2 Aerial Photography, Remote Sensing and Archaeology

2.1 Aerial survey for archaeology¶

This section presents a brief overview of the history and role of aerial photography and remote sensing in archaeology. It discusses the principal methods employed.

2.1.1 A brief history of the development of aerial photographs for archaeology¶

Air photographs may reveal archaeological sites directly, where they are extant, or as crop, soil or other surface indications where the site is buried. Taking photographs from the air began with balloons in the nineteenth century, and has developed as a means of landscape survey during the twentieth century using aircraft, mainly aeroplanes and helicopters, and more recently using satellite imagery. One of the earliest aerial photographs of an archaeological site was taken from a balloon in 1906, and shows Stonehenge. Between the two World Wars the technique was developed by a number of British, French and German scholars in Europe, the Middle East and the Mediterranean. In Britain, the major pioneers during the 1920s and 1930s were O.G.S. Crawford, and Major Allen. During the Second World War, aerial photography was developed for reconnaissance purposes, and air photo interpretation became a technique requiring highly skilled staff. Derrick Riley, bomber pilot and later a flying instructor at Stanton Harcourt, had the opportunity to fly extensively over the Thames Valley where he observed, and sometimes had the opportunity to photograph, archaeological sites. Many archaeologists were employed as air photo interpreters during the war, and Crawford was able to make successful requests to the RAF (during the war) for reconnaissance flights to record cropmarks (most notably Verulamium, St Albans in Hertfordshire).

For the majority of the twentieth century the majority of work focused on data capture – taking and collecting aerial photographs, although Crawford was a pioneer in making maps of archaeological sites from aerial photographs from the start. After the Second World War, and especially in the 1950s, the interpretation of the information derived from aerial photographs slowly became as important as aerial reconnaissance itself. Pioneering projects which examined the impact of gravel extraction on archaeological sites in river valleys resulted in the seminal publication A Matter of Time RCHME, 1960 and in mapping projects in the Thames Valley and other parts of Britain. The formation of the Cambridge University Committee for Aerial Photography (CUCAP, in 1949) and the RCHME’s Air Photography Unit (in 1965) provided a solid foundation for the discipline, going beyond data capture with a series of publications by St Joseph and colleagues.

Aerial reconnaissance and interpretation are now routine methods in the UK and in many parts of Europe for the discovery and understanding of archaeological sites. In England, a National Mapping Programme has been running since 1993 England, 2023, and there are targeted mapping projects in Scotland and Wales. Other European countries have had varying degrees of success with aerial survey, but France, Germany, Czech Republic, Poland, Italy and Britain have been particularly successful. Since 1990, and the end of the Cold War, many countries of the former Eastern Bloc are now able to undertake surveys, and aerial survey is becoming a much more widely used technique.

2.1.2 Types of aerial photograph¶

A variety of aerial photographs exist in terms of format, film types, and print sizes. The major distinction for archaeological purposes is between vertical photographs (taken by commercial companies for mapping purposes, census information, or other tasks, as well as historic verticals taken by the RAF) and oblique photographs, taken for specifically archaeological purposes.

There are many sources of vertical photographs (see Appendix 1), and national collections of historic vertical photographs are held by English Heritage, the Royal Commission on the Ancient Historical Monuments of Scotland (RCAHMS) and by the Welsh office for Wales. An organisation, NAPLIB, has also produced a directory of all holders of aerial photographs NAPLIB, 1993.There are also many collections of oblique photographs; the main ones for archaeological research are the three Royal Commissions (England, Scotland and Wales) as well as what was CUCAP but is now the Unit for Landscape Modelling. Many county councils also have collections of vertical and oblique photographs.

The distinguishing characteristics between vertical and oblique photographs are as follows:

• Verticals require less geometric transformation when creating archaeological maps but usually they have not been taken in optimum conditions for the archaeological interpretation of sites. Oblique photographs are more often taken for archaeological purposes, and therefore contain information relating to archaeological sites, but the oblique angle means that to create maps a more complex transformation system (either computerised or manual) is required.

• For both types of photograph the same air photo interpretation skills are required, but vertical photographs are more often capable of being used in stereo pairs. The stereo image enhances the opportunities for a confident interpretation. Oblique stereo images are taken by some aerial photographers but are less common.

The rapid developments in digital photography have meant that in the last few years there is little “film” photography done in the UK and it is predominantly digital; all the major aerial reconnaissance teams use only digital cameras. The majority of aerial photographs (until the 1980s) were taken on black-and-white film, but colour photography (for both verticals and obliques) became much more common (in either print or slide format). Infra-red (both black-and-white and false colour) films are also used but costs and difficulties in both storage and handling have meant that their use is not as widespread as it might be, given the good results. Apart from the commercial vertical survey companies, who use calibrated vertically mounted cameras, the majority of archaeological aerial photographers use either 35mm hand-held cameras or medium format cameras Bewley, 1993.

2.1.3 Mapping data derived from aerial photographs¶

Common practice is for digital rectification of the information contained on aerial photographs from the oblique (or vertical) photograph to a map or plan. There are a number of types of interpretative mapping data in use and for more details see the relevant websites. English Heritage is producing a manual for its National Mapping Programme, in the near future.

• paper or film formats of rectified plans derived from interpretations of aerial photographs. These have usually been created manually or via a computerised mapping system and drawn up by hand using ink on a translucent film. Scales used vary from 1:500 to 1:25,000. This type of plan represents the majority of data to 1998.

• digital (rectified) plans in an interim form derived from aerial photographic interpretations. These are generally used as the basis for the above.

• digital (rectified) plans which are in a final form, having been created using digital rectification methods and computer-aided design packages to create the final plan. This is standard practice throughout Britain, in English Heritage and the two Royal Commissions and in commercial mapping companies.

• (unrectified) sketch plots, where the archaeological features have been traced from a photograph but no rectification programme was applied. These are few in number and should not be used for archaeological evaluation purposes.

A variety of computers linked to stereo mapping machines are available for detailed photogrammetric survey (as used by the RCHME, and in various parts of Europe, especially Austria and Italy), but the majority of machines are stand-alone PCs attached to a digitiser and plotter. Details on the specification of the hardware and software used can be supplied on request from any of the organisations listed under “aerial photography mapping and interpretation” section of Appendix 1. For further information on commercial sources of both photographs and digital data relating to photographs, maps and plans, please also see Appendix 1. Further details regarding digital vector mapping can be obtained from the GIS and CAD Guides.

Finally, it is important to emphasise that the maps, either in digital or paper form, are best used in conjunction with archaeological records which contain the description of the sites mapped and interpreted. There are a number of databases which provide such information: the national records in England, Northern Ireland, Scotland, and Wales; local HERs (Historic Environment Records); and commercial archaeological mapping companies. The basic information required for any site is its location (physical and administrative), its date, a short description, and an interpretation.

2.2 Digital Remote Sensing¶

2.2.1 The development of digital remote sensing¶

Remote sensing for archaeology is still largely based on low altitude aerial survey techniques at optical and near infra-red wavelengths. The most important developments since the 1970s were the introduction of digital multispectral imaging sensors, thermal imaging radiometers,imaging RADAR and more recently Lidar. These sensors have increased the wavelength range at which images can be acquired into the short wave, middle (thermal) infra-red, and microwave range. The data are produced in a digital form that can be enhanced, rectified and reclassified using a wide variety of algorithms and specialist software. The output of such procedures can be used to identify a wide variety of features which may be of interest to researchers.

Until very recently available civilian optical satellite imagery has been of low spatial resolution and of limited use in archaeology. There have been a few notable studies that have demonstrated its value in particular circumstances. However, new developments have altered the value of spaceborne satellite imagery. Both the Russian and American military spy satellite programmes have released photography with up to 2m ground resolution and such digital imagery has good geometric fidelity and can easily be integrated into Geographical Information Systems. Other important developments have been made in RADAR imaging since Charles Elachi first showed that NASA’s Shuttle Imaging Radar (SIR) could penetrate very dry and smooth sand surfaces to reveal buried structures Elachi, 1982. The Shuttle radar system has been considerably enhanced and may soon be followed by an orbiting radar sensor.

2.2.2 Current developments¶

Remote sensing for archaeology has developed swiftly in the past decade and perhaps the greatest change has been in the use of lidar imagery using laser beams to record the landscape, (Light detecting and ranging) Crutchley & Crow, 2009. There are important new developments in space imaging that may see vertical aerial photography replaced by high resolution satellite digital images. Airborne sensors are improving all the time and becoming much cheaper to build and deploy. Thermal and RADAR imaging sensors have also improved recently and these offer considerable potential for detecting buried structures. Satellites also provide accurate positional information on the Earth’s surface using the US Navstar Global Positioning System (GPS) or the Russian GLONASS which is especially important for the accurate mapping of relatively small-scale archaeological features.

The science of image processing has also developed very rapidly since its birth in the 1960s at NASA’s Jet Propulsion Laboratory. It is now possible to convert prints and films from analogue to digital form relatively cheaply, and to use widely available processing techniques geometrically to correct or visually enhance the imagery. High quality image scanners are now relatively cheap to buy and so digital image processing can be used to improve the geometry or visual quality of oblique and vertical aerial photography as well as digital sensor data Scollar et al., 1990.

2.2.3 Sensor types and their use¶

Airborne multispectral scanners

Archaeological structures can usually be seen well from the air while crop and soilmarks are more difficult to detect with certainty. The visibility of cropmarks in particular depends on vegetation type, illumination and soil conditions. Multispectral remote sensing is able to look simultaneously at a wide range of different wavelengths, many of which are more sensitive to vegetation status than either the human eye or photographic film. The limited spectral range of photographic film (350-1100nm) is overcome by the use of photoelectric sensing devices that record their data in a digital form. These devices separate light into a number of discrete narrow wavebands, hence the term multispectral. This has the advantage of allowing scientists to look at particular parts of the wavelength range in isolation, or to combine different wavelength ranges that are of particular interest.

In the United Kingdom, research work has been undertaken by the Natural Environment Research Council (NERC) in part of the Fenlands Donoghue & Shennan, 1988Donoghue D.N. & Shennan, 1988Shennan & Donoghue, 1992, the Nottingham area Allsop & Greenbaum, 1989, the Vale of Pickering Pryor et al., 1992Powlesland et al., 1997Powlesland & Donoghue, 1993, Wroxeter (unpublished) and the Welland and Nene valleys (unpublished) to assess the archaeological potential of multispectral data. In the United States, a number of studies have been undertaken to record the archaeology and environment at Chaco Canyon in New Mexico and other sites Lyons & Avery, 1977Lyons & Mathien, 1980Avery & Lyons, 1981. These studies were undertaken in areas unsuitable for the formation of crop and soilmarks and have demonstrated the value of near infra-red and thermal infra-red imagery in enhancing the archaeological record.

Airborne thermography

Thermal prospection techniques have many important applications in geology, archaeology and environmental monitoring. Both pre-dawn and daytime thermal imagery have proved valuable in detecting or providing additional information on buried archaeological structures. Unfortunately, much of the information content in thermal imagery is hidden by the relatively uniform temperature of the ground surface. However, heat flow is governed by time, depth, density, heat capacity and thermal conductivity, all properties relevant to ground disturbance. Pre-dawn and mid-day images can be combined to compute the diurnal heat capacity (otherwise termed the thermal resistance to temperature change or apparent thermal inertia) of the ground. Apparent thermal inertia (ATI) offers considerable potential for the detection of buried archaeology. Seminal work by Tabbagh has illustrated the value of these techniques Tabbagh, 1976Tabbagh, 1979Scollar et al., 1990.

In the UK, NERC operates an airborne sensor capable of thermography. Other sources of thermal imagery include analogue sensors flown by the UK Royal Air Force and Police. Although not widely available, these data have been used to detect ground disturbance associated with archaeology. Airborne thermal radiometry can cover large areas at low unit cost and provide digital data that can be combined with other sources of information.

Optical imagery from space

Optical space photography dates back to the first manned space flights. However, systematic coverage of the Earth is obtained from orbiting satellites that use conventional film or digital imaging devices. Relatively high resolution panchromatic imagery can be obtained from the French SPOT-3 satellite (10m) and the Indian IRS system (5m). The best available Russian photography is from the KVR-1000 camera where the photographic film is digitised to provide 2m resolution over a 40km by 40km area. The best American data come from commercially available digital camera systems. Earlybird (3m), Carterra-1 (1m), Orbview-3 (1-2m) and Quickbird (0.82m) are either just recently launched or soon to be launched. An illustration of this type of imagery is given by Fowler 1996 for Wiltshire. Some of these new systems have a limited multispectral capability that may be of value.

Radar imagery from space

The value of spaceborne radar imagery has been its ability to penetrate very dry, smooth sand and tropical forest canopies to reveal archaeology beneath. Data from NASA’s Shuttle Imaging Radar (SIR) system have been used to demonstrate the existence of palaeo-channels in desert areas of the USA, north Africa and the Middle East McCauley et al., 1982. The same system has been used to reveal important archaeology hidden under dense forest canopy, such as Mayan canals in Mexico and Guatemala and around the city of Angkor in Kampuchea. There have only been four SIR missions. The last was in 1994, and so the data is only available for a part of the Earth. Spaceborne radar works independently of weather, day and night, because it supplies its own energy to illuminate the ground. Most radar sensors have poor spatial resolution and are not useful for directly detecting archaeology. However, they are good at many forms of environmental mapping.

Ground-based imagery

The archaeological examination of standing structures, such as historic buildings and monuments, may be aided by the use of ground-based remote sensing techniques to enhance or elucidate concealed information. Typical applications employ multispectral imaging to provide geological and construction phase differentiation in walls and related building features, laser contour profiling and thermal prospection to reveal features hidden under plaster.

In the UK, research work has been undertaken at the University of Nottingham Brooke, 1989, brooke1994a examining historic buildings and monuments over a wide area and timescale. In Germany, thermal prospection has been employed successfully for many years to reveal hidden building construction and concealed features Cramer, 1981.