Optimizing local regulating ecosystem services through nature-based urban retrofitting scenarios

Urban areas are facing increasing environmental pressures, particularly in densely built-up neighborhoods with limited opportunities for incorporating greenery. This study proposes a spatially explicit methodology to optimize Local Regulating Ecosystem Services (LRES), particularly air pollution removal, through nature-based retrofitting scenarios. First, we evaluate the baseline, integrating municipal tree inventory data with high-resolution LiDAR and orthophotography to estimate the complete urban forest, including trees on private lands, in a Mediterranean city context. Then, we develop a Simulated Optimal Tree (SOT) model from high-performing real trees. We use it to assess the potential of five scenarios: densification of trees in public parks (SC1) and private gardens (SC2), implementation of green roofs on flat-roofed buildings (SC3), conversion of marginal agricultural land into urban forests (SC4), and densification of tree-lined streets (SC5). Results show that reforesting agricultural land (SC4) delivers the highest ecosystem service gains, with a 237.4% rise in PM2.5 removal compared to the baseline. However, spatial constraints and closeness to pollution sources suggest that interventions like street trees (SC5), despite lower overall gains, remain essential to ensure alignment between demand and service delivery. The approach demonstrates how integrating field-based and remote-sensing data can guide urban greening investments and support decision-making for sustainable and resilient cities. The method is replicable, cost-effective, and adaptable for public administrations aiming to enhance LRES provision while considering spatial and structural constraints typical of high-density urban environments.