Abstract
Strongly asymmetrical spatial flows are analyzed by using as example the flows of news between major Canadian cities. Conceptually such flows can be modelled by combining a vertical dimension of dominance with two spatial dimensions. The conceptual model is interpreted mathematically by appropriate use of the attraction and emissivity terms in the spatial interaction model. A family of models is identified and calibrated using the Canadian data, allowing the positions of cities on the dominance dimension to be identified. By inverting the model it is possible to scale positions in the spatial dimensions to reflect interaction more correctly. The result is a three-dimensional image which gives a ‘fuzzy’ interpretation to the conventional topological tree model of a hierarchy. Relationships with Tobler's ‘winds' model are examined.
Get full access to this article
View all access options for this article.