University of Southampton OCS (beta), CAA 2012

Font Size: 
Open source geostatistics for archaeology: the fauna of Fumane cave
Enzo Cocca, Enrico Guastaldi, Marco Peresani, Matteo Romandini

Last modified: 2011-12-18


Analysis of spatial distribution of archaeological records integrated by geostatistical technique plays a primary role in prehistoric archaeology for a deeper knowledge of layers of human frequenting. In this work we present an approach by integrating statistics, database and GIS open source tools in an one-solution instrument based on QGIS desktop environment. Since archaeology is an intrinsically spatial discipline, we utilized a spatial quantitative approach like the multivariate geostatistics, which is typically dedicated to mining and environmental analysis, in order to the study of an important archaeological site in the northern Italy (Fumane cave, Verona, Italy), by managing the quantitative fieldwork information as regionalized variables. Therefore, we evaluating interesting spatial correlations by imposing linear coregionalization models on experimental semivariograms and cross-semivariograms, usual geostatistical tools. Besides, we make also spatio-temporal correlations among different layers of human frequenting. Moreover, we carried on such analysis by creating an tool that integrates several powerful R libraries such as Gstat for spatial distribution analysis of data implemented in  DBMS  PostgresSql with Postgis extension. Analyses performed on Fumane cave's fauna for several chronologic layers show interesting ways for spatially handling the cave regarding to the categories of studied fauna, opening new perspectives to quantitative archaeological studies.


Open source; Geostatistic; Spatial analysis; GIS