Temple landscapes : fragility, change and resilience of Holocene environments in the Maltese Islands

Geoarchaeological survey, test excavations and sampling on Gozo and Malta concentrated on the sites and landscapes associated with the Neolithic temple period of the fourth and third millennia bc. Targeted investigations were carried out at two Neolithic temple sites of Ġgantija and Santa Verna on the Xagħra plateau and the associated Ramla and Marsalforn valleys on Gozo. Sequences were also recovered from the excavations of the Neolithic Taċ-Ċawla settlement site in the modern town of Rabat and the later Bronze Age mesa-top site of In-Nuffara. On Malta, geoarchaeological work focused on the temple site of Skorba, and the nearby valley coring site of Xemxija, as well as the deep valley core sites of Wied Żembaq, Marsaxlokk and Salina (Figs. 2.4 & 5.1). In the context of the on-site investigations, test excavations at the Santa Verna, Ġgantija and Skorba temple sites and at the Taċ-Ċawla settlement site all revealed old land surfaces beneath mixed soil and cultural deposits. For the off-site geoarchaeological work, some 200 hand-augered boreholes were made during the 2014/15/16 field seasons. Most boreholes were in the Santa Verna to Ġgantija areas on the margins of the modern town of Xagħra, across the intervening Ramla valley to In-Nuffara and down-valley to the sea, and also in the Marsalforn valley from Rabat northwards to the sea (Fig. 5.1). The areas around the Ta’ Marżiena and Skorba temple sites were also investigated briefly for comparison using the hand auger, but no sample test pits were excavated. This geoarchaeological programme has provided sufficient soil/sediment sequence data to address several sets of aims as set out below, and in combination with the analysis of the deep valley cores (see Chapters 2 & 3), it is now possible to suggest a model for Holocene landscape development. It has always been assumed that the seasonally dry and hot Mediterranean climate made the Gozitan and Maltese landscapes quite ‘marginal’ in agricultural terms (Grima 2008a; Schembri 1997). As a consequence, it has also been presumed that terracing was adopted extensively from the Bronze Age onwards on both islands to conserve soils and moisture, and also to create a more suitable landscape for subsistence based agriculture (Grima 2004). Like many other parts of the southern Mediterranean, this landscape is prone to deforestation, drought and erosion combined with intensive human activity, and that this has been the case since Neolithic times (Bevan & Conolly 2013; Brandt & Thornes 1996; Hughes 2011; Grove & Rackham 2003). The FRAGSUS Project aimed to examine these assumptions and test them with a suite of archaeological science approaches that would shed new light on the nature and impact of Neolithic farming and on the degree of fragility of this island landscape. Within the overall project, the main objectives of the geoarchaeological work were to: 1) investigate the deposit and soil catena sequence of the Xagħra plateau and its associated Ramla and Marsalforn valleys for the Holocene; 2) identify floors, floor deposits, old land surfaces and palaeosols associated with the Neolithic monuments, concentrating on the Santa Verna, Ġgantija and Skorba temple sites, as well as the Taċ-Ċawla settlement site; 3) create a model for the Holocene land-use sequence for Gozo and Malta, focusing on the impact of Neolithic agriculture and later landscape terracing, and 4) establish if there is any correlation between observed soil properties and prehistoric activities and/or longer-term climate change. The results of the geoarchaeological analyses are discussed below, with the borehole logs and field profile descriptions (after FAO & ISRIC 1990) found in Appendix 6, the thin section descriptions in Appendices 7 and 8, the sample list in Table 5.1, the summary dating of the analysed profiles in Table 5.2, and the optically stimulated luminescence (OSL) dating report in Appendix 2. Note that the comprehensive radiocarbon dating study is discussed in this volume (see Chapter 2), and the site-based geoarchaeological and micromorphological studies at Santa Verna, Ġgantija, Skorba, In-Nuffara and Taċ-Ċawla are reported on separately in the FRAGSUS excavation Volume 2.This project has received funding from the European Research Council (ERC) under the European Union’s Seventh Framework Programme (FP7-2007-2013) (Grant agreement No. 323727).peer-reviewe