Abstract:

We present in this paper the results of the application of several visual methods on a group of locations, dated between VI and I centuries BC, of the ager tarraconensis (Tarragona, Spain) a hinterland of the roman colony of Tarraco. The diff iculty in interpreting the diverse results in a combined way has been resolved by means of the use of statistical methods as Principal Components Analysis (PCA) and K-means clustering analysis. These methods have allowed us to carry out site classif ications in function of the landscape visual structure that contains them and of the visual relationships that could be given among them. Also we have created and analized some visual networks in order to study the visual cohesion among the settlements.