The Historic Landscape Project - Background Information ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Archaeologists in Hampshire County Council's Planning Department are trying to assess the range and density of archaeological sites on council owned farm and recreational land, by taking systematic samples of the archaeological material incorporated in the plough soils in those areas. The basic units of information in this project are derived from artefacts gathered along systematically spaced transects. A variety of collection techniques, including quadrat provenance and shovel-testing methods, are used to cope with different and sometimes difficult landscape conditions. These data are entered into the Winchester Graphics System for analysis so that the data exploration procedure outlined above can be pursued. Sites are defined within the computer, on the basis of mapped density characteristics, as discrete high-density areas of artefacts. Originally two-dimensional plots would have been used. However, it was postulated that presenting the data as a two-dimensional plan disguised the major influence of topography on the observed artefact distributions. Clearly, the analysis of these data would improved if they could be mapped onto three-dimensional terrain models. This approach also facilitated the production of new useful questions concerning how much land usage and topography actively influenced distribution patterns. Terrain Modelling ~~~~~~~~~~~~~~~~~ To investigate the relationship between the local terrain and the distribution of finds in an area it was necessary to build models of the terrain. Not having the resources to perform level surveys for the large tracts of land involved in this project, contour data is digitised from maps and loaded into WGS where it can be managed more easily. The relevant contour data are extracted and then interpolated onto a regular grid as the basis of a terrain model. The various artefact data collected from sample area are also interpolated on to the coordinate grid. The terrain is represented by a shaded surface and the artefact distributions are represented by colour-coded or grey-scaled contours mapped on to the surface of the terrain-model. Although high quality shaded models are sometimes useful for exhibition purposes, the wire-framed models have the advantage of being susceptible to real-time viewing transformations, such as rotations, pans, or zooms. Once again, these vector models are excellent for analysis, and interesting activity areas can be recognised more quickly. For example, the distribution of prehistoric chipped flint flakes over a parcel of HCC owned land in Fair Oak, in Hampshire have been represented using this technique. On first inspection it would be fair to infer that three distinct activity areas have been revealed. However, if the same data are looked at from a different angle a significantly different interpretation is possible. The height readings, taken from the base map have been accentuated to emphasise the relief in the terrain model, but exactly the same data as before, have been mapped onto this surface. The same is true of the distribution of modern pottery sherds over the same piece of land. When looked at in plan three activity areas, or concentrations of material, are noticeable. However, two of the activity areas which previously looked discrete are in fact probably residues of a single concentration of material deposited at the crest of the slope. Seemingly some of this material has spilled down the gradient and collected in a slight hollow at the bottom of the slope. Instead of inferring two separate activity areas, or perhaps a single large activity area with two main focal points, this picture prompts the query as to whether the high density represents only a small site, material from which has been carried down the slope. One is also prompted to hypothesise a likely cause of the apparent down-slope movement. Such three-dimensional computer graphics are potentially very important tools with which to deal with the some of the problems associated with strategic planning and cultural resource management. Providing better quality summaries of survey data over wider areas can only help to improve the ability of planners to provide and allocate resources in the correct priorities.