Of particular importance to the study of large-scale phenomena in physical geography is the modifi able areal unit problem (MAUP). While often viewed as only a problem in human geography (particularly demographic studies), the MAUP is an issue for all quantitative studies in geography of spatial phenomena (Openshaw and Taylor, 1979). Increasingly, remote sensing and Geographic Information Systems (GIS) are being used to assess the distribution of phenomena from a large scale. These phenomena are modelled using areal units that can take any shape or size resulting in complications with statistical analysis related to both the scale and method used to create the areal units. In this paper, we defi ne the modifi able areal unit problem, present examples of when it is a problem in physical geography studies, and review some potential solutions to the problem. Our aim is to increase awareness of this complicated issue and to promote further discussion and interest in this topic.

No abstract