Skip to main navigation Skip to search Skip to main content

"Input-Data Biases: Meeting the Challenges of Building Interdisciplinary Predictive Models in a Multinational Maritime Context"- ARCHON – GIS days at the VU

Activity: Talk and presentation › Professional or public presentation › Professional

Description

With the implementation of the Malta Convention (signed in 1992), heritage scholars and managers have increasingly turned to predictive models using GIS techniques to predict the location and the state of preservation of archaeological sites in order to optimize both their protection and spatial planning policies.
Predictive models start to be utilized also in the field of underwater archaeology; nonetheless they are still underexploited, particularly in the Mediterranean basin, which highlights specific problems of data quality and an oversimplification of factors determining the locations of underwater
archaeological sites.
Aim of my ongoing PhD research is to build both a suitable theory and a new GIS-based methodology for the prediction of ancient routes and potential shipwreck locations in the Mediterranean Basin. Particularly, and in contrast with the current mainly environmental deterministic models, it aims to include as site predictors cultural and cognitive variables. These have -despite their undoubted historical importance- thus far been ignored or barely considered due to the difficulty of implementing qualitative insights into computable (GIS) actions.
The present lecture will debate the most important shortcomings and challenges of current archaeological predictive models in the Mediterranean context. Particularly, by looking at the distribution of the already known shipwrecks, the quality and quantity of the input data will be examined in order to identify the main biases in the available dataset and discuss the choice of the most suitable methodological approach.
Period15-Sept-2017
Held atVrije Universiteit Amsterdam, Netherlands