Differential crop growth has revealed a complex of plough-levelled sites at Aytonlaw in Berwickshire. Cropmarking, recorded from the air during aerial reconnaissance has completely revolutionised understandings of lowland areas, identifying sites that there is no other effective means of discovering and recording. Most of the known Iron Age monuments in lowland south-east Scotland, for example, are only known from this type of record.
Aerial reconnaissance and photography has been established for over 100 years, and its application for archaeology across Europe has demonstrated that it is one of the most effective methods of archaeological reconnaissance and recording. This is especially true in lowland areas where millennia of ploughing have levelled most archaeological sites, rendering them invisible except to the airborne surveyor as differential crop growth. In Scotland this survey record, comprising about 8000 sites concentrated into lowland arable areas, built up since the 1940s, has completely and utterly revolutionised the knowledge-base for these areas. There is currently no other effective method of recording these sites, and many periods and site-types would be almost unknown without this data. Thus most Neolithic cursus monuments and Roman temporary camps, for example, are known only because of aerial survey.
The aerial perspective has also shown to be very effective in recording upstanding, earthwork monuments, especially where the combination of low winter sunlight and low vegetation combine to throw even slight surface variations in topography into relief through a combination of highlights and shadow. This can be enhanced through a slight dusting of snow, but often snow-cover disguises as much as it reveals. Appropriately lit and targeted aerial photographs can be incredibly powerful tools for archaeological interpretation.
This oblique aerial view centred on the remains of Woden Law Fort in the Cheviots records the fine detail of the earthworks and adjacent cord rig in remarkable detail and provides a powerful interpretative tool. The low winter sun and slight dusting of snow reveal details that would be very time-consuming to understand on the ground.
Why aerial survey and photographs?
The aerial perspective has proved a highly effective tool for detecting and recording heritage assets, providing the only means of recording some sites (e.g. plough-levelled sites revealed as cropmarking), adding value to the records of upstanding sites as sources that permit rapid understanding and in providing the most textured records available to see the broader landscape and traditional sites in context. Aerial reconnaissance is the most effective method of finding sites and creating basic records, both using light aircraft for active prospection, but also in undertaking ‘survey’ in existing aerial photograph collections. These are very good sources of heritage information, often recorded fortuitously on photographs taken for other purposes, such as military intelligence and to provide map making data. In cases where landuse change has dramatically altered landscapes (e.g. afforestation, urban sprawl, upland pasture improvement) historic aerial photographs may be the only available record. Aerial photographs are a very powerful way of conveying complex information about the layout and contexts of sites, that have no immediate parallel and give real meaning to a picture being worth a thousand words. Aerial survey can be a rapid and effective method of monitoring condition and material change in monuments, though, while this approach is routine in England and Wales, and increasingly widely adopted in continental Europe, it is not applied in Scotland.
There are three main types of aerial photographs: over 100,000 oblique, generally site orientated images captured during observer-directed survey largely held by the RCAHMS, the majority taken since the mid-1970s; about 1.5 million vertical photographs, collected for a variety of non-archaeological purposes (mainly map making) largely since the mid-1940s (also held by the RCAHMS); and web-based vertical photographs (often wrongly referred to as satellite images) such as Google Earth, which add a valuable contemporary view, with limited ‘historic’ elements. All are valuable reservoirs of information on sites and monuments and at a landscape scale, and the collections of historic vertical photographs are a virtually un-tapped resource that archaeologists should use more. All require a basic level of understanding of purpose/application to ensure appropriate usage, including the basic difference between types of photographs and the survey methodology that lies behind them. Observer-directed survey and photography, whereby reconnaissance is undertaken in a small aircraft, with an observer looking for particular targets and only photographing what they are interested in, dominates archaeological recording. It produces vital results, but suffers from the fact that only sites that are noticed by the observer/photographer are recorded and so the resulting dataset suffers from the biases imposed by their own knowledge. So-called ‘vertical’ photographs are usually collected as a block-coverage exercise across a large area using cameras mounted on the aircraft automatically taking photographs at selected intervals to produce overlapping pictures that can be used for stereo-viewing (e.g. 60% overlap). These photographs are only truly vertical at the centre of the picture, with increasing distortion towards the edges of the frame. Generally, these photographs are not taken for archaeological purposes because of the greater expense involved with more expensive, purpose converted, aircraft and cameras, but they do not suffer from the observer bias that may be a problem with other types of imagery. They offer potentially greater scope to explore the unfamiliar that may not be in the mind of the airborne surveyor and therefore do not get recorded.
Understanding aerial photographs
To make best use of this resource it is necessary to have a good understanding of the type of photographs and what they may tell. For example, what is the scale of the photograph: At 1:5,000 a feature 5m across will measure 1mm across, or about the size of a blunt pencil point. At 1:10,000 the same feature will measure 0.5mm across, and at smatter scales (e.g. 1:25,000, 1:100,000) the feature will be vanishingly small. Small pits will not be visible on 1:25,000 scale photographs and if this is the type of archaeology expected then time should not be wasted on such photos. Equally, there is little point in targeting photographs taken during the winter if there is a desire to see cropmarking. This is a process of implementing your survey strategy and increasing the chances of getting good results.
Aerial photographs are raw material for archaeologist: they contain information and this requires interpretation and mapping. These processes are vital to creating reliable archaeological information that can facilitate the understanding and management of heritage assets. Exploration of prehistoric lowland settlement patterns, in East Lothian for example, requires accurate mapping to create robust settlement morphologies that can be analysed systematically. Here, over 40 years of sustained aerial survey, the results of which have now been systematically mapped, underpin the understanding of regional settlement patterns and contribute to the creation of settlement frameworks.
Aerial survey, especially observer directed reconnaissance, is a subjective process very much dependent on the perspectives, assumptions and prejudices of the observer (Palmer 2005). Effective use of aerial photographic data requires an understanding of these processes – which are not unique to aerial data. Indeed, all archaeological distribution maps are maps of recovery, more likely to reflect the present than any reality in the past, since known distributions are usually produced by transformations of what existed in the past, rather than directly conveying past patterns. So differential patterns of past and present land use may conspire to either hide or reveal sites, while the knowledge and interests of antiquarians and modern archaeologists may directly impact on the known distributions of sites. Furthermore, interpretation is not a self-evident process; it must be taught and practised so that experience must be developed based on knowledge of external factors such as post-depositional processes and the types of site that are likely to be encountered.
Aerial survey for Scottish archaeology
Aerial reconnaissance for archaeology in Scotland was initiated by OGS Crawford, who undertook pioneering flights here in the 1930s (Crawford 1930, 1939), principally exploring conjectured lines of Roman campaigns. However, it was not until 1945 that more regular survey was undertaken by JK St Joseph, under the Cambridge University Committee for Aerial Photography (CUCAP), initially facilitated by the RAF who supplied the aircraft and pilot. St Joseph continued an almost annual survey across most of the British Isles, chasing the harvest northwards, with a personally inspired, but by no-means exclusive emphasis on Roman archaeology. The CUCAP flying in Scotland tailed off under David Wilson in the 1980s after St Joseph retired, but also saw a shift to more vertical coverage taken for commercial contracts (e.g. Forestry Commission). The main research outputs from this period were in Roman archaeology, starting with the publication of Crawford’s Rhind lectures (1949) and continuing through a series of St Joseph papers (e.g. 1976).
The real expansion in aerial reconnaissance occurred in 1976 and 1977, two very dry summers that can now been seen as better than average years for cropmark formation. In 1976 RCAHMS commenced its aerial survey programme, following exploratory flights in 1975, and in 1977 Ian Shepherd and Ian Ralston initiated the Aberdeen Aerial Surveys (AAS). The initial focus of both survey programmes, supported from Central Government (Scottish Development Department), lay in exploring arable areas for plough-levelled monuments visible as differential cropmarking, but they soon expanded to take in upland areas (Cowley 2005). Both surveys have continued on an annual basis, though at the time of writing (2011) the future of AAS is unclear with the retirement of Moira Grieg. The remit of RCAHMS is national though until the early 2000s an emphasis on the south and east is evident, while AAS focused on the north-east, extending along the Moray Plain and overlapping considerably with RCAHMS in Angus. These sustained survey programmes were supplemented by regional fliers, often partly sponsored by SDD and later RCAHMS grants, who flew particular areas of interest. The main period of activity of the ‘sponsored fliers’ was from the mid 1980s to the mid 1990s, after which progressive difficulties in resourcing (time and funds) and tightening up of regulations for aerial survey by the Civil Aviation Authority saw this contribution to flying wither to virtually nothing. The contraction of regional fliers has been offset to some degree by RCAHMS expanding its remit to become truly national, ensuring that coverage extended to all areas practicable in a single engine aircraft, and occasionally employing helicopters.
These processes have placed many thousands of sites on record, especially in lowland areas, and these are now embedded in management policies (i.e. scheduling) and in academic discourse. However, throughout there has been a sustained emphasis on data collection, and relatively little analysis or publication of the results took place until the last 10-15 years (below). Indeed, in the early years many sites were added to the RCAHMS database with generic classifications, such as ‘enclosure’ or ‘cropmark’, on the rationale that with more information would come more reliable classifications. Predictably, the sheer mass of cropmark information that rapidly accrued soon ensured that any review of the generic classifications would be a daunting task. However, with the appointment of an Aerial Survey Mapping Manger to RCAHMS in 2005 (Kevin Macleod) the systematic mapping and classification of cropmarked sites is now underway.For example, detailed mapping of the Antonine Wall, undertaken as part of its nomination as a World Heritage Site, included the reinterpretation and mapping of the aerial photographic record. Reliable interpretation and accurate mapping was critical to the success of the nomination and the information gathered is actively used in management (Jones and McKeague 2011).
The analysis of the results has still some way to go, although the last 15 years has seen considerable developments. The publication of Roman material, principally by St Joseph, David Wilson and Gordon Maxwell, led the way for many years, supplemented by a few general papers(e.g Maxwell 1978, 1983, 1987), the RCAHMS publication on south-east Perth (RCAHMS 1994) and Lesley Macinnes’ PhD (1983). More recently a series of thematic PhD’s have been completed drawing principally on aerial information, namely Brophy (cursus monuments – 1999), Jones (Roman Temporary Camps – 2006) and Millican (Neolithic Pit-defined monuments – 2009). In tandem, a series of publications have addressed regional patterns (Cowley and Brophy 2001, Cowley 2008, 2009a), other specific themes (e.g. long cists – Cowley 2009b; maritime landscapes – Cowley and Martin 2011) and reflected on best practice (Cowley 2002 , Cowley and Dickson 2007, Jones 2005, Cowley and Gilmour 2003, 2005), addressing criticism of survey bias identified elsewhere (Hanson and Macinnes 1991).
Scotland has an international holding of aerial photographs, numbering many millions and covering large parts of the globe. The Scottish material is also substantial adding up to over 1.5 million images, spanning the period since 1946, and this represents a barely-tapped research resource. The PhD research projects already undertaken (above) demonstrate the potential of these resources. With the imminent release of aerial mapping data through web-mapping services this will be accessible, though access to other parts of the collection and digital browsing facilities are not all they might be.
Experience in interpretation of aerial photographs is vested in a small group of professionals, and the Department of Archaeology, University of Glasgow, offers taught postgraduate courses that include aerial photograph interpretation. Undergraduate and Masters students undertake dissertations on aspects of the aerial record intermittently, but there is certainly extra capacity in this area. The Scotland’s Rural Past initiative (http://www.scotlandsruralpast.org.uk/) has included basic familiarisation with aerial photographs in its training modules. The West Lothian Archaeology Group (http://www.armadale.org.uk/archaeologyintro.htm) is an active group experimenting with various survey techniques, including aerial photography, in a robust scientific framework.
Emerging opportunities, future research areas, and future needs
The systematic mapping of the cropmark record and its delivery as a web-mapping service is a great step forward and may encourage more routine engagement with this material. There are myriad cropmark record themes for which aerial photographs are the principal source, but amongst the key areas are the later prehistoric landscapes of the Tweed, the Moray Plain (where the uncatalogued Barri Jones archive of aerial photographs is a potentially invaluable resource).
The effective management of plough-levelled archaeology has attracted some research interest (Dunwell and Ralston 2008 a or b), but this too remains a rich seam, especially in the light of the adoption of the European Landscape Convention.
Airborne Multi-Spectral Prospection
Airborne Multi-spectral prospection can detect buried archaeological remains that are not visible, or readily apparent, in conventional aerial photography. It is a passive technique and records how the sun's energy is either absorbed, transmitted or reflected by the earth's surface depending on the physical and chemical properties of that surface.
Multispectral sensors record the reflected energy from a variety of wavelengths from visible light through to infrared. The infrared portion of the spectrum is extremely sensitive to subtle variations in surface vegetation, topsoil and topography caused by buried archaeological remains which are always detectable using to visible light.
Thermal images taken at pre-dawn and midday can be used to determine thermal inertia (the variation in these two temperatures) which can be influenced by archaeological remains with some retaining heat and others losing it more rapidly. Figures 38 to 40 are from a survey over an enclosure on the Isle of Coll. (see also Winterbottom and Dawson 2005).
Multi-spectral survey of an area on the Isle of Coll showing an enclosure in the centre of the image. This is a true colour composite using the 4, 3, 2 bands in the red, green and blue display respectively i.e. the visible light range. ©Dawson/Winterbottom/NERC
This is a false colour composite using the 9, 7, 3 bands in the red, green and blue display respectively i.e. the inferred light range. This combination maximises variations in vegetation and the structure, moisture content and nutrient content in the soil resulting in a shaper definition of the enclosure.©Dawson/Winterbottom/NERC
This is a daytime thermal image using band 11 of the spectrum i.e. infrared. This image enhances topographic variations and result in the turf banks and ridge and furrow being particularly visible. ©Dawson/Winterbottom/NERC