Undine LIEBERWIRTH
(Berlin, Germany)
How accurate must a model be? Counter-question: For what use is the model intended? In this paper, I would like to introduce one of the problems of uncertainty which arose during 3D GIS model building.
“The accuracy of reconstructed models of archaeological sites and stratigraphy is only as good as the data input.” (Green, D.A. 2003, Stratigraphic Visualisation for archaeological investigation. Thesis at Brunel University London, 160). In this example, drawings took on a pseudo-precision by converting them into digital data, models took on a pseudo-accuracy by including intuitive knowledge. Hence, models are not just calculated they are also a product of cognitive processes. How long is such a model reliable? Is it possible to express its accuracy by for example a “reliability index” (Hermon, S. et al. 2006, Deconstructing the VR – Data Tranparency, Quantified Uncertainty and Reliability of 3D Models. In: Ioannides, M. et al. 2006, The 7th International Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST)?
However, disregarding precision, a model will stay an image/copy of the real world. Hence, imperfection will always accompany model building as well as the fact that archaeology has to deal with a great amount of uncertain, estimated data. Should we therefore dismiss the whole process of model building?
At present no final solution to this case study has been found. Hence, colleagues are invited for inspiring proposals and discussions.
Keywords: 3D GIS, hypotheses, model accuracy