Nonlinear pathways to climate sustainability: the role of digital technologies, clean energy, and structural change
Abstract
Climate change remains one of the most pressing global challenges, particularly in highly industrialized and technologically advanced economies. This study examines how sustainable digital technologies, clean energy transition, industrial structural upgrading, and socio-economic dynamics influence climate change in the world’s five largest robot-adopting economies: China, Japan, Germany, the United States, and South Korea. The study uses panel data from 2000 to 2024, and a multilayer neural network (MNN) framework combined with surface slope analysis and Deep SHAP interpretation to capture nonlinear relationships between key variables and climate indicators. The findings reveal strong nonlinear effects across the four dimensions. Robotics adoption, environmental technologies, and ICT development contribute to emission reduction as technological maturity increases. Modern renewable energy and access to clean energy technologies significantly reduce emissions, while nuclear power shows weaker and heterogeneous effects. Industrial structural upgrading, represented by industrialization and research and development, improves environmental outcomes when technological innovation supports cleaner production systems. However, economic growth and urban population growth continue to exert upward pressure on emissions. The Deep SHAP results highlight renewable energy, robotics, and industrial innovation as the most influential predictors of climate outcomes. Theoretically, the findings align with Ecological Modernization Theory and Innovation and Technology Diffusion Theory, demonstrating that climate mitigation emerges from cumulative technological diffusion and structural transformation rather than isolated interventions. Policy implications highlight the need for coordinated industrial–energy–digital strategies, threshold-targeted green technology investment, and adaptive urban development frameworks.