With the growing awareness of public health, urban parks have increasingly become popular venues for physical activities due to their accessibility and pleasant landscapes, among which walking, jogging, and running dominate. This study innovatively integrates exercise trajectory data from the Strava platform and semantic segmentation technology to analyze the interaction mechanisms among park view elements, physical activities, and physiological responses, based on empirical data from three representative parks in Shanghai. This study includes the following: (1) acquiring hotspot exercise paths and physiological data (heart rate and speed) of walking, jogging, and running users through the open Strava platform; (2) conducting semantic segmentation on real-word photos of three case parks to extract 17 types of park elements; (3) applying Spearman’s correlation analysis to reveal the differential impacts of park elements on physiological responses under walking, jogging, and running behaviors, demonstrating that combinations of elements such as trees, water bodies, fences, and sky influence exercise performance; and (4) constructing scene modules for site attraction, training improvement, and restorative relaxation for walking, jogging, and running, and proposing phased landscape scene sequence strategies to provide quantitative guidance for health-oriented park planning and design. This study breaks through the limitations of traditional subjective evaluations by coupling objective physiological data with spatial elements, offering novel insights for optimizing the exercise functionality of urban green spaces.

Objective: To explore the correlation between park view elements and their combinations on the heart rate (HR) and speed of walkers, joggers, and runners in different groups of people’s profiles and walking types, provide suggestions for the planning and design of walking suitability of walking trails in parks, and guide people with different walking needs to scientifically choose walking trails in parks.Methods: Profile data and exercise data of users who recorded walking activities in Century Park are collected on Strava, and the park view images (PVIs) were taken and segmented semantically. Data are grouped according to gender, age, weight and exercise type, and the quantitative relationship between HR, speed and 17 park view elements is studied by Spearman correlation analysis.Results: (1) The influence of the same park view elements on the exercise physiological indicators of different genders is small; (2) Park view elements combination based on sky, grass-plant and tree can better stabilize the walking HR of the older adult; (3) Semi-enclosed trail dominated by tree can improve the walking HR and speed of people with larger body weight; (4) Natural routes dominated by sidewalk-path and supplemented by tree and sky elements are more suitable for walking, while the trails with larger sky area, no trees and wider trails are more suitable for running.

As high-density cities face severe conflicts between urban expansion and limited ecological space, exploring the potential drivers of carbon sequestration within urban green spaces is crucial. The overarching aim of this study is to establish an interpretable, model-based analytical framework for assessing the Annual Carbon Fixation Rate (ACFR). To address the limitations of linear models and “black-box” machine learning algorithms, this study analyzed 149 urban parks in Nanjing by coupling Sentinel-2 high-resolution data with a Random Forest algorithm and the SHAP framework to model Annual Carbon Fixation Rate. The model achieved high predictive reliability (R² = 0.9690). This study quantitatively identified critical nonlinear thresholds: an impervious surface proportion exceeding 30% is associated with a structural decline in carbon sinks, while the optimal Leaf Area Index interval for maximum efficiency is 2.5–3.5. Moreover, the present research identified a potential model-derived ecological compensation effect; optimizing vegetation community structures under high anthropogenic disturbance showed a 23.9% higher modeled marginal ACFR response compared to low-disturbance habitats. This explainable framework attempts to translate complex model-derived biophysical associations into referenced quantitative thresholds. Importantly, the ACFR and derived thresholds represent remote sensing-based model estimates rather than field-measured ecological causality, serving as exploratory planning references. These findings challenge the assumption that greening is inefficient in degraded environments and provide concrete control indicators for targeted micro-renewal and sustainable urban regeneration.

In the context of national fitness and ecological construction, urban water, as the core carrier of water sports, is increasingly being explored. Through empirical analysis of urban water bodies in Nanjing and Shanghai, the Spearman model and an XGBoost model, based on SHAP, are used to examine the correlation between the subjective perceptions of water-landscape images by various types of water-sports participants and objective evidence of urban water-landscape elements. The results show that amateurs prefer a high green-view index (>20%) + low water visibility (≤30%) + low sky visibility (≤30%) + shallow-water area (≤1.5 m), which enhances their satisfaction with the leisure experience; progressors prefer a moderate green-view index (10–20%) + moderate water visibility (30–40%) + moderate sky visibility (30–40%) + medium water-depth area (1.5–2.5 m), which helps them achieve better perception of skill improvement and training effectiveness; professionals prefer high water visibility (≥40%) + high sky visibility (≥40%) + low green-view index (≤10%) + deep-water area (≥2.5 m), which meets their training requirements for professional competitions. This study provides a scientific basis for urban water landscape planning and design in order to create urban water sports spaces that respond to the needs of various types of water sports participants.

Sustainable regeneration of historic districts requires small-scale interventions to be targeted carefully because conservation constraints, limited resources, and heterogeneous street conditions make blanket renewal inappropriate. We develop an uncertainty-audited GeoAI workflow that supports field-audit prioritization for conservation-sensitive micro-renewal. In a Nanluoguxiang, Beijing case study, 2144 directional street-view images represented 536 panoramas, of which 535 entered final perception analysis. Semantic segmentation and repeated vision-language-model scoring produced streetscape indicators and AI-estimated visual-impression constructs. On a blinded 20-panorama subset, ten independent raters showed excellent average-measures reliability for perceived vitality and oppressiveness (ICC(2,10) = 0.967 and 0.954), and human mean ratings were strongly associated with Qwen scores (Spearman rho = 0.951 and 0.921; both p < 0.001). Leakage-resistant, spatially nested validation explained 30.0% of held-out vitality variance and 40.0% of held-out oppressiveness variance. A 48-specification stochastic priority analysis yielded a median top-15 overlap of 13/15 with the reference specification. Joint propagation of measurement, segmentation-architecture, spatial/model, panorama-to-edge, and decision-weight uncertainty identified E0079, E0076, and E0129 as the strongest inner-core candidates, with conservative conditional selection frequencies of 0.608, 0.605, and 0.577. These frequencies are algorithmic acceptability measures under tested assumptions, not calibrated probabilities, causal effects, or construction recommendations. A separate local Gemma 3 12B check on the same 20 panels showed moderate positive vitality agreement with human means (Spearman rho = 0.752) and with Qwen scores (rho = 0.752), as well as positive but weaker oppressiveness agreement with human means (rho = 0.671) and Qwen scores (rho = 0.654). Gemma underestimated vitality by 0.700 points on average but had near-zero mean oppressiveness bias (+0.010), so cross-model validity remains construct-specific rather than established globally. The workflow supports transparent resource targeting and follow-up field investigation while preserving the need for stakeholder consultation, engineering assessment, cross-model validation, and cross-district replication. Here, conservation sensitivity is implemented as a constraint on subsequent field verification and intervention appraisal; heritage significance is not directly estimated by the ranking model.