Sampling the countryside of Great Britain: GIS for the detection and prediction of rural change

In 1990, the lnstitute of Terrestrial Ecology completed a survey of the land use and vegetation of Great Britain (GB), repeating and extending the earlier surveys of 1978 and 1984. The surveys, which were structured using the ITE Land Classification, were funded by Government departments and the information collected is being used to assist in central decision making. GIS is being used at several different levels: checking information recorded in the field; incrporating remotely sensed imagery; comparing individual survey sites; linking other census; and surveyed information and presenting the results. The survey uses the Ordnance Survey of GB (0S) 1 km square grid, and collects information on all types of land use. The three dominant forms of land use, agriculture, forestry and urban. are recorded in formats which are as compatible as possible with those used by other specialist organizations (e.g., Forestry Commission. Ministry of Agriculture, Food and Fisheries (MAFF) and Department of the Environment), which allows information collected by different groups to be integrated and data to be exported in accepted styles. The scale selected has proven successful for use at both regional and national levels. The Land Use Research Group at Merlewood has a working knowledge of the requirements of policy advisers and can present ecological information in a variety of relevant ways. Maps and summary statistics are easily produced but estimates of accuracy can raise doubts and lead to results being unfairly undervalued. It is important that GIS is not seen as a system which can only give absolute answers; it must be able to qualify its boundaries and statistics in a comprehensible way. There are a number of statistical methods which can be applied to land cover data, but the presentation of the cartographic results is always the most problematic aspect. The limits of the production and interpretation of error terms are described, detailing elements which are assessed and those which are omitted. The surveys are a good example of a practical way of collecting and passing information between a variety of sources and disciplines. Results and methods of presentation are discussed.