In order to study the leaching of exogenous nitrogen during green space management and maintenance, the parameters of the model were calibrated through field monitoring and grow box simulation experiments, and the Model for Studying Nitrogen Transport in Green Space Ecosystems was established by using HYDRUS-2D software. Results showed that the model is highly reliable for simulating nitrogen transport in microtopography, with R² values greater than 0.9 and RMSE values below 5. Slope gradient was positively correlated with horizontal nitrogen differences (ammonium and nitrate nitrogen) and negatively correlated with vertical differences (p < 0.05), while nitrogen application was positively correlated with both horizontal and vertical differences in nitrate nitrogen and negatively correlated with ammonium nitrogen (p < 0.05). The vertical differences of soil ammonium nitrogen exhibited a significant negative correlation with slope (−0.837 to −0.851), while the horizontal differences of nitrate nitrogen showed a significant positive correlation, with correlation coefficients of 0.965 and 0.967 for surface and subsurface soils, respectively. The increasing nitrogen application rate exacerbated these discrepancies, with the highest nitrogen treatment (0.312 g) exhibiting the most pronounced differential effects. Notably, the horizontal variation in nitrate nitrogen reached 6.9-fold that of ammonium nitrogen, while the vertical discrepancy demonstrated a 7.0-fold magnitude relative to ammonium nitrogen levels.

Against the backdrop of global warming and the urgent demand for sustainable development, blue–green spaces (BGSs) play a vital role in carbon reduction and sequestration, yet the multi-scale spatial mechanisms by which blue–green space patterns (BGSPs) regulate carbon storage (CS) remain unclear. Taking the Zhengzhou Metropolitan Area as the study area, this research clarifies the BGSP-CS correlations at both class and landscape levels and quantifies their spatial interaction mechanisms, providing scientific support for integrated BGS planning that aligns with sustainable development objectives. Using the InVEST model coupled with regional carbon density correction, the total CS of the area is estimated at 1112.27 × 10⁶ t. Spearman’s correlation analysis shows that at the class level, area–edge and shape complexity indicators (e.g., Landscape Shape Index, LSI: r = −0.427) are negatively correlated with CS, while connectivity indicators exert no significant effect. At the landscape level, Shannon’s Diversity Index (SHDI: r = −0.635) and area–edge indicators inhibit CS, whereas Shannon’s Evenness Index (SHEI: r = 0.602), Largest Patch Index (LPI: r = 0.618) and shape complexity indicators exert positive effects. A comparative analysis of three regression models reveals that the multi-scale geographically weighted regression (MGWR) model outperforms the ordinary least squares (OLS) and geographically weighted regression (GWR) models, with R² values of 0.505 (class level) and 0.484 (landscape level). It effectively captures the “west–strong and east–weak” spatial heterogeneity of BGSP impacts on CS. This study identifies key BGSP indicators regulating CS and their spatial mechanisms, providing scientific support for integrated BGS planning, regional carbon sink enhancement, the achievement of “dual carbon” goals, and the promotion of sustainable development in metropolitan areas. Future research may optimize model parameters through field surveys and explore the coupling mechanism between BGSPs, land surface temperature and CS to better align BGS management with sustainable development agendas.

In the context of global warming, blue-green spaces (BGSs) are important means for carbon reduction and storage. However, research on the influence of blue-green space patterns (BGSPs) on carbon storage (CS) and their spatial mechanisms is still limited. Therefore, this study focused on the Zhengzhou metropolitan area to analyze the correlation between the overall and internal landscape patterns of BGS and CS, along with key indicators. We quantified the spatial relationship between BGSP and CS, revealing spatial mechanisms that influence them. The results indicate that, at the class level, area-edge and shape complexity negatively affect CS, while connectivity has no significant effect. At the landscape level, Shannon’s Diversity Index, Area-edge, and Connectivity suppress CS, while Shannon’s Evenness Index, Largest Patch Index and Shape complexity promote it. Moreover, the impact of BGSP on CS is spatially heterogeneous. Therefore, we utilized multiscale geographically weighted regression (MGWR) to measure the intervention intensity on BGSs. Based on these findings, we proposed refined optimization suggestions for BGS systems to promote the integrated development of urban blue-green systems and provide references for metropolitan planning practices and resource management.