科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ Journal of environmental management2026-08-12

Urban green infrastructure suitability and its impacts on stormwater management: A geospatial and artificial intelligence-based assessment framework.

Khansa Gulshad, Marco Helbich, Niki Frantzeskaki, Bishawjit Mallick, Michał Szydłowski, S M Labib

原始摘要(英文原文)· Original abstract
As cities increasingly adopt nature-based solutions (NBS) to enhance urban resilience, green infrastructure (GI) can mitigate environmental challenges, including increased runoff. However, most studies focus on single interventions, rely only on greenery visibility, and often ignore feasibility constraints or stormwater performance under design variability. This study introduces an integrated framework using geographic information systems and artificial intelligence models to assess GI potential. Using Gdańsk, Poland, as a case study, the suitability of blue-green roofs was evaluated using building characteristics. Street-level imagery was used to estimate current greenery levels utilizing an image segmentation model. For streets with critically low greenery, hypothetical future greening scenarios were generated to estimate the number of additional trees required to achieve improved street-level greenery using machine learning, incorporating environmental variables and physical constraints (i.e., street length, tree spacing, and existing trees). Finally, hydrological performance of potential roof and street greening scenarios was estimated under varying rainfall intensities. Results showed that 20.65% of rooftops are suitable for greening, while 38.7% of streets have low greenery and could benefit from additional trees. The hypothetical greening scenarios indicated that approximately 48,000 additional trees would be required to achieve the target street-level greenery, with 75% of streets requiring seven or fewer additional trees. High-retention blue-green roof design mitigates 4.14% of citywide runoff, whereas street trees with medium-sized canopies intercept 0.21-0.22% of runoff. These findings provide a planning basis for prioritizing i) high-retention roofs on large, low-slope buildings, ii) target tree planting on low-greenery streets, and iii) integrate roof-tree greening strategies. The proposed framework is a transferable decision-support tool for sustainable urban planning, climate adaptation, and stormwater management within the NBS paradigm.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Urban green infrastructure suitability and its impacts on stormwater management: A geospatial and artificial intelligence-based assessment framework. — 科研速览 Science Skim