2026-08-21

7 arXiv categories· 96 journals· 2187 candidates — 9 worth your time· 83 without an open abstract

No headline today

nothing cleared the headline bar today — the day's papers are below

Today, together

tag shift
Transportation and Mobility Innovations on 3 items, against a 26-day average of 0.81
shares references
no items share references today counted over the 9 papers published above
on the same shoulders
4 of the day's papers cite Travel and the Built Environment (Reid Ewing, Robert Cervero 2010) counted over the 76 of this day's 92 named papers whose reference lists we have — standing on the same older work is not the same as being about the same thing
first appearance
European University Institute; Identités et Différenciation de l'Environnement des Espaces et des Sociétés; Institute for Urban and Regional Research; and 6 more none of them on a paper we published in the 70 days before this issue

Two counts, not one counted twice: every row here but “on the same shoulders” counts the papers published above, and that one counts every paper this day named — those and the ones under Also published today, whose abstracts we could not read.

published: Computational Urban Science

Shahfahad, Swapan Talukdar, Mohd Waseem Naikoo, et al.
Computational Urban Science

Banaras Hindu University

Urban Heat Island Mitigation · Flood Risk Assessment and Management · Remote Sensing and Land Use · random forest · gradient boosting · geemap

Predicting surface urban heat island intensity and its drivers using machine learning and spatial XAI for achieving SDG-11

The study modeled Surface Urban Heat Island (SUHI) intensity in the Mumbai Metropolitan Region from 2002 to 2024 using Google Earth Engine and four machine learning algorithms. LightGBM performed best with a test R² of 0.86 and an RMSE of 1.49 °C, while spatial cross-validation yielded a robust R² of 0.77 to 0.78. Explainable AI (XAI) and a novel spatial XAI model identified built-up intensity, bare land, night-time lights, and population density as the primary positive drivers of heat intensity, which shifted from 13.69% low-intensity coverage in 2002 to 6.31% in 2024.

Why it matters — It introduces a spatial explainable AI approach that maps exactly where and how different urban features drive heat accumulation across a metropolitan area. This shifts urban heat mitigation from uniform, city-wide policies to targeted, location-specific interventions based on localized driver patterns.

Caveat: The predictive models and spatial XAI relationships were developed and tested solely within a single metropolitan region, which may limit direct transferability to cities with different climatic or morphological characteristics.

Computational Urban Science · pdf · doi

published: Discover Cities

Ida Shaheera Bakhtiar, Mohd Zamreen Mohd Amin
Discover Cities

University of Technology Malaysia

Smart Cities and Technologies · Transportation and Mobility Innovations · Digital Economy and Work Transformation · University of Technology Malaysia · Universiti Teknologi MARA

The roles of institutional trust and spatial justice in the resilient smart mobility systems

This conceptual paper introduces the Just Smart Mobility Model (JSMM), a theoretical framework that integrates Mobility Justice with the Technology Acceptance Model (TAM) to analyze smart mobility systems. The framework redefines urban safety to include protection from algorithmic bias and spatial marginalization, positioning institutional trust as a key mediator for technology adoption. The authors apply this model to a spatial vignette of the Iskandar Rapid Transit (IRT) in Malaysia to demonstrate how historical planning decisions that bypassed low-socioeconomic communities create digital exclusion and trust deficits.

Why it matters — It shifts the evaluation of smart mobility systems from purely technical optimization to a socio-spatial perspective, demonstrating that digital exclusion and infrastructural neglect function as fundamental safety risks that can cause smart systems to fail through social disconnect rather than technical malfunction.

Caveat: The framework is conceptual and illustrated through a single qualitative vignette rather than being statistically validated across multiple empirical settings.

Discover Cities · doi

published: Cities

Sofia Borushkina, Антон Городничев
Cities

Monash University Prato Centre

Cross-Border Cooperation and Integration · Urbanization and City Planning · Regional resilience and development · Monash University Prato Centre · European University Institute · National Research University Higher School of Economics

Spatial governance under polycrisis: Reconfiguring agglomeration policy

This study traces how Russia's national spatial planning priorities shifted between its 2019 and 2024 Spatial Development Strategies, using policy documents, implementation materials, and expert discussions. It analyzes how the state's spatial planning rationalities pivoted away from economic growth and global integration toward territorial stabilization, population retention, and managing systemic vulnerability.

Why it matters — The research demonstrates that agglomeration policy is not a universally stable model but a contingent framework that, under conditions of polycrisis, can be repurposed from a growth-oriented governance technology into a mechanism for territorial anchoring and state control.

Caveat: The findings are based on a single-country case study of Russia, which limits their direct generalizability to other national contexts experiencing different types of crises.

Cities · pdf · doi

published: Cities

Daphne Wing Chan, Nancy Y. Liu, Helen X. H. Bao
Cities

University of Cambridge

Housing, Finance, and Neoliberalism · China · University of Cambridge

The gender dimension of homeownership: Evidence from China

This study analyzes the relationship between the gender of household heads and three housing asset indicators in China: primary homeownership, second homeownership, and housing asset value, while also testing if having a son alters these dynamics. The empirical analysis reveals that while male household heads have a higher probability of primary homeownership, female household heads are more likely to own a second home and female-headed households hold higher-value housing assets than their male counterparts.

Why it matters — The findings challenge the assumption of uniform male advantage in real estate accumulation by demonstrating that gender disparities in housing wealth operate differently across primary ownership, secondary property investment, and asset valuation.

Cities · doi

preprint

Mohamed Khalafalla, Doreen Kobelo, Thobias Sando, et al.
arXiv physics.soc-ph

Urban Transport and Accessibility · Older Adults Driving Studies · Transportation and Mobility Innovations

Rural Commute Patterns

Using data from the 2017 National Household Travel Survey (NHTS), this study analyzed rural commuting patterns using multinomial logistic regression to identify factors influencing transport mode choice. The analysis reveals a 92.1% dependency on personal vehicles for rural trips and demonstrates how socio-demographic, household, and trip characteristics alter transit and active travel choices.

Why it matters — The study quantifies the extreme reliance on private vehicles in rural areas and identifies specific demographic groups—such as older adults, women, and higher-income households—who are least likely to use public transit, providing a baseline for targeted rural transport planning.

Caveat: The analysis relies on self-reported survey data from a single year, 2017, which may not capture recent post-pandemic shifts in rural travel behavior.

arXiv · pdf

preprint

Victor Vignolles, Rémi Lemoy
arXiv physics.soc-ph

Economic Growth and Productivity · Identités et Différenciation de l'Environnement des Espaces et des Sociétés

Removing Population Size, World Cities Leave on Land a Footprint of Wealth

Using a dataset of over 1,800 global urban areas, this study identifies a scaling law where built-up footprint scales proportionally to total population, indicating that built-up area per capita remains constant regardless of city size. The analysis reveals that world cities share homothetic radial land use profiles and similar internal structures. Deviations from this baseline scaling law are strongly correlated with wealth, specifically gross domestic product per capita.

Why it matters — The findings establish that urban land consumption is not inherently more efficient in larger cities, challenging the assumption of economies of scale in land use and demonstrating that economic development directly drives higher per capita land footprints.

arXiv · pdf

preprint

Joan Perez, Giovanni Fusco
arXiv cs.CV

Urban Transport and Accessibility · Urban Design and Spatial Analysis · Human Mobility and Location-Based Analysis · street view imagery

From Street View Imagery to Street Quality Indicators: Vision Language Inference for the Suburban 15-minute City

The study evaluates streetscape quality across several thousand street-level observations in the north-eastern periphery of Nice, France, using SAGAI, an open-source workflow that leverages vision-language models on Google Street View imagery. The updated workflow assesses sidewalk presence, pedestrian entrance density, and vegetation, revealing that these pedestrian-friendly qualities are concentrated in compact developments and traditional suburban faubourgs, while largely absent on residential hills.

Why it matters — It demonstrates a scalable, automated method for evaluating pedestrian infrastructure across extensive suburban and peri-urban territories where manual field surveys are too resource-intensive to conduct.

Caveat: The assessment relies on street-level imagery proxies rather than direct physical measurements of pedestrian infrastructure quality.

arXiv · pdf

preprint

Narges Ahmadi, Yubo Jiao, Jônatas Augusto Manzolli, et al.
arXiv cs.AI

McGill University

Human Mobility and Location-Based Analysis · Persona Design and Applications · Transportation and Mobility Innovations · regression · random forest

An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction

This study developed a three-agent workflow that coordinates conversational data collection, structured processing, and behavioral prediction using a chatbot-administered, image-augmented survey of student commuters. The system collected 454 respondent-scenario observations across five weather conditions, evaluating travel mode choices using multinomial logit, logistic regression, random forest, and nine locally deployed large language models (LLMs) ranging from 2B to 35B parameters. The best text-only zero-shot LLM achieved 69.9% five-class accuracy, while the top vision-based LLM configuration reached 71.5% accuracy by processing the same weather images shown to respondents.

Why it matters — It demonstrates that zero-shot and vision-enabled LLMs can match or exceed the predictive accuracy of traditional machine learning models in travel behavior forecasting without requiring task-specific fitting, while establishing a fully automated, auditable pipeline from survey collection to demand prediction.

Caveat: The workflow was evaluated on a small sample of 454 observations from student commuters in a single, unspecified university context, which may limit its immediate generalizability to broader urban populations.

arXiv · pdf

preprint

Nils Bonfils, Christoph Becker
arXiv cs.CY

Green IT and Sustainability · Digital Marketing and Social Media · Digital Platforms and Economics

A Scoping Review of Methods to Measure the Energy and Carbon Footprint of Web Tracking and Advertising

This scoping review analyzed 15 papers selected from an initial pool of 46 on Google Scholar to categorize methods for measuring the energy and carbon footprint of web tracking and advertising. It identifies five distinct methodological approaches used in the literature: ad blocking, controlled environments, replaying ads, traffic flow analysis, and literature-derived estimation.

Why it matters — It establishes a structured taxonomy and methodological foundation for a research area that has previously been scattered across disciplines and lacked consistent terminology, enabling more systematic environmental accounting of the digital advertising ecosystem.

Caveat: The findings are based on a small final corpus of 15 papers retrieved through a title-based search on a single database.

arXiv · pdf

Still cited

Shushi Peng, Shilong Piao (2011), Surface Urban Heat Island Across 419 Global Big Cities 1 of today's items cite it · 12 of 3240 in the archive stand on it

Also published today

These appeared today in journals we track. Their abstracts are not openly available, so we cannot summarise them.