Introduction to the techniques
People change in many direct and indirect ways the landscapes they live in (e.g. Goudie 2006). Landscapes also constrain what people can do. Archaeologists need to know about and record evidence for the contemporary landscape context of a site and geomorphological and sedimentological evidence for landscape-shaping processes contemporary with and after occupation.
Many approaches are not the preserve only of the specialist. Data on landscape-scale geomorphology should be insisted upon by those overseeing the design of projects and collected as a matter of course by archaeologists during excavation or survey. All that changes is the recognition that an archaeological site has a larger landscape context. The most appropriate spatial scale for data collection depends on the archaeological work. Developer-funded excavation often constrains what can be done in a small area but desk-top survey can be undertaken and ‘natural’ contexts recorded in detail. The spatial scale of landscape archaeology most comfortably fits that of the geomorphology and truly synthetic work can emerge. Often the river catchment emerges as the natural scale for people in the landscape. There is value in looking afresh at the techniques developed by Jarman, Bailey and Jarman (1982).
Desktop survey should include reference to British Geological Survey mapping, particularly of ‘drift’ rather than solid geology [‘drift’ is an antique description for superficial sediments, of most interest to the archaeologist (www.bgs.ac.uk/geoindex)]. This web-site also contains borehole data, sometimes out-dated but often very detailed. Developer-funded work has the great advantage of often providing high-resolution contour surveys from which landforms can be mapped, and civil engineering borehole logs. Landscape context is defined through field mapping, aerial photo interpretation and, increasingly, from remote-sensed data. There are conventions for geomorphological mapping (e.g., Goudie 1981: Rose and Smith 2008). Mapping is made easier through remote-sensed satellite imagery (see Section 5) draped over high-resolution digital terrain models within a Geographic Information System (GIS) platform (Smith and Pain 2011). Such images define with stunning clarity even small-scale landforms and archaeological features as visually insignificant as broad rig (Smith, Rose and Booth 2006). The technique is expert-driven but is being explored for archaeological work, notably at the Royal Commission on Ancient and Historic Monuments of Scotland. Interpretation of geomorphological features can also be expert-driven but most can be identified and mapped with only a passing knowledge of physical geography. There should always be a route by which field archaeologists can contact geomorphologists, particularly in the academic community. Landform surfaces need to be surveyed not only in a relative sense (i.e. from a temporary bench mark) but to Ordnance Datum. A simple change in working methods would provide much valuable detail.
Landscape as palimpsest is a good analogy. To define the landscape contemporary with an archaeological site requires the peeling-away of later features, as well as the reconstruction of processes that may have distorted the archaeological record (Schiffer 1987). The approach is sediment-stratigraphic, supported by chronological tools (see Section 1). Sediment stratigraphies are recorded on-site with great care by archaeologists, but frequently not using the terms used by geomorphologists. Jones, Tucker and Hart (1999) is a useful attempt to standardise description. What is also needed is a change in the approach to interpretation. Alluvial and colluvial sediment fills on archaeological sites are all too frequently recorded only as archaeologically ‘sterile’ and thus uninteresting. They are not. Such features may represent major landscape-changing events. They may have been anthropogenic in origin, and so of equal significance as the material culture is to the archaeologist. They may have shaped the perceptions of people and altered the way in which the landscape was regarded. Such features need to be recorded, dated, evaluated and interpreted with the same care given to demonstrably anthropogenic sediments. Off-site sediment stratigraphies are usually seen as the preserve of the specialist but again they need not be. It is important that they are considered as inseparable from the archaeology. There is little additional training in recording and sampling these. The only general rule is to preserve the sediment stratigraphy: tins or gutters are much better than bags. Most sedimentological analyses, be they particle size through laser diffraction, non-destructive core scanning geochemical techniques are expert-driven and not readily available to the archaeological community, but SUERC may have a role to play here in providing greater access to these tools for the archaeological community.
The Published Literature
Recent thorough surveys of the literature of landscape change and driving forces pertinent to Scotland are by Ballantyne (2008) and Macklin et al. (2010). Geomorphologists have their own research concerns, introduced briefly here, but anthropogenic activity is one of these. Ballantyne (2008) reviews the evidence for postglacial geomorphic processes and rates of landscape change in the Scottish highlands. Many geomorphologists are drawn to upland areas because processes and their effects are more spectacular, and in the uplands, driving forces of change are perhaps easier to define in the absence of large farming populations, although a direct and proportional relationship between human population size, activities and geomorphic effects should not be assumed. Positive feedbacks abound in which small-scale change can be greatly magnified. This bias to upland areas markedly skews our understanding of regions more densely and continuously populated: a comparable survey to Ballantyne’s for lowland regions in Scotland has yet to be written, largely because there are too few data.
Holocene geomorphic processes in both upland and lowland regions are seen as episodic rather than constant, with low frequencies but high magnitudes, and to have been rapid rather than gradual. Again, too few landscape features in Scotland have been dated sufficiently to explore this: a major effort to define the chronology of debris flows (Ballantyne 2004) resulted in only equivocal evidence for episodicity. Episodic change is seen to have been prompted by rapid climate change (e.g. Mayewski et al 2004) or by infrequent intensities of anthropogenic activities. Macklin et al. (2010) have argued strongly for a climatic driver (increased precipitation) to explain accelerated fluvial activity in the Holocene, based on perceived synchroneity across Britain of events such as floodplain construction or incision/erosion. However, in too few river catchments has the timing and scale of anthropogenic activities been reconstructed. These two drivers are not mutually exclusive, of course. Attention has also focused in recent years on the connectivity in sediment supply between landscape features. They are less clearly connected than might be assumed. This probably means that describing change in one landscape element, such as in the example above of rivers, says far less about change in other elements in the catchment, like slopes above the river. Sediment is stored in parts of the landscape for substantial periods in a rather jerky ‘conveyor belt’. These current research concerns form an important context to analyses relevant to archaeologists.
Archaeological site locations and land uses are very often related to the underlying geology or to highly dynamic geomorphic processes. The association of early Neolithic timber halls with well-drained glaciofluvial terraces is clear (Murray, Murray and Fraser 2009). The association in the highlands between settlement and alluvial fans is also clear (e.g. Barclay 1995) though interpretation is not: do fans provide the best soil or do they provide locations which will not be eroded? Habitation in caves requires knowledge of past sea level change (Bonsall and Sutherland 1992). Sea level change is also fundamental in understanding sediment supply to areas of machair (Ritchie 1966, 1979; Gilbertson et al. 1999) or to explaining the setting of marine crannogs (Sands and Hale 2001). It should be noted that recent exciting approaches to landscape need to have a deep grounding in the physical as well as the experiental attributes of landscape (Tilley 1994; Barton et al. 1995; Richards 1996; Cummings and Whittle 2003). This need has led in recent years for closer links between archaeologist and geomorphologist in distinguishing natural from human-made features (Bradley 2000; Tilley et al. 2000).
Geomorphic processes also distort the archaeological record. Erosion is one obvious effect and impending erosion is a strong justification for rescue excavation. Over ten years Tom Dawson of the SCAPE Trust has developed a methodology and database for coastal archaeology in Scotland that is among the best in the world (http://www.scapetrust.org). Slope erosion will impact on site preservation (Wilkinson et al. 2006). Within coastal dune systems the archaeological record becomes exceptionally difficult to understand because wind erosion selectively erodes sand separating archaeological contexts, deflating sediments and conflating the archaeological sequence (Cowie 1996; Ritchie and Whittington 1994).
Archaeological site prospection often requires knowledge of landforms and geomorphological processes (e.g. Brown 1997; Howard and Macklin 1999; Timberlake 2001; Macklin, Howard and Passmore 2003). Threats from fluvial erosion are clear but work in Scotland has as yet to explore this problem in detail, or use data in resource management despite the lead on the English side of the border (Howard et al. 2008; Passmore and Macklin 1997; Passmore and Waddington 2009, 2011). Tipping et al. (2008) tried to predict the likelihood of subsurface engineering work impacting on such concealed sites along the River Kelvin from understanding the age of floodplain sediment: here the threat was only to Mesolithic or even Upper Palaeolithic deposits, if they existed, because the present floodplain had been constructed before the Neolithic period. Sediment in river terraces can be a product of human activity through soil erosion from forest clearance and cultivation as soils become wetter, driven by weaker evapo-transpiration. In extreme cases the entire river terrace can be seen as anthropogenic in origin (Tipping 1992), which creates a dilemma for archaeological resource managers. An understanding of the chronology of river terrace formation can thus greatly enhance archaeological interpretation. Environmental stress to communities is often mediated through accelerated geomorphic processes (e.g. Dawson et al 1990; Carter 1998; Foulds and Macklin 2006).
Lakes act as stores of sediment from the catchment up-valley. They are sources for innumerable palaeoecological techniques, discussed elsewhere in this section. They can preserve records of eroding soil that are more complete than in floodplains and river terraces because they are less likely to be eroded. Geochemical techniques (Engstrom and Wright 1984; Entwistle et al. 1995; Croudace et al. 2006; Vogel et al. 2008) can define sediment sources if catchment geology is varied. Geomagnetic signals in lake sediment can define whether topsoil was eroded, affecting large areas of farmed land, or subsoil and parent materials from deeply incised gullies (Oldfield et al. 2003). Other evidence for soil erosion comes from the distortion of 14C assays as old organic matter is washed in (Edwards and Whittington 2001), sometimes to curtail research programmes as at Buiston in Ayrshire (Tipping et al. in Crone 2000, 46). Sometimes apparent anthropogenic impacts on lakes have other causes (Wells, Hodgkinson and Huckerby 2000; Coles 2001).
Existing Research Capacity
All Scottish universities currently support activities related to geomorphology, though archaeologists should be aware of a broad differentiation in the discipline between contemporary, process-driven approaches and those concerned with environmental change. The SAGES initiative (Scottish Alliance for Geosciences and Society) has pooled the expertise in geoscience and environmental science from across Scotland’s research base, creating a multi-disciplinary alliance between the Universities of Aberdeen, Abertay, Dundee, Edinburgh, Glasgow, St Andrews, Stirling, SAMS UHI, SUERC, and the West of Scotland. http://www.sages.ac.uk/themes/landscape/ is the best way in to the current academic resource and interests (see also http://www.sages.ac.uk/sagesfacilities/). Archaeology and past societies have not, however, driven the output of this collaboration.
Geomorphologists are absent in commercial excavation units in Scotland. Often the pedologist, if one is employed, assumes responsibility, but the two disciplines require very different skills and knowledge.
The SAGES initiative demonstrates that Scotland contains an internationally significant research community into geomorphic processes and change. Future climate change has provided the impetus for the SAGES initiative but the focus needs to move to understanding the complex interactions over time between people and environment. Scotland is better placed in this regard than anywhere else in north west Europe because of its location at the eastern edge of the Atlantic Ocean where climatic changes were probably greatest. It is imperative, as indicated at the outset of this Panel report, that modern day researchers understand the vulnerability and resilience of past communities, and how coping strategies emerged and shifted. Many of the stresses faced by these communities will have been mediated through the physical landscape. Their solutions to crises are critical for a modern undetrstanding. This must be the most significant opportunity to establish new and deeper collaborations.
Archaeological resource managers have yet to work with geomorphologists in understanding the preservation of the historic environment, although the value of this has been demonstrated by recent work in England. This new work was funded almost in its entirety by the Aggregates Levy Sustainability Fund Grant Scheme 2008-2011. This financial source was not used for this work in Scotland and this opportunity was missed. It is to be hoped that others emerge in the near-future.
It cannot be expected in the current economic climate that commercial excavation units will employ geomorphologists, but accessing specialist workers in Scottish universities can be made much easier by a register of workers and their interests: that for the SAGES initiative is far from complete. Almost all specialists will offer advice and guidance by e-mail, telephone or a site visit. Very often what the archaeologist uncovers is of direct value to the specialist. Training in non-expert techniques should be facilitated. Short courses for field archaeologists should be developed, with recognition of new skills acknowledged by organisations such as the IFA. Masters courses need to be developed.
Landscape archaeology has to be collaborative. There is no value in an archaeological site distribution when landscape change has not been factored in. A change in legislation to allow geomorphological work to be developer-funded on archaeological sites would lead immediately to new data. There is almost no data for lowland regions of Scotland, where developer-led archaeology is centred. Geomorphologists have focused on particular parts of the landscape, such as hills and mountains, and river systems because their sediments and landforms are at least relatively easy to analyse and date, but work on slopes and aspects like soil erosion and colluviation which are probably more central to archaeological interpretations have been neglected (e.g. Brown 2009). Work on this is much more advanced on the continent (Lang 2003; Hoffmann, Lang and Dikau 2008; Dreibrodt et al. 2009). Because landscapes are less well connected than previously thought there is a need to establish independent chronologies of change for different landscape features.