Portable ED-XRF as a tool for optimizing sampling strategy: The case study of a Hellenistic amphora assemblage from Paphos (Cyprus)

Abstract

Provenance studies of archaeological ceramics based on their elemental composition illuminate the production and distribution of pottery vessels and, in the case of transport containers, of the commodities that they contained. A basic assumption is that the elemental composition of ceramics from a specific workshop or production area can be distinguished from other production groups, mainly because of the use of geochemically different clays, either singly or in combination. In some cases, however, the compositional differences between production groups are quite small. Thus, laboratory methods with high performance, in terms of precision and accuracy, such as neutron activation analysis (NAA) or wavelength dispersive X-ray fluorescence (WDXRF), are often preferred for analyzing archaeological ceramics, especially for effective comparison with reference groups and published data from other studies. Handheld portable energy dispersive XRF systems (pXRF), although increasingly used during recent years, offer lesser analytical performance, which may obscure compositional differences and currently do not offer the same potential for comparison to known reference groups. However, due to their potential for fast and non-invasive measurements, considerably larger numbers of samples can be analysed by pXRF, offering an array of advantages. We argue that pXRF offers the opportunity for an initial analytical survey of a large ceramic assemblage as the basis for efficient sample selection for laboratory analysis, covering a large number of samples and avoiding for the generation of redundant measurements. We present the application of such a stepped analytical approach to a well-studied assemblage of amphorae of the Hellenistic period at Nea Paphos, Cyprus. The analysis of 287 amphora fragments by pXRF, the grouping of that data to select samples for further analysis of 97 individuals by NAA, and the comparison of the grouping of data from both chemical techniques is presented. This leads not only to archaeological insights on the production and circulation of amphorae, but tests an innovative methodology that offers the chance to maximize and extend the application of geochemical techniques of high accuracy and precision on an assemblage-wide scale.