A dose-response curve describing the relationship between tree cover density and landscape preference

Does adding more and more trees to a residential street yield a reliable increase in preference? Or is there a point at which, in terms of preference, additional trees will have minimal effect, no effect, or even a negative effect? To address these questions, we selected 121 community streets in four Midwestern urban areas in the U.S. and produced a panoramic photograph of each site and then measured the density of tree cover visible at eye level (Panorama). We also collected Google Earth aerial photographs to measure the top-down tree cover density (Google) for the sites. Then, 320 individuals provided preference ratings for a randomized subset of the panoramic photographs (15 pictures per person). Through linear and curvilinear regression analysis, we found a power line model best describes the relationship between each measure of tree cover density and preference. The power lines have a similar shape: when sites are relatively barren, a slight increase in tree density yields a steep increase in preference. After tree cover density exceeded those values, however, higher tree densities yielded smaller, but still positive increases in preference. These findings suggest that to ensure a moderate level of preference, tree cover density should be not less than 41% as measured by panoramic photographs or 20% as measured by Google Earth aerial photographs. Planting trees in barren residential areas will result in considerably more impact than if the same trees were planted in already green areas. Still, the findings here demonstrate that, for preference, every tree matters.