7  Conclusion

This study investigated how the spatial configuration of green spaces influences pluvial flooding in urban areas by comparing Xi’an, China, which experienced a major pluvial flood on 11 August 2023, with Delft, the Netherlands, which served as a non flood reference city. The results show that the spatial configuration of green space has only a weak relationship with flooding, and that this relationship depends strongly on the hydrological setting of the city. In Xi’an, the observed relationships mainly reflect where large green spaces are located within natural flood accumulation zones. In Delft, where flooding is largely controlled by engineered drainage, green space configuration shows limited evidence of influencing how surface water is distributed.

The first sub question asked how green space configuration and flood related surface water patterns could be measured and compared between Delft and Xi’an. This study demonstrated that combining FWEI derived flood masks, landscape metrics, DEM variables, Pearson correlation analysis, and k-means typology provides a consistent framework for comparing green space and flood patterns between different cities. The workflow successfully detected extensive flood related surface water change in Xi’an and only limited change in Delft, confirming that the approach can distinguish between a flooded city and a non flood control.

The second sub question asked what spatial patterns emerge when green space, flood related, and topographic indicators are combined. Correlation analysis showed that most relationships between green space configuration and flooding are weak (|r| < 0.20). In Xi’an, a consistent spatial pattern emerged: the grid cells with the most compact, well connected, and spatially expansive green patches are also the cells where flooding is most extensive and where the largest continuous flood areas occur. These cells are concentrated in the open, lower lying areas of the study area where large heritage green spaces are situated. Rather than indicating that green space causes flooding, this pattern reflects the fact that these areas occupy low lying landscape positions where floodwater naturally accumulates. The typology reinforces this by identifying these locations as Type 3 — the class with the highest green cover and highest flood extent (flood share = 0.640 in Delft, 0.610 in Xi’an).

The third sub question examined how the identified patterns differ between Delft and Xi’an and what they imply for urban flood related planning. Xi’an experienced a real flood event, so its results primarily reflect the influence of topography and flood accumulation. Delft showed only 129 grid cells with detectable surface water change, but one important difference emerged: more connected green patches were associated with fewer separate flood patches (r = −0.200, p = 0.046). Although weak, this suggests that where engineered drainage already limits flooding, the arrangement of green space may influence how remaining surface water is organised. No comparable negative relationship was observed in Xi’an, where flood magnitude and topography dominated the spatial pattern.

Taken together, these findings show that simply increasing the amount of green space is unlikely to reduce flooding in areas that already occupy natural flood accumulation zones. Instead, effective flood management should integrate green infrastructure with drainage networks and local topography. The comparison between Xi’an and Delft demonstrates that the effectiveness of green space configuration depends not only on the arrangement of green patches, but also on the wider hydrological and engineering context. Future research should combine the FWEI derived flood detection workflow with hydrological modelling and observed flood records to validate these findings and further investigate how different forms of green infrastructure influence urban flooding under contrasting environmental and drainage conditions.

Brom, Peta, Kristine Engemann, Christina Breed, Maya Pasgaard, Titilope Onaolapo, and Jens-Christian Svenning. 2023. “A Decision Support Tool for Green Infrastructure Planning in the Face of Rapid Urbanization.” Land 12 (2): 415. https://doi.org/10.3390/land12020415.

Farhadi, Hadi, Hamid Ebadi, Abbas Kiani, and Ali Asgary. 2024. “A Novel Flood/Water Extraction Index (FWEI) for Identifying Water and Flooded Areas Using Sentinel-2 Visible and Near-Infrared Spectral Bands.” Stochastic Environmental Research and Risk Assessment 38: 1873–95. https://doi.org/10.1007/s00477-024-02660-z.

Meerow, Sara. 2019. “A Green Infrastructure Spatial Planning Model for Evaluating Ecosystem Service Tradeoffs and Synergies Across Three Coastal Megacities.” Environmental Research Letters 14 (12): 125011. https://doi.org/10.1088/1748-9326/ab502c.

Prokić, Marija, Stevan Savić, and Dragoslav Pavić. 2019. “Pluvial Flooding in Urban Areas Across the European Continent.” Geographica Pannonica 23 (4): 216–32. https://doi.org/10.5937/gp23-23508.

Sarabi, Shahryar, Qi Han, Bauke de Vries, and A. Georges L. Romme. 2022. “The Nature-Based Solutions Planning Support System: A Playground for Site and Solution Prioritization.” Sustainable Cities and Society 78: 103608. https://doi.org/10.1016/j.scs.2021.103608.

Shi, Qian, Mengxi Liu, Andrea Marinoni, and Xiaoping Liu. 2023. “UGS-1m: Fine-Grained Urban Green Space Mapping of 31 Major Cities in China Based on the Deep Learning Framework.” Earth System Science Data 15: 555–77. https://doi.org/10.5194/essd-15-555-2023.