures
Start Time
Figure 2: Archaeological time exploration. Top Left: Grouping by Start Time, Top Right:
Grouping by End Time, crossing time lines between groups. Bottom: Station view
illustrating the duration associated with specific cultures at specific sites.
References
Andrienko, N., Andrienko, G., Gatalsky, P., 2003. Exploratory spatio-temporal
visualization: an analytical review. Journal of Visual Languages & Computing,
Vol.14, No. 6, Dec., 2003, pp. 503-541.
Dean, J., G. Gumerman, J. Epstein, R. Axtell, A. Swedlund, M. Parker, S. McCarroll.
2000. Understanding Anasazi culture change through agent-based modeling. In
Agent-ased Modeling of Long-term Human Adaptive Systems, T. Kohler and G.
Gumerman (eds) Oxford University Press. pp. 179-205.
Diestel, R., 2000. Graph Theory. Third Edition. Springer-Verlag, Heilderberg Graduate
Texts in Mathematics.
Dykes, J. A., MacEachren, A. M. and Kraak, M.-J., 2005. Exploring Geovisualization,
Oxford: Elsevier.
H gerstrand, T. 1970. What about people in regional science? Papers of the Regional
Science Association 24: 7-21.
Kohler, T. A., J. Kresl, C. V. West, E. Carr, and R. H. Wilshusen. 2000. Be there then: A
modeling approach to settlement determinants and spatial efficiency among late
ancestral pueblo populations of the Mesa Verde region, U.S. Southwest. Pages
145-178 in T. A. Kohler and G. J. Gumerman, eds. Dynamics in Human and
Primate Societies. Oxford Univeristy Press, New York and Oxford.
Kraak, M-J. and O. Huisman, forthcoming. Beyond Exploratory Visualisation of Space-
Time Paths. In Miller, H. and J. Han (editors), Geographic Data Mining and
Knowledge Discovery, 2nd Edition. Taylor and Francis, London.
Kraak, M-J. and Koussoulakou, A., 2004. Visualization Environment for the Space -
Time Cube. In: SDH 2004: Proceedings of the 11th international symposium on
spatial data handling: advances in spatial data handling II. : 23-25 August 2004,
University of Leicester. / ed. by P.F. Fisher. - Berlin etc.: Springer, 2004. pp. 189-
200.
van Leusen, P. 1995 GIS and archaeological resource management : a European
agenda In Lock G. Stancic, Z. (eds) (1995): Archaeology and Geographical
Infromation Systems. A European perspective. London, Taylor and Francis.
Llobera, M. 2003. Extending GIS-based visual analysis: the concept of visualscapes,
International Journal of Geographical Information Science Vol.17 No.1, 2003,
pp.25- 48.
Madry, S. L. H., and L. Rakos 1996. Line-of-Sight and Cost-Surface Techniques for
Regional Research in the Arroux River Valley. In New Methods, Old Problems:
Geographic Information Systems in Modern Archaeological Research, edited by
H. D. G. Maschner, pp. 104 126. Southern Illinois University at Carbondale,
Occasional Paper No. 23. Centre for Archaeological Investigation, Carbondale,
Ill.
Peuquet, D.J., 1994. It s About Time: A Conceptual Framework for the Representation of
Temporal Dynamics in Geographic Information Systems. Annals of the
Association of American Geographers 84(3):441-461.
Ruggles C L N, Medyckyj-Scott D J, Gruffydd A, 1993, Multiple viewshed analysis
using gis and its archaeological application: a case study in northern Mull ', in
Computing the Past Eds J Andresen, T Madsen, I Scollar (Aarhus University
Press, Aarhus) pp. 125-132.
Ware, C., 2004. Information Visualization: Perception for Design, 2nd ed. San Francisco,
CA: Morgan Kaufmann.
White, A. M. 2002. Archaeological Predictive Modeling of Site Location Through Time.
Msc Thesis, University of Calgary.
Williams, M., 2004, Archaeology and GIS: Prehistoric Habitat Reconstruction ESRI User
Conference Papers. Access on 25/4/2008:
http://gis.esri.com/library/userconf/proc04/docs/pap1835.pdf
Contact this candidate