The Terramare and the surrounding hydraulic structures: a geophysical survey of the Santa Rosa site at Poviglio (Bronze Age, northern Italy)

Highlights

• Geophysical anomalies of a 7 ha site were mapped using FDEM and ERI methods.

• They revealed structures of a large Bronze Age Terramara site in northern Italy.

• They showed the shapes of the settlements, moats and the hydraulic network.

• These structures had not been fully surveyed with previous site excavations.

• The hydraulic structures are related to water management and irrigated agriculture.

Abstract

The Terramara Santa Rosa is a Middle and Late Bronze Age archaeological site located in the Po alluvial plain, northern Italy. It is constituted of two moated villages delimited by earthen ramparts. The peripheral structures of the site are sealed by fine-textured flood plain deposits and they have not been fully explored through excavation due to their large extent. Because the shape of the villages and their relation to moats and the fluvial network are of paramount importance to understanding the landscape management and the use of water resources in the Terramare civilisation, a geophysical survey was planned to extend the results of the existing archaeological excavations to the site scale. A frequency-domain electro-magnetic sounding (FDEM) and electrical resistivity imaging (ERI) covered an area of approximately 26 ha; ERI was conducted for a total length >9000 m. Despite the predominance of electrically conductive fine-textured sediments, electrical resistivity anomalies were observed and they were attributed to subtle lithological differences in the sedimentary context of the alluvial plain. The geophysical interpretation, after the calibration with the excavation data, revealed the structures of the Terramara and of the surrounding hydraulic network, which are not visible at the surface due to flood plain deposits. The Santa Rosa site was founded in a favourable geomorphological position, on the top of a crevasse splay lobe of the adjoining Po palaeo-channel, rising above the surrounding alluvial plain. The Terramara and their surroundings were delineated through an artificial modification of this pre-existing crevasse splay lobe and a well-targeted urban design, with the objective of diverting water, most likely from a palaeo-channel of the Po River, through the digging of peripheral moats used to collect water around the site and to distribute it to the surrounding fields for irrigation. The water management documented by this study in the Terramara Santa Rosa can be considered as paradigmatic for the whole Terramare civilisation, which is therefore responsible of the introduction of the irrigated agriculture into western Europe for the first time.

Introduction

The Terramare are the archaeological remains of banked and moated villages of the Middle and Recent Bronze Ages (1600–1150 BC), located in the central alluvial plain of the Po River. They are evidence of a complex society, whose subsistence was based on intensive agriculture, pastoralism, and long-range trade (Mutti, 1993, Barfield, 1994, Bernabò Brea and Cardarelli, 1997, Cardarelli, 1988). It reached its apogee at the beginning of the Recent Bronze Age and, at the end of this period, it suffered a societal collapse that led to the abandonment of the villages in a few generations (Cardarelli, 2010).

The shape and the structure of the villages have been a main topic of the archaeological research focused on this civilisation, since the first seminal studies (Strobel and Pigorini, 1864, Chierici, 1881) and also in recent research projects (Bernabò Brea and Cardarelli, 1997, Pearce, 1998). The villages were supposed to have been built following sophisticated urban planning, as they are square in shape, with houses on posts that are distributed in regular rows, and enclosed inside earthen ramparts. They were supposed to be surrounded by a moat connected to an adjoining river through a canal network (Strobel and Pigorini, 1864, Chierici, 1881). This scheme is still under discussion (Pearce, 1998), but is supported by recent research on aerial photographs (Fig. 1) which show the existence of moats around most of the sites (e.g., Terramara Gaione – Mutti, 1993) and, in several cases, a direct connection with a neighbouring water course (e.g., Terramara La Falconara – Calzolari, 1997; Terramara Fabbrica dei Soci – Cremaschi, 1997; Terramara Castello del Tartaro – Balista and De Guio, 1997). At present, it is hypothesised that the moats were designed not only with a defensive goal, but also to collect water from the local fluvial network and distribute it to cultivated fields around the site (Balista, 1997, Balista, 2002, Cremaschi and Pizzi, 2007, Cremaschi and Pizzi, 2011, Ferri, 1989, Ferri, 1996), to sustain the irrigated agriculture on which the subsistence of the Terramare was based. The availability of water played a vital role for this culture, and it has been proposed that a dry climatic crisis is a possible cause of the disappearance of the Terramare (Cardarelli, 2010, Cremaschi, 2010, Cremaschi et al., 2006). Consequently, exploring the shapes of the villages and their relation to the moats and the fluvial network is of paramount importance to understanding the level of land exploitation performed by the Terramara culture on the surrounding landscape. Up to now, the knowledge about the shape of the sites and the presence of moats and ditches was manly based on aerial photographs, because subsurface surveys and archaeological digs extended at the site scale was lacking.

An archaeological project focused on these problems has been on-going since 1984 at the Santa Rosa site (Poviglio, Reggio Emilia, Italy; Fig. 2). At present, it is one of the largest archaeological excavations of European prehistory (Bernabò Brea and Cremaschi, 1996, Bernabò Brea and Cremaschi, 2004a, Bernabò Brea and Cremaschi, 2004b, Cremaschi, 2004, Cremaschi, 2013). An area of nearly 1 ha has been unearthed from a larger 7 ha site, and a detailed geophysical survey was carried out over the site and in its closest surroundings to extend the information on the archaeological structures that have been directly investigated (rampart, moat, general shape of the settlement) to the entire site.

Ground-based archaeogeophysical surveys have been successfully applied in archaeological exploration (Scollar et al., 1990, Kvamme, 2003). They provide a time- and cost-effective tool to delimit promising excavation areas and reduce the costs of excavation campaigns. Active and passive methods, such as direct current (Arlsan et al., 1999, Cardarelli and Di Filippo, 2009, Cardarelli et al., 2008, De Domenico et al., 2006), electromagnetic (Carrozzo et al., 2003, Conyers et al., 2008, Piro et al., 2003, Thiesson et al., 2009) and magnetic (Drahor, 2006, Gondet and Castel, 2004, Maillol et al., 2004) methods, can be applied in a wide range of geological environments and permit the non-destructive mapping of archaeological features. The effective identification of geophysical anomalies produced by buried man-made structures relies on vertical and lateral contrasts in the physical properties (such as electrical resistivity, magnetic susceptibility and dielectric constant) of the archaeological target and of the hosting medium (Kvamme, 2003, Reynolds, 2011). In other words, detecting archaeological features is most likely when the physical property contrasts are large.

The Santa Rosa site is located in an environment of fine-textured sediments, which can be considered a highly electrically conductive environment (McNeill, 1980a, Reynolds, 2011). Only slight lithological differences occur between sandy-loamy sediments, which characterise most of the investigated archaeological structures and the substrate sediments of the site, and the widespread flood plain cover (silt and clays). Thus, the main geophysical challenge is to identify small variations in electrical conductivity related to slight changes of sand content between the archaeological features, the substrate sediments and the flood plain cover. To achieve this goal, frequency domain electro-magnetic (FDEM) and earth resistivity imaging (ERI) methods were applied for two reasons: the use of time-varying and stationary fields could yield complementary information for the characterisation of heterogeneities in electrical conductivity and their application could yield different horizontal and vertical resolutions. The FDEM method was adopted to map the lateral transitions between the settlements, the hydraulic structures and the surrounding areas at the site scale. The ERI method was used to validate the electrical response of the site at a finer scale and to integrate additional information about the depth of the archaeological features, which cannot be directly evaluated by the FDEM method.

A description of the archaeological site is given in the following section. The third section is devoted to data acquisition and processing, and the results are discussed in the fourth section. The discussion of the results obtained from the integration of the geophysical interpretation with the archaeological and geoarchaeological data are presented in the fifth section.