PubAg

Main content area

Predicting tree preferences from visible tree characteristics

Author:
Hofmann, Mathias, Gerstenberg, Tina, Gillner, Sten
Source:
European journal of forest research 2017 v.136 no.3 pp. 421-432
ISSN:
1612-4669
Subject:
conifers, landscapes, planning, plant characteristics, planting, prediction, regression analysis, residential areas, trees, variance
Abstract:
This paper presents a psychological perspective to the selection of trees for urban residential areas. Sixty tree species suitable for urban planting sites were rated by lay participants regarding preference. We then used outward tree features to predict the preference ratings. Twenty-five different plant characteristics served as possible predictors in a regression model for tree preference. We found that the distinction between conifers and deciduous trees, the maximum tree height, and the crown height-to-width ratio were valuable predictors for preference, explaining more than 70% of the variance. This adds support for evolutionary theories of landscape preference. The regression model presented in this paper can be applied to calculate a preference estimate for other tree species using their known physical data, which may facilitate tree selection tasks in green space planning. By specifying preference-relevant tree characteristics, our findings may also inform the process of selecting diverse species for sites where a homogenous overall appearance is a planning goal.
Agid:
5750039