Defining wild land in Scotland through G.I.S. based wilderness perception mapping

Abstract

In a country such as Scotland, there is little land, if any, that has escaped the influence of humankind in one way or another. The degree of human influence varies along a continuum from, for example, the city centre office block, to a very remote mountain hilside. Despite the effects of human influence, the landscapes at the latter end of the continuum are stil perceived by many as wild and are relatively untouched. Wild land is valued for both utilitarian and intrinsic reasons and provides a range of benefits. However, owing to the subjective nature of current wild land definitions, these values and benefits are rarely taken into account in current land use management and new developments. The aim of this study was to define in spatial terms the concept of wild land in Scotland using people's visual perceptions of the landscape. This was achieved through the development of a method to define objectively wild land by quantifying the wildness of a location based on the surrounding landscape attributes. The main objectives of the study were an assessment of the physical and perceptual characteristics of wild land, the examination of the current wildness of a range of Scottish areas which in turn enabled the stability of wild land perceptions over time to be evaluated in comparison with existing data. The perceptual nature of wild land necessitates a multidisciplinary approach and requires a broad range of opinion to be consulted in its definition. The use of a photographic questionnaire enabled the views of those living nearby and visiting potential wild land areas to be gathered. The photographs represented the range of characteristic landscape attributes within the two study areas of the Cairngorms and Wester Ross, and were rated for their wildness. The extent of visible landscape attributes was quantified using a geographical information system (G.I.S.) and was used along with wildness ratings to develop and test predictive wildness models using multiple linear regression techniques. Wildness models were then applied within each of the two study areas, producing maps of wild land that could then be used in decisions on future planning and conservation issues.