While technologies like renewable energy and low-carbon transportation are known to mitigate carbon emissions from urban expansion, achieving carbon neutrality during this process remains a critical unresolved challenge. This issue is particularly pressing for developing countries striving to balance urbanization with carbon reduction. Taking Qionglai City as a case study, this study simulated the territorial spatial functional patterns (TSFPs) and carbon emission distribution for 2025 and 2030. Based on the key drivers of carbon emissions from urban expansion identified through the Geographical and Temporal Weighted Regression (GTWR) model, carbon-neutral pathways were designed for two scenarios: urban expansion scenarios under historical evolution patterns (Scenario I) and urban expansion scenarios optimized under carbon neutrality targets (Scenario II). The results indicate that (1) urban space is projected to expand from 6094.73 hm² in 2020 to 6249.77 hm² in 2025 and 6385.75 hm² in 2030; (2) total carbon emissions are forecasted to reach 1.25 × 10⁶ t (metric tons) and 1.40 × 10⁶ t in 2025 and 2030, respectively, exhibiting a spatial pattern of “high in the central-eastern regions, low in the west”; (3) GDP, Net Primary Productivity (NPP), and the number of fuel vehicles are the dominant drivers of carbon emissions from urban expansion; and (4) a four-pronged strategy, optimizing urban green space vegetation types, replacing fuel vehicles with new energy vehicles, controlling carbon emissions per GDP, and purchasing carbon credits, proves effective. Scenario II presents the optimal pathway: carbon neutrality in the expansion zone can be achieved by 2025 using the first three measures (e.g., optimizing 66.73 hm² of green space, replacing 800 fuel vehicles, and maintaining emissions at 0.21 t/10⁴ CNY per GDP). By 2030, carbon neutrality can be achieved by implementing all four measures (e.g., optimizing 67.57 hm² of green space, replacing 1470 fuel vehicles, and achieving 0.15 t/10⁴ CNY per GDP). This study provides a methodological basis for local governments to promote low-carbon urban development and offers practical insights for developing nations to reconcile urban expansion with carbon neutrality goals.

Territorial spatial function layout optimization can alleviate the conflict between low-carbon development and economic growth by regulating the quantitative composition and spatial allocation of production, living, and ecological functions. However, existing spatial optimization models remain limited in their ability to simultaneously address quantitative uncertainty, spatial functional conflicts, and the coordination among spatial layouts, quantitative structures, and optimization objectives. This study develops a cellular automata (CA)-based Territorial Spatial Function Layout Optimization (TSFLO) model. The model integrates multi-objective fuzzy linear programming to optimize the target-year quantitative structure, a non-homogeneous Markov process to characterize functional transitions, and a game-theoretic approach to coordinate spatial conflicts. Furthermore, the model inversely derives the optimal quantitative structure of the base year through the Markov state transition equation, adjusts the base-year layout according to coordinated suitability to construct the CA initial state, and compares the TSFLO model with the PLUS model using Qionglai City, China, as a case study. Relative to the 2020 baseline layout, the proportions of areas without functional transitions under the TSFLO scenario were 81.40% and 73.60% in 2025 and 2030, respectively, which were 3.39 and 0.28 percentage points higher than those of the PLUS model. The maximum absolute relative errors between the TSFLO-optimized results and the target quantitative structure were 0.07% and 0.05%, respectively, representing reductions of 99.28% and 98.52% compared with the PLUS model. The minimum matching degrees between the seven territorial spatial functions and their corresponding high-coordinated-suitability areas were 71.86% and 69.51% in 2025 and 2030, respectively. From 2020 to 2030, the decoupling index of ecological regulation service functions was −2.8642, compared with −1.9128 for PLUS, indicating that the TSFLO scenario improved the decoupling intensity by 49.7%. The results demonstrate that TSFLO can integrate quantity optimization, spatial conflict coordination, and initial state construction within a unified CA framework, thereby improving quantitative allocation accuracy and spatial suitability matching. This provides methodological support for coordinating low-carbon development and economic growth through territorial spatial function layout optimization.