Impacts and threshold effects of urban human settlements upon job-housing balance in megacities: A case study of Wuhan metropolitan development area

Entering the stage of high-quality development, the problem of job-housing imbalance in megacities has intensified, triggering systemic "urban diseases" such as traffic congestion and resource misallocation, which (in)directly affect residents' well-being and sustainable development. Interventions in the human settlement environment can reconstruct the job-housing relationship, playing a significant role in addressing the challenge of job-housing imbalance. However, existing studies primarily focus on the linear relationship between the built environment and job-housing balance, making it difficult to capture the nuanced effects of multidimensional human settlement environment factors. Therefore, a nonlinear approach is essential to deconstruct these dynamic interactions and thereby promote urban governance. Taking the Wuhan metropolitan development area, a typical megacity, as a case study, this research employs commuting time as a proxy metric to assess job-housing balance and discloses its spatial patterns. Meanwhile, a machine learning model is applied to identify the nonlinear effects of urban human settlement on job-housing balance and quantify the threshold values of key variables. The results show that: (1) The contributions of the built, social, and natural environment to job-housing balance vary significantly. The most influential drivers, ranked by contribution, are: the "proportion of young people" and the "proportion of elderly population" in the social environment, "intersection density" in the built environment, "habitat quality" in the natural environment, and "road network density" in the built environment. (2) The nonlinear relationships between different human settlement environment factors and job-housing balance exhibit various forms, including approximate linear relationships, inverted U-shaped curves, and threshold effects. Based on these findings, we call for optimizing resource allocation according to the characteristics of key human settlement indicators and implementing differentiated policies. These new measures can effectively enhance urban livability in megacities.