Automating the analysis of remotely-sensed data

Land cover is a complex phenomenon, its appearance and transition influenced by the soil, topography, climate, politics, ecology, and other aspects of land use. An operational Study of land-cover change must consider such factors, and, if automation is to be employed, a new approach to software architecture and reasoning is required. The Macaulay Institute has developed the concept of task orientation as a means of supporting such work, and is currently investigating its utility in the update of the Land Cover of Scotland (1988) dataset. This paper describes the motivation behind the development of ETORA-II, an operational toolkit providing a task-orientated capability, and SYMOLAC-II, a proof-of-concept application. The final aim of this work is to produce an environmental information system for Scotland's land cover.