A number of problems exist with surveying the coastal to shelf sites in Scotland, some of which are particular to the geographical and historical development of the coast and some of which are generic to all coastal landscapes.
2.4.1 Surveying in Hostile Waters
The current depth (up to 150m of water depth) and hostility (storms, wave, currents) of the North Sea has limited archaeological investigations to date. However, recent developments in survey techniques, such as with remotely operated vehicles (ROVs) and autonomous underwater vehicles (AUVs), are starting to make significant progress to addressing the difficulties and allowing high quality surveying in marginal conditions.
2.4.2 Glacial History
The glacial history of Scotland and the Scottish shelf show a complex sequence of advances and retreats over the last 500,000yrs with field evidence for at least two of the major periods of glacial maximum, the Anglian and Late Devensian (LGM). At their maximum extent the impact of glaciations was not only to cover and scour the land but also to cause significant compression of the land under the weight of the ice with a complimentary bulge in the land beyond the ice limits. However, debate still exists among glaciologists as to whether the last Scottish ice sheet encompassed the Orkney Islands. On retreat, isostatic readjustment of the land surface combined with global eustatic sea level rises resulted in a complex history of flooding and emergence of both seafloor and landmass. Recent investigation and modeling of ice cover has provided a new understanding of palaeo-landmass areas that were available for occupation at different time periods. In particular, low relative sea levels around 10,000 years ago create one large land mass in Orkney and extensive land areas along the entire coast of the Western Isles (SEA4 and SEA7).
2.4.3 Key-hole investigations
A major issue results in the scale problem associated with the large area under investigation, an area that is significantly larger than the onshore Scottish landmass, and the isolated nature of the few finds that have been reported. Placing these small numbers of finds within a larger picture rather than a set of discrete observations requires paradigmatic modelling and an understanding of palaeo-landscapes on a European-wide scale - models that currently may need to be developed. There is a very real danger, due to the lack of contextual information, that isolated finds are interpreted as potential hotspots of activity and thus disproportionately awarding high significance to the finds and the position of the finds within any palaeo-landscapes.
2.4.4 Use of the Landscape
It is speculated that humans have occupied the UK landmass for at least the last 700,000yrs (Parfitt et al.., 2005) with strong evidence for presence from at least 500,000yrs. Little hard evidence exists for human settlement in Scotland before approximately 11,000yrs ago, however the lack of direct evidence does not mean that humans did not occupy and use the Scottish land mass and the now submerged land surrounding Scotland. Sites have been reported elsewhere that experienced the over-running of ice and potential loss of evidence, for example at High lodge, Mildenhall (Ashton and Cook 1990). Furthermore, sites have been described in similar northern provinces, such as Russia, where occupation is recorded as far back as 40,000yrs ago (Pavlov et al. 2001). At these sites, clear occupation and use of areas very close to the ice margin are reported that demonstrate the importance of not only the coastal corridor but also the bountiful seas that are typically associated with marine ice margins. It could be argued that it is precisely these ice margins that are the most likely areas to find evidence of occupation.
2.4.5 Buried Surfaces
Many of the surfaces that are of interest are deeply buried. This applies particularly to the Palaeolithic surfaces, while the Mesolithic surfaces possibly lie closer to the present day seafloor. Reconstructing palaeo-landscapes requires spatial data derived not only from the surface, but from lower sediments as well (Gaffney et al. 2007). However the methods typically used for data collection tend to be based on point sampling or line transects. It is therefore always necessary that some form of extrapolation is undertaken between the measurements. All forms of extrapolation and modeling require assumptions and so the resulting surfaces are products only as good as the assumptions. The assumptions leading to the final model are typically based on similar landscapes elsewhere, however it is important to remember that these might not be wholly appropriate to the ones being mapped. It is recommended that a Scottish-based context model should be established for future interpretations.
2.4.6 Preservation potential
Preservation of archaeology is a consequence of a number of factors including geomorphology, in particular island archipelagos, gully formation, sediment type (gravel, clay, silt, peat etc), exposure of setting.
Sediment type: The sediment type will affect the survival of archaeological remains. Wenban-Smith (2002) indicates that the burial of intercalated organic deposits within fine-grained and cohesive materials (clays and silts) presents a situation that is more resistant to exposure and erosion. Gravel has also been highlighted as an important sediment with examples of well preserved artifacts recovered from both terrestrial and marine units.
Flooding event: The speed at which a material is covered by a flooding event influences the preservation, in particular if the material is not subject to the effects of wave impact within the surf zone for prolonged periods. It may well be the case that very rapid burial of landscapes by the rising sea level during the early Holocene would lead to a likelihood of survival and not erosion in sheltered coastal localities. The preservation is particularly dependent on the elements of erosion subsequent to the deposition of the units. This however is difficult to assess, rather it is the end results, that is if a deposit exists or it has been eroded that is measured. Scotland has a complex sealevel history that has significant variation both temporally and laterally, thus an increased number of site specific sealevel curves are necessary in order to better map the variation.
Morphology: The morphology of the palaeo-landscape can dictate not only the type of material upon which the archaeology is initially deposited but also the likelihood of that material being subsequently eroded. For example, very good preservation potential sites are offered by caves where subsequent erosive environments such as ice erosion often passes the cave system by. This is particularly the case when the cave openings are in the lee of the advancing ice fronts. An example of this is the Pontnewydd Caves in North Wales where sequences with Neanderthal material are preserved in a glaciated landscape. Ulva cave, off the Isle of Mull provides an example of this type in Scottish waters.
Other situations where preservation potential is enhanced due to the local topography are small erosive gullies subsequently in-filled by material and the sides of river or estuary systems, and areas where large local topographic changes result in sheltered waters such as coastal indentations, lagoons and near-shore island archipelagos.