A tool to predict perceived urban stress in open public spaces

This article presents an exploratory framework to predict ratings of subjectively perceived urban stress in open public spaces by analysing properties of the built environment with GIS and Space Syntax. The authors report on the findings of an empirical study in which the environmental properties of a sample of open public spaces in the city of Darmstadt, Germany were constructed and paired to users’ ratings. The data are analysed using different types of multivariate analyses with the aim to predict the ratings of perceived urban stress with a high explained variance and significance. The study finds that open public space typologies (park, square, courtyard, streets) are the best predictors for perceived urban stress , followed by isovist characteristics, street network characteristics and building density. Specifically, the isovist visibility , vertices number and perimeter , previously related to arousal and complexity in indoor spaces, show significant relation to perceived urban stress in open public spaces, but with different direction of effects. A model is presented that achieves a predictive power of R 2 = 54.6%. It extends existing models that focused on green spaces and streetscapes with a first exploratory attempt to predict more complex reactions such as perceived urban stress .