Carbon emission reduction in urban green space construction: A multi-module optimization framework for multi-scale applications

Urban green spaces serve as critical spatial carriers for mitigating the urban heat island effect and sustaining urban carbon balance. Their materialization phase spans material production, transportation, and on-site construction, and the associated carbon emissions are largely predetermined by planning and design decisions, yet no systematic optimization pathway has been established. Conventional life cycle assessment relies on static databases, while experience-driven design adjustments struggle to accommodate multiple constraints imposed by policy, functionality, and spatial distribution simultaneously. This study presents a multi-module optimization framework that sequentially links carbon emission calculation, optimization algorithms, and spatial layout adjustment and applies it to three real-world green space projects at different spatial scales. A performance comparison of six algorithms across the three project scales showed that particle swarm optimization (PSO) consistently obtained optimal or near-optimal solutions and was therefore selected as the core solver. Under the engineering constraint of a maximum adjustment magnitude (MA) of 0.10, the mean emission reduction rates for the three sites were 7.34%, 8.48%, and 6.06%, respectively. Under different constraint scenarios, the framework further translates optimization outputs into implementable land-use adjustment schemes. By advancing urban green space carbon management from post-construction accounting to upstream intervention at the planning and design stage, this framework provides a quantifiable decision-support tool for carbon neutrality goals in urban green space planning.