High-resolution satellite imagery in archaeological application: a Russian satellite photograph of the Stonehenge region.

Fowler, Martin J.R.

The spy satellites - by repute of the thriller writers - have such good image-resolution that they can read the letters on a vehicle licence-plate. A generation after LANDSAT imagery vividly showed broad ecological zones, higher resolution pictures are now being released of a quality to allow practical archaeological application. The example printed here illustrates the Stonehenge landscape - a little patch of southern England that is among the most photographed archaeologically anywhere.

Archaeological potential of satellite imagery

Since the monument at Stonehenge was photographed 90 years ago from a balloon (Capper 1907), a light aircraft flying at low altitude has become the usual platform for aerial photography for archaeological studies. With the launch in 1972 of the first of the American LANDSAT series of satellites, a new source of remote-sensed data became available to the archaeologist in the form of satellite imagery from low Earth orbit.

The first LANDSAT satellites produced low resolution digital imagery with a pixel size approximating to 80 m on the ground. Ten years later, with the launch of LANDSAT 4 carrying the Thematic Mapper (TM) sensor, the resolution improved to 30 m and in 1986 the launch of the French SPOT satellite provided products with resolutions of 20 and 10 m. Since the break-up of the Former Soviet Union, high-resolution imagery has become available with nominal ground resolutions of 3-4 m or better (Fowler 1995a) and imagery from the early US spy satellite programme is also becoming available (McDonald 1995).

With early and low-resolution imagery, it was possible to see only very large archaeological structures such as the Pyramids at Giza (Quann & Bevan 1977); archaeological applications have attempted to exploit the multi-spectral nature of the LANDSAT and SPOT products through the use of image classification and modelling techniques (Ebert & Lyons 1980; Ebert 1989; Allen et el. 1990). In Britain, LANDSAT and SPOT images have been used to study areas of archaeological potential in the Cumbrian peatlands (Cox 1992), archaeological features in the eastern Fenlands (Shennan & Donoghue 1992), linear features such as Roman roads 'fossilized into the modern landscape around the Danebury hill-fort (Fowler 1994a), and the circular earthworks of the Figsbury Rings hill-fort on the Wessex chalklands near Salisbury (Fowler 1994b).

Satellite imagery of the Stonehenge environs

The immediate Stonehenge region includes upstanding features, crop- and soil-marks in relatively uncluttered grassland and arable. This region is well documented archaeologically (RCHM(E) 1979; Richards 1990; Chippindale 1994; Cleal et el. 1995) and represents a good area to see what archaeological features can be visible on satellite imagery.

An earlier study (Fowler 1995b) demonstrated that LANDSAT TM Band 4 imagery, covering the near-infrared region of the spectrum at a resolution of 30 m, shows areas of dark tone associated with round barrows in reserved areas of older grassland which have different spectral signatures to the surrounding fields. These areas are also visible to an extent on SPOT Panchromatic imagery (10-m resolution), where the highly-reflective visitor footpath around Stonehenge itself is most apparent. Apart from these examples, the low resolutions of the LANDSAT and SPOT products used in the study precludes the direct identification of archaeological features present in the area. Recently released Russian imagery, however, is comparable in detail with medium-scale vertical air photography and hence has the potential of directly detecting archaeological features.

KVR-1000 image and processing

Russian KVR-1000 satellite imagery is taken from an altitude of 210-235 km with a camera recording an area of 40x300 km on film at a nominal ground resolution of 3-4 m or better (Fowler 1995a). The panchromatic film, covering the spectral range 0.51-0.76 m, is subsequently scanned for distribution as digital image products.

An image covering an area some 4.7x6.3 km to the east of Stonehenge photographed one early morning in June 1993 was kindly provided by Nigel Press Associates Ltd (Edenbridge, Kent) for the cost of writing the digital image to film and making a 1:25,000 photographic print. Scanned at a resolution of 300 dots per inch, the resulting images have a ground pixel size of about 2 m (compared with 1.4 m for the original digital product). The digital images were displayed and contrast enhanced using CorelDraw! 3.0[TM], with overlays of features manually transcribed and exported into Map Viewer[TM] for the production of interpretative maps. Two extracts of the image are interpreted below.

Stonehenge and immediate environs to the east [ILLUSTRATION FOR FIGURES 1 AND 2 OMITTED]

The circular bank and ditch of Stonehenge are readily traced as highlights and shadows and the central circular stone setting can be discerned, but not its individual stones. Leading away to the northeast, the Avenue is seen as parallel dark and light lines over some 500 m. Within it, close by the modern road, the Heel Stone is seen as a highlight, with bell barrow 11 upstanding to the east. The light-toned visitor footpath is again prominent, and traces of older tracks to the monument are apparent as dark lines (they are visible on the original 1906 air photographs: Capper 1907).

Northeast of Stonehenge, across the open ground, are the New King Barrow group, standing monuments within woodland; four are seen as highlights and shadows. East of Stonehenge, a circular crop-mark is the site of Coneybury henge, totally flattened by ploughing. Other circular and linear crop-mark features can be spotted in the vicinity - where ridge-and-furrow is reported visible on some air photographs (RCHM(E) 1979: 13). Of more recent origin are the 18th-century earthworks of an unfinished causewayed road crossing Stonehenge Bottom visible in relief. To the northeast of these earthworks, a square of light streaks is characteristic of fodder for grazing animals as seen on air-photographs (cf. Wilson 1982: figure 110).

Boreland Hill [ILLUSTRATION FOR FIGURES 3 & 4 OMITTED]

Prominent on this image extract are the 'Celtic' field systems, visible as soil marks and in relief, that are adjacent to two enclosures on Boreland Hill. The northern enclosure, a dark oval crop-mark some 100x150m oriented north-south, is visible on air-photographs taken in 1976 (National Monuments Record SU 1137/3/145-47); the southern, a similar dark oval oriented east - west, was photographed as a soil-mark that same dry summer (SU 1137/1, SU 1137/2/141-4).

Discussion

Considerably better than the low resolution LANDSAT and SPOT images, the KVR-1000 images are comparable with orthodox medium scale vertical air photographs and show both standing monuments and - for the first time - crop- and soil-marks. As the satellite records an area of 40x300 km, a single image theoretically can cover a whole study-area (e.g. Palmer's (1984) around Danebury). At a cost of some [pounds]2500 for a 40x40-km image, this could be a cost-effective means of carrying out a first pass investigation of a new study area. However, it should be recognized that the Stonehenge area is relatively uncluttered and the particular KVR-1000 image may be of a lucky quality. Further work is required to assess the quality of the typical Russian product and its suitability to those many regions of the world not explored by orthodox air photography.

Acknowledgements. I thank Rog Palmer and Chris Cox for their encouragement of this work, Bob Bewley for identifying the aerial photographs of the enclosures on Boreland Hill, Hilary Needham for commenting on an earlier version of this note and Nigel Press Associates Ltd for kindly providing the satellite image.

References

ALLEN, K.M.S., S.W. GREEN & E.B.W. ZUBROW (ed.). 1990. Interpreting space: GIS and archaeology, London: Taylor a Francis.

CAPPER, J. 1907. Photographs of Stonehenge, as seen from a war balloon, Archaeologia 60: 571, plates 69-70.

CHIPPINDALE, C. 1994. Stonehenge complete, London: Thames & Hudson.

CLEAL, R.M.J., K.E. WALKER & R. MONTAGUE. 1995. Stonehenge in its landscape: twentieth-century excavations. London: English Heritage. Archaeological Report 10.

COX, C. 1992. Satellite imagery, aerial photography and wetland archaeology, World Archaeology 24: 249-67.

EBERT, J.I. 1989. Techniques, methods and theoretical goals in American archaeological remote sensing: 'Predictive modelling' as an example, in D. Kennedy (ed.), Into the sun: essays in air photography in archaeology in honour of Derrick Riley: 86-101. Sheffield: J.R. Collis Publications.

EBERT, J.I. & T.R. LYONS. 1980. Remote sensing in archaeology, cultural resources treatment and anthropology: the United States of America in 1979, Aerial Archaeology 5: 1-19.

FOWLER, M.J.F. 1994a. Danebury and its environs from 830 kilometres, Hampshire Field Club and Archaeological Society Section Newsletters new series 21: 26-30.

1994b. Satellite image processing for the archaeologist, Archaeological Computing Newsletter 39: 2-8.

1995a. High resolution Russian satellite imagery, AARGnews 11: 28-32.

1995b. Detection of archaeological features on multispectral satellite imagery, AARGnews 10: 7-14.

MCDONALD, R.A. 1995. Opening the Cold War sky to the public: declassifying satellite reconnaissance imagery, Photogrammetric Engineering and Remote Sensing 61: 385-90.

PALMER, R. 1984. Danebury, an Iron Age hillfort in Hampshire: an aerial photographic interpretation of its environs. London: RCHM(E).

QUANN, J. & B. BEVAN. 1977. The Pyramids from 900 kilometres, MASCA Newsletter 13: 12-14.

RCHM(E). 1979. Stonehenge and its environs: monuments and land use. Edinburgh: University Press.

RICHARDS, J. 1990. The Stonehenge Environs Project. London: English Heritage. Archaeological Report 16.

SHENNAN, I. & D.N.M. DONOGHUE. 1992. Remote sensing in archaeological research, Proceedings of the British Academy 77: 223-32.

WILSON, D.R. 1982. Air photo interpretation for archaeologists. London: Batsford.