Estimating the “memory of landscape” to predict changes in archaeological settlement patterns

Philip Verhagen (1) , Laure Nuninger (2, 3) , Frédérique Bertoncello (4) , Angelo Castrorao Barba (5)

1 CLUE - Cultural Landscape and Urban Environment
2 LCE - Laboratoire Chrono-environnement (UMR 6249)
3 MSHE - Maison des Sciences de l'Homme et de l'Environnement Claude Nicolas Ledoux (UAR 3124)
4 CEPAM - Culture et Environnements, Préhistoire, Antiquité, Moyen-Age
5 UNISI - Università degli Studi di Siena = University of Siena

In this paper, we present a method to calculate a “land use heritage map” based on the concept of “memory of landscape”. Such a map can be seen as one variable among others influencing site location preference, and can be used as input for predictive models. The computed values equate to an index of long-term land use intensity. We will first discuss the method used for creating the land use heritage map, for which kernel density estimates are used. We will then present the use of these land use heritage maps for site location analysis in two study areas in SE France. Earlier analyses showed that the influence of the natural environment on settlement location choice in the Roman period is limited. In contrast, land use heritage seems to have a stronger influence on the placement of new settlements. We will discuss the implications for predictive modelling of settlement patterns.

Mots clés: settlement pattern analysis; socio-cultural variables; memory of landscape; heritage map; Predictive modelling

Domaines: Archéologie et Préhistoire

Philip Verhagen, Laure Nuninger, Frédérique Bertoncello, Angelo Castrorao Barba. Estimating the “memory of landscape” to predict changes in archaeological settlement patterns. 1, Archaeopress (Oxford), pp.623-636, 2016, KEEP THE REVOLUTION GOING, Proceeding of the 43rd Computer Applications and Quantitative Methods in Archaeology Conference (CAA 2015 SIENA). ⟨hal-01252715⟩