Improving the Urban Thermal Environment in Chengdu: A Multi-Objective Land-Use Optimization Framework Integrating Remote Sensing, Numerical Simulation, and NSGA-II

This study examines how the city’s evolving spatial structure shapes its thermal environment. Using Google Earth Engine (GEE) and the Local Climate Zone (LCZ) scheme, we tracked structural changes across Chengdu and its central districts (Jinjiang and Wuhou) in 2017, 2021, and 2025. We then combined the Weather Research and Forecasting (WRF) model with the NSGA-II algorithm. This allowed us to explore links between LCZ patterns and the Universal Thermal Climate Index (UTCI) in the urban center. Results confirm a strong but non-linear relationship between built form and the local climate. Optimized scenarios, respecting practical planning constraints, show that rebalancing LCZ proportions can reduce peak temperatures in the core area by 1.72–2.75 °C. Future plans for Chengdu should therefore limit high-risk compact types (LCZ 1, 3, 8), expand mid-rise and open arrangements (LCZ 4, 5), and preserve or restore natural surfaces (LCZ A–C) to achieve a more thermally equitable urban landscape.