Quantifying the urban gradient: Linking urban planning and ecology

Abstract

Ecologists have suggested that ecological conditions in urbanizing landscapes can be described by a complex urban-to-rural gradient (McDonnell and Pickett 1990). The gradient paradigm offers a useful framework to test hypotheses on the impacts of urban development on ecological processes. These studies, however, tend to simplify the actual urban structure into monocentric agglomerations characterized by concentric rings of development surrounding a dense core. The assumption of gradient analysis is that the overall urban exposure changes predictably with distance from the urban core. Due to such simplification, current gradient studies fail to capture the effects of alternative urban development patterns on ecological processes. In this paper we argue that urban-to-rural gradients cannot be represented by the distance from the urban core. Rather they can best be described using a series of pattern metrics that link urban development to ecological conditions. Based on an analysis of land-use and land-cover patterns in the Seattle metropolitan area we propose a strategy to quantify urban patterns. We examine the behavior of various pattern metrics and propose a set of metrics useful to test formal hypotheses on the relationships between urban patterns and ecological disturbances. Finally, we discuss the implications of this empirical study for gradient analysis of metropolitan areas and for future urban ecological research.