Road Network Distance Research Articles

Projected urban flood risk assessment under climate change and urbanization based on an optimized multi-scale geographically weighted regression

Previous research has primarily focused on evaluating flood risks influenced by single factors, often neglecting the combined impacts of climate change and urbanization on historical and future flood risks. This study utilized 12 assessment indicators across three criteria layers. It employed the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) and an optimized Multi-scale Geographically Weighted Regression (MGWR) to develop a comprehensive assessment framework for historical and projected urban flood risks under combined scenarios of RCPs and SSPs. The risk of future extreme precipitation events increased under 4 RCPs. Compared with 2010, Beijing's urban land area is expected to increase by an average of 53.64 % in 2060 under 5 SSPs, accompanied by rapid socioeconomic development. MGWR based on K-means optimization improves operational speed by reducing the number of patches. The areas classified as high and highest flood risk levels increased by 38.87 % and 60.17 % from the 2010s to the 2060s, indicating an upward trend in future flood risk. Higher flood risk is observed under the high emissions combination scenario. Structural equation modeling (SEM) revealed that road network density (RND), runoff depth (RD), and distance to road (DRO) were the main influencing factors contributing to flood risks in Beijing.

Coupling graph neural networks and travel mode choice for human mobility prediction

Travel mode serves as the link for communication among people, between people and objects, and between people and places. The human mobility is closely related to travel mode. Scientifically predicting human mobility can help alleviate traffic congestion and provide flexible travel choices. However, in current predictions, only the travel demand or human mobility is taken into account, while the influence of residents on the travel mode choice is neglected. Therefore, taking Wuhan City as an example, this paper proposes a new method for predicting human mobility by employing graph neural network techniques and travel mode choice behavior. The prediction model established in this study utilizes network analysis to measure the accessibility time and road network distance of traffic travel mode. The graph neural network method is employed to capture the dynamic temporal and spatial relationships underlying human mobility. Furthermore, the performance of the prediction model is evaluated. The results indicate that at a fine spatial scale, the new method can more accurately reveal the spatial patterns of changes in human mobility, significantly improving the accuracy of predicting human mobility.

Public Bicycle Dispatch Method Based on Spatiotemporal Characteristics of Borrowing and Returning Demands

Public bicycle systems (PBSs) serve as the ‘last mile’ of public transportation for urban residents, yet the problem of the difficulty in borrowing and returning bicycles during peak hours remains a major bottleneck restricting the intelligent and efficient operation of public bicycles. Previous studies have proposed reasonable models and efficient algorithms for optimizing public bicycle scheduling, but there is still a lack of consideration for actual road network distances between stations and the temporal characteristics of demand at rental points in the model construction process. Therefore, this paper aims to construct a public bicycle dispatch framework based on the spatiotemporal characteristics of borrowing and returning demands. Firstly, the spatiotemporal distribution characteristics of borrowing and returning demands for public bicycles are explored, the origin–destination (OD) correlation coefficients are defined, and the intensity of connections between rental point areas is analyzed. Secondly, based on the temporal characteristics of rental point demands, a random forest prediction model is constructed with weather factors, time characteristics, and rental point locations as feature variables, and station bicycle-borrowing and -returning demands as the target variable. Finally, bicycle dispatch regions are delineated based on actual path distances between stations and OD correlation coefficients, and a public bicycle regional dispatch optimization method is established. Taking the PBS in Ningbo City as an example, the balancing optimization framework proposed in this paper is validated. The results show that the regional dispatch optimization method proposed in this paper can achieve optimized dispatch of public bicycles during peak hours. Additionally, compared with the Taboo search algorithm (TSA), the genetic algorithm (GA) exhibits a 11.1% reduction in rebalancing time and a 40.4% reduction in trip cost.

Disadvantaged groups have greater spatial access to pharmacies in New York state

BackgroundThe accessibility of pharmacies has been associated with overall health and wellbeing. Past studies have suggested that low income and racial minority communities are underserved by pharmacies. However, the literature is inconsistent in finding links between area-level income or racial and ethnic composition and access to pharmacies. Here we aim to assess area-level spatial access to pharmacies across New York State (NYS), hypothesizing that Census Tracts with higher poverty rates and higher percentages of Black and Hispanic residents would have lower spatial access.MethodsThe population weighted mean shortest road network distance (PWMSD) to a pharmacy in 2018 was calculated for each Census Tract in NYS. This statistic was calculated from the shortest road network distance to a pharmacy from the centroid of each Census block within a tract, with the mean across census blocks weighted by the population of the census block. Cross-sectional analyses were conducted to assess links between Tract-level socio demographic characteristics and Tract-level PWMSD to a pharmacy.ResultsOverall the mean PWMSD to a pharmacy across Census tracts in NYS was 2.07 Km (SD = 3.35, median 0.85 Km). Shorter PWMSD to a pharmacy were associated with higher Tract-level % poverty, % Black/African American (AA) residents, and % Hispanic/Latino residents and with lower Tract-level % of residents with a college degree. Compared to tracts in the lowest quartile of % Black/AA residents, tracts in the highest quartile had a 70.7% (95% CI 68.3–72.9%) shorter PWMSD to a pharmacy. Similarly, tracts in the highest quartile of % poverty had a 61.3% (95% CI 58.0-64.4%) shorter PWMSD to a pharmacy than tracts in the lowest quartile.ConclusionThe analyses show that tracts in NYS with higher racial and ethnic minority populations and higher poverty rates have higher spatial access to pharmacies.

Association of spatial proximity to fixed-site syringe services programs with HCV serostatus and injection equipment sharing practices among people who inject drugs in rural New England, United States

BackgroundHepatitis C virus (HCV) disproportionately affects rural communities, where health services are geographically dispersed. It remains unknown whether proximity to a syringe services program (SSP) is associated with HCV infection among rural people who inject drugs (PWID).MethodsData are from a cross-sectional sample of adults who reported injecting drugs in the past 30 days recruited from rural counties in New Hampshire, Vermont, and Massachusetts (2018–2019). We calculated the road network distance between each participant’s address and the nearest fixed-site SSP, categorized as ≤ 1 mile, 1–3 miles, 3–10 miles, and > 10 miles. Staff performed HCV antibody tests and a survey assessed past 30-day injection equipment sharing practices: borrowing used syringes, borrowing other used injection equipment, and backloading. Mixed effects modified Poisson regression estimated prevalence ratios (aPR) and 95% confidence intervals (95% CI). Analyses were also stratified by means of transportation.ResultsAmong 330 PWID, 25% lived ≤ 1 mile of the nearest SSP, 17% lived 1–3 miles of an SSP, 12% lived 3–10 miles of an SSP, and 46% lived > 10 miles from an SSP. In multivariable models, compared to PWID who lived within 1 mile of an SSP, those who lived 3 to 10 miles away had a higher prevalence of HCV seropositivity (aPR: 1.25, 95% CI 1.06–1.46), borrowing other used injection equipment (aPR: 1.23, 95% CI 1.04–1.46), and backloading (aPR: 1.48, 95% CI 1.17–1.88). Similar results were observed for PWID living > 10 miles from an SSP: aPR [HCV]: 1.19, 95% CI 1.01–1.40; aPR [borrowing other used equipment]:1.45, 95% CI 1.29–1.63; and aPR [backloading]: 1.59, 95% CI 1.13–2.24. Associations between living 1 to 3 miles of an SSP and each outcome did not reach statistical significance. When stratified by means of transportation, associations between distance to SSP and each outcome (except borrowing other used injection equipment) were only observed among PWID who traveled by other means (versus traveled by automobile).ConclusionsAmong PWID in rural New England, living farther from a fixed-site SSP was associated with a higher prevalence of HCV seropositivity, borrowing other used injection equipment, and backloading, reinforcing the need to increase SSP accessibility in rural areas. Means of transportation may modify this relationship.

Navigating the Maps: Euclidean vs. Road Network Distances in Spatial Queries

Using spatial data in mobile applications has grown significantly, thereby empowering users to explore locations, navigate unfamiliar areas, find transportation routes, employ geomarketing strategies, and model environmental factors. Spatial databases are pivotal in efficiently storing, retrieving, and manipulating spatial data to fulfill users’ needs. Two fundamental spatial query types, k-nearest neighbors (kNN) and range search, enable users to access specific points of interest (POIs) based on their location, which are measured by actual road distance. However, retrieving the nearest POIs using actual road distance can be computationally intensive due to the need to find the shortest distance. Using straight-line measurements could expedite the process but might compromise accuracy. Consequently, this study aims to evaluate the accuracy of the Euclidean distance method in POIs retrieval by comparing it with the road network distance method. The primary focus is determining whether the trade-off between computational time and accuracy is justified, thus employing the Open Source Routing Machine (OSRM) for distance extraction. The assessment encompasses diverse scenarios and analyses factors influencing the accuracy of the Euclidean distance method. The methodology employs a quantitative approach, thereby categorizing query points based on density and analyzing them using kNN and range query methods. Accuracy in the Euclidean distance method is evaluated against the road network distance method. The results demonstrate peak accuracy for kNN queries at k=1, thus exceeding 85% across classes but declining as k increases. Range queries show varied accuracy based on POI density, with higher-density classes exhibiting earlier accuracy increases. Notably, datasets with fewer POIs exhibit unexpectedly higher accuracy, thereby providing valuable insights into spatial query processing.

Changing in fire-inextinguishable areas (FIAs) in Tokyo based on Euclidean and road network distances

Changing in fire-inextinguishable areas (FIAs) in Tokyo based on Euclidean and road network distances

Proximity of alcohol outlets and presentation to hospital by young people after self-harm: A retrospective geospatial study using the integrated data infrastructure.

There is a well-established association between alcohol use, misuse, intoxication and self-harm, the latter of which is associated with suicide. This study aimed to better understand the association between proximity to alcohol outlets and the likelihood of young people presenting to hospital following self-harm. This was a nationwide retrospective geospatial study using data from the New Zealand Integrated Data Infrastructure using population-level data for 10-29-year-olds for the 2018 and 2017 calendar years. Presentations to hospital following self-harm were identified using the national minimum data set. Proximity to alcohol outlets was defined in road network distance (in kilometres) and ascertained using Integrated Data Infrastructure geospatial data. Alternative measures of proximity were employed in sensitivity analyses. Complete-case two-level random intercept logistic regression models were used to estimate the relationship between alcohol outlet proximity and hospital presentation for self-harm. Adjusted models included sex, age, ethnicity, area-level deprivation, urbanicity and distance to nearest medical facility. Analyses were also stratified by urbanicity. Of the 1,285,368 individuals (mean [standard deviation] age 20.0 [5.9] years), 7944 (0.6%) were admitted to hospital for self-harm. Overall, the odds of presenting to hospital for self-harm significantly decreased as the distance from the nearest alcohol outlet increased, including in adjusted models (adjusted odds ratio 0.980; 95% confidence interval = [0.969-0.992]); the association was robust to changes in the measure of alcohol proximity. The effect direction was consistent across all categorisations of urbanicity, but only statistically significant in large urban areas and rural areas. The findings of this study show a clear association between young people's access to alcohol outlets and presentation to hospital for self-harm and may provide a mandate for government policies and universal interventions to reduce young people's access to alcohol outlets. Further research regarding causative mechanisms is needed.

Learning Region Similarities via Graph-Based Deep Metric Learning

Region similarity learning plays an essential role in applications such as business site selection, region recommendation, and urban planning. Earlier studies mainly represent regions as bags of points of interest (POIs) for region similarity comparisons, which cannot fully exploit the spatial features of the regions. Recently, researchers propose to use deep neural networks to exploit spatial features such as POI geo-coordinates and categories, which have produced more accurate and robust region similarity learning results. However, many useful features such as the height and size of a POI, and the distance and relative importance between the POIs, are still overlooked in these methods. To take advantage of such features, we propose to represent regions as graphs, where nodes are POIs with rich features such as height, size, and hexagonal coordinates, while edges are the relationships between POIs formulated by their road network distances. To capture POIs' importance, we weigh them by their height and size. Since there is limited availability of ground-truth region similarity data, we propose a contrastive learning-based multi-relational graph neural network (C-MPGCN) for region similarity learning based on the graph representations. To generate data for model training, we propose a soft graph edit distance (SGED) based algorithm to generate triples of similar and dissimilar graphs of a given graph (representing a given region) based on the POI weights. Experimental results show that C-MPGCN outperforms the state-of-the-art methods for region similarity learning consistently with an improvement of at least 8.6% and 9.4% in terms of MRR and HR@1, respectively.

Automated external defibrillator location and socioeconomic deprivation in Great Britain

ObjectiveThe early use of automated external defibrillators (AEDs) improves outcomes in out-of-hospital cardiac arrest (OHCA). We investigated AED access across Great Britain (GB) according to socioeconomic deprivation.MethodsCross-sectional observational study using...

Exploring the Robustness of Alternative Cluster Detection and the Threshold Distance Method for Crash Hot Spot Analysis: A Study on Vulnerable Road Users

Traditional hot spot and cluster analysis techniques based on the Euclidean distance may not be adequate for assessing high-risk locations related to crashes. This is because crashes occur on transportation networks where the spatial distance is network-based. Therefore, this research aims to conduct spatial analysis to identify clusters of high- and low-risk crash locations. Using vulnerable road users’ crash data of San Francisco, the first step in the workflow involves using Ripley’s K-and G-functions to detect the presence of clustering patterns and to identify their threshold distance. Next, the threshold distance is incorporated into the Getis-Ord Gi* method to identify local hot and cold spots. The analysis demonstrates that the network-constrained G-function can effectively define the appropriate threshold distances for spatial correlation analysis. This workflow can serve as an analytical template to aid planners in improving their threshold distance selection for hot spot analysis as it employs actual road-network distances to produce more accurate results, which is especially relevant when assessing discrete-data phenomena such as crashes.

Temporal Metrics Based Aggregated Graph Convolution Network for traffic forecasting

Traffic forecasting is one of the most well-studied problems in the Intelligent Transportation Systems (ITS). However, existing studies mainly utilize Euclidean distance or road network distance as the metrics at the spatial level, where the distance between two points is calculated based on the GPS coordinates. When using the spatial level metrics to construct the road network’s topological and indexing structure, two completed unrelated road segments such as the upper and lower levels of the viaduct will be contiguous in the adjacent matrix and consequently be indexed in the same block. However, it may taking a long time to drive through these two road segments, which makes the spatial proximity meaningless, and this is pretty common during the morning and evening rush hours. Thus, it is very important to take into consideration of the temporal metric along with the spatial distance metric when dealing with the traffic forecasting problem.Motivated by the above observation, we propose a novel Temporal Metrics Based Aggregated Graph Convolution Network (TMAGCN), which utilize the time reachability based road network weight metric, and reveal its unique characteristics. TMAGCN mainly consists of three components: spatio-temporal reachability calculation, region clustering(road segments aggregating), and traffic flow forecasting. We explore the cluster of road segments and the reconstruction of the spatio-temporal reachability graph with temporal-level information rather than spatial-level information for better capturing the latent region information. To the best of our knowledge, this paper is the first attempt on this problem. The comprehensive experiments indicate that TMAGCN achieves higher accuracy and efficiency than other off-the-shelf models.

ROAD INFRASTRUCTURE DISTANCE INDICATORS AND RELATIONSHIP WITH AFFORDABILITY OF HOUSING

This paper looks at the road infrastructure distance indicators and explores the relationship with the affordability of housing. This paper focuses on the aspect of the built environment in Malaysia, with some review of ASEAN and Asian regions. The paper aims to outline the road infrastructure indicators based on distance and find out its relationship with housing affordability. The objective of this paper is to provide road infrastructure indicators based on distance and affordability of housing. The methodology of this review paper is the implementation of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), and two main journal databases: Web of Science and Scopus. Based on the SLR conducted, the researchers have extracted seven main themes: traffic number, commute cost, vehicle mobility, travel time, workplace distance, road network distance and house desirability. The findings were then utilised in determining affordable housing relationships with the indicators and within the Malaysian context. The themes were then expected to become the basis for future data collection processes by the researchers.

Urban densification over 9 years and change in the metabolic syndrome: A nationwide investigation from the ORISCAV-LUX cohort study

A growing body of evidence suggests that urban densification may be protective against obesity, type 2 diabetes, and cardiometabolic diseases, yet studies on how built environmental features relate to metabolic syndrome (MetS) and its components are scarce. This longitudinal study examines the associations of baseline urban density and densification over 9 years with MetS and MetS components, among 510 participants enrolled in both waves of the ORISCAV-LUX study (2007–2017) in Luxembourg. A continuous MetS score (siMS) was calculated for each participant. Six features of residential built environments were computed around participants’ home address: street connectivity, population density, density of amenities, street network distance to the nearest bus station, density of public transport stations, and land use mix. A composite index of urban densification (UDI) was calculated by averaging the six standardized built environment variables. Using adjusted generalized estimating equation (GEE) models, one-SD increase in UDI was associated with a worsening of the siMS score (β = 0.07, 95% CI: 0.02, 0.13), higher triglyceride levels (β = 0.05, 95% CI: 0.02, 0.09), and lower HDL-c levels (β = −1.29, 95% CI: −2.20, −0.38). The detrimental effect of UDI on lipid levels was significant only for participants living in dense areas at baseline. Higher baseline UDI, as well as increased UDI over time among movers, were also associated with greater waist circumference. There were no associations between UDI, fasting plasma glucose and systolic blood pressure. Sex and neighborhood socio-economic status did not moderate the associations between UDI and the cardiometabolic outcomes. Overall, we found limited evidence for an effect of urban densification on MetS and its components. Understanding urban dynamics remains a challenge, and more research investigating the independent and joint health effect of built environment features is needed to support urban planning and design that promote cardiometabolic health.

How Can We Measure Alcohol Outlet Density Around Schools? A Comparison Between Two Buffer-Based Methods

Measuring the density of alcohol outlets around schools is a critical step towards understanding the drivers of drinking among adolescents. Different methodologies have been used in the literature for this purpose, but the implications of using one methodology or another have not been clearly assessed. Our aim was to compare different methods to measure alcohol outlet density and highlight under which characteristics of the environment might be best using each approach. We used Geographic Information Systems to geolocate schools (n = 576) and alcohol outlets (n = 21,732) in Madrid. We defined the density of alcohol outlets as the number of establishments within an area of 400 m around schools measured using two buffering methods: crow flies’ and street network distances. We evaluated the agreement between both methods visually and through regression models, including street connectivity, population density, and density of recreational venues as predictors of disagreement. The density of alcohol outlets around schools was higher using crow flies’ distances compared to street network distances. The differences between methodologies were wider in areas of higher density of outlets, especially in the downtown areas, where there are higher population density and street connectivity. Our results suggest that the spatial characteristics and morphology of the study area (e.g., street connectivity and population density) should be considered when deciding the methodology to be used to measure alcohol outlet density. Future studies should explore the implications of different exposure measures in their association with drinking prevalence and consumption patterns among different geographical contexts.

Road network distances and detours in Europe: Radial profiles and city size effects

The form and the size of cities influence their social, economic and environmental outcomes. The form of a city is itself influenced by the shape of its road network, but this relationship and how it is affected by city size are unclear. We analyse how road distances to the main centre vary across 300 European cities and how radial physical detours (i.e. the distance on the road network compared to the Euclidean distance) are affected by city size and extent. We use landuse and population data to sample potential residences and compute the fastest routes to the main centre. We find a linear relationship between road and Euclidean distances, and for the first time document an average radial physical detour of 1.343 across Europe. We then rescale distance bands so to make cities of different population size comparable and show the effect of different urban delineations. We find that physical detour ratios increase when core cities only are considered without suburbs. At the urban region scale, radial physical detours increase with city size, especially when other significant geographical factors (latitude, longitude, elevation change and proximity to coast) are controlled for. When the central part of cities only is considered, larger cities have smaller radial physical detours.

Green space type and healthy ageing in place: An Australian longitudinal study

Research on the health benefits of people’s long-term exposure to green space is lacking. Addressing this emerging topic, this study uncovers the association of green space types with a person’s healthy ageing score (HAS). We investigated the association with HAS by continuous exposure to various green space types for two years. Green space types include tree canopy, low-lying vegetation and open grass percentage within 1.6-km road network distance buffers at baseline and second follow-up. Healthy ageing outcomes included functional capacity, resilience and HAS while accounting for relocation. This is a longitudinal study of a cohort of 22,715 New South Wales residents aged over 45 who participated in the Sax Institute’s 45 and Up Study living in Sydney, Wollongong and Newcastle. Longitudinal models of healthy ageing on green space types were fit with controls for socioeconomic confounders. Women who did not relocate were associated with a lower functional capacity (β; 95%CI: −0.10; −0.15,−0.05) and higher resilience (0.11; 0.08, 0.14) compared to those who relocated. Apart from age, personal characteristics did not explain the variation in healthy ageing outcomes for participants who moved. For participants who did not relocate, 30% grass cover was associated with decreased functional capacity (−0.22; −0.41,−0.04) and HAS (−0.31; −0.56,−0.05). Also, 5–9% low-lying vegetation was associated with a decline in functional capacity (−0.09; −0.15, −0.03) and HAS (−0.09; −0.17, −0.01) of participants who stayed but improved resilience (0.28; 0.01, 0.55) of participants who relocated. Green space comprising over 30% tree cover appeared most beneficial for functional capacity (0.42; 0.31, 0.53), resilience (0.19; 0.13, 0.25) and HAS (0.60; 0.45, 0.75). For participants who relocated, over 30% tree canopy was associated with improved functional capacity (0.33; 0.54, 0.62) but not resilience or their HAS. Increased neighbourhood tree canopy supports healthy ageing. Older people should be supported to live in-home or move into nearby residential care and maintain regular contact with green spaces and trees, to maximise potential benefits for health and wellbeing.

Are the Effects of Opening New Mass Rapid Transit Segments in Taiwan on Nearby Housing Prices Positive?

This study evaluates the effects of opening new segments of mass rapid transit (MRT) lines on housing prices near the MRT stations in Tucheng District and Xinzhuang District, New Taipei City, Taiwan. The effect of proximity to each MRT station is estimated separately with difference-in-differences regressions integrated with spatial econometrics with heteroscedasticity-robust standard errors. The opening of the new segment of the Blue Line, also known as Bannan Line of the Taipei Metro, does not significantly influence housing prices within 600-metre road network distance of the MRT stations, compared to prices outside the distance range. In contrast, and also unlike the findings of prior studies, although the segment and the stations are underground structures, the opening of the new segment of the Orange Line, also known as the Zhonghe-Xinlu Line of the Taipei Metro, significantly decreases housing prices within 600-metre road network distance of the MRT stations, compared to prices outside the distance range, perhaps because the opening of the stations is delayed about one and a half years to be used as a temporary storage area for MRT trains. The findings have implications for homebuyers, investors, mortgage lending institutions and tax assessment authorities.

Bidirectional k-nearest neighbor spatial crowdsourcing allocation protocol based on edge computing.

Spatial crowdsourcing refers to the allocation of crowdsourcing workers to each task based on location information. K-nearest neighbor technology has been widely applied in crowdsourcing applications for crowdsourcing allocation. However, there are still several issues need to be stressed. Most of the existing spatial crowdsourcing allocation schemes operate on a centralized framework, resulting in low efficiency of crowdsourcing allocation. In addition, these spatial crowdsourcing allocation schemes are one-way allocation, that is, the suitable matching objects for each task can be queried from the set of crowdsourcing workers, but cannot query in reverse. In this article, a bidirectional k-nearest neighbor spatial crowdsourcing allocation protocol based on edge computing (BKNN-CAP) is proposed. Firstly, a spatial crowdsourcing task allocation framework based on edge computing (SCTAFEC) is established, which can offload all tasks to edge nodes in edge computing layer to realize parallel processing of spatio-temporal queries. Secondly, the positive k-nearest neighbor spatio-temporal query algorithm (PKNN) and reverse k-nearest neighbor spatio-temporal query algorithm (RKNN) are proposed to make the task publishers and crowdsourcing workers conduct two-way query. In addition, a road network distance calculation method is proposed to improve the accuracy of Euclidean distance in spatial query scenarios. Experimental results show that the proposed protocol has less time cost and higher matching success rate compared with other ones.

What types of green space disrupt a lonelygenic environment? A cohort study.

Associations between green space type and social loneliness (a scarcity of people one feels they can depend on) were investigated in city-living participants in the Sax Institute's 45 and Up Study. Availability of green space, tree canopy and open grass were measured as a percentage of land-use within 1.6 km road-network distance buffers using high-resolution data. Multilevel logistic regressions adjusted for confounding tested associations between each green space indicator with the odds of social loneliness at baseline (prevalence) and follow-up (incidence), adjusted for demographic and socioeconomic confounders. The prevalence of social loneliness at baseline was 5.3% (n = 5627 /105,498). Incidence of social loneliness at follow-up was 3.4% (n = 1772/51,365). Adjusted regressions indicated reduced odds of prevalent (OR = 0.95, 95%CI = 0.92-0.98) and incident social loneliness with 10% more green space (OR = 0.92, 95%CI = 0.90- 0.96). Similar associations were found with a 10% increase in tree canopy for both prevalent (OR = 0.92, 95%CI = 0.88-0.95) and incident social loneliness (OR = 0.92, 95%CI = 0.88-0.97). Two-way interaction terms indicated effect modification by sex but not couple status. Among women, a 10% increase in total green space was associated with lower odds of prevalent (OR = 0.95, 0.91-0.95) and incident (OR = 0.89, 0.85-0.95) social loneliness. A 10% increase in tree canopy among women was associated with lower odds of prevalent (OR = 0.89, 085-0.92) and incident (OR = 0.85, 0.80-0.92) social loneliness. Meanwhile, a 10% increase in open grass among women was associated with higher odds of prevalent (OR = 1.08, 1.01-1.15) and incident (OR = 1.19, 1.03-1.35) social loneliness. Associations for men were statistically significant for a 10% increase in total green space (OR = 0.96, 0.92-0.99) and tree canopy (OR = 0.93, 0.90-0.97) for prevalent social loneliness only. Urban greening and tree canopy restoration may reduce risks of social loneliness, perhaps especially in women.