Predictive Landscape Portrayals: A Selective Research Review

Abstract

Landscape portrayals are widely used to simulate the appearance in perspective of proposed modifications to the landscape. This paper reviews some simulation-related research which may contribute to an analytical framework for creating and validating landscape portrayals. A number of criteria for evaluating simulations are drawn from two important approaches. The first is to analyze people's responses — that is, the equivalence of responses to simulations with responses to reality. The second is to analyze the simulation image — the accuracy with which a portrayal replicates actual visual properties. Various studies have examined simulation validity by using several response dimensions, and have found both high and low response equivalence. Relatively few studies have attempted to appraise image properties, although methods of doing this appear to be available. Sources of inaccuracy and the poor representativeness of portrayals are postulated as threats to simulation validity. Research directions of practical significance are suggested by the potential benefits of relating image properties to responses; this is the crucial step towards estimating the validity of landscape portrayals before the depicted development is constructed.