Valuing the characteristics of urban green spaces to delineate their differential economic benefits based on housing value

Urban green spaces (UGS) offer significant economic benefits that are well captured by housing value. Traditional valuation research focuses on the impact of UGS size and proximity on housing prices, neglecting the contribution of other green space characteristics such as quality, type, and shape, as well as how to guide the balanced construction of UGS. To address this gap, we use Wuhan, a Chinese city with high housing sales volume, as a case study. We use multi-source geographic big data and machine learning to comprehensively evaluate the characteristics and value added by UGS, deploying a combined random forest and multi-scale geographically weighted regression model. We find that the quality, type, and shape of UGS have far greater ability to add value to housing than do UGS scale and distance. In particular, quality-related characteristics exhibit an extremely large and stable positive effect on housing value. In addition, we also discovered threshold effects of scale, type, shape, and quality characteristics, beyond which the value-added benefits of green spaces will be greatly reduced. Based on these findings, we group green space characteristics by value added and propose a method of regulating green spaces in Wuhan. This method will help urban planners and policy makers better identify the differences in value added by the multiple characteristics of green spaces and provide new guidance for optimising their features. • The composite characteristics of UGS are quantified. • The quality characteristics of UGS has the greatest values. • UGS demonstrate threshold effects in housing value enhancement. • Proposed a balanced strategy to regulate UGS value-added.