Skip to main navigationSkip to main content
The University of Southampton
ArchaeologyPart of Humanities

Research project: 'Inherent visibility' indices for archaeological landscapes - Dormant - Dormant

Currently Active: 

In recent years there has been an upsurge in landscape archaeology studies in which human landscape (mostly visual) perception is a central component (eg Barrett 1993, Tilley 1994). While these studies have brought to light the complexity of the process of human-landscape interaction they have not precipitated new methodological developments to support them.

Most of the interpretations included in these studies are based on skilful but often limited experience of the investigator during fieldwork. He or she simply cannot move through all possible paths or easily compare views from different locations. Many of the observations are aimed at establishing the significance of monument settings particularly in relation to their visual properties (ie whether they dominate the landscape, can be seen from many locations). To answer some of these questions, and to obtain useful insight into other ones, requires retrieving the visual properties inherent in the landscape independently of where the monuments were located. In spite of advances in computer technology to determine what might be in sight, ie the calculation of a viewshed, still remains a computationally expensive process. The intensity of computation is most noticeable when large number of viewsheds need to be repeatedly calculated (in our case we consider the calculation from all possible locations, ie cells, in a DEM).

Specific aim
The aim of this project is to produce the total viewshed for a given DEM. That is, to generate a raster image in which each cell contains the number of other non-obstructed cells, ie number of cells that can be connected by an uninterrupted line-of-sight.

Given a DEM (a raster data structure) there are several strategies that may be adopted in order to optimise the calculation of a total viewshed. Because of time constraints, the project will concentrate on a single one: the possibility of distributing the load, ie viewpoints, among a pool of computers. To this end, we will be using the CONDOR HIGH-THROUGHPUT software, a low-cost cluster computing middleware. CONDOR is a specialised workload management system for compute-intensive jobs freely available from the Computer Science department at the University of Wisconsin, USA. Like other full-featured batch systems, it provides a job queuing mechanism, scheduling policy, priority scheme, resource monitoring, and resource management. Users submit their serial or parallel jobs to CONDOR, CONDOR places them into a queue, chooses when and where to run the jobs based upon a policy, carefully monitors their progress, and ultimately informs the user upon completion. One of the main benefits of CONDOR is that it will work with a non-dedicated pool of computers. The pool may be part of a local area network (eg within a department). This makes CONDOR very attractive for departments that do not have the benefit of a dedicated set of computing resources, frequently the case among arts and humanities departments.

Related research groups

Archaeological Computing
Centre for Applied Archaeological Analyses
Share this research project Share this on Facebook Share this on Twitter Share this on Weibo
Privacy Settings