Nonlinear effects and spatial spillovers of emerging industry clustering on welfare performance under renewable energy transition in China''s urban agglomerations

The global energy transition is currently hindered by inefficient welfare conversion and regional disparities, which have precipitated a "new resource curse" across Chinese urban agglomerations. The resulting welfare divergence between production and consumption zones underscores the urgency of optimizing industrial spatial allocation—a prerequisite for fulfilling the nation''s "dual carbon" objectives. Based on 2014–2023 POI data from eight emerging sectors, this study characterizes the spatial patterns of 19 urban agglomerations using Kernel Density Estimation. It employs an improved enhanced group-frontier dynamic SBM model with super-efficiency to gauge the welfare performance under renewable energy transition (WPRE) and develops an integrated XGBoost-SHAP-GEO model to reveal the non-linear associations and spatial spillover effects of industrial agglomeration. The study reveals: ①A "plateaued fluctuation and volatile growth" trajectory defines WPRE of Chinese urban agglomerations; ②emerging industry agglomerations exhibit polycentric evolution and a strengthening core-periphery architecture; ③the association between emerging industry agglomeration and WPRE is non-monotonic, consistent with diminishing rather than reversing returns; ④Geospatial characteristics, together with their interaction effects, constitute the primary predictors of WPRE, with associations exhibiting substantial stage-specific heterogeneity. Uncovering the nonlinear associations and spatial heterogeneity of industrial agglomeration, this work offers a scientific basis for breaking the ''new resource curse'' and enhancing industry-energy-welfare synergy. The conclusions provide significant policy implications for global resource-rich regions striving to promote inclusive welfare distribution within the renewable energy transition.