Over the years, Cotswold Archaeology has been involved in many projects that have included geoarchaeological investigations, and I’ve picked out a few different sites to demonstrate the range of geoarchaeological work we undertake at CA. While geoarchaeology often forms a component of our archaeological projects, occasionally the geoarchaeological work itself represents the principal focus of the work. For instance, we have recently been working on a site at Saxonvale in Frome, Somerset, where the main aim of the project was to collect data on alluvial sediments and peats associated with the River Frome. The sediments encountered in both boreholes and test pits were logged and sampled in order to create a deposit model for the River Frome catchment and set it within a wider archaeological and environmental context.
Geoarchaeological analysis can contribute significantly to the way we interpret sites, allowing us to distinguish between various site formation processes, such as colluvial (material deposited at the base of slopes) and alluvial (material deposited by running water) sequences. These processes can vary significantly in their complexity. On a recent site at Whitchurch near Bristol, for example, we encountered a relatively simple sequence of slope wash deposits that covered and partially filled all of the archaeological features. The geoarchaeological assessment enabled us to recognise a thick layer of redeposited occupation debris derived from the area upslope, which was then sealed by redeposited oxidized clay derived from the local bedrock, which had eroded from the slope above.
One of our most interesting projects involving more complex geoarchaeology was undertaken at East Challow, Oxfordshire. Here, the geoarchaeological investigation conducted by Nick Watson at ARCA revealed that layers initially interpreted as natural bedrock were in fact a thick Devensian solifluction deposit, covered by Early Holocene colluvium which contained a buried soil with mud cracks, filled with the colluvial deposits (Fig. 1). Closer examination of the monolith samples in the laboratory helped to confirm the presence of the buried soil that formed at the very beginning of Holocene (around 9500 cal BC) and developed over 900 years before being buried.
The range of different geologies we work on can be challenging at times and can make the recognition of natural features such as paleochannels and periglacial features tricky. One such site is at Swavesey, Cambridgeshire, where the stripped area was described as being like one massive ‘lava lamp’ landscape by the project leader. The geoarchaeological assessments of these ‘blobs’ allowed us to establish that the features formed under cold conditions during the Devensian period (although they contained some intrusive artefactual material just to confuse the issue!). The periglacial condition affected the top fraction of the bedrock and the overlying braidplain gravels, resulting in the formation of a thermokarstic landscape (an irregular surface of hollows and hummocks) which underlay the entire site (Fig. 2).
We have also been working on a long term geoarchaeological project, doing some test pitting in Kent. The Swanscombe area in the Ebbsfleet Valley is one of the most important locations for Paleolithic archaeology in Britain; our work there has revealed a rich sequence of deeply buried Pleistocene deposits containing flint tools and paleoenvironmental remains, which is contributing to the research programme led by Francis Wenban-Smith of the University of Southampton.
Although geoarchaeological work is often linked with Pleistocene sediments and buried soils, we also often use it to help answer questions about specific archeological features. Recently, the geoarchaeological examination of sediments in monolith samples from a defensive ditch, thought to be part of an Early Roman fortlet, from St Sidwell’s Point in Exeter aimed to characterise the origin of the ditch`s fills and to determine if there was any evidence for eroded rampart material within them. The examination revealed that the fills derived from a bank, which may have been hardened using turf and stones. It also demonstrated that this bank was situated on the outside of the ditch, not on the inside (Fig. 3).