Coupling between the layout pattern of green space and the function of urban wetland biomes in the context of urbanisation

Abstract

Accelerated global urbanisation has led to a sharp decrease in the area of natural wetlands and degradation of biological functions, and the morphological heterogeneity of urban green space layout has become a key variable affecting the sustainability of wetland ecological services. Based on the multi-source data fusion of satellite remote sensing (Landsat-8, Sentinel-2), biometric sensor network (infrared camera, acoustic recorder, eDNA) and microclimate hydrological monitoring, we constructed a machine learning-driven coupled model of morphology-function, and analyzed the nonlinear response mechanism of the green space layout parameters and wetland biological communities. We constructed a machine learning-driven 'form-function' coupled model to analyse the nonlinear response mechanism between green space layout parameters and wetland biological communities. Empirical evidence shows that the density of green space patches in the core area (27.4 patches/km²) is 3.4 times higher than that in the suburban area, but the connectivity index (0.15) is less than 1/4 of that in the suburban area, which leads to a 58% reduction of heron nesting area; the insect diversity in the area of nighttime light intensity >50 lux decreases by 63%, and the 'ecological blind zone' revealed by the sensor data accounts for The 'ecological blind zone' revealed by sensor data accounted for 78% of the built-up area-wetland interface zone. The prediction accuracy of the random forest model (R²=0.83) was significantly better than that of the traditional method, and the connectivity index (32.7% contribution) and water proximity (28.5%) were identified as the key driving factors. The study proposes a 'multi-centre + corridor' resilience planning framework, which improves the success rate of biotic migration by 57% and delays the peak of stormwater runoff by 0.8 hours through the implantation of 1-3 ha greenbelt nodes and 30-metre ecological corridors. The results provide dynamic assessment tools and spatial intervention targets for ecological restoration in high-density cities, and promote the transformation of urban planning from form-fitting to process synergy.