Archaeological geophysical prospection in peatland environments.

Abstract

Waterlogged sites in peat often preserve organic material, both in the form of artefacts and pa1aeoenvironmenta1 evidence as a result of the prevailing anaerobic environment. After three decades of excavation and large scale study projects in the UK, the subdiscipline of wetland archaeology is rethinking theoretical approaches to these environments. Wet1and sites are generally discovered while they are being damaged or destroyed by human activity. The survival in situ of these important sites is also threatened by drainage, agriculture, erosion and climate change as the deposits cease to be anaerobic. Sites are lost without ever being discovered as the nature of the substrate changes. A prospection tool is badly needed to address these wet1and areas as conventional prospection methods such as aerial photography, field walking and remote sensing are not able to detect sites under the protective over burden. This thesis presents research undertaken between 2007 and 2010 at Bournemouth University. It aimed to examine the potential for conventional geophysical survey methods (resistivity, gradiometry, ground penetrating radar and frequency domain electromagnetic) as site prospection and landscape investigation tools in peatland environments. It examines previous attempts to prospect peatland sites, both in archaeology and environmental science. These attempts show that under the right circumstances, archaeological and landscape features could be detected by these methods, but that the reasons why techniques often fail are not well understood. Eight case-study sites were surveyed using a combination of conventional techniques. At three of the sites ground truthing work in the form of excavations, bulk sampling and coring was undertaken to validate the survey interpretations. This was followed up by laboratory analysis ofthe physical and chemical properties ofthe peat and mineral soils encountered. The key conclusion of the case study work undertaken is that conventional geophysical prospection tools are capable of detecting archaeological features in peat1and environments, but that the nature of the deposits encountered creates challenges in interpretation. Too few previous surveys have been adequately ground truthed to allow inferences and cross comparisons. The upland case studies demonstrated that geophysical survey on shallow types ofupland peat using conventional techniques yields useful information about prehistoric landscapes. The situation in the lowlands is more complex. In shallow peat without minerogenic layers, timber detection is possible. There are indications that in saturated peat the chemistry ofthe peat and pore water causes responses in the geophysical surveys, which could be developed as a proxy means to detect or monitor archaeological remains. On sites where the sediments are more complex or affected by desiccation, timbers were not detected with the methods attempted. However, important landscape features were and there are indications that geophysical surveys could be used as part of management and conservation strategies. This thesis concludes that geophysical prospection contributes to theoretically informed wet1and archaeology as a tool for site detection, landscape interpretation, and conservation. Future research should aim to further our understanding of the relationship between geophysical response and peat1and geochemistry, alongside a more extensive programme of surveys and ground-truthing work to improve survey methodologies and archaeological interpretations.