Digital surface reconstruction of biominerals under a quantitative perspective: implications in the taxonomy and preservation of archaeomalacological material

Keywords: archaeomalacology; three-dimensional reconstruction; surface morphology; taxonomy; preservation

Biomineralisation is the natural process by which living organisms form minerals. During biomineralisation, accumulations of organic and inorganic compounds with inhomogeneous distributions of diverse shapes and sizes are formed (Mann, 2001; Weiner, 2010). One important category of biominerals is the exoskeleton of molluscs (terrestrial, marine και freshwater invertebrate animals). The phylum Mollusca is able to live in remarkably diversified habitats over time and is divided into seven classes, four of which are of relevance to the field of archaeology being of central interest in human economies over the millennia (Bar-Yosef Mayer, 2005): Gastropoda, Bivalvia, Scaphopoda and Cephalopoda (Morton and Yonge, 1964; Ponder and Lindberg, 2008). Molluscs appear in a variety of complex three-dimensional architectures (Addadi et al., 2006; Thompson, 1917) and their skeleton composition is usually aragonite and sometimes calcite (Lowenstam and Weiner, 1989; Stenzel, 1963), alongside traces of organic matter (Addadi et al., 1991; Degens et al., 1967). The use of molluscs is reported by several ancient writers including Aristotle ca. 384-322 BC, (Historia Animalium, De Generatione Animalium, De Motu Animalium, De Incessu Animalium, De Partibus Animalium) (Voultsiadou and Vafidis, 2007) and their presence is often reported in a range of iconographic sources (e.g. pottery, mosaics, painting) from prehistory up to the present (Cattaneo-Vietti et al., 2016; Pernet et al., 2019).

The study of molluscs includes a wide array of applications in the natural sciences and archaeology, offering useful information on aspects of taphonomy and site formation (Law and Thew 2017; Robins and Stock, 1991; Walker et al., 1999), paleoeconomy and paleodiet (Bailey, 1975; 1981; Veropoulidou, 2011), on gathering, preparation, consumption, disposal and recycling practices (Bar-Yosef Mayer, 2005; Erlandson, 2001; Krahtopoulou and Veropoulidou, 2017; Powell, 1996, Waselkov, 1987), on the use of natural dyes (Koren, 2005; Ruscillo, 2005), and the creation of ornaments, adornements, tools, musical instruments and other artefacts for rituals and everyday life (Halstead, 1993; Miller, 1997; Neto et al., 2012; Veropoulidou, 2011, Reese, 1983). Furthermore, several molluscan species have been recorded for their therapeutic properties and uses since ancient times (Benkendorff 2010; Gopal et al., 2008; Voultsiadou, 2010). Since molluscs are sensitive to climatic changes they may also be used as indicators for paleoecology and seasonality studies (Bailey et al. 1983; Coutts, 1970; Schöne and Gillikin, 2013). Archaeomalacology is the discipline of environmental archaeology that principally focuses on the study of the molluscan exoskeletons that have been recovered either complete or fragmented from archaeological sites (Bar-Yosef Mayer, 2005; Karali, 1979; Milner et al., 2007). In archaeomalacology, the parameters of form, dimension, texture, colour and composition of the exoskeleton are among others essential for the study of molluscs in all their multiple facets (i.e. taxonomy, distribution, morphometry, quantification, characterisation, and preservation).

Each exoskeleton has characteristic morphological and anatomical features that are used diagnostically for the recognition of mollusc species (Ponder and Lindberg, 2008). The preservation of their exoskeleton in relation to the burial environment is critical for their reliable archaeological interpretation, since they are frequently found affected by the depositional conditions, displaying alterations of their morphological characteristics. There are a number of biogeochemical conditions over time along with bioturbation, trampling, burning and sampling that may have an impact on the (micro)structure, composition and behaviour of the archaeomalacological material, causing anything from the weathering of structures to partial dissolution and final degradation (Bar-Yosef Mayer, 2005; Goldberg and Macphail, 2006; Robins and Stock, 1991; Schiffer, 1987; Weiner, 2010). The three-dimensional (3D) representation of the surface of malacological material may contribute towards a deeper understanding of its surface morphology and potential weathering, and leads to conclusions related to the natural and cultural pre- and post-depositional processes that affect the archaeological evidence. It may also contribute to the design of mechanical models and three-dimensional digital documentation of ethnographical and archaeological collections of molluscs. The study of modern molluscan specimens may provide an important framework for evaluating the fossil assemblages. Along these lines, malacological material requires delicate handling and its 3D representation may prove useful for the creation or the improvement of comparative reference collections, constituting a primary tool for the study of archaeomalacological material.

Three-dimensional digital reconstruction is an established practice in the fields of geomorphology and topographic mapping, while its use in the (archaeo)malacological research is quite limited (Golding and Jones, 2007; Hawe et al., 2013; Pedrouzo et al., 2019; Troisi et al., 2015). In the present study, we propose the non-destructive imaging and 3D representation of the surface of molluscan exoskeletons using digital photogrammetry principles. The main obstacles for using image analysis techniques to reconstruct the digital surface model of small sized ecofacts, is the sensitivity of the image geometry originating from common cameras and the absence of a reliable coordinate system to provide a robust scale reference for the studied object. Towards this end, in this study, digital reconstruction of a series of modern exoskeleton molluscan specimens from Eastern Mediterranean is performed. A relative coordinate system is implemented to provide quantitative information, while camera calibration partially undertakes intrinsic geometric errors of the camera system. Complete and fragmented specimens are examined, while exoskeleton flakes are also considered. The main challenges in the course of the proposed methodology, is first to establish a quantifiable model that may offer measurable properties including distances, areas and volumes and second to geometrically correlate unknown flakes with complete ones in order to identify their species. In addition, morphology indices are proposed to establish a set of inventory attributes to further support species taxonomy, including volume to area ratios and fractal-space indices of flakes and their origin structures.

The proposed study demonstrates a new insight in the optimisation of the three-dimensional reconstruction through commonly used digital photogrammetry tools. It also delivers an approach to reveal unknown flake identities and propose novel classification indices to enrich archaeomalacological records. The aforementioned methodology may find future applications in the study of other biogenic mineralised materials with archaeological or palaeontological significance, such as teeth, bones, eggshells, otoliths and phytoliths.