This paper is the product of three years of PhD research investigating the spatial patterning of Prehistoric Aboriginal archaeological sites in and around the Melbourne (Victoria). In conjunction with an industry partner (Aboriginal Affairs Victoria), the research was conducted as a component of a PhD Program, based in the Archaeology Department of La Trobe University.
The main focus of my research has been to assess the utility of vast amounts of existing archaeological and CRM data, and to determine the 'useability' of this data in a predictive modelling exercise. As predictive modelling using GIS becomes more and more commonplace, it is vital that these models are scrutinized adequately, and that the models are indeed based upon sound and rigouress scientific methods.
The author found that the majority of CRM reports that have been completed in the study area over the past 25 years are more or less unusable in a formal probability-based predictive model. This leaves us with a number of dilemmas, least of which is how to create useable data in the future.
Keywords: archaeology, predictive modelling, GIS, CRM, Aboriginal