Urban waterfront spaces are key venues for residents’ physical activity, and their spatial environment significantly impacts usage efficiency. Existing studies predominantly employ linear models and focus on single waterfront types, making it difficult to reveal differences across various types and the nonlinear mechanisms of influencing factors. To address this, this study investigates three types of waterfront spaces in Shenzhen—riverfront, lakeshore, and seafront spaces—integrating multi-source data and machine learning techniques to systematically analyze the differential impacts of the same elements on physical activity. The results indicate: (1) In terms of transportation accessibility, public transport is the most important factor for riverfront and lakeshore spaces, while road network accessibility is most critical for seafront spaces. (2) Regarding natural landscapes, the dominant factors are normalized difference vegetation index (NDVI) for riverfront spaces, green view index for lakeshore spaces, and distance to the shoreline for seafront spaces. (3) For facility services, the core factors are building density (riverfront), number of sports facilities (lakeshore), and number of leisure facilities (seafront). (4) The study further reveals nonlinear relationships and threshold effects of multiple elements. For instance, a turning point in physical activity intensity occurs when the distance to a subway station reaches 2–2.5 km. The green view index shows a threshold of 30% in the overall model, while dual-threshold phenomena are observed in the lakeshore and seafront models. (5) Synergistic effects between elements vary by waterfront type: in riverfront and seafront spaces, activity is more vibrant when areas are close to subway stations and have a low sky view index, whereas the opposite pattern is observed in lakeshore spaces. A combination of a high green view index and greater distance to the shoreline promotes activity in lakeshore spaces, while a high green view index combined with proximity to the shoreline has the most significant promotional effect in riverfront and seafront spaces. This study provides a scientific basis for health-oriented, precise planning and design of urban waterfront spaces.

The multifunctionality of urban parks significantly enhances citizen well-being. Uncovering the potential spatiotemporal non-stationary relationships between the built environment and park vibrancy will provide more effective information for urban green space governance. This study estimates park vibrancy based on Baidu heatmap data and depicts the spatiotemporal distribution of vibrancy across three types of urban parks in Chengdu. A Geographically and Temporally Weighted Regression (GTWR) model was applied to explore the dynamic spatiotemporal impacts of internal and external built-environment variables on park vibrancy. The results indicated that the GTWR model demonstrated superior goodness-of-fit compared to the linear regression model. The associations between multidimensional built-environment factors and park vibrancy exhibited notable spatiotemporal heterogeneity. Furthermore, the influence of greenery on park vibrancy was highly context-dependent, varying across perspectives (2D vs. 3D) and granularities (overall vs. local). For instance, normalized difference vegetation index (NDVI), reflecting overall green volume, was negatively associated with park vibrancy, whereas the coverage and fractal dimension of tree canopy, characterizing local vegetation structure, showed predominantly positive associations. These findings provide rich analytical perspectives and effective methodologies for examining and revealing the mechanisms influencing urban-park vibrancy, and can guide policymakers and urban planners in formulating more targeted park-management strategies.