A Hybrid Simulation-Optimization Model for Assessing and Enhancing Carbon Sequestration in Urban Parks

Urban parks play a vital role in mitigating the negative effects of urbanization by serving as essential carbon sinks. As cities face increasing pressure to reduce greenhouse gas emissions, it is crucial to precisely measure and optimize the carbon sequestration ability of these green spaces. This research proposes a hybrid simulation-optimization model designed to evaluate and enhance the carbon sequestration potential of urban parks. The simulation component incorporates spatial data from various sources, including vegetation indices, satellite imagery, and structural factors obtained from LiDAR. To ensure the accuracy of the data, the Savitzky-Golay filter is applied to remove noise from the sensing data. The Hybrid Coral Reefs Optimizer-driven Scalable Random Forest (HCRO-SRF) algorithm is used to classify biotopes within the park. This is followed by the estimation of net primary productivity and biomass to quantify carbon sequestration from 2019 to 2024. The dynamic assessment captures spatial-temporal patterns and the influence of vegetation changes over time. The CRO model identifies optimal configurations for species selection, planting density, and spatial layout, aimed at maximizing carbon sequestration while maintaining ecological diversity and park functionality. The results demonstrate that the proposed model can improve carbon sequestration by a higher MAE of 10.16, RMSE of 12.03, and R ² of 0.94, providing actionable strategies for policymakers, urban ecologists, and landscape architects. This research contributes to advancing climate-resilient urban design through integrated environmental modeling and optimization.