The Stour Valley Project - Aerial Archeology Research Group

The Stour Valley Project, England: a cropmark landscape in three dimensions

Introduction

The river Stour forms the boundary between counties of Essex and Suffolk in south-east England (Fig. 1). It flows along a band of alluvium, terraced valley gravels and glacial sands and gravels, in an area that is predominately Boulder Clay with pockets of underlying London Clay. The dense concentration of cropmark sites along the river valley was regularly flown from the 1950’s on by the Cambridge University Committee for Aerial Photography (CUCAP); the Royal Commission on the Historical Monuments of England (RCHME); and the Archaeology Section of Essex County Council. The latter included survey carried out in the exceptional conditions in 1995-6, which not only discovered new sites, but also afforded important additional detail at a number of important sites (Strachan 1996 and 1997).

Figure 1: The location of the river Stour running between Essex and Suffolk, and in relation to London and the East Anglian counties. Parts of the study area lie within the Suffolk Rivers Environmentally Sensitive Area (ESA) and the Dedham Vale Area of Outstanding Natural Beauty (AONB). The area also includes two Sites of Special Scientific Interest (SSSI). Management of the cropmark landscape is one of the long-term aims of the project.

Figure 2: An oblique image of a nucleated barrow cemetery appearing as a cropmark complex at Cavendish, Suffolk. The site includes a long mortuary enclosure and several ring-ditches, including two dual concentric examples, and is situated on a meander of the river. The photograph was taken in the June 1996 (photo: D.Strachan. Copyright Essex County Council).

The area was re-mapped between 1997-8 (Strachan and Ingle 1998 and 1999) by the Essex Mapping Project (Ingle and Strachan 1994) as part of the National Mapping Programme (Bewley 1995 and 1998), revising the existing 1:10,560 cropmark plots maintained by the SMRS. This process illustrated the extent and diversity of cropmark sites along the flood-plain, and in particular, the prehistoric “Monument Complexes” consisting of ring-ditch cemeteries (including large dual concentric examples); elongated enclosures (interpreted as long mortuary enclosures and long barrows); and the two cursus monuments at Wormingford and Stratford St. Mary. This current project arises from the preparation of a regional research framework for the eastern counties (Glazebrook 1997) and has been funded by English Heritage (EH) as part of the implementation of the Monuments at Risk Survey. The first stage builds on the work described above and involves the large-scale mapping of the cropmark landscape, and the use of a digital terrain model (DTM) within a GIS environment, as the basis for study and interpretation. This summary describes the methodology employed by the project, and it is hoped that results will appear in a forthcoming AARGnews.

Methodology

Air Photographic Sources

Air photographic collections held by the Essex Heritage Conservation Record (EHCR) and the SMR at Suffolk County Council were the primary sources used for image rectification. These collections contain a variety of sources including images by CUCAP, local flyers, RCHME survey and Essex County Council (Fig. 2). The CUCAP specialist oblique collection was also consulted, in addition to black and white Aerofilms vertical runs, at a scale of 1:12,000 (dating from June 1960 and September 1990) which are held by the Essex County Council Information Resource Centre.

Image rectification, interpretation and mapping

Selected images were scanned at a resolution of 300 dpi and imported into Aerial 5.5 software (Haigh 1996) for rectification based on OS land-line data accessed via the Essex County Council Arcview GIS. The residual error for rectification was kept below 5m, allowing future work, such as geophysics or excavation to be added to the GIS and confidently correlated with identified cropmark features. When insufficient control data existed on photography for a site, the NMP plot (accurate to less than 10m) was imported into GIS and used. The resulting rectified images were then geo-referenced and imported into the GIS, to be viewed with OS land-line data and allowing mapping of identified features (Fig. 3).

Figure 3. The rectified image imported into GIS and viewed with OS land-line data, including 5m contours. The interpreted cropmark plot of the barrow cemetery is shown, while the local geology and rectilinear landscape plots are present but not visible. The Drift Geology layer is also inactive. © Crown copyright 2000. All rights reserved. Use of this image is limited to viewing on-line and printing one copy. LA 076619.

The cropmark landscape consists of three main elements: circular, sub-circular and elongated enclosures (both rectilinear and curvilinear); the rectilinear landscape; and areas of local drift geology and palaeo-channels. These features were mapped on screen at a scale of 1:500, producing three polygonal vector layers (“Monument Complexes”, the “Rectilinear Landscape” and “Geology”) which can be viewed independently or in combination. In addition to these, a number of other data sets, from a variety of sources, have been added to the GIS. These include 1:25,000 drift geology data (partly made available by Wessex Archaeology) and find-spot information from the SMRs (divided by period: Mesolithic, Neolithic, Bronze Age, and Iron Age). A total of 334 individual sites were recorded in the “Monuments Complex” class.

Viewing sites in the landscape and view-shed analysis

The resulting plots and associated data sets can be viewed at a variety of scales (ranging from individual sites at e.g. 1:1,250 to distributions at around 1:200,000). A site, or group of sites, can therefore be viewed against geology and find-spots, and in relation to other sites and topography (in the form of the 5m contours). In addition, however, the 5m contour data can be used to create a Triangulated Irregular Network (TIN), or surfaced elevation model, on which the data can be draped. The 3D Analyst and Spatial Analyst extensions of Arcview allow the inter-visibility of monuments to be studied by selecting a position on, or above the DTM, and carrying out “line of sight” (LOS) or view-shed analysis. The former determines what is visible on a surface, looking in a certain direction from a single point, while the latter shows areas on a surface visible from one or more observation points. View-shed analysis is generally more useful as an aid to studying an archaeological landscape as it reveals other sites which share visibility with the observation point. The process results in a “flood diagram” which shows areas visible from that position and areas which are obscured by terrain (Fig. 5). Cumulative view-shed analysis (CVA) can be carried out using a number of different positions, or sites, to create a model of inter-visibility. While the model does not take into account vegetation cover, it is a valuable tool for the analysis of the relationship between sites and their topographical setting. The principles and methodology for CVA have been described in a study of the inter-visiblility of long barrows in the Stonehenge and Avebury areas (Wheatley 1995). The technique requires only a suitable DTM and the location of sites in the landscape, and can therefore be easily applied, and is likely to produce archaeologically significant, and often unexpected, results. Complex studies can be quickly carried out over large areas and on contours where it would be difficult to judge inter-visibility by contour lines alone. It is also invaluable in areas where modern development may obscure visibility and remove the possibility of checking visibility on the ground.

Figure 4. The Cavendish complex viewed with neighbouring sites, including a complex at A, along with the 5m contours, the river, and the rectilinear landscape. © Crown copyright 2000. All rights reserved. Use of this image is limited to viewing on-line and printing one copy. LA 076619.

Figure 5. As per Fig. 4 but including view-shed analysis from a position 3m above a central ring-ditch in the Cavendish complex. The black areas indicate were visibility from that point is obscured by terrain, including the complex at A. Other sites are shown to be inter-visible. © Crown copyright 2000. All rights reserved. Use of this image is limited to viewing on-line and printing one copy. LA 076619.

In addition, TINs can be viewed used as the basis for 3D visualisation, and viewed in three dimensions, along with the cropmark plots and other data, in perspective from a selected static 3D position, or in rotation (Fig. 6). This feature is useful for viewing sites in their topographical setting, and the software allows for vertical exaggeration of the DTM (which can reveal topographical subtleties) and control over the suns azimuth and altitude. In addition, when viewed from ground level, the models display the spatial organisation of monument complexes within their topographical setting. The potential for this technique as an aid to landscape interpretation is evident, but is more likely to be significant when combined with a morphological study of the sites mapped. The project has involved simple morphological analysis of sites within the “Monument Complexes” class, allowing distributions and cumulative view-shed analysis to be carried out on selected site-types. It is hoped to include details of this process, along with results, in a future AARGnews and the AARG web-site.

References

Bewley, R. 1995 “A National Mapping Programme for England.” In Luftbildarchäologie in Ost- and Mitteleuropa. Forschungen zur Archäologie im Land Brandenburg, 3, pp 83-92. Potsdam.

Bewley, R. 1998 “England’s National Mapping Programme: A Lincolnshire Perspective.” In Lincolnshire’s Archaeology from the Air (ed.) Bewley, R. (1998) Occasional Paper in Lincolnshire History and Archaeology 11. pp 9-17. Gainsborough.

Edis, J., MacLeod, D. and Bewley, R. 1989 “An archaeologist’s guide to classification of cropmarks and soilmarks.” Antiquity 63, pp 112-26.

Fassbinder, J. and Irlinger, W. 1999 Archaeological Prospection. Third International Conference on Archaeological prospection. Arbeitshefte des Bayerischen Landesamtes für Denkmalpflege, Band 108.

Glazebrook, J. 1997 Research and Archaeology: A framework for the Eastern Counties 1. Resource Assessment E. Anglian Archaeol. Occ. Pap. 3.

Haigh, J.G.B. 1996 “Another member of the AERIAL software family.” AARGnews 12, newsletter of the Aerial Archaeology Research Group.

Ingle, C. and Strachan, D. 1994 National Mapping Project 1993. in “Work of the E.C.C. Archaeology Section” (ed.) Bennett, A. Essex Archaeology and History 25, pp 233-8.

Strachan, D. 1996 Aerial Survey in 1995. In “Work of the E.C.C. Archaeology Section” (ed.) Bennett, A Essex Archaeology and History 27,

Strachan, D. 1997 Aerial Survey in 1996. In “Work of the E.C.C. Archaeology Section” (ed.) Bennett, A Essex Archaeology and History 28,

Strachan, D. and Ingle, C. 1998 Essex Mapping Project 1997. in “Work of the E.C.C. Archaeology Section” (ed.) Bennett, A. Essex Archaeology and History 29, pp 186-8.

Strachan, D. and Ingle, C. 1999 Essex Mapping Project 1998. in “Work of the E.C.C. Archaeology Section” (ed.) Bennett, A. Essex Archaeology and History 30.

Wheatley, D. 1995 “Cumulative viewshed analysis: a GIS-based method for investigating intervisibility, and its archaeological application.” in (ed.’s) Lock and Stancic Archaeology and Geographical Information Systems: A European Perspective. London.