Urban Currents

Assessing urban mobility behaviors and inequalities in Morocco’s major cities through socio-economic indicators: Cases of Casablanca and Rabat

This study analyzes mobility inequalities in Casablanca and Rabat, Morocco, using household travel survey data covering more than 36,000 individuals. It combines descriptive socioeconomic profiling across 18 demographic configurations with multivariate inferential models to measure how income, gender, age, employment, and residential location shape travel patterns. The analysis reveals that low-income households rely on slower transport modes, women experience higher immobility rates and smaller spatial reach, and peripheral residency compounds these disadvantages.

Why it matters — It provides empirical, intersectional evidence of urban transport disadvantages in North African cities, demonstrating that the combination of being older and female creates a risk of immobility that exceeds the additive effects of age and gender alone.

Caveat: The study relies on self-reported household travel surveys, which may underreport short trips or informal mobility practices.

Journal of Urban Mobility · doi

Machine learning downscaling of satellite derived land surface temperature in the historic urban heritage area of São João del-Rei Brazil

The study downscaled Landsat 8/9 thermal imagery from 30 m to 10 m resolution for the historic center of Sao Joao del-Rei, Brazil, covering wet summers (January) and dry winters (July) between 2019 and 2025. A CatBoost Regressor was trained on Sentinel-2 spectral indices, revealing that vegetation and moisture indices dominate land surface temperature variability in summer, while the built-up index becomes the primary predictor in winter. The downscaled maps showed lower estimated temperatures near vegetated areas, water bodies, and compact street segments compared to open, impervious spaces.

Why it matters — It provides a reproducible, high-resolution spatial downscaling framework that allows planners to identify micro-scale thermal variations within historic urban fabrics using freely available satellite data.

Caveat: The downscaled maps represent model-derived estimates rather than direct street-scale thermal observations, and they do not provide direct evidence of thermal inertia, urban canyon effects, or pedestrian comfort.

Discover Cities · pdf · doi

Development and validation of a smart destination perception scale for tourists and residents (SDPS): Evidence from Madrid

The study develops and validates the Smart Destination Perception Scale (SDPS), a 29-item instrument measuring how tourists and residents perceive smart cities. Using survey data from 1,200 participants in Madrid (split equally between 600 residents and 600 tourists), exploratory and confirmatory factor analyses identified four core dimensions: Innovation, Connectivity, Market Intelligence, and Social and Environmental Responsibility.

Why it matters — It provides a standardized, validated tool that allows municipal authorities to directly compare smart city performance across different destinations while simultaneously capturing the distinct viewpoints of both visitors and local inhabitants.

Caveat: The scale was validated using data from a single, highly mature smart destination, which may limit its immediate generalizability to cities in earlier stages of smart development.

Cities · doi

A return ticket for transport investment: How Land Value Capture supported London's Crossrail

This study evaluates the success of London's Mayoral Community Infrastructure Levy (MCIL), a land value capture mechanism that has raised over £1 billion to help fund the Crossrail transit project. The analysis applies Marsh and McConnell's multi-dimensional policy success framework to assess how the levy balanced revenue generation against the financial viability of private property developments and local borough infrastructure needs.

Why it matters — It provides empirical evidence on how a major global city successfully implemented a large-scale land value capture program without stalling private real estate development or stripping local municipalities of funding for affordable housing.

Caveat: The evaluation is based on a single, highly specific institutional and economic context in London, which may limit its direct application to cities with different planning and taxation powers.

Cities · pdf · doi

Citizen engagement in practice: Prioritizing and monitoring factors that influence citizen engagement

The study developed and validated a Structured Methodological Framework to measure and sustain resident participation in urban regeneration. The framework was built using a four-cycle Participatory Action Research approach that prioritized 36 engagement factors with 20 experts, validated a diagnostic survey with 160 residents, mapped Human-Centered Design tools to these factors, and tested the system through a three-year longitudinal case study.

Why it matters — It provides urban managers with a diagnostic system to establish community baselines and strategically select targeted interventions, transforming citizen engagement from a static metric into an actionable, trackable process.

Caveat: The framework's longitudinal validation is based on a single case study of an urban regeneration project, which may limit its immediate generalizability to other urban governance contexts.

Cities · doi

Conditions influencing City-level adaptation and mitigation: A systematic review across high- and low-income country contexts

This systematic review analyzed 150 peer-reviewed studies published between 2003 and 2023 to identify the governance and contextual conditions shaping urban climate action. Following the PRISMA 2020 protocol, the study categorizes these conditions across vertical, horizontal, internal, and contextual domains, while distinguishing between mitigation and adaptation pathways in both high-income countries and low- and middle-income countries (LMICs).

Why it matters — It establishes distinct, configurational pathways for urban climate action, showing that mitigation is driven by norms and incentives, adaptation is driven by mandates and extreme weather risks, and LMIC contexts often rely on donor-led or co-benefit pathways due to centralized governance and limited vertical support.

Caveat: The findings are constrained by existing literature biases, including conceptual ambiguity, a heavy focus on planning rather than implementation, and a general lack of empirical evidence from LMIC contexts.

Cities · pdf · doi

Exploring cargo bike viability for last-mile delivery in European capitals: A mass-participation multi-city Multi-Criteria Satisfaction Analysis

This study evaluated public acceptance of a hypothetical cargo bike delivery service, EcoDeliver, using a survey of approximately 2,500 citizens across six European capital cities. The researchers applied a multi-city Multicriteria Satisfaction Analysis (MUSA) model to analyze how residents prioritize and evaluate five key criteria: environmental impact, noise, traffic congestion, safety, and shipping cost. The results show generally positive public reception, with environmental benefits, noise reduction, and congestion relief scoring the highest.

Why it matters — It demonstrates that while overall acceptance of cargo bike logistics is consistently high across different capitals, the underlying priorities and demographic preferences vary significantly by location. This provides planners with a method to diagnose local resistance or support structures rather than assuming a uniform public response to green logistics initiatives.

Caveat: The findings are based on citizen evaluations of a hypothetical delivery service scenario rather than an active, deployed cargo bike system.

Cities · doi

Beyond personal factors: How external factors are associated with readiness for mobility transition in older drivers

This study analyzed how external factors influence older drivers' readiness for driving cessation using hierarchical regression models on the 16-item Assessment of Readiness for Mobility Transition (ARMT-16). Adding external variables to a baseline model of demographic, physical, cognitive, and personality traits significantly increased the explained variance in readiness scores from 47.6% to 51.1%. Specifically, individuals with limited ride options from family and friends scored 0.22 points higher (indicating lower readiness) than those with constant access, while those in multi-person households scored 0.20 points lower than solo dwellers.

Why it matters — It demonstrates that an older adult's preparedness to stop driving is shaped not just by personal health and cognitive capacity, but by tangible social and household support structures, allowing counselors to better tailor transition planning.

Caveat: The analysis is based on a relatively small sample size of 264 participants, as indicated by the regression degrees of freedom.

Journal of Transport & Health · doi

Road traffic noise analysis in urban roundabouts: a systematic review

This systematic literature review synthesizes research on road traffic noise at urban roundabouts, analyzing studies that primarily use field measurements at a height of 1.5 meters across three daily periods alongside noise prediction models like NMPB-Routes, RLS-90, and CNOSSOS-EU. The synthesized findings, concentrated in Europe and Asia, show that roundabouts generally produce lower noise levels than signalized intersections due to smoother traffic flow and fewer acceleration and deceleration events.

Why it matters — It establishes that while roundabouts offer acoustic benefits over traffic lights, these advantages can be diminished or reversed by the introduction of traffic signals that alter vehicle dynamics, highlighting a critical trade-off for urban intersection design.

Caveat: The synthesized findings are limited by a lack of methodological standardization in measurement and modeling procedures across the reviewed literature.

Journal of Transport & Health · doi

Public transport challenges and technology-assisted accessibility for visually impaired elderly residents in urban environments

This study used a mixed-methods approach to analyze public transport accessibility in Edinburgh, combining spatio-temporal machine learning analyses of live-recorded transit data with qualitative thematic analysis of semi-structured interviews with visually impaired elderly residents. The analysis mapped network coverage, service patterns, and density profiles, while the interviews explored spatial cognition, memory-based navigation, and the use of assistive technology.

Why it matters — It demonstrates that while vulnerable elderly residents are willing to adopt artificial intelligence and digital navigation tools, their independence is restricted by a highly centralized transit network, a lack of real-time information in accessible formats, and a systemic failure to collect and store granular accessibility data for physical facilities.

Caveat: The empirical findings are based on a single-city study of Edinburgh, which has a highly historic and complex urban landscape that may not represent modern or grid-based transit environments.

Journal of Transport Geography · pdf · doi

Nonlinear effects of tree cover on crime in urban parks: A generalized additive modeling approach

The study analyzed 2,081 urban parks in Seoul, South Korea, using generalized additive models and semi-parametric piecewise linear terms to examine the non-linear relationship between tree canopy cover and crime risk. The analysis revealed that crime risk initially increases by 1.9% for every 1 percentage-point increase in tree cover, but reverses to a 0.8% decline once canopy cover exceeds 48%, and drops further by 2.9% above an 89% canopy threshold. Additionally, the research identified that low tree cover paired with moderate grass coverage correlates with elevated crime risk.

Why it matters — It demonstrates that the relationship between park vegetation and crime is non-linear and threshold-dependent, overturning previous linear assumptions that led to conflicting conclusions about whether trees deter or facilitate crime.

Caveat: The findings are based on a single-city analysis of Seoul, which may limit generalizability to cities with different urban layouts, park management practices, or crime dynamics.

Urban forestry & urban greening · doi

Street-view measures of urban greenness, area-level deprivation, and the risk of firearm-related homicide: A matched case-control study

A matched case-control study in Cook County, Illinois, analyzed 899 firearm-related homicides and 2,256 natural deaths to evaluate the protective effects of urban vegetation. The study measured greenness using satellite-derived NDVI, tree canopy coverage, park proximity, and two machine learning-derived metrics from Google Street View images. The green view index from street-level imagery (OR = 0.693), walkable street greenness (OR = 0.787), and tree canopy coverage (OR = 0.792) were all significantly associated with lower odds of firearm homicide.

Why it matters — It demonstrates that the protective association of urban greenness against violent crime is highly dependent on neighborhood socioeconomic context, showing that tree canopy and segmented green space provide substantially stronger protective effects in less deprived areas.

Caveat: The study relies on a retrospective case-control design in a single county, meaning the observed protective effects of greenness may not generalize to other geographic regions with different urban layouts or socioeconomic dynamics.

Urban forestry & urban greening · pdf · doi

Developing business ecosystems around urban nature-based solutions: Evidence from Finland

This study identifies the key actors, barriers, and solutions for integrating nature-based solutions (NBS) into urban business ecosystems by analyzing two Finnish cities participating in the EU Mission on Climate-Neutral and Smart Cities. Based on interviews and focus groups with 21 experts in 2023, the research outlines seven thematic solution areas for scaling NBS, including economic instruments, policy tools, and collaborative business models.

Why it matters — It demonstrates that the success of urban nature-based solutions depends on a highly active city-business interface, positioning municipal governments as multi-role facilitators rather than just regulators in local green business ecosystems.

Caveat: The findings are based on qualitative data from a small sample of 21 experts in two frontrunner cities, which may limit their direct applicability to cities with less municipal capacity or different regulatory contexts.

Urban forestry & urban greening · doi

Run to the hills: Car ownership as a shield against ageing anxiety in China

Using double/debiased machine learning on a nationally representative survey of 2,068 respondents in China, this study measures the relationship between household car ownership and ageing anxiety. The analysis reveals that owning a car significantly reduces anxiety related to ageing, though this protective effect is concentrated among men, urban dwellers, and younger adults, while showing no significant impact for women, rural residents, peasants, and middle-aged individuals.

Why it matters — The study establishes a direct link between personal automobility and psychological resilience to ageing, demonstrating that transport access serves as a psychosocial buffer rather than just a utility for physical transit.

Caveat: The findings rely on self-reported survey data to measure anxiety levels, and the observed mitigating effects of car ownership do not extend to the demographic groups reporting the highest baseline levels of ageing anxiety.

Journal of Urban Mobility · pdf · doi

WildCity: A Real-World City-Scale Testbed for Rendering, Simulation, and Spatial Intelligence

The researchers introduced WildCity, a real-world multimodal dataset consisting of 18 trajectories that average 83.7 kilometers each, captured by autonomous vehicle fleets. The dataset preserves real-world perception challenges like dynamic objects, lighting changes, and imperfect camera poses. Using this data, they built an urban-tailored reconstruction baseline and converted the reconstructed environments into a closed-loop simulator.

Why it matters — It provides a city-scale testbed that allows researchers to train and evaluate spatial intelligence models, moving beyond small-scale scene reconstructions to environments spanning tens of square kilometers.

arXiv · pdf

Reward-Density Heuristic for Dynamic Multi-Vehicle Routing: Performance and Computational Efficiency

The study introduces and evaluates the 'Efficiency heuristic', a reward-density heuristic designed for dynamic, online multi-vehicle routing and orienteering problems where fleets must maximize reward under fixed time limits. The method was tested against four classical construction heuristics and three metaheuristics (Adaptive Large Neighbourhood Search, Genetic Algorithm, and Simulated Annealing) across autonomous drone task allocation and urban taxi dispatch scenarios with varying fleet sizes.

Why it matters — It demonstrates that a simple, greedy reward-density heuristic can match the solution quality of complex metaheuristic algorithms while reducing computational planning time by two to three orders of magnitude. This establishes that sophisticated search procedures are not necessary for real-time, online fleet dispatch systems where rapid replanning is required.

arXiv · pdf

STAGformer: A Spatio-temporal Agent Graph Transformer for Micro Mobility Demand Forecasting

The STAGformer model forecasts station-level bike-sharing demand by utilizing a two-step agent attention mechanism that reduces standard self-attention complexity to linear O(NT). The architecture integrates a spatio-temporal encoder with weather, time, and point-of-interest data, alongside graph propagation and temporal convolution modules. It was validated using real-world trip datasets from NYC Citi-Bike and Chicago Divvy-Bike, outperforming existing baseline models in RMSE and MAE across multiple prediction horizons.

Why it matters — It enables computationally efficient, global spatio-temporal modeling for large-scale micro-mobility networks without the quadratic processing costs that typically limit standard transformer models.

arXiv · pdf

Regulation, Power, and the Compliance 1 Paradox: A Longitudinal Study of Smart Homes

This longitudinal study tracks smart home adoption in Jordan across two phases, combining a secondary analysis of 30 interviews from 2022 with 28 new interviews conducted in 2025. The sample spans household members, domestic workers, policymakers, and civil society advocates to examine how domestic surveillance practices interact with cultural norms and Jordan's 2023 Data Protection Law.

Why it matters — It documents a compliance paradox where new privacy laws actually facilitate domestic exploitation because the regulations exclude the household sphere, shifting legal responsibility to families and intensifying the surveillance of domestic workers.

Caveat: The findings are based on qualitative interview data from a single national context, which may limit generalizability to regions with different legal frameworks and domestic labor structures.

arXiv · pdf