Although exposures to noise or greenery in isolation have been widely studied, limited research has explored the nonlinear associations between spatiotemporal contexts and compound disadvantages from a daily mobility perspective. Drawing on data from mobile sensors, GPS, satellite imagery, points of interest, and activity diaries, this study measures real-time noise and greenery exposures along relevant spatiotemporal contexts. Methodologically, daily mobility is delineated through a spatiotemporal threshold approach, modeling them as stay and move event sequences that underpin the development of a compound disadvantage index. A novel framework, integrating GPBoost with residual-based bootstraps and SHAP, subsequently accommodates time-sequence dependencies and uncertainties to investigate nonlinear relationships between spatiotemporal contexts and compound disadvantages. The main findings indicate that (i) the neighborhood effect averaging problem is apparent in noise and greenery measurements; (ii) socioeconomic status coupled with spatiotemporal contexts can infer individual compound disadvantages; (iii) employment status constitutes the foremost socioeconomic predictor of compound disadvantages; and (iv) the bootstrap-SHAP methodology pinpoints spatiotemporal contexts with statistically significant effects on compound disadvantages, concurrently disclosing thresholds and interactions with socioeconomic statuses.

Extensive research on the relationships between environmental factors and older adults' health can inform urban planning for healthy and age-friendly cities. Past studies have predominantly considered such relationships as stable over space, with less emphasis paid to spatial non-stationarity and space-based rationale. Within Chinese cities' socio-economic spatial patterns, the different potential environment-health pathways corresponding to different space-based populations remain unclear. Further, limited types of environmental factors have been investigated, while diverse combinations of multiple environments have received less attention. This study thus examined how salient environmental factors may vary across spatial types and the spatial non-stationarity in the health-environmental relationship. Multi-source data characterized Guangzhou's multiple environments, and older adults' health patterns were mapped using self-rated health data. Subsequently, relationships between multi-environmental factors and older adults' health across spatial types were explored by ridge models. The results indicated that (i) There were obvious spatial differences in older adults' health, and significant relationships between multi-environmental factors and older adults' health were observed. (ii) Spatial nonstationarity in such relationships was evident, driven by space-based physicosocial contexts. (iii) Dominant factors notably varied across spatial types. The findings provide space-based planning insights for healthy and age-friendly cities.

Urban villages, a unique residential typology of China's rapid urbanization, and their residents' sentiments have become a prominent issue in the development and management of Shenzhen. While numerous studies have investigated the impact of the urban environment on residents' sentiments, most have focused on the general population and used administrative units, neglecting environmental impacts across spatial scales based on daily mobility. To address these research gaps, this study employed multi-source big data and a large language model to assess residents' sentiments and evaluate environmental features within 5-, 10-, and 15-min community life circles in the urban villages of Shenzhen. Hierarchical regression and ridge regression were used to explore the key environmental features affecting urban village residents' sentiments. The findings indicated that the sentiment scores of urban village residents were significantly lower than those of the general population, while low sentiment scores were concentrated in suburban areas. Further, the environmental variables within the 10-min community life circle had the greatest influence on sentiments for urban villagers. Specifically, subway stations and green spaces contributed positively to residents' sentiments, while industrial parks and building density/height had a negative impact. These findings advance urban research by integrating activity-space theory with big data analytics, deepening the understanding of how spatial environments shape marginalized groups' sentiments. The results also provide an empirical foundation for optimizing urban renewal policies to improve the quality of life and mental well-being of urban village residents through the development of more livable and healthier urban environments.

People's emotional states and associated environmental exposure in geographic and social science are often assessed through recall measurements, such as the Day Reconstruction Method (DRM), which can overlook the need for real-time understanding and may encounter recall bias. This study examines recall bias in the relationship between mobility-based greenery and people's short-term happiness across activity contexts, including homes, workplaces, recreational activity sites, and travel routes. It compares Geographic Ecological Momentary Assessment (GEMA) with the DRM. GEMA is a method that integrates Ecological Momentary Assessment (EMA) with Global Positioning Systems (GPS). The results indicate that people's short-term happiness and mobilitybased greenery exposure were underestimated as measured via the DRM's recall data, particularly for outdoor recreational activity sites and travel routes. The GEMA effectively mitigates the recall bias issues inherent in the DRM.

While growing attention has been paid to audio-visual interactions, the findings on the role of greenery exposure in noise perception are inconsistent. The inconsistent conclusions may stem from the uncertain geographic context problem. Building on two distinct mechanisms identified in past studies (i.e., restorative and masking effects and the audio-visual congruency effect), this study aims to unveil how greenery exposure influences noise perception across geographic contexts. Employing portable devices and subjective sensing tools, we collect people’s real-time sound exposure, greenery exposure, and noise perception during their real-life contexts. Subsequently, interpretable machine learning methods are used to investigate the global and local effects of greenery exposure on noise perception. The results include: i) a significant and positive association between real-time sound level and perceived noise level was observed, with greenery exposure significantly moderating the association; ii) notable non-linearity in such relationships was also identified, with consistent sound level and greenery exposure thresholds across geographic contexts. While the restorative and masking effects dominate, their magnitudes vary, and the audio-visual congruency effect can be identified locally. The prevalence of the two distinct mechanisms is associated with specific urban functional contexts. The findings can serve as scientific references for policymakers on noise governance and greenery design.

Learning from the COVID-19 pandemic is vital for future global health crises. Forecasting pandemic spread across countries in the presence of viral mutations, such as SARS-CoV-2 variant strains, remains a challenge. Previous studies indicate spatial and temporal regularities in SARS-CoV-2 variant distribution, yet limitations persist: (1) static geographical patterns overlook dynamic changes and their causes, and (2) the association between temporal variations in SARS-CoV-2 lineages and epidemic outbreaks is underexplored. To address these gaps, we propose an analysis framework using more than 10 million SARS-CoV-2 genome sequence metadata from 100 countries. Our framework identifies spatial heterogeneity patterns in the relative frequency of SARS-CoV-2 variants through clustering, examines factors influencing spatial heterogeneity using explainable machine learning, and analyzes the time lag effect of temporal variation on COVID-19 infection waves based on spatial patterns. Our findings demonstrate the following: (1) The distribution of SARS-CoV-2 variant strains among 100 countries exhibits a spatial pattern characterized by geographic proximity, with observed dynamic changes. (2) Bilateral distance consistently dominates, and factors related to globalization and response policy contribute to nongeographic proximity and temporal variation of the spatial pattern. (3) Leveraging this spatial pattern allows for the forecast of time lags (mainly ranging from twenty to thirty-eight days) between the emergence of new strains and subsequent infection waves. Our study clarifies the spatial pattern and time lag effect of the dynamic distribution of SARS-CoV-2 variant strains. It offers a spatial basis and temporal reference for early warning in future global public health crises from a geographical perspective.

Equal exposure to quality-built environments fosters livable, inclusive cities. The neighborhood effect averaging problem (NEAP) suggests that daily mobility plays a crucial role in shaping environmental exposure. This study aims to unveil the NEAP in built-environment quality exposure. Street-view image data and mobile phone signaling data are coupled to measure built-environment quality in people's residential space and activity space. Subsequently, a conditional process analysis model is employed to investigate how daily mobility, income, and built-environment quality in residential space influence built-environment quality in activity space. The results indicate that (1) there is a significant disparity in built-environment quality exposure, although the disparity in activity space is smaller than that in residence; (2) income exerts a dual influence on built-environment quality in activity space through direct and indirect pathways, and the pathways could be moderated by high mobility; and (3) neighborhood effect averaging is evident at the individual level and manifests as a derived phenomenon associated with income groups. The findings provide insights for better serving environmental equality.

When investigating the relationship between the acoustic environment and human wellbeing, there is a potential problem resulting from data source self-correlation. To address this data source self-correlation problem, we proposed a third-party assessment combined with an artificial intelligence (TPA-AI) model. The TPA-AI utilized acoustic spectrograms to assess the soundscape's affective quality. First, we collected data on public perceptions of urban sounds (i.e., inviting 100 volunteers to label the affective quality of 7051 10-s audios on a polar scale from annoying to pleasant). Second, we converted the labeled audios to acoustic spectrograms and used deep learning methods to train the TPA-AI model, achieving a 92.88 % predictive accuracy for binary classification. Third, geographic ecological momentary assessment (GEMA) was used to log momentary audios from 180 participants in their daily life context, and we employed the well-trained TPA-AI model to predict the affective quality of these momentary audios. Lastly, we compared the explanatory power of the three methods (i.e., sound level meters, sound questionnaires, and the TPA-AI model) when estimating the relationship between momentary stress level and the acoustic environment. Our results indicate that the TPA-AI's explanatory power outperformed the sound level meter, while using a sound questionnaire might overestimate the effect of the acoustic environment on momentary stress and underestimate other confounders.

Ecological conservation red line is an important initiative for ecological civilization construction in China. Studies and works have been conducted to delineate the ecological conservation red line using remote sensing and manual survey methods. However, the traditional ecological assessment methods ignore the impact of dynamic elements on the ecological environment, and a permanent monitoring system for ecological conservation red line is still necessary. Therefore, this paper aimed to build a daily monitoring system of the ecological conservation red line from the perspective of ecological soundscape, combining deep learning and traditional Geographic Information System (GIS) methods. We analyzed the main factors affecting the ecological soundscape to provide a benchmark for selecting sites for permanent soundscape monitoring. Initially, this study developed the first classification standard of soundscape elements for daily monitoring of ecological conservation red line concerning the "ecological-production-living spaces" concept and existing literature on soundscape and built a deep learning model for training, with training and validation accuracies of 89.88% and 72.41%, respectively. The identification results of this model confirmed that the solution of intelligent monitoring through soundscape is reliable in daily monitoring. Then, the deep learning model was employed to classify and predict the soundscape collected from the case sites and effectively identify the areas with severe ecological encroachment by calculating the proportion of ecological soundscape elements, performing the simulation of daily monitoring, which is important guidance for the actual ecological conservation red line monitoring. Finally, a regression model was used to analyze the spatial characteristics of ecologically vulnerable areas, with an R 2 of 0.472. The regression model showed that nighttime light intensity had the greatest influence on ecological soundscape, followed by the distance from residential areas, spatial centrality, and road distance. These factors should be considered when implementing permanent soundscape monitoring sites. In addition, Normalized Difference Vegetation Index (NDVI) and topography were not significant in the regression model, which confirmed the shortcomings of considering only the ecological background and ignoring the dynamic biological elements. Daily monitoring should capture dynamic encroachments so that major ecological damage can be detected before it occurs. Therefore, the innovative integration of soundscape elements into the monitoring system meets the basic needs of monitoring ecological conservation red lines daily.

With the rapid growth in both urbanization and the ageing of the population, elderly migrants have become a more prominent group in urban China. Previous studies have shown that elderly migrants are a vulnerable group in terms of subjective well-being (SWB) and studies have emphasized the role of their socioeconomic status (SES) and family-related attributes (FRA) on their SWB. However, there is less attention on whether the perceived residential environment (PRE) to which elderly migrants are exposed and their social interactions have some effect on their SWB. To fill this research gap, street view images and questionnaire survey data from Guangzhou, China were collected. The association between PRE, social interactions, and elderly migrants' SWB was examined within a comprehensive framework using structural equation models. The results indicated that elderly people who had migrated within China had lower levels of SWB than Guangzhou-born elderly adults. Social interactions mediated the effects of SES, FRA, and perceived environment on the SWB of elderly migrants. After controlling for SES and FRA, a livelier and safer PRE was directly positively associated with elderly migrants' SWB (coefficient = 0.181 at the 1% level) and indirectly associated with elderly migrants' SWB through social interactions with local friends (coefficient = 0.035 at the 1% level) and with neighbours (coefficient = 0.014 at the 5% level). These results suggest that increased social interactions and the creation of a better PRE would benefit elderly migrants' SWB in the context of active ageing.

A tremendous amount of research use questionnaires to obtain individuals' fear of crime and aggregate it to the neighborhood level to measure the spatial distribution of fear of crime. However, the cost of using questionnaires to measure the large-scale spatial distribution of fear of crime is high. The built environment is known to influence people's perceptions, including fear of crime. This study develops a machine learning model to link built environment extracted from street view images to fear of crime obtained from questionnaires, and then applies this model to extrapolate fear of crime for neighborhoods without the questionnaires. Using massive street view images and a survey among 1,741 residents in 80 neighborhoods in Guangzhou, China, this study developed a novel systematic approach to measuring large-scale spatial fear of crime at the neighborhood level for 1,753 neighborhoods. This is the first study to measure fear of crime at the neighborhood level for a metropolitan area of nearly 20 million people. The integration of survey data and street view images provides an opportunity to develop a more effective way to measure the spatial distribution of fear of crime. This approach could be applied to map other types of perceptions at a spatial resolution of the neighborhood level.

While there are plenty of studies on the effects of neighborhood and park greenness on personal overall satisfaction and walking behavior, the relationship between street greenness exposure and walking satisfaction has received limited attention. Also, the possible pathways by which street greenness exposure affects walking satisfaction need to be further examined. To fill these research gaps, we measured eye-level street greenness using street view images, machine learning techniques and global position systems. A structural equation model was used to examine the mediating effects of objective noise and PM2.5 exposure and related subjective annoyance, on the relationship between street greenness exposure and people's walking satisfaction. The results showed that street greenness exposure not only had a significant direct effect on walking satisfaction, but also has a significant indirect effect on walking satisfaction through subjective environmental annoyances (including noise and PM2.5 annoyances) rather than through objective noise and PM2.5 exposures. Besides physical activity and social interaction, the indirect effect of street greenness exposure on walking satisfaction through subjective environmental pollution annoyance accounted for about 17.39% of the total effect and cannot be ignored. These results suggest that the urban greenness layout policy should not only consider residential greenness but should improve people's environmental perception and walking satisfaction by allocating more greenness on streets with high noise and PM2.5 levels.

近年来,国内关于空间活力影响因素的研究层出不穷,但缺乏对人本导向下公众感知的街道空间品质等非物质层面因素的考虑.文章利用手机信令数据测度空间活力,基于深度学习和公众街景感知的评分结果测度空间品质,并将POI丰富度、时间可达性、房屋租金等因素作为控制变量,系统分析了空间品质与空间活力的关系.研究结果显示:从全局看,空间品质对空间活力的分布具有一定的解释力,但相关性较弱;从局部看,不同维度的空间品质对空间活力的影响具有较强的空间异质性,在城市空间中呈现较为复杂的关系;在不同类型的空间中,空间品质对空间活力影响作用的差异主要与区位相关,且空间品质优化对空间活力的提升作用需要具备一定的物质先决条件.因此,在实际的空间改造工作中,应根据空间现有的物质条件判定是否可以采取品质优化的方式提升活力;需考量空间安全感、热闹感、美丽感、脏乱感等维度的品质在不同类型的空间中对居民活动的差异化影响,并因地制宜地采取改造策略.

Previous literature has examined the relationship between the amount of green space and perceived safety in urban areas, but little is known about the effect of street-view neighborhood greenery on perceived neighborhood safety. Using a deep learning approach, we derived greenery from a massive set of street view images in central Guangzhou. We further tested the relationships and mechanisms between street-view greenery and fear of crime in the neighborhood. Results demonstrated that a higher level of neighborhood street-view greenery was associated with a lower fear of crime, and its relationship was mediated by perceived physical incivilities. While increasing street greenery of the micro-environment may reduce fear of crime, this paper also suggests that social factors should be considered when designing ameliorative programs.