ABSTRACT
The former are mainly restricted to analogue processe and will not be di cussed further. In this chapter we outline some of the ways in which the latter can be used to improve our ability to model geographical distributions in urban areas. Any such use of remotely-sensed data in urban analysis must take into account the nature of these data . The limits to the integration of remotely-sensed data in spatial analysis within a Geographical Information System (GIS) have often been highlighted in the literature (Mace 1991 , Ehlers 1995). The e result from :
• the relevance of the attainable land cover/land use classification to the purposes of the study;
• the format of the digital data, which may create difficulties relating to its amount ( ize) and to its grid (raster) discretization of space; and,
• the measurement scale of the studied attribute. For example, land use/land cover i defined on a qualitative scale for which the number of possible mathematical and statistical operations is restricted .
