Posts | Urban Demographics

Trip to Boston and DC in March

Thu, 05 Feb 2026 00:00:00 +0000

I’ll be in the US in the first two weeks of March. Agenda below. More importantly above all, I’m super excited to meet old and new friends.

Boston, 1st week of March

March 2: I’ll be giving a lecture at MIT in Fabio Duarte’s class on “Comparative land use and transportation planning”
March 3: I’ll be presenting a seminar the Harvard Graduate School of Design
March 4: I’m also super excited I’ll be meeting Esteban Moro’s team at Northeastern, where I’ll be presenting a seminar as well
March 5 and 6: If you’re in Boston, don’t miss the Urban Mobility and Climate Change Conference. I’ll be speaking on urban form and sustainable mobility, alongside a stellar lineup of speakers addressing key issues at the intersection of urban mobility and climate change. More info and registration available here

Washington, DC, 2nd week of March

March 10, 11 and 12: I’ll attending Transforming Transportation, where I’ll be presenting at the “Transport As a Catalyst for Jobs” session

CfP Decolonising Research in Transport Geography

Thu, 11 Dec 2025 00:00:00 +0000

We have a new Call for Papers for a special issue on “Decolonising Research in Transport Geography”, in the Journal of Transport Geography. Please help us spread the word!

Submission deadline: 30 September 2026
See more info below and on this link.

Guest editors:

Qiyang Liu, School of Urban Planning and Design, Peking University, Shenzhen, China
Zihao An, University of Leeds, Leeds, UK
Rafael H. M. Pereira, Institute of Applied Economic Research, Brasilia, Brazil
Zahara Batool, Institute for Transport Studies, University of Leeds, Leeds, UK
Tim Schwanen, University of Oxford, Oxford, UK

Special issue information:

Transport geography, often dominated by Global North narratives, frequently applies established paradigms uncritically to diverse global contexts, particularly the Global South. This risks obscuring unique transport processes and their distinct local causes, implications, and responses. While foundational work has initiated crucial discussions on decolonising transport research by highlighting power imbalances and knowledge hegemonies, the past five years have witnessed intensified debates and theoretical advancements, necessitating a deepened and broadened engagement.This Special Issue offers an urgent and novel platform to advance decolonisation by fostering genuine knowledge pluralism. It challenges prevailing frameworks that inadvertently universalise not only which issues are deemed important but also how they are conceptualised and addressed. We advocate for methodological innovation and direct engagement with alternative epistemologies, moving beyond problem identification to proactive engagement. A key focus is on integrating indigenous and local knowledges into transport and infrastructure planning, promoting bottom-up approaches that truly reflect global complexity.

The issue places significant emphasis on the rapidly evolving urbanisation patterns and socio-economic transformations prevalent in the Global South. It addresses contemporary challenges such as the dual nature of informal transport systems (e.g., women’s safety, congestion) and the complex implications of emerging future transport systems (e.g., autonomous vehicles, MaaS) in potentially exacerbating inequalities. These complexities demand specific, contextually nuanced understandings that move beyond universal solutions. Ultimately, this collection fundamentally rethinks disciplinary tools and assumptions, critically reconsidering existing transport models and priorities from a decolonial lens to foster a more inclusive, equitable, and context-specific understanding of transport futures.

This Special Issue invites studies that proactively decolonise the discipline, moving beyond the reactive application of established frameworks. We encourage contributions that critically assess underlying assumptions, integrate multiple epistemologies, and engage with diverse conceptualisations of transport. We welcome theoretical, empirical, and analytical contributions that critically engage with the legacies in transport research and propose pathways towards knowledge pluralism and equitable transport futures. Papers that interrogate power relations in transport knowledge production, centre indigenous and local knowledges, and explore innovative decolonial narratives are especially encouraged.

Potential topics may include, but are not limited to, the following:

Socioeconomic roles and spatial dynamics of context-specific formal and informal transport in the Global South
Emerging future transport systems and their societal implications in the Global South
Social and environmental implications of informal and active transport modes in developing regions
Governance and policy challenges in regulating and integrating diverse mobility systems
Assessing Marginalised Communities’ Mobility Needs in Global South Contexts
Indigenous and local knowledge contributions to sustainable urban transport and infrastructure planning
Critiques of transport models and priorities: reconsidering paradigms in emerging economies
Methodological innovations for decolonial transport research, such as participatory and context-sensitive approaches

Accessibility and spatial interaction models

Tue, 18 Nov 2025 00:00:00 +0000

A few months ago, I share the preprint of a study where we discuss the missing link between spatial accessibility and spatial interaction models. The paper has now been published, and the manuscript became more concise and better thanks to peer-review, which is not always the case lol.

Soukhov, A., Pereira, R. H., Higgins, C. D., & Páez, A. (2025). A family of accessibility measures derived from spatial interaction principles. PloS one, 20(11), e0335951. https://doi.org/10.1371/journal.pone.0335951

Abstract:

Transportation planning has long prioritized the efficiency of movement. However, the concept of accessibility represents a more comprehensive evolution, shifting focus from movement (i.e., trips) to the potential to spatially interact with desired destinations. Despite growing recognition of accessibility-based planning approaches, the concept remains fragmented, with inconsistent definitions and unclear interpretations. To this end, this paper makes a methodological contribution by specifying a family of accessibility measures that are grounded in the shared ‘gravity-based’ theoretical roots of spatial interaction models, particularly their balancing factors. From this foundation, we outline four members of the family: the ‘unconstrained’ measure (i.e., Hansen-type accessibility), the ‘total-constrained’ measure (i.e., a constrained version of the Hansen-type accessibility), the ‘singly-constrained’ measure (i.e., related to the popular two-step floating catchment approach – 2SFCA), and the ‘doubly-constrained’ measure representing realized access (i.e., equal to the doubly-constrained spatial interaction model). These measures can be interpreted as either the number of accessible opportunities or accessible population (i.e., market potential). A toy example illustrates how they produce interpretable unit-based values, offering a clearer and more coherent basis for accessibility analysis.

Crime and school access

Wed, 05 Nov 2025 00:00:00 +0000

Knoblauch, S., et al. (2025). Crime-associated inequality in geographical access to education: Insights from the municipality of Rio de Janeiro. Cities, 160, 105818. https://doi.org/10.1016/j.cities.2025.105818

Abstract:

Education is a fundamental right, supported by initiatives like Education for All (EFA) and the Millennium Development Goals (MDGs). Despite progress, full educational access remains challenging, particularly in highly criminal areas. This paper examines the impact of crime on school access in the municipality of Rio de Janeiro. Using ancillary data and geospatial artificial intelligence (GeoAI), we downscaled official police crime records to street level. By considering different levels of crime tolerance in school path choices, we simulated how crime can force students to walk longer distances to avoid violence. Our findings indicate a 48.60% average increase in travel time to the closest school for students whose shortest routes intersect with high-crime areas. This adjustment reduces mean crime exposure by 44.10 % and maximum exposure by 81.94 %. Both individual crime risk aversion and no-go areas of criminal disputes significantly (p 0.05) impacted educational access. Estimating street-level crime exposure was challenging due to spatial bias in official and crowdsourced crime reporting. These methods and insights are crucial for improving educational access in high-crime areas, providing a better understanding of barriers to equitable education, and being applicable to other cities and accessibility studies for various societal needs.

Fare-free transit and the travel behavior of older adults

Wed, 22 Oct 2025 00:00:00 +0000

Can free public transport really get people out of their private cars and onto public transit? We have a new paper showing that the answer is no. Link to the paper with ungated PDF.

We analyzed data from 11 household travel surveys across 7 metropolitan regions in Brazil (covering 25% of the national population) to measure the causal effect of fare-free public transit on people’s travel behavior. To do this, we took advantage of Brazil’s rule granting free public transport to people aged 60 or 65 and used a quasi-experimental design (regression discontinuity) that compares travel patterns of individuals just below and just above the age thresholds for fare eligibility.

🧩 Two key findings:

1️⃣ Fare-free transit increases the number of public transport trips among eligible seniors by about 7%.

2️⃣ This increase comes mainly from a reduction in walking trips. The policy has no measurable effect on car use.

🚦 In other words, fare-free transit can be defended on other grounds—such as promoting social inclusion—but it cannot be justified on the assumption that it will shift people from cars to public transport and thereby reduce congestion or emissions. Almost all studies around the world (for people of all age groups) point to the same conclusion.

💡 For cities seeking to curb car dependence, the message is clear: fare-free transit alone is not enough. It must be combined with policies that discourage car use and improve the quality and integration of public transport systems.

📄 The full paper provides robust evidence and international comparisons. The article is currently available in English, and a Portuguese version will soon be published by the Institute for Applied Economic Research (Ipea).

Proximity-centred accessibility

Mon, 08 Sep 2025 00:00:00 +0000

I’m glad to share a new paper where we surveyed 1300 transport and planning practitioners from 22 countries to ask their understanding and normative views of proximity-centred accessibility. The full data set of the survey is available as open data here. To me, the most amazing results that emerged from this paper are that:

There is a seemingly universal understanding of “proximity” (up to 1600 m), with small variations depending on city size and country.
Despite context and cultural differences across countries, there is also a consensus that the “adequate distance” to activities related to basic needs and caregiving is consistently shorter, even though the “reasonable access time” varies significantly across activities.

ps. this was a great collaboration with over 30 authors, but with the fantastic lead of Cecilia Silva and Benjamin Büttner.

Silva, C., Büttner, B. et al. (2025). Proximity-centred accessibility–A conceptual debate involving planning practitioners worldwide. Cities, 167, 106376. https://doi.org/10.1016/j.cities.2025.106376

Abstract

In recent years, the concept of proximity has garnered increasing attention in both transportation research and practice, albeit under various terms and interpretations. Among these, the concept of the 15-minute city has catalysed attention in planning practice, with recent evolution to the x-minute city and city of proximities. In research, proximity-centred accessibility has been offered as an umbrella term to express the ability to reach activities and destinations at short distances. Regardless of the terminology used, the essence of proximity lies in the ease with which one can access desired activities and destinations within reasonable travel times, independent of speed-enhancing transport modes most notably through walking. This research investigates the nuanced meanings ascribed to proximity-centred accessibility by planning practitioners globally, spanning diverse regional and local contexts. For this, we used an online survey, disseminated among over 9000 practitioners from 22 countries across 5 continents, which generated over 1300 responses. The survey explored the preferred terms for proximity-centred accessibility and their definitions, specifically emphasizing time and distance thresholds and the identification of relevant activities. By juxtaposing our findings with an earlier survey of accessibility researchers, this study also contributes to the groundwork for a conceptual framework for proximity-centred accessibility. Our findings affirm a relatively consistent interpretation of proximity among global planning practitioners, predominantly extending up to 1600 m, in accordance with earlier results for accessibility researchers. Despite some relevant dissimilarities among practitioners from megacities compared to their smaller city counterparts, or in specific countries (most notably the Netherlands), the distance that is considered proximate is the attribute that generates the most consistent results across different contexts. Also consistent was the relevance of short distances (up to 15 min walking) for activities such as primary and pre-primary schools, playgrounds, parks, food shopping, and pharmacies, reinforcing the importance of proximity to basic and caregiving activities. No term was found to be consistently meaningful across different contexts, although terms like local and neighbourhood accessibility and walking/pedestrian, or cycling accessibility, show higher preference in the global sample.

The social and spatial effects of fare cuts on public transport

Thu, 28 Aug 2025 00:00:00 +0000

We have a new study where we used smartphone geolocation data from 11 million devices (11.7 billion data points) to examine the impact of transit subsidies on travel behavior. Using a time-shifted difference-in-difference model, we assessed how the Deutschlandticket policy in Germany affected visitation volumes and trip distances across the country.

Liao, Y. et al (2025) Uncovering the Social and Spatial Effects of Fare Cuts on Public Transport with Mobile Geolocation Data. Transportation Research Part A: Policy and Practice. https://doi.org/10.1016/j.tra.2025.104647

Abstract

Subsidizing public transit fares is a common policy tool for promoting sustainable mobility and reducing car dependency. Nonetheless, few studies have been able to investigate the causal impact of large fare subsidies on travel behavior patterns. This study investigates the impacts of a nationwide fare reduction policy in Germany: the Deutschlandticket (DT), which priced regional and local transit at 49 euros per month, effective from May 2023 through December 2024. Using large-scale mobile geolocation data from over 11.1 million mobile phone devices, covering 11.7 billion geolocation records in March, April, and May for 2022 and 2023, we employed a time-shifted difference-in-difference model to assess changes in visitor volumes and distance of trips to various locations across Germany. Our results indicate that the D-Ticket increased visit numbers (+26.2%) and increased travel distances (+11.8%) in the first month. Moreover, we found that the impact varied spatially and socioeconomically: urban centers such as high-activity hubs experienced the highest increase in visits and travel distance. Areas visited by a higher share of the foreign population (residents w/o German citizenship) and people from low-rent areas benefited the most, seeing more substantial increases in trips and distances. These results contribute to understanding the effectiveness of transit policy interventions by offering large-scale, high-resolution, and previously unobserved evidence of how they influenced mobility in Germany. Our study provides valuable insights into the broader impacts of public transit pricing, informing equitable and effective fare subsidy policies.

Exposure to extreme heat in public transit

Thu, 14 Aug 2025 00:00:00 +0000

Policy makers and scientists are increasingly concerned with how extreme heat events impact cities and their populations. In a new study that came out published this week, we propose the new Transit Heat Exposure Index (THEI) to gauge high-fidelity heat exposure for transit riders. We demonstrate the proposed method using using 1 m-by-1 m microclimate simulations and door-to-door transport network analysis in a case study of Miami, one of the hottest US cities.

We believe this could be a valuable tool to enhance transit resilience to heat and develop effective mitigation strategies across cities. I hope some of you’ll find it useful! The paper was written with amazing co-authors Xiaojiang Li, Xiang ‘Jacob’ Yan and led by the brilliant Luyu Liu

Liu, L., Li, X., Pereira, R. H., & Yan, X. (2025). Measuring exposure to extreme heat in public transit systems. Journal of Transport Geography, 128, 104383.

Geocoding Brazilian data with {geocodebr}

Tue, 08 Jul 2025 00:00:00 +0000

I’m super glad to share {geocodebr}, our new R package for geocoding Brazilian data.

Two key contributions

Truly open and public-sector ready: {geocodebr} is the first fully free and open-source geocoder built entirely on official Brazilian address data. Because all source data and code are public, results can be audited, reproduced, and improved by anyone—an essential feature for government workflows and academic research.
Blazing performance: The package is written in high-performance R streaming data through Arrow and DuckDB backends. In our benchmarks we geocoded the entire Cadastro Único (CadÚnico) register — over 43 million addresses — in about 65 minutes. That’s orders of magnitude faster than traditional approaches using Google Maps or ArcGIS and comes with zero per-request fees.

Why another geocoder if one can use Google Maps, ArcGis or Nominatim? Brazil already has rich, official address cadaster (CNEFE) but there was no free, programmatic tool that could leverage it at scale. Existing commercial APIs are costly, impose usage limits, and often lack transparency about how results are produced. We needed something better for our research at Ipea’s Access to Opportunities Lab (AOP-Lab), where we routinely process tens of millions data points of Brazilian administrative records.

An invitation to the community

Our initial motivation was to accelerate our own research — geocoding Brazil’s administrative records so we can study how location shapes access to jobs, schools, and public services. By releasing {geocodebr} on CRAN we hope to empower other government institutions, researchers, and civic-tech practitioners, who will be able use this package to address (pun intended) several other issues that require geocoded information.

The package is live on CRAN, the source is on GitHub, and extensive examples are in the vignette. Give it a spin, file issues, and let us know what you build. Hopefully, we can help raise the bar for spatial data quality across Brazil.

Happy geocoding!

Reuniting spatial interaction and accessibility

Wed, 04 Jun 2025 00:00:00 +0000

For a long time I felt there was a missing link between spatial accessibility and spatial interaction models. We finally have a paper that helps us understand the common historical roots of these two streams of literature and why they diverged. Our new study (preprint) goes down memory lane to show what spatial accessibility research can learn by looking into the past.

The paper shows that the proportionality constant in spatial interaction models was dropped in the early days of spatial accessibility research. This omission created fundamental problems that undermine the interpretation, communication, and comparability of accessibility analysis, particularly gravity-based metrics ~~thanks Hansen~~.

The study shows that bringing the proportionality constant back into accessibility allows us to think of a new “family of accessibility measures” derived from spatial interaction principles. In practical terms, this fixes those fundamental problems, making sophisticated gravity-based and competitive accessibility more interpretable, widely adoptable, and policy-relevant

The paper was written with a stellar team led by the incredible Anastasia Soukhov.

ps. It’s been a while since I’ve learned so much from a collaboration.

Soukhov, A., Pereira, R. H. M., Higgins, C. D., & Paez, A. (2025). A family of accessibility measures derived from spatial interaction principles. OSF Preprints https://doi.org/10.31219/osf.io/a9dxb_v1

Abstract:

Transportation planning has long prioritized the efficiency of movement, or mobility. However, the concept of accessibility represents a more comprehensive evolution, shifting focus from mere movement to the potential to reach (i.e., spatially interact) with desired destinations. Despite growing recognition of accessibility-based planning approaches, the concept remains fragmented, with inconsistent definitions and unclear interpretations. This work’s aim is to clarify and unify the concept of accessibility by connecting it into spatial interaction modeling. We demonstrate that widely used mobility and accessibility models, such as gravity-based accessibility and spatial interaction models, share common theoretical roots. From this foundation, this paper offers three contributions: (A) we introduce a family of accessibility measures within the principles of spatial interaction, and (B) formally define four members of the family, namely the ‘unconstrained’ measure (i.e., Hansen-type accessibility), the ‘total constrained’ measure (i.e., a constrained version of the Hansen-type accessibility), the ‘singly constrained’ measure (i.e., related to the popular two step floating catchment approach - 2SFCA), and the ‘doubly constrained’ measure representing realized interactions or ‘access’, effectively equal to the doubly constrained spatial interaction model; and (C) we demonstrate the interpretability advantages of the family, as these constrained accessibility measures yield values in units of the number of potential “opportunities for spatial interaction” or “population for spatial interaction” for each zone and zonal flow. The family of accessibility measures proposed here clarifies the concept of ‘potential’ in accessibility, demonstrates theoretical and formulaic linkages across popular accessibility and spatial interaction models, and reintroduces measurement units into accessibility measures. By doing so, we believe this family of measures can unlock a clearer, more interpretable, and cohesive foundation for accessibility analysis.

Advancing Urban Accessibility for Inclusive Cities

Tue, 20 May 2025 00:00:00 +0000

I’ve been back in Brasilia for a few weeks now, and I just wanted to say I had a blast during my time in Toronto. Wonderful city, great university and fantastic colleagues and students. Canada, and particularly Toronto, is one of the best places to be if you are a student, a researcher or practitioner working with transportation justice and accessibility. Without a doubt.

In one of many special moments during my stay, I had the honor to give a Bousfield Lecture. I’ve tried to give a general overview on urban accessibility research, why it matters, the advances we’ve seen over the last decades and a few potential research avenues for the future.

ps. I’m in great debt to the hospitality of Steve, Chris, Antonio, Lea, Anastasia, Nacho, and Joões Parga and Bazzo. Thanks a bunch, y’all. I’m also pretty excited about some collaborations we have in the oven.

Toronto visit

Tue, 18 Mar 2025 00:00:00 +0000

I’m in Toronto between March and April and I’ll be involved in several activities as part of my Bousfield visiting professorship. Here’s a list of the public activities those in Toronto might wanna join.

Bousfield Lecture: Advancing Urban Accessibility for Inclusive Cities
- Where: University College, 15 King’s College Circle, Paul Cadario Conference Centre
- When: April 2, 2025, from 6 PM to 9 PM
- Registration here
Seminar “A scalable bottom-up model to estimate public transport emissions”
- Where: Institute of Transportation Engineers, at ITS Lab (SF3103)
- When: March 21, 2025, from 11:00 AM to 12:00 PM
- Registration here
- Paper here
- R package {gtfs2emis}
Bousfield Urban accessibility R Crash Course
- Where: Sidney Smith Hall 1071
- When: April 8, 2025 from 9:00 AM to 12:00 PM
- Registration here
- R packages {r5r} and {accessibility}
Seminar Spatial Data Science for Urban Accessibility for Inclusive Cities
- Where: MacMaster University, School of Earth, Enviroment & Society (SEES) - Michale DeGrotte Centre for Learnin, MDCL 1309
- When: April 4, 2025, from 4 PM to 5 PM
- No registration needed

The AAG is just around the corner

Thu, 06 Mar 2025 00:00:00 +0000

The annual meeting of the American Association of Geographers (AAG) is just around the corner and I’m counting the days. I’m particularly excited this year for a few reasons:

the program this year includes a mini-conference on transportation justice, key-note presentations by Harvey Miller, Antonio Paez, and Song Gao, and several interesting-looking sessions including ‘The Interesting Data and Methods Session’ and GeoAI topics.
the AAG is a great opportunity to meet old friends and connect with new people
this year, I’ll be making a road trip from Toronto to Detroit with colleagues from the Uni of Toronto.
My schedule is almost ready. Here are a few sessions I’m more directly or indirectly involved:

Session - Transportation Justice 1: Theory, politics and democratic planning
- Alex Karner will be presenting our work in progress on ‘New Directions for Assessing Transportation Inequality and Poverty’
Session - Transportation Justice 4: Dimensions of inequitable accessibility
- I’ll be presenting our paper on talking about the elephant in the room (that very few people look at monetary costs in accessibility analysis) and how the trade-off between time and money can significantly shape accessibility conditions and equity

Panel - Driving Change: Community Perspectives on Transportation Justice in Detroit
- Alex Karner and Steven Farber will be leading a panel discussion on Detroit’s journey toward transportation equity from community-centered lens
Session - Transportation & Environmental Exposure
- Luyu Liu will be presenting our paper on ‘Measuring Extreme Heat Exposure in Public Transit Systems’. Preprint here

ps. If you’re around the AAG this year, feel free reach out ! I’d love to connect.

Highlights of 2024

Wed, 05 Mar 2025 00:00:00 +0000

“We’re already in March 2025, and only now this guy is posting about his highlights of 2024?!". Yes, and I hope it’s not too late lol.

I’ve started a Bousfield Visiting Professor position (2024-2025) at the Geography and Planning Department at the University of Toronto
We received the David Levinson Best Paper Award at the 2024 WSTLUR (World Symposium on Transport and Land Use Research) for the paper ‘Ride-hailing and transit accessibility considering the trade-off between time and money’
Peer-reviewed 13 papers
Published 6 papers and got a number of desk rejections that I lost count.
Finished 4 working papers, one of which had a major role in helping the Brazilian government react to the mass floodings in the south of the country
Our book on Introduction to Urban Accessibility got to be a Finalist of the Academic Jabuti Award
One new R package
We welcomed new colleagues and RAs in our team at Ipea
I got listed among the top 2% cited scientists in the world for in the years 2020, 2021, 2022, 2023 and 2024 by Stanford & Elsevier. Full list here
Graduated one masters student
I had a great time teaching four workshops at academic conferences and one at the EIT Doctoral Training Network Annual Forum
Delivered invited lectures at MIT, Cornell, UFRJ, University of Twente, UFC, FGV
Delivered one Keynote at the annual AGILE conference
For the first time, I applied to acadmic jobs. Three in fact. Never heard from one, got short listed in another and went the full experience in the third one. The good news, I got accepted at the Department of Urban Planning at Berkeley. The bad news, I got 2nd in the selection and there was one spot only. Bittersweet.
Hands down, the peak of 2024 really was taking my daughter to swimming classes

ps. ~~procratinating~~ Writting this list down has not helped me cope with imposter syndrome.

Visiting Professor at Toronto

Sat, 16 Nov 2024 00:00:00 +0000

I haven’t been very disciplined in feeding this website. It’s been over three months since my last post, but this is because life has been a bit more hectic than usual. Dear Universe, do better, please.

There is one piece of news that I should not leave unnoticed, though. I’ve started a Bousfield Visiting Professor position (2024-2025) at the Geography and Planning Department at the University of Toronto. This is very exciting because this means spending much more time in person and online collaborating on projects with colleagues in Toronto, and some quality time teaching and engaging with students.

ps. I wasn’t be able to move the entire family to Toronto for the entire academic year but I’ll still get to enjoy the ~~freezing~~ chilly temperatures of Toronto.

Finalist in the Academic Jabuti Award

Mon, 05 Aug 2024 00:00:00 +0000

The Jabuti award is one of the most prestigious literary awards in Brazil. The award includes a broad range of categories, and this year they have created the Academic Jabuti Award to recognize the excelence of academic books in various fields. I’m thrilled to share that our book “Introduction to urban accessibility: a practical guide with R” has been shortlisted as a finalist in the ‘Architecture, Urbanism, Design, and Urban and Regional Planning’ category.

For many decades now, scientists and policy makers recognize the importance of evidence-informed public policies in areas such as health, education, and others. This should be no different in urban and regional planning. Open data and computational methods are essential to develop evidence-informed policies that help us promote more integrated, inclusive, and sustainable urban planning. This book makes a contribution in this direction.

In many ways, this is not a conventional academic book, though. This book provides a very hands-on approach to teaching/learning urban accessibility analysis, which has become a key element in urban data science and urban analytics. All the material in the book is presented with reproducible examples using open datasets and the R programming language. In fact, the book was entirely written with open source code in R and Quarto ( Github repo here). Now the icing on the cake 🍰 you can run the code used in the book online in your browser with Binder. Also, although there’s a PDF of the book that can be downloaded, the book is a native online website, which is quite nice because it makes it possible for us to track the readership of the book. Since the launch of the book, we’ve had already 19K unique “users”, which is really not bad for an academic book in such a small niche of urban and spatial data science.

Fingers crossed.

Exposure to extreme heat in public transit

Wed, 17 Jul 2024 00:00:00 +0000

Excited to share a new preprint! Here we introduce a novel methodology to measure the heat exposure of public transit riders at high spatial resolution. The method combines microclimate simulation using meteorological and LIDAR data with household travel survey and realistic transit network routing methods. The proposed method allows one to estimate heat exposure at the network link level during walking along street links and waiting at bus stops.

Liu, Luyu and Li, Xiaojiang and Yan, Xiang and Pereira, Rafael H.M., (2024) Measuring Exposure to Extreme Heat in Public Transit Systems. SSRN preprint: http://dx.doi.org/10.2139/ssrn.4869682

Abstract:

Public transit users are among the most vulnerable to extreme heat due to urban heat island effect and longer outdoor exposure. However, few studies have investigated transit riders’ heat exposure and considered exposure time and travel behavior when calculating heat exposure. This paper introduces a holistic measurement system – Transit Heat Exposure Index (THEI) – to gauge high-fidelity heat exposure for transit riders and apply the method to Miami, one of the hottest US cities. By using high-resolution meteorological and built environment data, we calculate 1m-by-1m feels-like temperature for Miami-Dade Transit with microclimate simulation techniques. Then, we calculate the detailed travel time between all census block groups in Miami with realistic transit routing technique. We then adopt a total time-degree approach to calculate heat exposure and aggregate THEI at different levels. Our results show that despite higher local feels-like temperature, downtown Miami has lower heat exposure due to better transit access. Source analysis also shows that walking is the primary source of heat compared to waiting, and a small portion of streets and bus stops contribute most heat exposure. This method provides first-hand evidence for future heat planning, enabling more effective strategies to mitigate extreme heat’s impact on transit riders.

How experienced segregation is shaped by homophily and limited travel

Thu, 11 Jul 2024 00:00:00 +0000

We have a new preprint where we use smartphone data and some clever simulations to examine how the segregation levels experienced by Native & Foreign-born individuals are shaped differently by homophily and limited travel (related to accessibility constraints).

Liao, Y., Gil, J., Yeh, S., Pereira, R. H., & Alessandretti, L. (2024). The Uneven Impact of Mobility on the Segregation of Native and Foreign-born Individuals. arXiv preprint https://doi.org/10.48550/arXiv.2407.00404

Abstract:

Segregation is a key challenge in promoting more diverse and inclusive cities. Research based on smartphone data has revealed that segregation can extend beyond residential areas into everyday activities like visiting shops and restaurants. The impact of these activities on segregation, however, is unclear. Some studies suggest that they promote mixing, while others indicate they reinforce segregation. Here, we elucidate how day-to-day mobility shapes overall segregation levels, looking at the distinctive segregation experienced by native and foreign-born individuals. Our study is based on ~320,000 smartphone trajectories collected in Sweden, where immigration creates profound divides. We find that while mobility levels generally promote mixing for native-born individuals, foreign-born individuals remain segregated in their out-of-home activities. Using counterfactual simulations, we show that this heterogeneous effect of mobility on experienced segregation results mainly from two mechanisms: homophily and limited travel, i.e., foreign-born individuals (i) prefer destinations visited by similar individuals, and (ii) have limited mobility ranges. We show that homophily plays a minor role, while limited mobility, associated with reduced transport access, limits opportunities for foreign-born to diversify their encounters. Our findings reconcile conflicting literature and suggest that enhancing transport accessibility in foreign-born areas could reduce social segregation.

WSTLUR workshop: a crash course on urban accessibility with R

Sat, 15 Jun 2024 00:00:00 +0000

I’m super excited that next week I’ll be in Bogota, Colombia, attending the 2024 WSTLUR (World Symposium on Transport and Land Use Research). I’ll get the chance to present our recent paper on “Ride-hailing and transit accessibility considering the trade-off between time and money”.

Perhaps even more exciting, I’ll be teaching the workshop “A crash course on urban accessibility with R”. Here’s the [website with all the teachig material and code of the workshop](https://ipeagit.github.io/access_workshop_bogota_2024/.

Soundtrack:

Equity implications of Transit Oriented Development

Fri, 14 Jun 2024 00:00:00 +0000

We have a new paper out looking at the equity implications of Transit Oriented Development (TOD) in Curitiba. The city of Curitiba is internationally celebrated for its TOD planning based on Bus Rapid Transit (BRT). However, few studies have examined how TOD projects are associated with socioeconomic and spatial inequalities (in other words, who actually benefits from TOD?).

In this paper we examine how the spatial organization of Curitiba’s TOD is associated with the distribution of population densities, socioeconomic groups, and real-estate values and its implications in terms of inequalities of access to employment opportunities and health services. The paper is part of André Turbay’s PhD thesis, which I co-supervised with Rodrigo Firmino.

Turbay, A. L., Pereira, R. H., & Firmino, R. (2024). The equity implications of TOD in Curitiba. Case Studies on Transport Policy, 16, 101211. https://doi.org/10.1016/j.cstp.2024.101211

As usual, you can download the ungated PDF of the paper on the Publications section of this website.

Measuring transportation equity

Mon, 04 Mar 2024 00:00:00 +0000

I’m very glad to (finally be able to) share our new paper on measuring transportation equity. In this paper we present a critical and theoretically grounded reflection on different methods to measure transportation inequality and transportation poverty. There are several important takeaways in the paper ~~please stop using the Gini coefficient!!!~~.

All the methods discussed in this paper have also been implemented in the {accessibility} package in R. On this link you can get the ungated PDF, the code and data used in the study.

Karner, A., Pereira, R. H. M., & Farber, S. (2024). Advances and pitfalls in measuring transportation equity. Transportation. https://doi.org/10.1007/s11116-023-10460-7

Abstract

Transportation systems play a pivotal role in facilitating access to out-of-home activities, enabling participation in various aspects of social life. But because of budgetary and physical limitations, they cannot provide equal access everywhere; inevitably, some locations will be better served than others. This realization gives rise to two fundamental concerns in transportation equity research and practice: (1) accessibility inequality and (2) accessibility poverty. Accessibility inequalities may rise to the level of injustice when some socioeconomic groups systematically have lower access to opportunities than others. Accessibility poverty occurs when people are unable to meet their daily needs and live a dignified and fulfilling life because of a lack of access to essential services and opportunities. In this paper, we review two of the most widely used approaches for evaluating transport justice concerns related to accessibility inequality and accessibility poverty: Gini coefficients/Lorenz curves and needs-gap/transit desert approaches, respectively. We discuss how their theoretical underpinnings are inconsistent with egalitarian and sufficientarian perspectives in transport justice and show how the underlying assumptions of these methods and their applications found in the transportation equity literature embody many previously unacknowledged limitations that severely limit their utility. We substantiate these concerns by analysing the equity impacts of Covid-19-related service cuts undertaken in Washington, D.C. during 2020. The paper also discusses how alternative methods for measuring transportation equity both better comport with the known impacts of such changes and are consistent with underlying moral concerns.

ps. The original version of the paper was published a couple of months ago, but it took a long time for Springer to correct an error on the author list.

A summary of my 2023, and happy 2024

Mon, 01 Jan 2024 00:00:00 +0000

I’ve been less active around here but for good reasons. I’ve been ~~procrastinating~~ busy with other things. Here’s a summary. This past year, I:

Proudly saw two of my students finish their PhD and Master degrees. I also started co-supervising my first undergraduate student 🎓
Published 10 papers 🥳 and had at least 5 desk rejections lol 🥸
Proposed two new accessibility metrics ( here and here)
Crossed the mark of 7K citations and got listed among top 2% cited scientists in the world for three consecutive years
Published one book 📘 in both English and Portuguese. My first online book written in R with Quarto 🤓 I also singed the contract of another book (co-authored with wonderful colleagues) and which will be published by CRC Press
Published 8 working papers / reports
Published one editorial of the first special issue I co-edited (in great company)
Started co-editing another special issue
Co-organized a workshop at an international conference
Created 3 R packages and one Twitter bot
The r5r package crossed the mark of 700K downloads, and all our packages have together crossed 1 million downloads 👯‍♀️
Became a member of the National Consultative Committee on Urban Mobility in Brazil
Became a member of the Editorial Advisory Board of the Journal of Transport Geography
Became Head of Data Science at Ipea
Presented at over 16 events (seminars, conferences etc), including 2 keynote presentations
Peer-reviewed 8 papers. This is down from 17 and 16 I did in 2022 and 2021, but for a very good reason!
By far, the most remarkable of all events this year was the arrival of my daughter. Becoming a father is definitely one of the most important and joyful life changes I’ve ever had.

I have some really exciting plans for 2024. Can’t wait to see what’s ahead.

Happy 2024, everyone !!!

ps. I was quite reluctant in writing this post because some people may see it as too self-congratulatory. It is a bit, but from time to time we need to do this to shake our imposter syndrome off, so I think it’s Ok.

ps2. Today I received an email I had sent myself 3 years ago, with questions and reflections I had about my life and the plans I had back then. Got a bit emotional 🥰 . It’s amazing how in only 3 years some things change completely and others do not change at all.

Route-based accessibility and its association with transit ridership

Thu, 21 Dec 2023 00:00:00 +0000

I’m super glad to share we have a new paper out of the oven! In this study we:

propose a new route-based accessibility measure
propose a Directed Acyclic Graph (DAG) to map the causal paths of key variables that influence transit ridership, and
compare the performance of our proposed indicator against similar metrics by examining their association the ridership of transit routes.

Among all of the papers I’ve been involved in the past year or so, this is certainly one of the papers I’m most proud of. For different reasons. First, the paper was led by João Albuquerque-Oliveira as part of his Master’s dissertation. João is incredibly dedicated I’m super proud of him for what he has achieved. Second, the study makes various methodological contributions and I think the route-based accessibility metric proposed could really help practioners engage more direcly with accessibility-oriented planning. Finally, I believe this paper opens up a lot of new research questions and avenues for investination. This is a true mark of good research.

Albuquerque-Oliveira, J. L., Oliveira-Neto, F. M., & Pereira, R. H. M. (2024). A novel route-based accessibility measure and its association with transit ridership. Transportation Research Part A: Policy and Practice, 179, 103916. https://doi.org/10.1016/j.tra.2023.103916

🔓 Ungated PDF here.

The intersection of racial and income inequalities in access to healthcare

Wed, 06 Dec 2023 00:00:00 +0000

We have a new paper published where we look at the intersection between racial and income inequalities in access to healthcare in Brazilian cities. I’m very glad to see one more output from the Access to Opportunities Project.

Tomasiello, D. B., Vieira, J. P. B., Parga, J. P. F., Servo, L. M., & Pereira, R. H. (2024). Racial and income inequalities in access to healthcare in Brazilian cities. Journal of Transport & Health, 34, 101722. https://doi.org/10.1016/j.jth.2023.101722

🔓 Ungated PDF and code here.

The trade-off between time and money in ride-hailing and transit accessibility

Sun, 26 Nov 2023 00:00:00 +0000

Super glad to share the news that our paper on accessibility by ride-hailing has been published in the journal Cities.

Pereira, R. H., Herszenhut, D., Saraiva, M., & Farber, S. (2024). Ride-hailing and transit accessibility considering the trade-off between time and money. Cities, 144, 104663. https://doi.org/10.1016/j.cities.2023.104663

The study covers many interesting results comparing accessibility by ride-hailing Vs transit Vs ride-hailing integrated with transit. However, in a world where most studies estimate accessibility based solely on travel times, one of the key contributions of our paper is to show how to, and how important it is to consider trade-offs between travel time and money in accessibility analyses.

🔓 Ungated PDF and code here.

Advances in geospatial approaches to transport networks and sustainable mobility

Mon, 30 Oct 2023 00:00:00 +0000

Really glad to see our Special issue on “Advances in Spatial and Transport Network Analysis” is now online, publised in Environment and Planning B: Urban Analytics and City Science.

You can check all papers in the special issue here, and our editorial here.

Thanks to many authors contributing with high quality papers and thanks Henrikki Tenkanen, Elsa Arcaute, and Marta C. Gonzalez for a wonderful colaboration!

Online workshop: Getting to Services

Fri, 13 Oct 2023 00:00:00 +0000

Exciting upcoming workshop on “Getting to Services: Transport and Accessibility”, organized by OECD and European Commission. The event will feature presentations by the European Commission, OECD, and researchers in Brazil, Canada and the Netherlands, sharing their experiences on some of the recent advances and persistent challenges in accessibility research and planning.

The workshop will be entirely online on Thursday, the 26th of Oct/2023 from 15:00-17:00 CET. Click here for more info on the agenda and how to register.

Update: The recording of the workshop is now available online!

Free public transit and voter turnout

Fri, 29 Sep 2023 00:00:00 +0000

Glad to share the news that our paper on the impact of free public transit on voter turnout and mobility levels has now been publised! On this link you can get the ungated PDF, the code and data used in the study.

Pereira, R. H. M., Vieira, R. S., Bizzarro, F., Barbosa, R. J., Dahis, R., & Ferreira, D. T. (2023). Free public transit and voter turnout. Electoral Studies, 86, 102690. https://doi.org/10.1016/j.electstud.2023.102690

Abstract:

Transportation costs are an under-studied barrier to political participation. In many elections worldwide, subsidies to voter transportation are already provided or are under discussion. However, these types of incentives have not been rigorously evaluated. Here we examine possibly the world’s largest-ever intervention to lower these costs, the adoption of a fare-free public transit policy in Brazil during the 2022 national election, when about half of Brazilian voters were granted the right to use public transit for free on election days. However, while some cities adopted the benefit for both rounds of the election, others adopted it only for the second round. Using an event study design, we exploit this difference in adoption timing to examine the policy’s causal impact on voter turnout rates and human mobility levels. We find that fare-free transit increased ridership between 7.2% and 17.5% on election days, however, we estimate a precise and robust null effect of the policy on voter turnout (Coef. −0.03p.p. with standard error of 0.22p.p.). Our results illustrate that monetary transport costs may not always be a critical factor behind non-participation. Although reducing transportation costs improves access to polling places, we show that even a full transit subsidy may not be sufficient to increase voter turnout..

censobr: Download Data from Brazil’s Population Census

Fri, 15 Sep 2023 00:00:00 +0000

I’ve always been a heavy user of Brazilian census data. This is of one the key data sets I use in most of my reseach projects on cities, urban and regional development, urban transport and accessibility. Like in many other countries, the population census in Brazil is the most comprehensive data collection process, covering various different topics and using a consistent high-quality method for the entire country at fine spatial resolution.

Nonetheless, getting access to census data has never been easy or convenient. Although Brazil’s official statistics and geography institute (IBGE) makes the census data publicly available, the data is not shared in a ready-to-use format. Moreover, census data sets are often larger-than-memory, what puts critical barriers for most users with limited computational resources.

To help overcome these two problems, I’ve created censobr, an R package to make it easy for anyone to download data from Brazil’s population census. You may install censobr from CRAN or the dev version from Github.

The package currently includes 5 main functions to download Census microdata:

read_population()
read_households()
read_mortality()
read_families()
read_emigration()

At the moment, censobr only includes microdata from the 2000 and 2010 censuses, but it is being expanded to cover more years and data sets.

As I mentiond, microdata of Brazilian census are often be too big to load in users’ RAM memory. To overcome this problem, censobr is built on top of Apache Arrow, which allows users to analyze larger-than-memory data sets like they would with a regular data.frame simply using common functions from dplyr. More info in this vignette.

Note: all data sets in censobr are enriched with geography columns following the name standards of the {geobr} package to help data manipulation and integration with spatial data from {geobr}.

ps. I also have the census in a special place in my heart because of my demography background.

The Causal Quartet

Wed, 23 Aug 2023 00:00:00 +0000

The “Anscombe’s quartet” is a famous way to illustrate how datasets can have nearly identical descriptive statistic but very different distributions. It serves as a warning, calling researchers to carefully visualize their data in any data analysis process.

More recently, Lucy D’Agostino McGowan has proposed the “Causal quartets”: a group of four datasets all with the same statistical summaries & visualizations but different true causal effects. See her blog post and paper preprint.

Takeaway: theory-informed data analysis is key. A good theoretical understanding of the causal mechanisms behind a phenomenon is critical if we want to correctly draw causal paths & estimate true causal effects.

**Anscombe's quartet and the DAG of the causal quartet**

Accessibility within planetary boundaries

Mon, 07 Aug 2023 00:00:00 +0000

Our paper on accessibility within planetary boundaries has now been published (open access):

Willberg, E., Tenkanen, H., Miller, H. J., Pereira, R. H. M., & Toivonen, T. (2023). Measuring just accessibility within planetary boundaries. Transport Reviews, (0), 1–27. https://doi.org/10.1080/01441647.2023.2240958

I’m obviously biased, but Elias Willberg has done a great job leading the paper and I think the study makes a great contribution to research at the big area of intersection between transportation and sustainable development. I hope the paper will spark new discussion around sustainable accessibility.

Abstract:

Our societies struggle to provide a good life for all without overconsuming environmental resources. Consequently, scholarly search for approaches to meet environmental and social goals of sustainability have become popular. In transport research, accessibility is a key tool to characterise linkages between people, transport, and land use. In the current paper, we propose a conceptual framework for measuring just accessibility within planetary boundaries. We reviewed transport studies and discovered a substantial literature body on accessibility and social disadvantage, much vaster compared to the literature around environmental and ecological impacts of accessibility. We also show a gap in approaches that have integrated these two perspectives. Building on the review, we suggest a conceptual framework for incorporating environmental and social sustainability goals in accessibility research. We conclude the paper by pointing to key challenges and research avenues related to the framework, including (i) dealing with uncertainty and complexity in socio-ecological thresholds, (ii) integrating environmental limits into the conceptualisations of transport equity, (iii) measuring accessibility through other costs than travel time, and (iv) integrating both quantitative and qualitative data.

Accessibility and space-time differences in when and how people travel

Tue, 01 Aug 2023 00:00:00 +0000

Glad to share our new (open access) paper:

Ryan, J., Pereira, R. H., & Andersson, M. (2023). Accessibility and space-time differences in when and how different groups (choose to) travel. Journal of Transport Geography, 111, 103665. https://doi.org/10.1016/j.jtrangeo.2023.103665

In this study, we examine how:

space-time differences in accessibility are associated with the time period in which people commute to work;
the flexibility of individuals’ time constraints influence their commute mode choice

Abstract:

Abstract A fair distribution of accessibility to key activities is a central concern for distributive justice in transport planning. This implies that disparities in accessibility and the negative effects associated with a lack of accessibility should be mitigated. However, even though accessibility is not purely spatial or static, it is conventionally treated as if it were. Several studies have significantly advanced the spatio-temporal representation of accessibility. However, there is still a lack of focus on how these dynamics affect different groups of people with differing socio-economic and demographic characteristics and how these dynamics might influence their travel mode choice. The aim of this paper is to analyse how differences in accessibility to the workplace by public transport during multiple times of the day are associated with the time period in which different groups of people commute to work, and how the flexibility of individuals’ time constraints might influence whether they can and do commute using public transport. Using a transport justice-focused time geography perspective, this study draws on the spatio-temporal aspects of people’s real activities based on travel survey data for the Stockholm Region. This study contributes to the literature by highlighting the differences in accessibility to the workplace by public transport across multiple departure time periods; showing how these results compare with mode choice (if there is a choice); and revealing which socio-economic, demographic and geographical factors characterise these differences. This study also illustrates how a transport justice-focused time geography approach could help researchers have a more detailed and nuanced understanding of the relationship between accessibility and sustainable travel behaviour, and how this relationship might change at different times of the day.

New paper: time interval metric for cumulative opportunity accessibility

Thu, 15 Jun 2023 00:00:00 +0000

Chuffed to see our new paper online.

Tomasiello, D. B., Santos, D. H. M., Oliveira, J. L. A., Braga, C. K. V., & Pereira, R. H. M. (2022). A time interval metric for cumulative opportunity accessibility. Applied Geography, 157, 103007. https://doi.org/10.1016/j.apgeog.2023.103007
You can check the ungated PDF and code of the paper here.

Contribution

In this paper we introduce the new time interval cumulative accessibility measure, which mitigates the impact of arbitrary choices of trip duration on cumulative accessibility analyses.

Context of the problem

Cumulative opportunity measures have become the most commonly used accessibility metric used by researchers and practitioners. This is largely because it is simple to calculate and to communicate, as they tell the number of opportunities (e.. jobs or schools) that can be reached within a given travel time threshold.

But what threshold should we choose? Traditional cumulative accessibility requires an arbitrary choice of a single travel time cutoff, but this choice can significantly impact the conclusions of transport policy assessments, inequality analyses & subsequent policy recommendations. This is known as the boundary effect of the modifiable temporal unit problem #MTUP. It’s a statistical bias related to the arbitrary selection of a phenomenon’s duration (max trip duration in this case), and it’s documented in the transport literature here

While conducting sensitivity analysis with multiple cutoffs is important, it generates a greater number of results that can be hard to communicate/interpret. It’s also not obvious which cutoff should be considered when results disagree. What should we do then?

The proposed metric

To overcome this problem above, we introduce a new time interval cumulative accessibility measure. The proposed metric calculates the average number of opportunities that can be reached over multiple min-by-min thresholds within a given travel time interval.

The main advantage of the proposed time interval approach is to mitigate the impacts of such ad-hoc choices on the conclusions of accessibility estimates and accessibility inequality analyses. The proposed metric can also capture the accessibility impacts of transportation and land-used interventions with more nuance, as traditional threshold-based cumulative metrics do not capture any changes in accessibility conditions below or above the selected threshold. The time interval cumulative metric also shares important strengths with the traditional cumulative opportunity accessibility: it is straightforward to calculate and easy to calculate.

To make the new measure readily available for researchers and practitioners, we have implemented the cumulative_interval() function in the accessibility package in R.

Empirical analysis and tests

To demonstrate the proposed metric, we assess how a planned subway expansion will likely impact employment accessibility levels + inequalities in Fortaleza (Brazil). We compare the results using both the time threshold- and interval-based cumulative accessibility metrics.

In summary, we find that accessibility estimates and inequality levels can vary a lot depending on the time threshold you choose with traditional cumulative accessibility. Meanwhile, we find that the time interval metric generates much more robust results that are less sensitive to ad-hoc methodological choices.

Takeaway

The time interval cumulative accessibility metric shares the same advantages of traditional cutoff-based cumulative measures. It is easy to communicate and to calculate. Plus, it is able to mitigate the impacts of arbitrary choices of trip duration (MTUP). In sum, the time interval cumulative accessibility could become easily incorporated into transport research & planning and help prevent advocates from opportunistically cherry-picking the time cutoff that will lead to the results that support their agenda.

Calculating Urban Centrality in R

Tue, 13 Jun 2023 00:00:00 +0000

uci is an R package to measure the centrality of cities and regions. The package implements the Urban Centrality Index (UCI) originally proposed by Pereira et al. (2013).

UCI measures the extent to which the spatial organization of a city varies from extreme monocentric to extreme polycentric in a continuous scale, rather than considering a binary classification (either monocentric or polycentric). UCI values range from 0 to 1. Values closer to 0 indicate more polycentric patterns and values closer to 1 indicate a more monocentric urban form.

obs. UCI was originally proposed in the Geographical Analysis Journal. This was my 1st paper (10 years ago!), what makes this a special package to me.

Here’s a simple reproducible example of how it works ( for more info, check the package vignette):

Installation

uci is available on CRAN, so you can install it running:

install.packages('uci')

Data input

The uci package comes with a sample data for demonstration and test purposes. The data is a small sample of the spatial distribution of the population, jobs and schools around the city center of Belo Horizonte, Brazil. This sample data set is a good illustration of the type of data input required by uci.

library(uci)
library(sf)
library(ggplot2)

# load data
data_dir <- system.file("extdata", package = "uci")
grid <- readRDS(file.path(data_dir, "grid_bho.rds"))

head(grid)
#> Simple feature collection with 6 features and 4 fields
#> Geometry type: POLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -43.96438 ymin: -19.97414 xmax: -43.93284 ymax: -19.96717
#> Geodetic CRS:  WGS 84
#>                id population jobs schools                       geometry
#> 1 89a881a5a2bffff        439  180       0 POLYGON ((-43.9431 -19.9741...
#> 2 89a881a5a2fffff        266  134       0 POLYGON ((-43.94612 -19.972...
#> 3 89a881a5a67ffff       1069  143       0 POLYGON ((-43.94001 -19.972...
#> 4 89a881a5a6bffff        245   61       0 POLYGON ((-43.9339 -19.9728...
#> 5 89a881a5a6fffff        298   11       0 POLYGON ((-43.93691 -19.971...
#> 6 89a881a5b03ffff        555 1071       0 POLYGON ((-43.96136 -19.970...

The data is an object of class "sf" "data.frame" with spatial polygons covering our study area and a few columns indicating the number of activities (e.g. jobs, schools, population) in each polygon. Our particular sample data is based on a spatial hexagonal grid (H3 index). While there are advantages of using regular spatial grids to calculate spatial statistics, uci also works with non-regular geometries, such as census tracts, enumeration areas or municipalities.

We can visualize the spatial distribution of jobs using ggplot2:

ggplot(data = grid) +
  geom_sf(aes(fill = jobs), color = NA) +
  scale_fill_viridis_c() +
  theme_void()

Calculating UCI

We can easily calculate how mono/polycentric our study area is using the uci()function. In the example below, we consider the spatial distribution of jobs. The output is a data.frame with the value of UCI (0.253) and the components we use to calculate UCI. For more info, check the package vignette.

df <- uci(
  sf_object = grid,
  var_name = 'jobs'
  )

head(df)
#>         UCI location_coef spatial_separation spatial_separation_max
#> 1 0.2538635     0.5278007           3880.114               7475.899

Quote of the day

Sun, 11 Jun 2023 00:00:00 +0000

I heard this one today, and it really resonated with me. Not sure who the author is, though.

CfP: Environmentally and socially sustainable mobility

Wed, 07 Jun 2023 00:00:00 +0000

Quick heads up that we have an open Call for Papers for a special issue on “Environmentally and socially sustainable mobility”, to be published in Computers, Environment and Urban Systems (CEUS).

Scope

Decarbonising transport brings about several challenges as social equity and environmental sustainability are often in tension with one another. For instance, without guaranteeing viable and affordable sustainable transportation alternatives, some policy interventions that seek to reduce transport emission can disproportionately harm the accessibility and mobility of low-income communities and increase social exclusion.

This special issue aims to continue the discussions and push the boundaries on how advances in geospatial data science can be leveraged to promote development that balances social and environmental sustainability in transport and mobility to build more equitable neighborhoods and sustainable communities. We invite papers that develop new methods and technologies as well those applying existing geospatial analysis techniques to examine research questions at the intersection of social equity and sustainable mobility including but not limited to:

How do we reach socially and ecologically sustainable mobility that considers both people and the environment at the same time?
How can novel geographic artificial intelligence (GeoAI) methods contribute to the development of more inclusive and sustainable transport systems?
Can we develop accessibility measures that move beyond travel time and monetary costs to incorporate energy consumption and environmental costs?
How should we measure the trade-offs between environmental and social sustainability by investigating accessibility levels of different transport modes and mobility patterns?
How do urban form and land use patterns affect transportation choices and accessibility to different transportation options, and induce more or less sustainable mobility patterns?
Can we combine mobility and accessibility measures to better understand the drivers of sustainable transportation adoption and uncover inequalities based on gender, income, age, race etc.?
To what extent will the adoption of new transportation technologies, such as electric and autonomous vehicles create unintended conflicts between social equity and environmental goals?

Submission deadline

November 30, 2023

Guest editors

Grant McKenzie (McGill University, Canada)
Alessia Calafiore (University of Edinburgh, United Kingdom)
Henrikki Tenkanen (Aalto University, Finland)
Eun-Kyeong Kim (Luxembourg Institute of Socio-Economic Research, Luxembourg)
Rafael H. M. Pereira (Institute for Applied Economic Research, Ipea, Brazil)

Prior to the special issue deadline, the guest editors of the special issue are co-organizing two workshops related to this topic at the 12th International Conference on Geographic Information Science (GIScience ’23). These workshops will take place on September 12, 2023 and all are invited to participate. More details:

The impact of travel time inaccuracy and variability on accessibility

Wed, 17 May 2023 00:00:00 +0000

New paper, this one led by Kaue Braga as part of his PhD thesis.

Braga, C. K. V. Loureiro, C. F. G., Pereira, R. H. M. (2023). Evaluating the impact of public transport travel time inaccuracy and variability on socio-spatial inequalities in accessibility. Journal of Transport Geography, 109, 103590.

You can download the ungated version of this paper here.

It’s always best to read the full paper, but here’s a quick summary:

We show that public transport accessibility analysis based on GTFS data can lead to biased estimates because of inaccuracies in scheduled levels of service and because it ignores day-to-day travel time variability.
We also show that such biases affect some neighborhoods and socioeconomic groups more than others, which raises important implications for transport research and policy concerned with equity.
Finally, the study introduces a method to use historical GPS data to account for inaccuracy and day-to-day variability issues in public transport GTFS data in order to overcome those issues in transportation research and policy guided by public transport accessibility analysis.

Estimating public transport emissions from GTFS data

Fri, 12 May 2023 00:00:00 +0000

Super glad to see this study finally published:

Vieira, J. P. B., Pereira, R. H., & Andrade, P. R. (2023). Estimating public transport emissions from General Transit Feed Specification data. Transportation Research Part D: Transport and Environment, 119, 103757.

You can download the ungated version of this paper (and all my publications) from this very website, here.

In this paper we introduce the gtfs2emis model, a bottom-up method to estimate environmental emissions of public transport systems at high spatial and temporal resolutions. The model only requires requires public transport data in the standard GTFS format and a table with general characteristics of the vehicle fleet profile, which makes the method easily applicable across the globe even for cities with limited data. The paper also demonstrates the gtfs2emis model with a case study estimating the total emissions & emissions per passenger of the public transport system of Sao Paulo. We also show that the city could emit ~25% less CO2 if buses with Euro III stage alone were electrified.

The model is available as an R package 📦. The package currently supports estimates of hot exhaust emissions of over 16 pollutants and energy consumption based on emission factor models from Europe, the United States, and Brazil. We’re expanding the 📦 to include more regions and emission sources such as cold-start, tire & brake wear.

What is most exciting is not this particular paper, but all the possibilities for research & policy applications opened up by gtfs2emis. It could help inform policies for low-carbon transition in several ways, for example estimating the environmental benefits of different electrification policy scenarios or fleet renewal investments. It could also be used to advance environmental justice research projects concerned with air quality and health exposure.

ps. In this previous post I mentioned a bit about the story behind the idea of the paper and the fact that the gtfs2emis model was shortlisted as a Finalist in the 2022 ITF Decarbonising Transport Award.

Open source transport network analysis

Mon, 08 May 2023 00:00:00 +0000

Several people use ArcMap (ArcGis) to run transport network analysis to calculate network distances, travel time matrices, accessibility etc etc. ArcMap is nice but it is also quite expensive $$$. I would like to highlight a few open source alternatives:

Using R:

Using Python:

Other languages:

PgRouting

As you see, there are several open source alternatives for running transport network analysis. Mind you that only a few of these tools allow for multimodal routing including public transport, such as r5r, m4ra, OpenTripPlanner, r5py and urbanaccess.

This paper by Christopher Higgins et al. compares the performance and results of a few of these options. It would be great to have a more comprehensive benchmark, though

Celebrating a milestone of the Access to Opportunities Project

Mon, 01 May 2023 00:00:00 +0000

As many of you will know, one of the things that makes me really excited these days is the Access to Opportunities Project (AOP), which I started with Kaue Braga and Bernardo Serra in 2019. The AOP project has had a really productive phase between 2020 and 2022 thanks to a partnership with the Secretariat for Urban Mobility of the Brazilian Federal government (Semob). Thanks to this partnership, I was able to hire a few really talented research assistants who have made substantial contributions to expanding the reach and the impact of the AOP project.

Changing gears

The first partnership with Semob ended in 2022, and so the AOP project is now entering a new phase. For the next couple of years, we will be working with a smaller team and less focused on data collection and generation, and more on using the data and methods we have developed to conduct new studies and policy evaluations. We are going to continue our work trying to advance transport accessibility methods and we’ll do a bit more work on capacity building, helping students, researchers and policy makers to use computational methods to analyze accessibility with open source software and data.

Celebrate

I’m writing this post to register and celebrate what our team and collaborators have achieved in these amazing tree years between 2020 and 2022. I write this post knowing that my future self will be happy to look back and see some the incredible work we did with AOP during this period, despite the most horrible circumstances under COVID-19. So here’s a dropdown list with some of our main outputs:

Several data sets📊

- Annual estimates of accessibility to jobs, health, education and social protection services by transport mode 🚶‍♂️ 🚲 🚌 🚗 for the 20 largest cities in Brazil for the years 2017, 2018 and 2019.
- Detailed data on the spatial distribution of population and land use activities at a fine spatial resolution for all cities included in the project
- A Shiny app to explore interactive visualizations of our data in maps 🌎 and charts 📈

All data sets from AOP are available for download from the project website 🌐 or through the aopdata package in R.

6 papers published 📑

1. The impact of transit monetary costs on transport inequality
2. Geographic access to COVID-19 healthcare in Brazil using a balanced float catchment area approach
3. r5r: Rapid Realistic Routing on Multimodal Transport Networks with R5 in R
4. Urban access across the globe: an international comparison of different transport modes
5. Analisando a variabilidade de estimativas de acessibilidade por transporte público a partir de dados de GPS
6. Evaluating the impact of public transport travel time inaccuracy and variability on socio-spatial inequalities in accessibility

5 papers currently under review 📄

1. Ride-hailing and transit accessibility considering the trade-off between time and money
2. A time interval metric for cumulative opportunity accessibility
3. Racial and income inequalities in access to health in Brazilian cities
4. The equity implications of TOD in Curitiba
5. A novel route-based accessibility measure and its association with transit ridership

Plus four other manuscripts that we need to finish editing / translating to English before submitting them to journals.

1 book available in both Portuguese and English 📔

- PT Introdução à acessibilidade urbana um guia prático em R
- EN Introduction to urban accessibility: a practical guide with R

12 reports / working papers 📒

of which three focus on the assessment of governmental projects and policies. All reports are available on the project website 🌐

Estimativas de acessibilidade a empregos e serviços públicos via transporte ativo, público e privado nas 20 maiores cidades do Brasil em 2017, 2018, 2019
Distribuição espacial de características sociodemográficas e localização de empregos e serviços públicos das vinte maiores cidades do Brasil
Transporte urbano e insuficiência de acesso a escolas no Brasil
Desigualdades raciais e de renda no acesso à saúde nas cidades brasileiras
Tendências e desigualdades da mobilidade urbana no Brasil II: Características e Padrões de Consumo da Mobilidade por Aplicativo
Tendências e desigualdades da mobilidade urbana no Brasil I: o uso do transporte coletivo e individual
Forma urbana e mobilidade sustentável: evidências de cidades brasileiras
Impactos da expansão do metrô de Fortaleza sobre o acesso a oportunidades de emprego, saúde e educação
As Condições de Acessibilidade Urbana dos Terrenos Vazios da União nos Centros Urbanos
Mobilidade urbana e o acesso ao Sistema Único de Saúde para casos suspeitos e graves de COVID-19 nas 20 maiores cidades do Brasil
Relatório de avaliação: Programa Minha Casa Minha Vida (PMCMV)
Desigualdades socioespaciais de acesso a oportunidades nas cidades brasileiras, 2019

2 working papers currently in progress ✏️

- Urban mobility trends and inequalities in Brazil III: the use of active travel
- The impacts of traffic congestion on urban accessibility

7 graduate dissertations co-supervised 🎓

- 3 PhDs dissertations in progress
- 2 Masters dissertations in progress
- 2 concluded Masters dissertations

🎉 Two of our research assistants have been approved for fully funded PhDs, both of them at the University of Toronto but at different departments (Geograhy and Civil Engineering)

Outreach 📣

- A six-day capacity-building workshop teaching government officials how to run accessibility analysis in R to assess the impact of transportation projects
- Over 70 presentations for local governments, at international conferences, webinars and graduate programs

5 R packages and 1 Python package 📦 that today sum over 235K downloads

- aopdata: Data from the Access to Opportunities Project
- r5r: Rapid realistic routing on multimodal transport networks with R5
- r5py: Rapid Realistic Routing with R5 in Python
- accessibility: Transport Accessibility Metrics
- gtfstools: General Transit Feed Specification (GTFS) Editing and Analysing Tools
- gtfsio: Read and Write General Transit Feed Specification (GTFS) Files

ps. 1: These outputs have received modest 268 citations. Not bad for a short period of three years, but I like to think that this is just the tiny tip of the iceberg, and that the impact and reach of the Access to Opportunities Project run deeper than what citation metrics can tell us. I wouldn’t mind seeing citation numbers going up in the next few years though.

ps. 2: While I’m very glad to celebrate our accomplishments so far, I’m even more excited about the next steps we’re taking in our research project / agenda. Stay tuned.

A new special friend

Sun, 30 Apr 2023 00:00:00 +0000

This happened. We’ve become parents of two, and I’ve made a new special friend. I’m among the happiest tired men on Earth ❤️

Trading off costs with time in accessibility by ride-hailing and transit

Tue, 28 Mar 2023 00:00:00 +0000

I’m excited to share the preprint of our new study comparing accessibility by ride-hailing Vs transit Vs the combined use of ride-hailing + transit.

Perhaps the most important contribution of the study is to demonstrate how accounting for the trade-offs between travel time and monetary costs is a critical aspect in accessibility analysis and which should be considered by future research on transportation accessibility and equity. Apart from that, the study also brings robust evidence that policies that promote an integration between mass transit and on-demand rideshare could bring substantial accessibility gains. Nonetheless, our findings show that such accessibility gains are likely not going to benefit low-income communities without some form of subsidized fare discounts.

Pereira, R. H. M., Herszenhut, D., Saraiva, M., & Farber, S. (2023, March 23). Ride-hailing and transit accessibility considering the trade-off between time and money. OSF Preprints. https://doi.org/10.31219/osf.io/pesjk

Abstract:

Ride-hailing services have the potential to expand access to opportunities, but out-of-pocket costs may limit the benefits of ride-hail for low-income individuals. This paper examines how ride-hailing services can shape spatial and socioeconomic differences in access to opportunities while accounting for the trade-off between travel time and monetary costs. Using one year of aggregate Uber trip data for Rio de Janeiro in 2019 and a new multi-objective optimization routing method, we analyze the potential for ride-hailing services to improve employment accessibility when used as a standalone transportation mode and in conjunction with transit as a first-mile feeder service. We compare the accessibility Pareto frontiers of these transport mode alternatives with cumulative opportunity measures considering multiple combinations of travel time and monetary cost thresholds. We find that, compared to transit, ride-hailing can significantly expand accessibility as a standalone transport mode for relatively short trips (between 10 and 40 minutes), and as a first-mile feeder to transit in trips longer than 30 minutes. In both cases, the accessibility advantages of ride-hailing are mostly limited by relatively higher out-of-pocket costs. When we account for different affordability thresholds, the accessibility benefits of ride-hailing services accrue mostly to high-income groups. These findings suggest that policy efforts to integrate rideshare with transit are likely not going to benefit low-income communities without some form of subsidized fare discounts to alleviate affordability barriers. The paper also highlights how accounting for trade-offs between travel-time and monetary costs can importantly influence the results of transportation accessibility and equity studies, suggesting that this issue should be addressed in future research.

English version of the book on introduction to urban accessibility with R

Fri, 10 Mar 2023 00:00:00 +0000

In November 2022, Daniel Herszenhut and I released online the Portuguese version of our book “Introdução à acessibilidade urbana: um guia prático em R”. We are now glad to share the book is also available in English:

📖 Introduction to urban accessibility: a practical guide with R

A quick summary of the book:

The book aims to equip readers with the fundamental concepts, data analysis skills & tools needed to perform accessibility analyses and assess the impact of transp. projects. The book presents several reproducible examples using open source data sets and code in R. The publication is entirely online, but the Institute for Applied Economic Research (Ipea) will be publishing PDF and e-PUB versions of the book soon.

The book is divided into 5 sections:

The 1st section presents the concept of urban accessibility, clarifies the differences between accessibility and mobility, and presents the main indicators found in the literature to measure urban accessibility.
The 2nd section teaches how to compute urban accessibility estimates in R using the {r5r} and {accessibility} packages from open public data, and how to visualize results with maps and charts.
The 3rd section introduces the GTFS specification for public transport data, and shows how to work with and analyze GTFS data using the {gtfstools} package. It also shows where to find public transport data in GTFS format for cities across the globe, including Bazil
The 4th section brings a case study to illustrate how the combined knowledge of previous chapters can be used to assess the impact of transportation policies on urban accessibility conditions and on inequalities in access to opportunities
Finally, the 5th section shows how to download, analyze & visualize the data of the Access to Opportunities Project (AOP) using the {aopdata} package, which brings detailed data on landuse patterns & accessibility for different transport modes and years for Brazilian cities #aopdata

The book was entirely written with open source code in R and Quarto ( Github repo here) Now the icing on the cake 🍰 you can run the code used in the book online in your browser with Binder.

Daniel I wrote this book as a teaching material. We’ve already used it in a workshop with policy makers in Brazil & received some really nice feedback. We hope other workshops will follow in Brazil and abroad, and that the book will help students, researchers & practicioners improve urban policies.

Impact of network centrality and income on slowing infection spread

Mon, 27 Feb 2023 00:00:00 +0000

Very glad to share our new paper, this one led by Shiv Yücel, in collaboration with Pedro Peixoto, and Chico Camargo. In case you’d like a pocket version of the paper, early last year Chico posted a really nice thread on Twitter summarizing some of the key contributions and findings of the paper. The paper is open access.

Yücel, S.G., Pereira, R.H.M., Peixoto, P.S. et al. Impact of network centrality and income on slowing infection spread after outbreaks. Applied Network Scince. 8, 16 (2023). https://doi.org/10.1007/s41109-023-00540-z

Abstract:

The COVID-19 pandemic has shed light on how the spread of infectious diseases worldwide are importantly shaped by both human mobility networks and socio-economic factors. However, few studies look at how both socio-economic conditions and the complex network properties of human mobility patterns interact, and how they influence outbreaks together. We introduce a novel methodology, called the Infection Delay Model, to calculate how the arrival time of an infection varies geographically, considering both effective distance-based metrics and differences in regions’ capacity to isolate—a feature associated with socio-economic inequalities. To illustrate an application of the Infection Delay Model, this paper integrates household travel survey data with cell phone mobility data from the São Paulo metropolitan region to assess the effectiveness of lockdowns to slow the spread of COVID-19. Rather than operating under the assumption that the next pandemic will begin in the same region as the last, the model estimates infection delays under every possible outbreak scenario, allowing for generalizable insights into the effectiveness of interventions to delay a region’s first case. The model sheds light on how the effectiveness of lockdowns to slow the spread of disease is influenced by the interaction of mobility networks and socio-economic levels. We find that a negative relationship emerges between network centrality and the infection delay after a lockdown, irrespective of income. Furthermore, for regions across all income and centrality levels, outbreaks starting in less central locations were more effectively slowed by a lockdown. Using the Infection Delay Model, this paper identifies and quantifies a new dimension of disease risk faced by those most central in a mobility network.

Measuring just accessibility within planetary boundaries

Wed, 22 Feb 2023 00:00:00 +0000

Glad to share this preprint of our study led by Elias Willberg, in collaboration with Henrikki Tenkanen, Harvey Miller and Tuuli Toivonen.

Willberg, E., Tenkanen, H., Miller, H. J., Pereira, R. H., & Toivonen, T. (2023). Measuring just accessibility within planetary boundaries. OSF Preprints. https://doi.org/10.31219/osf.io/3h6wn

Abstract:

Our societies are wrestling with the problem of how to provide a good life for all without overconsuming environmental resources. Consequently, the search for approaches that simultaneously inspect the environmental and social goals of sustainability have become increasingly popular in many disciplines. In transport research, accessibility is a key tool to characterize linkages between people, transport, and land use. In the current paper, we propose a conceptual framework for measuring just accessibility within planetary boundaries. We first carry out a review of transport studies and discover a substantial body of literature on accessibility and social disadvantage, much vaster compared to the body of literature around environmental and ecological impacts of accessibility. We also show a gap in approaches that have integrated these two perspectives. Building on the review, we move forward by suggesting a conceptual framework for incorporating environmental and social sustainability goals in accessibility research. We conclude the paper by pointing to some key challenges and research avenues related to the framework, including (i) dealing with uncertainty and complexity in socio-ecological thresholds, (ii) developing ways to measure accessibility through other metrics than travel time, and (iii) integrating both quantitative and qualitative data.

Major release of r5r v1.0.0

Mon, 30 Jan 2023 00:00:00 +0000

A new major release of r5r v1.0.0 is now on CRAN. This version has many new goodies for those doing routing and accessibility analysis in multimodal transport networks in #rstats.

Read this post for 6 highlights or the full changelog here:

Perhaps the most exciting news: Now you can use #r5rstats to account for💰 monetary costs💰in the functions travel_time_matrix(), accessibility() and detailed_itineraries(). Full details here. This is a game changer for transportation equity research

New pareto_frontier() function to find multiple path alternatives considering trade-offs between travel time ⏱️ & monetary costs 💰, making it easy to use the method developed by Matthew Bhagat-Conway and Anson Stewart in this paper. Check the detailed vignette on how to use the function here.

If you want more info for your travel matrix, the new expanded_travel_time_matrix(breakdown = T) function also returns the transit routes taken + several other info such as number of transfers & total access, waiting, in-vehicle & transfer times. Details here.
The detailed_itineraries() function is much faster 🚀 and substantially improved ! It now includes the parameters:
- suboptimal_minutes to find a larger set of suboptimal routes
- all_to_all to routes between all origins to all destinations
- max_fare to account for monetary costs

The output of the street_network_to_sf() function now includes more information on street segments, such as osm_id, level of traffic stress (LTS), max car speed, and type of street
Finally, we’ve added the output_dir parameter to all routing & accessibility functions. It allows one to save the outputs as .csv files (one file per origin), which is particularly useful for computers with limited RAM. It also makes r5r even faster 🚀🔥

Please mind you that the new r5r v1.0.0 also has a few breaking changes, which are all documented in the changelog. Huge thanks to all co-developers, specially Marcus Saraiva and Daniel Herszenhut for their major contributions to this new release!

bot: Esquina do Brasil

Wed, 04 Jan 2023 00:00:00 +0000

I needed a bit of time off from work during the holidays and decided to learn something new. I’ve created “Esquina do Brasil” @esquinadobrasil, a bot 🤖 that tweets the map 🛰🌎 of a random census tract in Brazil every 15 minutes. The tweet shows the satellite imagery data and includes additional information on resident population, area (km²), name of the neighborhood and whether that census tract is classified as an urban or rural area. The bot was created in R and the code is available on GitHub.

The purpose of this bot is to ~~procrastinate~~ help people ‘visit’ every single corner of Brazil and have a glance of the country’s beautiful places and inequality issues. The 🤖 is also meant to have an educational purpose to arouse people’s curiosity and interest on spatial data science and programming in R. Here’s the example of one of the early tweets from @esquinadobrasil.

Municipio: Matão - SP
Setor censitário: 352930205000047
População: 793
Área (Km2): 0.17
Zona: urbana
🗺 https://t.co/gGsBT1HOfq pic.twitter.com/RW0tDwbFR4
— Cada esquina do Brasil (@esquinadobrasil) January 4, 2023

ps. I had this idea a couple years ago but only now I had the mental space to learn how to do it. It was fun!

Free public transit and voter turnout

Sat, 17 Dec 2022 00:00:00 +0000

Super glad to share the preprint of our new study. Approx. 50% of the Brazilian electorate live in cities that provided fare-free public transit on the 2022 national election days. We examine the impacts of this large-scale fare-free policy on voter turnout rates, election outcomes, and human mobility levels.

Pereira, R. H. M., Vieira, R. S., Bizzarro, F., Barbosa, R. J., Dahis, R., & Ferreira, D. T. (2022). Free public transit and voter turnout. OSF Preprint. DOI 10.31219/osf.io/fzwgq

Abstract:

Transportation costs are an under-appreciated barrier to political participation. Here we examine whether a large-scale intervention to lower these costs, the adoption of a fare-free transit policy on election day in Brazil, increases voter turnout. Taking into account the different timing of when municipalities adopted a fare-free transit policy between the first and second rounds of the country’s 2022 presidential election, we use different event study designs to examine the policy impact on voter turnout rates, election outcomes, and human mobility levels. We find no effect of the policy on turnout or election outcomes, but we find a positive effect, between 7.2% and 17.5% increase, on mobility levels on election day. While reducing transportation monetary costs may improve people’s access to polling places, our findings suggest it is not sufficient on its own to increase voter turnout.

gtfs2emis: Estimating public transport emissions from GTFS data

Wed, 23 Nov 2022 00:00:00 +0000

Very glad to share gtfs2emis, a new bottom-up model to estimate the environmental emissions of public transport systems using GTFS data. It allows one to estimate the emissions of each public transport vehicle at high spatial and temporal resolutions with little data requirements.

gtfs2emis is available as an R package on CRAN, and you can read the detailed methods in this preprint, where we demonstrated the model by analyzing public transport emissions in São Paulo.

Bazzo Vieira, J. P., Pereira, R. H. M., & Andrade, P. R. (2022). Estimating public transport emissions from General Transit Feed Specification data. OSF Preprints. https://doi.org/10.31219/osf.io/8m2cy.

The model was co-developed with João Bazzo (Ipea) and Pedro Andrade (INPE), and it was shortlisted as a Finalist in the 2022 ITF Decarbonising Transport Award.

ps 1. This idea occurred to me over 3 years ago, but it took a long and incredible collaboration with João Bazzo and P.Andrade to make it happen. João Bazzo made gtfs2emis possible thanks to his great expertise with transport emissions. He’s applying for PhD this year so keep an eye on him!

ps 2. Our next step is to use gtfs2emis to conduct an international benchmark comparing the environmental performance of public transport networks across several global cities. This will take a huge effort but we’re hoping to count with new collaborators. Stay tuned!

Introduction to urban accessibility: a practical guide in R

Sat, 05 Nov 2022 00:00:00 +0000

For those interested in spatial analysis and urban transportation, Daniel Herszenhut and I have written a book about urban accessibility and data analysis. The book is currently only available in Portuguese, but we are planning to share a version in English in the next few months. More details about the book below:

📖 Introduction to urban accessibility: a practical guide with R

The book is intended for teaching purposes, serving as a practical guide to model transport networks, calculate accessibility and evaluate the impact of transportation projects on urban accessibility. The book provides reproducible examples based on open data and software using the R programming language and tools. The publication is entirely online, but the Institute for Applied Economic Research (Ipea) will be publishing PDF and e-PUB versions of the book soon.

The book is divided into 5 sections:

The first section introduces the concept of urban accessibility, clarifies the differences between accessibility and mobility, and presents the main indicators used to measure urban accessibility.
The second section teaches how to compute urban accessibility estimates in R using the {r5r} and {accessibility} packages from open public data.
The third section introduces the GTFS specification for public transport data, and shows how to work with and analyze GTFS data using the {gtfstools} package
The fourth section presents a case study to illustrate how the combined knowledge from the previous chapters can be used to evaluate the impact of transportation projects on urban accessibility conditions
Finally, the fifth section shows how to download and analyze data from the Access to Opportunities Project (AOP) with the {aopdata} package to analyze the spatial and socioeconomic distribution of the conditions of access to opportunities in Brazilian cities. #aopdata

The book was entirely written with open source code in R and Quarto ( Github repo here) Now the icing on the cake 🍰 The code used in the book can be run in the cloud using Binder.

A few weeks ago, Daniel and I taught this book as hands-on workshop for the staff of the urban mobility secretariat in Brazil’s federal government. We hope the book and the workshop will help develop the necessary critical thinking and evidence-informed urban policy to promote more inclusive and just cities.

Update of the Access to Opportunities Project

Sun, 23 Oct 2022 00:00:00 +0000

The Access to Opportunities Project (AOP) is a research initiative led by the Institute for Applied Economic Research (Ipea) with the aim to study transport accessibility conditions in Brazilian cities. In this project, we generate annual estimates of access to employment, health, education, and social protection services for each block and mode of transport (🚗 🚌 🚴‍♂️ 🚶‍♀️) for the 20 largest cities in Brazil. The first results of the project were published in 2020.

This week, we published the new results of AOP. What’s new?

Accessibility estimates for the years 2017, 2018 and 2019
Access by car 🚗 considering historical traffic conditions (congestion levels)
Access to social protection services
The entire metropolitan area of Goiânia
Data on the spatial distribution of jobs
New app for interactive visualization of results

All data can be downloaded:

from the AOP website
or in R using the {aopdata} package

The data and methods used in the project are detailed in these two working papers. For now, these reports are only available in Portuguese, but we’re planning to have a paper in English soon.

geobr is back on CRAN

Mon, 22 Aug 2022 00:00:00 +0000

I’m very glad to share that the geobr package is back on CRAN!!!

geobr is a computational package (available in R and Python) to download official spatial data sets of Brazil. The package includes a wide range of geospatial data in geopackage format (like shapefiles but better), available at various geographic scales and for various years with harmonized attributes, projection and topology (see detailed list of available data sets below).

The new R version of geobr v1.7.0 now simultaneously stores the data on two independent servers. In case one server connection fails, geobr will automatically download the data from the second server. This means the package is much more stable and less vulnerable to instabilities in data server connection.

Improving future travel demand projections

Thu, 18 Aug 2022 00:00:00 +0000

Very glad to share we have a new open access paper out, this one led by the wonderful Sonia Yeh (Chalmers University). In this position paper we call for more open science and interdisciplinary collaborative open data projects to advance science and policy towards improving the sustainability performance of future transport systems.

Yeh, S., Gil, J., et al (2022). Improving future travel demand projections: a pathway with an open science interdisciplinary approach. Progress in Energy 4 043002. https://doi.org/10.1088/2516-1083/ac86b5

Abstract:

Transport accounts for 24% of global CO2 emissions from fossil fuels. Governments face challenges in developing feasible and equitable mitigation strategies to reduce energy consumption and manage the transition to low-carbon transport systems. To meet the local and global transport emission reduction targets, policymakers need more realistic/sophisticated future projections of transport demand to better understand the speed and depth of the actions required to mitigate greenhouse gas emissions. In this paper, we argue that the lack of access to high-quality data on the current and historical travel demand and interdisciplinary research hinders transport planning and sustainable transitions toward low-carbon transport futures. We call for a greater interdisciplinary collaboration agenda across open data, data science, behaviour modelling, and policy analysis. These advancemets can reduce some of the major uncertainties and contribute to evidence-based solutions toward improving the sustainability performance of future transport systems. The paper also points to some needed efforts and directions to provide robust insights to policymakers. We provide examples of how these efforts could benefit from the International Transport Energy Modeling Open Data project and open science interdisciplinary collaborations.

A new time interval cumulative accessibility metric

Mon, 01 Aug 2022 00:00:00 +0000

I’m glad to share the preprint of a new study where we propose “A time interval metric for cumulative opportunity accessibility”. check the abstract below. If you’d like a non-academic summary, here is a Twitter thread summarizing the key findings of the paper.

Tomasiello, D. B., Santos, D. H. M., Oliveira, J. L. A., Braga, C. K. V., & Pereira, R. H. M. (2022). A time interval metric for cumulative opportunity accessibility. SocArXiv preprint https://doi.org/10.31235/osf.io/ux5ah

Abstract:

Cumulative accessibility measures allow one to estimate the number of opportunities that can be reached within a given travel time threshold. They have become the most commonly used metric in transport research and planning because of how simple it makes to calculate and communicate accessibility results. However, cumulative opportunity measures require an ad-hoc choice of a single travel time threshold, which can importantly influence the conclusions of transport project evaluations and equity analyses. In this paper, we introduce the time interval cumulative accessibility measure, a new accessibility metric that mitigates the impacts of arbitrary choices of trip duration on cumulative accessibility analyses while keeping its computation and communicability advantages. The proposed indicator estimates the average or the median number of opportunities that can be reached considering multiple minute-by-minute travel time cutoffs within a given travel time interval. We demonstrate the new metric in a case study assessing how a planned subway expansion will likely impact employment accessibility in Fortaleza, Brazil. Using sensitivity analyses with Monte Carlo simulations, we demonstrate that the selection of travel time thresholds can substantially bias the results of accessibility levels and that such biases are not evenly distributed in space, having important implications for equity analyses. We also show that the proposed time interval cumulative opportunity metric makes the results of accessibility estimates and inequality analyses significantly less sensitive to ad-hoc methodological choices while yielding results that are very similar to those found with traditional threshold-based cumulative measures. Future accessibility-oriented transport research and planning could greatly benefit from the way in which the proposed time interval cumulative opportunity measure provides more robust accessibility estimates without compromising the communicability of results.

ps. the time interval cumulative accessibility is super easy to calculate. To make it readily available for researchers and practitioners, we have added the cumulative_interval() function to the accessibility package in R.

Official release of the accessibility R package on CRAN

Tue, 26 Jul 2022 00:00:00 +0000

Glad to annouce the official release on CRAN of the R package accessibility v1.0.0. The package is coauthored with (and led by) Daniel Herszenhut ( Twitter).

The accessibility 📦 offers a set of fast & convenient functions to calculate multiple transport accessibility measures. Given a pre-computed travel cost matrix and some land use data, users can easily calculate transport access to opportunities considering multiple accessibility measures and decay functions. See list of accessibility measures and decay functions below. Please have a look at the package website for reproducible example.

Accessibility functions currently available

cost_to_closest() Minimum travel cost to closest N number of opportunities
cumulative_cutoff() Cumulative access based on a travel cost cutoff
cumulative_interval() Cumulative access based on maximum travel time interval
floating_catchment_area() Floating catchment area accessibility
gravity() Gravity-based accessibility measure

Decay functions currently available

For convenience, the accessibility package includes some of the most frequently used decay functions in the literature. Additionally, users can pass any custom function to convert travel costs into weights to be applied to opportunities (see vignette).

decay_binary() Binary (a.k.a. step) decay function
decay_exponential() Negative exponential decay function
decay_linear() Linear decay function
decay_power() Inverse power decay function
decay_stepped() Stepped decay function

Accessibility and space-time differences in sustainable mobility

Thu, 07 Jul 2022 00:00:00 +0000

Glad to share the preprint of a new study where we analyze how space-time differences in accessibility are associated with how sustainably different groups travel.

Ryan, J., Pereira, R. H. M., & Andersson, M. (2022). Accessibility and space-time differences in when and how sustainably different groups (choose to) travel. OSF Preprints, https://doi.org/10.31219/osf.io/mvjx9

Abstract:

A fair distribution of accessibility to key activities is a central concern for distributive justice in transport planning. This implies that disparities in accessibility and the negative effects associated with a lack of accessibility should be mitigated. However, even though accessibility is not purely spatial or static, it is conventionally treated as if it were. Severals studies have significantly advanced the spatio-temporal representation of accessibility. However, there is still a lack of focus on how these dynamics affect different groups of people with differing socio-economic and demographic characteristics and how they might influence their travel mode choice. The aim of this paper is to analyse the differences in the distribution of accessibility to the workplace by public transport combined with walking during multiple departure time periods. This study draws on the spatio-temporal aspects of people’s real activities based on travel survey data for the Stockholm region. With this study, we contribute to the body of literature by (1) highlighting the differences in accessibility to the workplace by public transport across multiple departure time periods; (2) showing how these results compare with modal choice; and (3) revealing which socio-economic, demographic and geographical factors characterise these differences. This study is conducted from a transport justice-focused time geography perspective. This approach shows how this perspective could help researchers have a more detailed and nuanced understanding of the relationship between accessibility and sustainable travel behaviour, and how this relationship might change at different times of the day.

Urban picture

Fri, 20 May 2022 00:00:00 +0000

Venice the city of love. Beautiful art picture by Lennart Pagel.

image credit: Lennart Pagel via Cityscape Discovery.

Exploring the time geography of public transport networks with gtfs2gps

Mon, 09 May 2022 00:00:00 +0000

A while ago, we created gtfs2gps, an R package that converts public transport data from GTFS format to GPS-like records. After a lot of package improvements, different versions and over 15K downloads on CRAN, we finally have a preprint of the paper that introduces gtfs2gps to the world. In this paper we explain the methods behind gtfs2gps and show with a reproducible example how the package can help us understand and visualize the time geography of public transport systems.

Pereira, R. H. M., Andrade, P. R., & Bazzo, J. P. V. (2022). Exploring the time geography of public transport networks with the gtfs2gps package. SocArXiv preprint https://doi.org/10.31235/osf.io/qydr6.

Abstract:

The creation of the General Transit Feed Specification (GTFS) in the mid-2000s provided a new data format for cities to organize and share digital information on their public transport systems. GTFS feeds store geolocated data on public transport networks, including information on routes, stops, timetables, and service levels. The GTFS standard is now widely adopted by thousands of transport authorities and a wide variety of software applications for different purposes, including trip planning, timetable creation and accessibility analysis. Yet, there is still a lack of tools to parse GTFS data in a way that allows one to analyze the complex spatial and temporal patterns of public transport systems. This paper presents gtfs2gps, a new computational tool to easily process GTFS data that allows one to analyze the space-time trajectories of public transport vehicles at fine spatial and temporal resolutions. gtfs2gps is an open-source R package that employs parallel computing to convert GTFS feeds from relational text files into a data table format similar to GPS records with the timestamps of vehicles in every single trip. This paper explains the package functionalities and demonstrates how gtfs2gps can be used to articulate key concepts in time geography to explore and visualize the spatial and temporal patterns of public transport networks. The paper is accompanied by a computational notebook in R Markdown that allows one to easily reproduce the results in this paper or even replicate the analysis and data visualizations for other contexts where GTFS data is available. Given the widespread use of GTFS by transport agencies, gtfs2gps opens new possibilities for researchers to examine the time geography of public transport systems in urban areas across the globe.

Figure 4: Bundling of space-time paths of routes/trips that arrive at a selected bus stop within a 10-minute window. São Paulo, Brazil.

The impact of COVID-19 on air passenger demand and CO2 emissions

Tue, 15 Mar 2022 00:00:00 +0000

Very glad to share that we have a new paper where we analyze the impact of COVID-19 on air passenger demand and CO² emissions in Brazil. Here for more info about the paper (including PDF, code, data and preprint).

Bazzo, J., Braga, K., Pereira, R. H. M. (2022). The impact of COVID-19 on air passenger demand and CO² emissions in Brazil. Energy Policy. https://doi.org/10.1016/j.enpol.2022.112906

TL;DR:

COVID-19 caused a 63% drop in CO² emissions in the Brazilian aviation market.
4.7 million metric tons of CO² were avoided by the Brazilian aviation sector in 2020. This is equivalent to one year of CO² emissions from all domestic flights in France.
The pandemic also led to lower fuel efficiency with more emissions per capita due to lower occupancy rate on flights.
Despite such a sharp reduction in aviation demand, passenger demand recovered to 64.2% of pre-pandemic levels by the end of 2020.
This fast recovery suggests that the emissions could soon return to pre-pandemic levels, demonstrating how hard it can be to achieve significant and sustained emission reductions in the aviation sector.

ps. Having this paper accepted makes me particularly happy. This study was previously desk rejected by 5 transport and/or environmental journals. We’ve heard from editors that the study was not relevant enough or that the focus on Brazil was of too narrow interest. Seriously? 🤡 Had we written the exact same paper looking at data for the United States, would we have received the same feedback? Anyways, after so many frustrating attempts, we aimed higher for a much better journal and it got accepted

Fig. Daily number of passengers and CO2 emissions observed in the data and in counterfactual estimates for domestic flights in Brazil, 2020.

Message of the day

Mon, 07 Mar 2022 00:00:00 +0000

Resharing this, originally posted by Silvia Surinova.

flightsbr: Download Flight and Airport Data from Brazil

Wed, 02 Mar 2022 00:00:00 +0000

Over the past couple of years, I got involved in various research projects related to COVID-19 where we needed to analyze human mobility data in different scales and transport modes. Quite often, we used data from Brazil’s Civil Aviation Agency (ANAC) to get information on international flights to and from Brazil as well as domestic flights within the country.

We have used these data in three papers ( 1, 2 and 3) that touched on issues related to the role of human mobility in the spread of COVID-19. In a more recent study, we used flight data to estimate the impact of COVID-19 on air passenger demand and CO2 emissions in Brazil.

After using these data quite a few times, I decided to create flightsbr, an R package to make it easy for anyone to download flight and airport data from Brazil’s Civil Aviation Agency (ANAC). The data includes detailed information on all aircrafts, aerodromes, airports, and airport movements registered in ANAC, and on every international flight to and from Brazil, as well as domestic flights within the country. You may install flightsbr from CRAN or the dev version from Github.

The package currently includes the four main functions below. These functions are kind of self-explanatory. In any case, I would strongly encourage users to check the package documentation and vignettes in the flightsbr website.

read_flights()
read_airports()
read_aircrafts()
read_airport_movements()

new paper: the impact of transit monetary costs on transport inequality

Thu, 24 Feb 2022 00:00:00 +0000

Glad to share our new paper showing how monetary costs can have profound effects in transport accessibility analysis and transport inequality estimates. The PDF of the paper can be downloaded here. The paper was led by Daniel Herszenhut as part of his master’s thesis.

Herszenhut, D., Pereira, R. H. M., Portugal, L. da S., & Oliveira, M. H. de S. (2022). The impact of transit monetary costs on transport inequality. Journal of Transport Geography, 99, 103309. https://doi.org/10.1016/j.jtrangeo.2022.103309

TL;DR: Most studies in the transport accessibility and equity literature only look at travel time restrictions. This implicitly holds the unrealistic assumption that all individuals equally face no transport budget constraints. In this study we show how simultaneously incorporating both travel time and monetary costs into accessibility measurement can significantly skew the conclusions of transportation equity analyses in a non-trivial manner. Although ignoring monetary costs leads to overestimated accessibility levels, it can either overestimate or underestimate inequality levels depending on the combination of temporal and monetary restrictions considered in the analysis and the spatial configuration of cities and transport networks.

ps. Daniel has passed his thesis defense with no correction and he has been a really important member of our team at the Access to Opportunities Project. Daniel has been doing some really nice work co-developing R packages for transport data and analysis, including gtfstool, r5r and gtfsio. We should keep an eye on his next steps.

Truchet tiles in R

Sun, 20 Feb 2022 00:00:00 +0000

Antonio Páez has recently created the truchet R package, which makes it super easy to create Truchet tiles and to arrange them in beautiful mosaics. Some of these Truchet tiles remind me a lot the work of Athos Bulcão, a famous artist who created some of the most iconic tile murals in Brasília, Brazil.

One of things I really liked about the truchet package is that it can also be used to create Truchet mosaic versions of photos and images, like this beautiful Frida created by Antonio Páez (code here).

image credit: Antonio Páez.

What is a preprint? and why you should probably be sharing yours

Thu, 17 Feb 2022 00:00:00 +0000

Tl;dr

A preprint is a research manuscript before formal peer review and publication.
Pretty much all of the mos important editing companies / journals allow preprints.
Sharing preprints helps promoting open science and increases your research visibility and impact.

What is a preprint?

A preprint is simply your research piece (e.g. manuscript of a paper or a chapter) before formal peer review and publication.

Most editing companies / academic journals keep publications with restricted access behind paywalls.
It takes several months (often over a year) between the time you submit your manuscript and the time it eventually gets published online.

Sharing your preprint online addresses both of these problems. It will make your study open access and immediately available to anyone at the distance of a click. Another advantage is that preprint servers assign a Digital Object Identifier (DOI) to each preprint. With a DOI, your study can get properly cited and it becomes visible for searching engines like Google Scholar without having to wait so long for editorial decisions.

There are several platforms to share preprints. I’ve used OSF and SocArXiv several times and I would highly recommend them. There are many others options such as arXiv and medRxiv that cater to different fields like hard sciences and medical sciences.

Do journals allow preprints?

Are journals Ok with you sharing a preprint? I’m glad you asked. Pretty much all of them allow you to share preprints before your journal submission. Here are the ‘terms and policies’ of some of the most important editing companies saying it’s Ok.

Any downsides?

Having a preprint online makes it easier for reviewers to spot your identity. However, even without preprints, it’s usually not that difficult to determine authors’ identity these days because a study would probably have been presented already at conferences or online seminars. Unless in very particular cases, I don’t think this is a strong reason to stop one from sharing a preprint.

Takeaway message

In summary, sharing preprints helps promoting open science and increases your research visibility and impact. There are lots of great and free preprint server options and journals. Most editing companies / journals allow preprints.

ps. 1: SSRN used to be one of most important preprint servers in the social sciences. However, they were bought by Elsevier in 2016.

ps. 2: What about ResearchGate or Academia.edu? They are really not preprint servers. They are commercial social networking websites for academics. Personally, I’m not a big fan but they can be useful publicizing your work.

The equity implications of Transit Oriented Development in Curitiba

Thu, 13 Jan 2022 00:00:00 +0000

We have a new preprint of a study analyzing the equity implications of Transit Oriented Development (TOD) in Curitiba, co-authored with André Turbay and Rodrigo Firmino..

Turbay, A. L. B., Pereira, R. H. M., & Firmino, R. (2022). The equity implications of TOD in Curitiba. SocArXiv Papers https://doi.org/10.31235/osf.io/cj87q

Abstract:

In this paper we analyze how socio-spatial inequalities have been shaped by transport and land-use planning in Curitiba (Brazil), a city internationally recognized for its Transit Oriented Development (TOD) planning based on Bus Rapid Transit (BRT). We examine how the spatial organization of the BRT system is associated with the distribution of population densities, socioeconomic groups, and real-estate values and its implications in terms of inequalities of access to employment and health services. The results show that Curitiba’s TOD has had limited influence on population densities, but has shaped the concentration of high-income classes and premium real-estate along its main BRT corridors. These effects contribute to the peripheralization of low-income communities with limited accessibility benefits from the transit system. Our findings suggest that Curitiba’s success story should be seen as a cautionary tale about the consequences of TOD planning, which perpetuate the spatial concentration of resources and reinforce inequalities of access to opportunities.

AI generated art

Wed, 15 Dec 2021 00:00:00 +0000

I recently came across this fun tool app.wombo.art that uses AI to generate art pieces based on text inputs by a user. Needless to say this is a procrastination trap. Here are a few AI-generated art pieces based on some key words of in my research agenda.

Access to opportunities
Cities
Equity
Spatial data science
Urban mobility
Transport
Networks

Spatial data science for transport planning

Mon, 06 Dec 2021 00:00:00 +0000

Last week, I delivered the closing keynote presentation at the III Brazilian Symposium on Transport Engineering. I was invited to present about spatial data science in the context of transport planning, so I took it as a good opportunity to talk about the computational tools my team and I have been developing in to conduct our research projects. These include a few R packages like:

Here’s the video of my presentation. I presented it in Portuguese, but the slides are in English.

Animating public transport networks on 3D stacked maps

Thu, 11 Nov 2021 00:00:00 +0000

As much as I would like to, I can’t ~~procrastinate~~ participate in all 30 days of the #30DayMapChallenge. So here is the reproducible code to create a 3D image with stacked maps and an animated public transport network. This should cover some topics of the 30DayMapChallenge.

2: Lines
4: Hexagons
11: 3D
12: Population
20: Movement

install.packages('easypackages')
easypackages::packages('geobr', 'aopdata', 'gtfs2gps', 
                       'data.table', 'sf', 'viridis',
                       'magrittr', 'dplyr', 'ggnewscale',
                       'ggplot2', 'gganimate', 'av')



###### Functions to tilt sf ---------------------------------------------
# more info at https://www.urbandemographics.org/post/figures-map-layers-r/
rotate_data <- function(data, x_add = 0, y_add = 0) {
  
  shear_matrix <- function(){ matrix(c(2, 1.2, 0, 1), 2, 2) }
  
  rotate_matrix <- function(x){ 
    matrix(c(cos(x), sin(x), -sin(x), cos(x)), 2, 2) 
  }
  data %>% 
    dplyr::mutate(
      geometry = .$geometry * shear_matrix() * rotate_matrix(pi/20) + c(x_add, y_add)
    )
}

rotate_data_geom <- function(data, x_add = 0, y_add = 0) {
  shear_matrix <- function(){ matrix(c(2, 1.2, 0, 1), 2, 2) }
  
  rotate_matrix <- function(x) { 
    matrix(c(cos(x), sin(x), -sin(x), cos(x)), 2, 2) 
  }
  data %>% 
    dplyr::mutate(
      geom = .$geom * shear_matrix() * rotate_matrix(pi/20) + c(x_add, y_add)
    )
}


###### DATA: city boundary ---------------------------------------------
code_muni <- geobr::lookup_muni(name_muni = 'Fortaleza')$code_muni
city <- read_municipality(code_muni = code_muni)




###### DATA: population from aopdata ---------------------------------------------

#' more info at https://www.ipea.gov.br/acessooportunidades/en/
aop <- aopdata::read_access(city = 'fortaleza', mode = 'public_transport', geometry = T)

# select areas with population
aop <- subset(aop, P001 > 0)




###### DATA: pubic transport data ---------------------------------------------

# read GTFS data

# I'm using a local file, but you can use a sample GTFS by uncommenting the line below 
gtfs_file <-  './GTFS_fortaleza_20211020.zip'
#gtfs_file <- system.file("extdata/fortaleza.zip", package = "gtfs2gps")
gtfs_raw <- gtfs2gps::read_gtfs(gtfs_file)

# select trips btween 7am and 8am
gtfs <- gtfs2gps::filter_day_period(gtfs_raw, period_start = '07:00:00', period_end = '08:00:00')

# select a sample of n routes
set.seed(42)
nroutes <- 80
myroutes <- sample(x = unique(gtfs$routes$route_id), size = nroutes, replace = F)
gtfs <- gtfs2gps::filter_by_route_id(gtfs, route_ids = myroutes)

# get route geometries
routes <- gtfs2gps::gtfs_shapes_as_sf(gtfs)

# convert GTFS to GPS-like data.table
gps <- gtfs2gps(gtfs, parallel = T, spatial_resolution = 200)
gps <- gps[ between(departure_time, as.ITime('07:00:00'), as.ITime('08:00:00'))]

# Convert "GPS" points of vehicle positions into sf
gps_points <- sfheaders::sf_point(gps, x = "shape_pt_lon" , y = "shape_pt_lat", keep = T)
sf::st_crs(gps_points) <- 4326

# rotate vehicle positions and back to points
gps_points_r <- rotate_data(gps_points, y_add = 0.3)
gps_points_df <- sfheaders::sf_to_df(gps_points_r, fill = T)

# remove missing
gps_points_df <- subset(gps_points_df, !is.na(departure_time))


###### Plot ---------------------------------------------

# annotate parameters
x = -80.92
clr = 'gray40'
sz = 4

# city boundary
temp1  <- ggplot() +
              geom_sf(data = city %>% rotate_data_geom(y_add = .03), color='gray90', fill='gray90', show.legend = FALSE) +
              annotate("text", label='City boundary', x=x, y=9.10, hjust = 0, color=clr, size=sz) +
              labs(caption = "Image by @UrbanDemog")

temp2 <- temp1 +
  
  # Population Income
  geom_sf(data = subset(aop, P001>0) %>% rotate_data(y_add = .1), aes(fill=R001 ), color=NA, show.legend = FALSE) +
  scale_fill_viridis_c(option = 'E') +
  annotate("text", label='Income', x=x, y= 9.18, hjust = 0, color=clr, size=sz) +
  
  # Employment accessibility
  new_scale_fill() +
  geom_sf(data = subset(aop, P001>0) %>% rotate_data(y_add = .2), aes(fill=CMATT60 ), color=NA, show.legend = FALSE) +
  scale_fill_viridis_c(option = 'inferno') +
  annotate("text", label='Job access', x=x, y= 9.25, hjust = 0, color=clr, size=sz) +

  # public transport routes
  geom_sf(data = routes %>% rotate_data(y_add = 0.3), color='gray80', alpha=.4, show.legend = FALSE) +
  annotate("text", label='Public transport \nNetwork', x=x, y= 9.35, hjust = 0, color=clr, size=sz) +

  # public transport trips
  new_scale_color() +
  geom_point(data = gps_points_df, aes(x = x, y=y, color = shape_id), size= .8, alpha = 0.5, show.legend = FALSE) +
  # annotate("text", label='Public transport \nvehicles', x=x, y= 9.35, hjust = 0, color=clr, size=sz) +
  scale_colour_viridis(discrete = T) +
  theme_void() +
  
  # gganimate specificatons
  labs(title = 'Time: {frame_time}') +
  transition_time(as.POSIXct(departure_time) + 10800) +  # need to fix issue with time zone
  shadow_wake(wake_length = 0.015, alpha = FALSE) +
  ease_aes('linear') +
  coord_sf(xlim = c(-81.5, -80.8)) +
  theme(plot.background = element_rect(fill = 'white', color='white'),
        plot.caption = element_text(color = 'gray30'))



###### Save ---------------------------------------------

# save gif
anim_save(animation = temp2, "stacked_map_gtfs2gps222.gif", fps = 10, width=550, height=400)

# save mp4
anim <- animate(temp2, duration = 10, fps = 10, renderer = av_renderer(), width=550, height=400)
anim_save(animation = anim, "stacked_map_gtfs2gps222.mp4", width=550, height=400)

Mapping walking time on OSM data with r5r

Fri, 05 Nov 2021 00:00:00 +0000

The topic of day 5 in the #30DayMapChallenge is ‘OpenStreetMap’. So here is a quick reproducible example showing how one can use OpenStreetMap data and the r5r package to calculate travel times.

options(java.parameters = "-Xmx4G")

library(r5r)
library(geobr)
library(sf)
library(ggplot2)
library(here)
library(osmextract)
library(data.table)
library(magrittr)


# create subdirectories "data" and "img"
dir.create(here::here("data"))
dir.create(here::here("img"))


# get city boundaries
city <- 'Rio de Janeiro'
city_code <- lookup_muni(name_muni = city)
city_boundary <- read_municipality(code_muni = city_code$code_muni, simplified = F)


# define city center
city_center_df <- data.frame(id='center', lon=-43.182811, lat=-22.906906)
city_center_sf <- sfheaders::sfc_point(obj = city_center_df, x='lon', y='lat')
st_crs(city_center_sf) <- 4326

# define buffer area of analysis 3 Km
buff <- st_buffer(city_center_sf, dist = 3000)

# crs
city_boundary <- st_transform(city_boundary, 4326)
city_center_sf <- st_transform(city_center_sf, 4326)
buff_bb <- st_bbox(buff)


# get OSM data
osmextract::oe_download(provider = 'openstreetmap_fr',
                        file_url = osmextract::oe_match("Rio De Janeiro")[[1]],
                        download_directory = here::here("data"), force_download = T)


# build routing network
r5r_core <- r5r::setup_r5(data_path = here::here("data"), verbose = FALSE)

# get street network as sf
street_network <- r5r::street_network_to_sf(r5r_core)

# drop network outside our buffer
edges_buff <- street_network$edges[buff, ] %>% st_intersection(., buff)
vertices_buff <- street_network$vertices[buff, ] %>% st_intersection(., buff)
city_boundary_buff <- st_intersection(city_boundary, buff)
plot(city_boundary_buff)

# add id to vertices
vertices_buff$id <- vertices_buff$index


# calculate travel times to city center 
tt <- r5r::travel_time_matrix(r5r_core,
                              origins = vertices_buff,
                              destinations = city_center_df,
                              mode = 'walk')


# add travel time info to street network
tt$fromId <- as.numeric(tt$fromId)
setDT(edges_buff)[tt, on=c('from_vertex'='fromId'), travel_time := i.travel_time]
edges_buff <- st_sf(edges_buff)

Now we only need to plot and save the figure.

# figure
temp <- ggplot() + 
        geom_sf(data=buff, fill='lightblue', color=NA)  +
        geom_sf(data=city_boundary_buff, fill='white', size=.3)  +
        geom_sf(data=buff, fill='gray90', color=NA, alpha=.5)  +
        geom_sf(data=edges_buff, aes(color=travel_time, size=length))  +
        geom_sf(data=city_center_sf, color='red', size=.5) +
        scale_size_continuous(range = c(0.1, .8), guide='none') +
        scale_color_viridis_c(direction = -1) +
  labs(title='Walking time to city center',
       subtitle = 'Rio de Janeiro',
       caption = 'Image by @UrbanDemog using r5r and osmextract',
       color='Minutes') +
  coord_sf(xlim = c(buff_bb[[1]], buff_bb[[3]]), 
                 ylim = c(buff_bb[[2]], buff_bb[[4]]), expand = FALSE) + 
  theme_void() +
  theme(plot.background = element_rect(fill = "white", color='white'),
        plot.title = element_text(color = "gray20"),
        plot.subtitle = element_text(color = "gray40"),
        plot.caption = element_text(color = "gray"))
        
# save figure
ggsave(temp, file=here::here("img", "aa32k.png"), 
       dpi=300, width = 14, height = 14, units = 'cm')

Mapping accessibility with data from the Access to Opportunities Project

Thu, 04 Nov 2021 00:00:00 +0000

The topic of day 4 in the #30DayMapChallenge is ‘Hexagons’. I thought this could be a good opportunity to show how one can download and map the data we generate in the Access to Opportunities Project.

So here is a quick code snippet to map employment accessibility by public transport in some of Brazil’s cities. The data can be easily downloaded from the aopdata package.


library(aopdata)
library(ggplot2)

# download data
cities_list <- c('Fortaleza', 'Curitiba', 'Sao Paulo', 'Porto Alegre')
a <- read_access(city=cities_list, mode = 'public_transport', geometry = T)

# function to generate plots
  save_plot <- function(i){ # i="Sao Paulo"
    
    # select city
    temp_df <- subset(a, name_muni== i)
    temp_df <- subset(temp_df, P001 > 0)
    
    temp_fig <- ggplot() +
                  geom_sf(data=temp_df, aes(fill=CMATT60), color=NA) +
                  scale_fill_viridis_c(option = "cividis", labels=scales::percent) +
                  labs(subtitle='Access to jobs by public transport in less than 60 min.',
                        title=paste0(i) ,
                       fill="Jobs \naccessible",
                       caption = 'Image by @UrbanDemog using aopdata') +
                  theme_void() +
                  theme(plot.title = element_text(color = "gray20"),
                        plot.subtitle = element_text(color = "gray30"),
                        plot.caption = element_text(color = "gray")
                  )
    # save
    ggsave(temp_fig, filename = paste0(i,'.png'), width = 7, height = 7, dpi=300)
  
  }

# run :)
lapply(X=cities_list, FUN=save_plot)

Creating a map of cycling, walking and driving routes in R

Tue, 02 Nov 2021 00:00:00 +0000

Here’s a quick contribution to the #30DayMapChallenge, day 2 ‘Lines’: a quick reproducible example showing how to use the r5r package to estimate cycling, walking and driving routes in R. In this example, we generate the cycling, walking and driving routes from 2 thousand random locations to the city center in the city of Curitiba, Brazil.

Prepare R environment

options(java.parameters = "-Xmx4G")

library(r5r)
library(geobr)
library(sf)
library(ggplot2)
library(here)
library(osmextract)
library(magrittr)


# create subdirectories "data" and "img"
dir.create(here::here("data"))
dir.create(here::here("img"))

Download the data

# get OSM data
osmextract::oe_download(
  file_url = osmextract::oe_match("Curitiba, Brazil")[[1]],
  download_directory = here::here("data"), force_download = T)

# get city boundaries and city center
city <- 'Curitiba'
city_code <- lookup_muni(name_muni = city)
city_boundary <- read_municipality(code_muni = city_code$code_muni)
city_center <- read_municipal_seat() %>% subset(code_muni == city_code$code_muni)
city_center$id <- 'center'

# get sample of origins in city
set.seed(42)
origins <- st_sample(city_boundary, size = 2000) %>% st_sf()
origins$id <- 1:nrow(origins)

origins <- st_transform(origins, 4326)
city_center <- st_transform(city_center, 4326)

Routing analysis

# build network
r5r_core <- setup_r5(data_path = here::here("data"), verbose = FALSE)

# routing
df <- detailed_itineraries(r5r_core,
                           origins = origins,
                           destinations = city_center,
                           mode = 'bicycle',
                           departure_datetime = as.POSIXct("13-03-2019 14:00:00", format = "%d-%m-%Y %H:%M:%S"),
                           max_trip_duration = 1440,
                           shortest_path = T)

Plot

fig <- ggplot() +
        geom_sf(data = city_boundary, color='gray70', fill=NA) +
        geom_sf(data = df, color='red', alpha=.1, size=.3) +
        labs(title='Cycling in Curitiba',
             subtitle = '2000 routes to the city center',
             caption = 'Image by @UrbanDemog using r5r') +
        theme_void() +
        theme(plot.title = element_text(color = "gray20"),
              plot.subtitle = element_text(color = "gray40"),
              plot.caption = element_text(color = "gray")
              )

# save figure
ggsave(fig, file=here::here("img", "cycling.png"), 
       dpi=300, width = 14, height = 14, units = 'cm')

A metaphor for the best writing tip I've heard

Wed, 27 Oct 2021 00:00:00 +0000

Writing never comes easy to me. Oftentimes my procrastination is fueled by the anxiety I feel when I need to sit down and write a manuscript. During my PhD I heard this great writing tip that really stroke a chord within me, but only a few days ago I came across a tweet by David Perell with a very eloquent metaphor that illustrates that writing tip.

Perell: How should you write? Skip the Printer Method and choose the Pixel Method instead.

New book: Social Issues in Transport Planning

Tue, 05 Oct 2021 00:00:00 +0000

I’m very glad to share that our book Social Issues in Transport Planning is finally published! Info about where to get the book here.

The book, co-edited by Geneviève Boisjoly ( École Polytechnique de Montréal) and me, was published by Elsevier as part of the Advances in Transport Policy and Planning series, organized by Bert van Wee.

We have gathered established scholars and rising stars to present an up-to-date and critical review of some of the most pressing and sometimes overlooked social issues in transport planning. The book covers a diverse range of topics such as transport poverty and car dependence, transit-induced gentrification, accessibility, gender, race, children’s mobility, the governance of paratransit, equity in project appraisal as well as customer satisfaction with public transport, and the potential of research methodologies in promoting more inclusive and participatory transport planning (see list of chapters and authors below).

In the preface we present a summary of the book, and reflect on what makes a social issues worthy of public attention in the transport planning process. In essence, any social issue only becomes recognized as a social problem inasmuch as its manifestation goes against our understanding of how ethics should be applied to transport planning and against our aspirations of justice with regards to equity, democracy and diversity.

Preface: Social issues in transport planning, by Rafael H.M. Pereira and Geneviève Boisjoly Preprint

The Roots of Racialized Travel Behavior, by Jesus M. Barajas Preprint
Gender gaps in urban mobility and transport planning, by Tanu Priya Uteng
The Social Dimensions of Children’s Travel, by E. Owen D. Waygood, Pauline van den Berg and Astrid Kemperman
Transport poverty and car dependence: a European perspective, by Giulio Mattioli
Making the links between accessibility, social and spatial inequality, and social exclusion: a framework for cities in Latin America, by Daniel Oviedo
Transit-Induced Gentrification and Displacement: The State of the Debate, by Elizabeth C. Delmelle Preprint
From “Para-Transit” to Transit? A Review of the Politics of Popular Transport, by Jacqueline M. Klopp
What are we doing with all that satisfaction data? Evaluating Public Transport customer satisfaction data collection and analysis techniques, by Cherise Roberts, Emily Grisé and Dea van Lierop
Social and distributional impacts in transport project appraisals, by Ruth Shortall and Niek Mouter
Innovative field research methodologies for more inclusive transport planning: review and prospect, by Gina Porter and Claire Dungey

Citizens Collective on Urban Inequalities

Wed, 29 Sep 2021 00:00:00 +0000

Juliana Gonçalves and Trivik Verma from TU Delft have recently launched the Citizens Collective, a series of webinars to discuss and reflect on research and innovation in cities. The theme of the first collection is Urban Inequalities.

Rachel Franklin (Newcastle) kicked off the seminar series a few days ago with a thought provoking presentation asking: Who Counts in the Smart City? The recording of Rachel’s presentation is available here.

The seminar series continues with many other speakers and themes, covering feminist cities, renewable energy, green areas, urban and transportation systems resilience, energy poverty and participatory mapping.

I’ll be speaking about Transport Justice, Equity & Inequalities in Access to Opportunities on 20 Oct 2021 at 4pm CET. You’re all invited ~~you too mom~~.

100 years of John Rawls

Thu, 23 Sep 2021 00:00:00 +0000

John Rawls is widely recognized as the most influential political philosopher of the 20th century. To celebrate the 100th anniversary of Rawls birth and the 50th anniversary of A Theory of Justice, Routledge is putting open access a series of studies that critically engages with his work. HT Jane.

If you’re interested in how Rawls’ ideas (and other theories of distributive justice) can help us frame and advance equity in transport policies, ~~self promotion alert~~ you might wanna have a look at this paper below. Ungated version here.

Pereira, R. H., Schwanen, T., & Banister, D. (2017). Distributive justice and equity in transportation. Transport Reviews, 37(2), 170-191.

This was definitely one of the most challenging and transformative research experiences I’ve been through. Even after 4 years of the publication, I’m still am quite happy with the result and I’m chuffed that the study has become one the most cited papers in the Transport Reviews journal.

credit: illustration by Ben Jones

Urban Picture

Mon, 20 Sep 2021 00:00:00 +0000

The Bay Bridge, by David Hanjani.

credit: David Hanjania

Interlinks between my research career and life trajectory

Fri, 10 Sep 2021 00:00:00 +0000

A few weeks ago I was interviewed for the podcast “Cientista também é…” (which could be translated to ‘Scientists are also…'), organized by Mettzer. I’ve been interviewed for other podcasts before (see below), but this one was quite special because it covers both professional and biographical aspects. It was really nice to step back a little bit and reflect about how my research career and agenda are intertwined with my personal life trajectory, and what lessons one can get out of this.

Here is the interview (recorded in Portuguese):

* Appearances in other podcasts (in Portuguese):

Map of the day

Sun, 05 Sep 2021 00:00:00 +0000

The city of Rio de Janeiro (Brazil) seen through OpenStreetMap data.

image credit: Rafael Pereira @UrbanDemog

Brazilian scientists right now

Wed, 01 Sep 2021 00:00:00 +0000

The fact that this tweet went viral tells a lot about:

the ditch Brazil has dug itself into
the unshakable ability of Brazilians to laugh at their own misery

Brazilian scientists right now pic.twitter.com/1ECKno2Uqj
— Rafael H. M. Pereira 🚡 Urban Demographics (@UrbanDemog) August 29, 2021

I feel attacked

Sat, 28 Aug 2021 00:00:00 +0000

I feel attacked!

image credit: Steve Gisselbrecht

Quote of the day

Tue, 17 Aug 2021 00:00:00 +0000

Wise words about scientific and philosophical creativity.

Highway concessions and road safety

Mon, 16 Aug 2021 00:00:00 +0000

I’m glad to share we have a new paper in Research in Transportation Economics, where we analyze the road safety benefits of road concessions. The ungated PDF of the paper can be downloaded here. The code we used in the data analysis is availabe on GitHub.

Alves, P.J., Emanuel, L. & Pereira, R. H. M. (2021). Highway concessions and road safety: Evidence from Brazil. Research in Transportation Economics DOI: https://doi.org/10.1016/j.retrec.2021.101118

Abstract:

Reducing road fatalities is a key policy goal in several countries and there is a vast literature on what factors affect road safety performance. Nonetheless, there is limited evidence on whether highway concessions and Public Private Partnerships (PPP) can bring road safety benefits, despite the growing number of countries adopting this type of policy to finance and manage road infrastructure. In this paper, we use a difference-in-differences approach to examine the causal effect of highway concessions on road safety outcomes using daily data from Brazilian Federal highways over 11 years between 2007 and 2017. We exploit the transition from public to private management in some but not all Brazilian states to provide both within- and between-states comparison. We find that concessions promote a small but significant reduction in the number and fatality of road crashes as well as and the number of people and vehicles involved in crashes. Between 2007 and 2017, procured roads had on average 16 fewer deaths then publicly managed highways for every 1000 crashes each year. These road safety benefits only become statistically significant a few years after a concession implementation, but they are marginally larger for every additional year of concession. Finally, our results suggest that including safety-based incentives in concession contracts can substantially improve road safety performance. The findings of the paper have important implications for the social and economic evaluation of road concessions and for road infrastructure policy more broadly.

Research grant proposals be like

Tue, 03 Aug 2021 00:00:00 +0000

original image via Car porn

The impact of public transport monetary costs on transportation equity

Wed, 21 Jul 2021 00:00:00 +0000

We have a new study preprint looking at “The impact of transit monetary costs on transport equity analyses”. The manuscript is currently under review, but one can already read/cite it from OSF Preprints. The paper was led by Daniel Herszenhut (Twitter: @dhersz), who has been doing some great work in his master’s thesis and in the Access to Opportunities Project.

TL;DR: Most studies in the transport equity / accessibility literature only look at travel time restrictions. This implicitly holds the unrealistic assumption that all individuals equally face no transport budget constraints. In this study we show how simultaneously incorporating both travel time and monetary costs into accessibility measurement can significantly skew the conclusions of transportation equity analyses in a non-trivial manner. Although ignoring monetary costs leads to overestimated accessibility levels, it can either overestimate or underestimate inequality levels depending on the combination of temporal and monetary restrictions considered in the analysis and the spatial configuration of cities and transport networks.

Herszenhut, D., Pereira, R. H. M., da Silva Portugal, L., & de Sousa Oliveira, M. H. (2021, July 14). The impact of transit monetary costs on transport equity analyses. https://doi.org/10.31219/osf.io/e3tac

Abstract:

Transport equity analyses are often informed by accessibility estimates based solely on travel time impedance, ignoring other elements that might hinder access to activities, such as the monetary cost of a trip. This paper examines how and to what extent simultaneously incorporating both time and monetary costs into accessibility measures may impact transport equity assessment. We calculate job accessibility by transit in the city of Rio de Janeiro, Brazil, using cumulative opportunity measures under distinct combinations of temporal and monetary thresholds, and compare how inequality levels vary across different scenarios. We find that the most common research practice of disregarding monetary costs tends to overestimate accessibility levels. However, stricter monetary constraints do not necessarily result in less equitable scenarios. How accessibility inequality is affected by monetary costs is highly dependent on what combinations of temporal and monetary cut-offs are considered in the analysis. In the case of Rio, opting for lower monetary thresholds when looking at shorter trips leads to inequality levels lower than those found in the no monetary threshold scenario, but results in higher inequality levels when allowing for longer trips. We find that the impact of monetary costs on transport equity analyses depend on a complex interaction between fare policies, the spatial organisation and operational characteristics of transit systems, and the spatial co-distribution of jobs and residential locations. The paper thus highlights that conclusions and policy recommendations derived from transport equity analyses can be affected in non-intuitive ways by the interplay between temporal and monetary constraints. Future research should investigate how different combinations of travel time and monetary costs thresholds affect equity analyses in different contexts.

How accessibility inequality between the wealthy and the poor vary considering different monetary cost and travel time thresholds. Rio de Janeiro, 2018.

Creating a figure of map layers in R

Fri, 16 Jul 2021 00:00:00 +0000

I like figures of map layers to illustrate the many different types of data sets we combine to do urban and transport modeling. And oftentimes I ~~get obsessed with~~ like making maps that are reproducible with code in R. In this post I’ll be sharing a reproducible example showing how to create a figure of stacked maps like this one below.

Quick background: In 2014, I was trying to find a way to create map layers in R. This was before the sf library was created. Most of us were using the sp library for handling spatial data and Barry Rowlingson was super helpful, as usual. I used Barry’s suggestion to create a reproducible example so I could use it latter, but then sf was created and it completely changed how we do spatial analysis in R. Since then, Lauren O’brien proposed a simple way to tilt and stack sf objects and Stefan Jünger created an elegant function to do this. I’ll be using Stefan’s function in my example below.

Load libraries

library(easypackages)
easypackages::packages("sf",
                       "raster",
                       "stars",
                       "r5r",
                       "geobr",
                       "aopdata",
                       "gtfs2gps",
                       "ggplot2",
                       "osmdata",
                       "h3jsr",
                       "viridisLite",
                       "ggnewscale",
                       "dplyr",
                       "magrittr",
                       prompt = FALSE
                       )

Functions to tilt sf

Original function created by Stefan Jünger.

rotate_data <- function(data, x_add = 0, y_add = 0) {
  
  shear_matrix <- function(){ matrix(c(2, 1.2, 0, 1), 2, 2) }
  
  rotate_matrix <- function(x){ 
    matrix(c(cos(x), sin(x), -sin(x), cos(x)), 2, 2) 
  }
  data %>% 
    dplyr::mutate(
      geometry = .$geometry * shear_matrix() * rotate_matrix(pi/20) + c(x_add, y_add)
    )
}

rotate_data_geom <- function(data, x_add = 0, y_add = 0) {
  shear_matrix <- function(){ matrix(c(2, 1.2, 0, 1), 2, 2) }
  
  rotate_matrix <- function(x) { 
    matrix(c(cos(x), sin(x), -sin(x), cos(x)), 2, 2) 
  }
  data %>% 
    dplyr::mutate(
      geom = .$geom * shear_matrix() * rotate_matrix(pi/20) + c(x_add, y_add)
    )
}

Load data

We’ll be using a few data sets available from the packages used here. The first thing we need to do is to load the data and crop them to make sure they have the same extent.

### get terrain data ----------------

  # read terrain raster and calculate hill Shade
  dem <- stars::read_stars(system.file("extdata/poa/poa_elevation.tif", package = "r5r"))
  dem <- st_as_sf(dem)
  
  # crop
  bbox <- st_bbox(dem)


### get public transport network data ----------------

  gtfs <- gtfs2gps::read_gtfs( system.file("extdata/poa/poa.zip", package = "r5r") )
  gtfs <- gtfs2gps::gtfs_shapes_as_sf(gtfs)
  
  # crop
  gtfs <- gtfs[bbox,]
  gtfs <- st_crop(gtfs, bbox)
  plot(gtfs['shape_id'])


### get OSM data ----------------

  # roads from OSM
  roads <- opq('porto alegre') %>%
           add_osm_feature(key = 'highway',
                           value = c("motorway", "primary","secondary")) %>% osmdata_sf()
  
  roads <- roads$osm_lines
  
  # crop
  roads2 <- roads[bbox,]
  roads2 <- st_crop(roads2, bbox)
  plot(roads2['osm_id'])


### get H3 hexagonal grid ----------------

  # get poa muni and hex ids
  poa <- read_municipality(code_muni = 4314902 )
  hex_ids <- h3jsr::polyfill(poa, res = 7, simple = TRUE)
  
  # pass h3 ids to return the hexagonal grid
  hex_grid <- h3jsr::h3_to_polygon(hex_ids, simple = FALSE)
  plot(hex_grid)
  
  # crop
  hex_grid <- hex_grid[bbox,]
  hex <- st_crop(hex_grid, bbox)
  plot(hex)


### get land use data from AOP project ----------------
#' more info at https://www.ipea.gov.br/acessooportunidades/en/

  landuse <- aopdata::read_access(city = 'poa', geometry = T, mode='public_transport')
  
  # crop
  landuse <- landuse[bbox,]
  landuse <- st_crop(landuse, bbox)
  plot(landuse['CMATT30'])
  
  # hospitals
  # generate one point per hospital in corresponding hex cells
  df_temp <- subset(landuse, S004>0)
  hospitals <- st_sample(x = df_temp, df_temp$S004, by_polygon = T)
  hospitals <- st_sf(hospitals)
  hospitals$geometry <- st_geometry(hospitals)
  hospitals$hospitals <- NULL
  hospitals <- st_sf(hospitals)
  plot(hospitals)
  
  # schools
  # generate one point per schools in corresponding hex cells
  df_temp <- subset(landuse, E001>0)
  schools <- st_sample(x = df_temp, df_temp$E001, by_polygon = T)
  schools <- st_sf(schools)
  schools$geometry <- st_geometry(schools)
  schools$schools <- NULL
  schools <- st_sf(schools)
  plot(schools)

Plot

### plot  ----------------

# annotate parameters
x = -141.25
color = 'gray40'

temp1 <- ggplot() +
          
        # terrain
        geom_sf(data = dem %>% rotate_data(), aes(fill=poa_elevation.tif), color=NA, show.legend = FALSE) +
        scale_fill_distiller(palette = "YlOrRd", direction = 1) +
        annotate("text", label='Terrain', x=x, y= -8.0, hjust = 0, color=color) +
        labs(caption = "image by @UrbanDemog")

temp2 <- temp1 +
  
        # pop income
        new_scale_fill() + 
        new_scale_color() +
        geom_sf(data = subset(landuse,P001>0) %>% rotate_data(y_add = .1), aes(fill=R001), color=NA, show.legend = FALSE) +
        scale_fill_viridis_c(option = 'E') +
        annotate("text", label='Population', x=x, y= -7.9, hjust = 0, color=color) +

        # schools
        geom_sf(data = hex %>% rotate_data(y_add = .2), color='gray50', fill=NA, size=.1) +
        geom_sf(data = schools %>% rotate_data(y_add = .2), color='#0f3c53', size=.1, alpha=.8) +
        annotate("text", label='Schools', x=x, y= -7.8, hjust = 0, color=color) +
        
        # hospitals
        geom_sf(data = hex %>% rotate_data(y_add = .3), color='gray50', fill=NA, size=.1) +
        geom_sf(data = hospitals %>% rotate_data(y_add = .3), color='#d5303e', size=.1, alpha=.5) +
        annotate("text", label='Hospitals', x=x, y= -7.7, hjust = 0, color=color) +
  
        # OSM
        geom_sf(data = roads2 %>% rotate_data(y_add = .4), color='#019a98', size=.2) +
        annotate("text", label='Roads', x=x, y= -7.6, hjust = 0, color=color) +
  
        # public transport
        geom_sf(data = gtfs %>% rotate_data(y_add = .5), color='#0f3c53', size=.2) +
        annotate("text", label='Public transport', x=x, y= -7.5, hjust = 0, color=color) +
  
        # accessibility
        new_scale_fill() + 
        new_scale_color() +
        geom_sf(data = subset(landuse, P001>0) %>% rotate_data(y_add = .6), aes(fill=CMATT30), color=NA, show.legend = FALSE) +
        scale_fill_viridis_c(direction = 1, option = 'viridis' ) +
        theme(legend.position = "none") +
        annotate("text", label='Accessibility', x=x, y= -7.4, hjust = 0, color=color) +
        theme_void() +
        scale_x_continuous(limits = c(-141.65, -141.1))

  
# save plot
ggsave(plot = temp2, filename = 'map_layers.png', 
       dpi=200, width = 15, height = 16, units='cm')

Urban Picture

Wed, 14 Jul 2021 00:00:00 +0000

A section of the Pinheiros River in São Paulo, Brazil. Beautiful photo by the talented Sérgio Souza, who you may follow on Twitter or Instagram.

credit: Sérgio Souza

New version of r5r v0.5.0 published on CRAN

Mon, 05 Jul 2021 00:00:00 +0000

I’ve posted here before about r5r, an R package for rapid realistic routing on multimodal transport networks. I’m glad to announce we have released a new version of r5r on CRAN. More info here. This update has some really nice new capabilities. Here are four of the most exciting updates of r5r v0.5.0:

New accessibility() function. It makes it super easy (and fast) to calculate accessibility with only two lines of code. It includes many options of decay function. Code with reproducible example of how to calculate accessibility with r5r.

New find_snap() function that indicates where origin/destination points are snapped to the road network and the distance between input points and snapped locations

This update of r5r also allows users to account for terrain elevation when running analysis for walking and cycling trips

Finally, r5r now includes a progress counter for routing and accessibility functions. You only need to set verbose=F and you’ll be able to follow the progress of the computation.

Creating a rainbow map of Brazil for Pride month in R

Tue, 29 Jun 2021 00:00:00 +0000

To celebrate Pride month with my friends, I’d like to share this piece of code in R to create a rainbow map of Brazil using data from the geobr package.

library(geobr)
library(ggplot2)
library(sf)
library(sfheaders)

df <- read_municipality(year=2018)
centrd <- st_centroid(df)

centrd <- sfheaders::sf_to_df(centrd, fill=T)
df$position <- centrd$y + centrd$x

ggplot() +
  geom_sf(data=df, aes(fill=position), color=NA) +
  scale_fill_gradientn(colours = rainbow(10)) +
  theme_void() +
  theme(legend.position = "none",
        plot.title = element_markdown(color='#fe6700', size=7, hjust = 0.5),
        plot.caption = element_markdown(color='gray70', size=4)) +
  labs(title= '#PrideMonth', 
       caption='map by @UrbanDemog #geobr')
  
ggsave('pridemonth.png', dpi=300, width = 5, height=5, units='cm')

11th Anniversary of Urban Demographics!

Sun, 27 Jun 2021 00:00:00 +0000

Quick post to celebrate the 11th anniversary of the Urban Demographics blog!

Not sure people still read blogs. Most of the info I get nowadays comes from Twitter, academic journals and LinkedIn (surprisingly). Over the past couple of years, I’ve also been more focused on ~~trying to survive working on too many projects at the same time~~ doing research. This means less procrastination time, and hence fewer posts on the blog as some of you will have noticed.

This is a bit reflected on the stats of Urban Demographics over the past year, so here you go:

45 posts, an average of ~0.8 posts per week
22,290 visits, an average of ~72 visits per day
13,578 followers on Twitter
3,516 likes/followers on Facebook
new Instagram account with 298 followers.
new LinkedIn account with 917 followers. Is this a thing?
702 RSS feed subscribers

The 5 most popular posts:

and 5 of my favorite posts:

Where do readers come from? (136 countries | 1543 Cities)

Brazil (30.0%)
United States (24.2%)
United Kingdom (05.0%)
France (04.0%)
Canada (04.0%)
India (02.0%)

Here’s simple density map showing where our visitors come from. And here is the code to create a similar map using Google analytics data.

Density of Sessions, data from Google Analytics

Follow Urban Demographics on Instagram

Thu, 24 Jun 2021 00:00:00 +0000

A few people ~~mostly Millennials and zoomers~~ often ask me why I’m not on Instagram, so I’ve decided to give it a try. You can now also find me @urbandemographics on instagram sharing ~~dog pictures~~ posts and publications. I still don’t know “how to Instagram”, though. Bear with me.

These are all the online platforms where you can find me ~~procrastinating~~.

500 thousand deaths in Brazil due to COVID-19

Sat, 19 Jun 2021 00:00:00 +0000

This weekend we reach the depressing mark of 500 thousand deaths in Brazil due to COVID-19. The current government is much to blame for this. Fora bolsonaro.

Transportation Equity

Mon, 14 Jun 2021 00:00:00 +0000

I’m very glad to share that Alex Karner and I have written the entry on Transportation equity for the International Encyclopedia of Transportation. The encyclopedia itself is a massive effort with almost 600 entries! I am thankful to Maria Börjesson and Roger Vickerman for their invitation to contribute to this incredible collective work.

In our chapter (ungated PDF), Karner and I define what transportation equity is and why it matters. We also present a literature review on how equity relates to broader concerns over transport and mobility justice and inclusive urban planning. We cover how a full understating of transportation equity needs to account for both egalitarian and sufficientarian moral principles, and how these complementary perspectives are fueled by concerns with structural inequalities and social exclusion of marginalized groups. Finally, we share some reflections on how transportation equity research is particularly important to reflect on the societal changes that might emerge with the technological innovations around automated vehicles, ride hailing, microtransit and on-demand transit services .

You can download the ungated PDF from here.

Pereira, R. H. M., & Karner, Alex. (2021). Transportation equity. In R. Vickerman, International Encyclopedia of Transportation (1st Edition, Vol. 1, p. 271–277). Elsevier. DOI: https://doi.org/10.1016/B978-0-08-102671-7.10053-3

Abstract:

Transportation equity is a way to frame distributive justice concerns in relation to how social, economic, and government institutions shape the distribution of transportation benefits and burdens in society. It focuses on the evaluative standards used to judge the differential impacts of policies and plans, asking who benefits from and is burdened by them and to what extent. Questions of transportation equity involve both sufficientarian and egalitarian concerns with both absolute levels of wellbeing, transport-related poverty and social exclusion as well as with relative levels of transport-related inequalities. Ultimately, the study of transport equity explores the multiple channels through which transport and land use policies can create conditions for more inclusive cities and transport systems that allow different people to flourish, to satisfy their basic needs and lead a meaningful life. Transportation equity issues broadly encompass how policy decisions shape societal levels of environmental externalities and what groups are more or less exposed to them, as well as how those decisions affect the lives of different groups in terms of their ability to access life-enhancing opportunities such as employment, healthcare, and education. Equity is a crucial part of a broader concern with transport and mobility justice. The call for transport justice goes beyond distributive concerns, and yet justice cannot be achieved without equity.

Interactive elevation maps

Sun, 13 Jun 2021 00:00:00 +0000

Quick reminder about this fantastic website to create interactively ridgeline elevation maps of any place on Earth. The website was created by Andrei Kashcha, who has created a number of other hypnotizing interactive apps ~~that will drag you into a rabbit hole and consume your entire day before you remember you had that project deadline~~.

World

South America

The coast of Brazil

Rio de Janeiro

Some food for thought on immigration

Sun, 06 Jun 2021 00:00:00 +0000

Some food for thought on immigration, by Louis C.K.

World Bicycle Day

Thu, 03 Jun 2021 00:00:00 +0000

Amartya Sen wishes you a happy World Bicycle Day

Global comparative study of transport accessibility

Mon, 31 May 2021 00:00:00 +0000

Glad to share our new paper comparing accessibility to jobs by transport mode across 117 cities from 16 countries and 6 continents. This is the largest global comparative study of transport accessibility conditions conducted so far. The study was led by Hao Wu in collaboration with David Levinson and many others. It is the combined result of huge efforts from different groups, including the Access to Opportunities Project in Brazil. The paper is open access, and you can download it in the link below:

Wu, H. et al. Urban access across the globe: an international comparison of different transport modes. npj Urban Sustainability 1, 16 (2021). https://doi.org/10.1038/s42949-021-00020-2

Abstract:

Access (the ease of reaching valued destinations) is underpinned by land use and transport infrastructure. The importance of access in transport, sustainability, and urban economics is increasingly recognized. In particular, access provides a universal unit of measurement to examine cities for the efficiency of transport and land-use systems. This paper examines the relationship between population-weighted access and metropolitan population in global metropolitan areas (cities) using 30-min cumulative access to jobs for 4 different modes of transport; 117 cities from 16 countries and 6 continents are included. Sprawling development with the intensive road network in American cities produces modest automobile access relative to their sizes, but American cities lag behind globally in transit and walking access; Australian and Canadian cities have lower automobile access, but better transit access than American cities; combining compact development with an intensive network produces the highest access in Chinese and European cities for their sizes. Hence density and mobility co-produce better access. This paper finds access to jobs increases with populations sublinearly, so doubling the metropolitan population results in less than double access to jobs. The relationship between population and access characterizes regions, countries, and cities, and significant similarities exist between cities from the same country.

Figure: Number of jobs accessible by public transport within 30 min. and population size Transit access to jobs and metropolitan population

Estimating public transport emissions from GTFS

Thu, 27 May 2021 00:00:00 +0000

Apologies for the short notice. Today May 27 I’ll be presenting our new project developing a bottom-up model to estimate public transport emissions based on GTFS data. The method will soon be available as an R package and it’s easily replicable to any city with GTFS plus some general info on fleet characteristics. This is a project Pedro Andrade (INPE), João Bazzo and I have been working for almost 2 years now. I’m glad it will finally see daylight and I plan to post more updates about the project soon.

The webinar will be in English, and it will be today at 8pm UTC on YouTube. Click here for more info about the event. The webinar is organized by Sergio Ibarra Espinosa at USP: https://lnkd.in/dwmQdGp

A transport network model of the Roman Empire

Tue, 25 May 2021 00:00:00 +0000

A long time ago in 2014, I posted about Orbis: The Stanford Geospatial Network Model of the Roman Empire. Orbis is a free online tool that allows one to estimate transportation costs in the Roman Empire. The tool indicates what would have been the likely route and duration of a journey given the selected season, transport mode, etc. It allows one to interactively explore transport routes, path trees, isochrones, cartograms etc. Orbis was developed by Walter Scheidel and Elijah Meeks.

I had forgotten about this project but a tweet from Jessica Dalton (ht Renato Colisete) reminded me how amazing this tool is.

There are a few short videos showing how to export data and to navigate the interactive maps of the project.

Comparing calculated and self-reported accounts of accessibility

Wed, 19 May 2021 00:00:00 +0000

Glad to share a new paper with Jean Ryan (Lund University and K2) on what we might be missing when we measure accessibility in different ways. In this paper we examine the ways in which “objective” and “subjective” accounts of access differ and can complement one another. The study includes a comparison of (1) ‘objective’ indicators of accessibility to key activities by various modes of transport; and (2) individuals’ own perceptions of their capability to access valuable out-of-home activities and the modal options available to them.

We show that conventional transport models, by overlooking people’s perceptions of their accessibility, overestimate accessibility levels & underestimate inequalities. We also show how perceived accounts of access can be incorporated to improve conventional accessibility models. The paper is also good illustration of how one can use the Capability Approach in transport studies to operationalize the notion of accessibility as capability. The findings of the paper bring valuable insights for transport accessibility analysis, but it also has some important implications for researchers/policy makers concerned with social exclusion, transportation equity and mobility justice more broadly.

The paper is open access, and you can download it in the link below:

Ryan, J., & Pereira, R. H. M. (2021). What are we missing when we measure accessibility? Comparing calculated and self-reported accounts among older people. Journal of Transport Geography, 93, 103086. https://doi.org/10.1016/j.jtrangeo.2021.103086 [open access]

Figure: Comparing calculated and self-reported accounts of accessibility.

Assorted Links

Mon, 17 May 2021 00:00:00 +0000

paper: ‘Impacts of transportation network companies on urban mobility’. The authors find that ride-hailing companies increase congestion, have little impact on car ownership rates and discourage people from using public transportation and walking.
OpenCellID, the world’s largest open database with the GPS positions of aprox. 40 million cell towers. Interactive map. HT Daniel Herszenhut.

Choosing my outfit for the next conference in a post-pandemic future
Self-recommending: ‘Ten simple rules for structuring papers’
The cost of COVID-19 vaccines via Max Roser
A 3D LiDAR scan of Brazil’s largest favela to explore the morphology of informal settlements. Project webpage. Paper pre-print. Nice summary of the project, which is led by the MIT Senseable City Lab in collaboration with Rio’s City Planning Commissioner Washington Fajardo.

Low-income and black communities are more vulnerable to COVID-19 in Brazil

Thu, 29 Apr 2021 00:00:00 +0000

I’m glad to share our new paper looking income and racial inequalities in vulnerability to COVID-19 in São Paulo, Brazil. In this study we combine epidemiological, serological, household survey, NPI & mobility data to 1) Estimate the social and racial inequalities in the risk of hospitalization and death by COVID-19, and 2) Show how vulnerability to COVID is shaped by preexisting social and health inequalities.

Li SL, Pereira RHM, Prete Jr CA, et al. (2021) Higher Risk of Death from COVID-19 in Low-Income and Non-White Populations of São Paulo, Brazil. BMJ Global Health. https://doi.org/10.1136/bmjgh-2021-004959 [open access]

The study finds that low-income and black communities have:

↘️ access to quality healthcare
↘️ ability to self-isolate
↗️ comorbidities
↗️ probability of being hospitalized and dying from SARI

One of the key takeaways from the study is that disadvantaged groups (alongside healthcare workers) should be prioritized in COVID19 vaccination campaigns. This would be optimal both to prevent and slow down community transmission and to help reduce health inequities. See this Twitter thread for a more elaborate summary and discussion of the paper results. Please note that some figures and numbers have been updated from preprint tp publication.

Acknowledgements: The study was led by Sabrina Li, Carlos Prete and I, with the collaboration of a fantastic team from the Cadde Project, the Institute for Applied Economic Research (Ipea) and from the universities of Oxford, USP and Imperial College.

Figure Differential risk based on varying ability to self-isolate in São Paulo Metro area. (A) Relative risk of severe acute respiratory infection (SARI) hospitalisation. (B) Seven-day moving average of daily isolation levels by race. (C) Seven-day moving average of daily isolation levels by income. (D) Difference in daily social isolation by race after the introduction of non-pharmaceutical intervention (NPI). (E) Difference in daily social isolation by income after the introduction of NPIs.

End the rainbow! Oh, wait.

Wed, 21 Apr 2021 00:00:00 +0000

It’s been a while now since there is a growing understanding that rainbow color palettes can negatively impact how accurately we perceive and interpret data visualizations. Rainbow maps have been linked to lower accuracy and are not readable by color-blind readers. This gave rise to the #EndTheRainbow movement among many scientists across various fields.

Some people might not know this, but a team at Google has come to rescue the rainbow lovers. A couple of years ago, they created Turbo, An Improved Rainbow Colormap for Visualization. It seems turbo is distinguishable and smooth for all types of color blindness conditions, except Achromatopsia (total color blindness).

The turbo color palette is now available in R with the viridisLite package. Here is a quick demonstration:

library(viridisLite)
library(ggplot2)
library(patchwork)
library(dplyr)
library(reshape2)

plot_list <- lapply(X=c('viridis', 'inferno', 'cividis', 'turbo'), 
                   FUN=function(i){
                                   volcano %>% 
                                     melt() %>%
                                     ggplot() +
                                     geom_tile(aes(x = Var1, y = Var2, fill = value)) +
                                     scale_fill_viridis_c(option = i) +
                                     labs(subtitle = paste0(i), fill='') +
                                     #coord_fixed() +
                                     theme_void()
                                     }
                                  )
# patch plots together
plot_list[[1]] +
plot_list[[2]] +
plot_list[[3]] +
plot_list[[4]]

ggsave(filename='turbo_test.png', dpi=200, 
       width = 16, heigh=10, units='cm')

Before we get too excited, even the turbo rainbow palette exaggerates the colors a bit. See below. If your data visualization involves really smooth and detailed transitions, you should probably stick to the #EndTheRainbow movement.

library(imager)

my_photo <- load.image('https://www.urbandemographics.org/img/post_images/urban_picture_rio_lucasmagno.jpg')

# color palettes
p_viridis <- scales::gradient_n_pal(viridisLite::viridis(n=1000))
p_inferno <- scales::gradient_n_pal(viridisLite::inferno(n=1000))
p_turbo <- scales::gradient_n_pal(viridisLite::turbo(n=1000))

# plot
png("turbo_test_rio.png", units="cm", width=14, height=17, res=200)
# insert ggplot code

par(mfrow=c(2,2), mai = c(.05, 0.05, .05, 0.05))

plot(my_photo, axes=FALSE)
grayscale(my_photo) %>% plot(colourscale=p_viridis,rescale=FALSE, axes=FALSE)
grayscale(my_photo) %>% plot(colourscale=p_inferno,rescale=FALSE, axes=FALSE)
grayscale(my_photo) %>% plot(colourscale=p_turbo,rescale=FALSE, axes=FALSE)

dev.off()

Open positions for research assistant

Wed, 14 Apr 2021 00:00:00 +0000

We have 3 open positions for Research Assistant to work with me in the Access to Opportunities Project, at the Institute for Applied Economic Research (Ipea). The work involves a number of activities, including evaluating the accessibility impacts of transportation projects, and conducting research on inequalities in access to opportunities.

More info here. Please feel free to reach out if you have any questions.

obs. The position is based in Brasília/Brazil, though we will be working remotely during the COVID-19 pandemic.

The effect of Uber on road safety

Mon, 05 Apr 2021 00:00:00 +0000

New paper looking at the impact of Uber on road safety. Given the variation in the timing of when Uber entered in different municipalities, the authors use an event study model (and several robustness checks) to determine the causal impact of Uber on road safety. The authors find that the introduction of Uber reduced traffic fatality by aprox. 10% and traffic-related hospitalization rates by aprox. 17% in Brazilian cities.

Barreto, Y., Neto, R. D. M. S., & Carazza, L. (2021). Uber and Traffic Safety: Evidence from Brazilian Cities. Journal of Urban Economics, Volume 123, May 2021, 103347.

Abstract

In this paper we estimate the causal effect of the introduction of Uber, an app designed to provide urban transportation services, on traffic-related fatalities and hospitalizations in Brazilian cities. Exploiting the staggered rollout of the company across cities, our research applies a difference-in-differences design and an event study specification using quarterly panel data from 2011 to 2016. Our results indicate that Uber’s introduction reduced by roughly 10% and 17%, respectively, traffic fatality and hospitalization rates in Brazilian cities. The results are robust to different specifications and falsification tests, and to the best of our knowledge, are the first estimates of this effect in a developing country.

Call for papers: Advances in Spatial and Transport Network Analysis

Mon, 22 Mar 2021 00:00:00 +0000

We have a new call for papers for a special issue on “Advances in Spatial and Transport Network Analysis”, to be published in Environment and Planning B: Urban Analytics and City Science. This special issue is edited by Henrikki Tenkanen (Aalto University), Elsa Arcaute (UCL), Marta Gonzalez (Berkeley) and myself.

This cfp comes as a great opportunity to create a dialogue between network and social scientists, transport geographers, engineers etc working on transport and mobility networks. This dialogue raises new challenges, though, as discussed in this thoughtful recent paper by Tim Schwanen.

Here is a short snippet of the cfp:

This special issue is dedicated to papers focusing on recent advances in the development of new measures and methodologies to evaluate and analyze the performance of transportation networks. These measures might include, but are not limited to, environmental costs or exposures (e.g., CO2, noise, pollution); monetary costs (the price of access), complexity; and resilience of multimodal transportation networks or focus on qualitative aspects of travel, where travel might be seen as a “gain” instead of cost (such as exposure to aesthetic or green environments). Methodologically, we welcome papers using novel ways to measure transport network connectivity, performance and accessibility, including recent advances in machine learning and artificial intelligence. Special attention will be given to papers studying transport related questions with interdisciplinary approaches, bridging fields such as network science, transport geography, urban and regional planning, economics, history and environmental studies

ps. Some of you will remember we initially started organizing this special issue to be published in the International Journal of Geo-Information, edited by MDPI. Along the way, we noticed how the way MDPI was handling the review process and the growing reputation of MDPI as predatory publisher could harm the quality of the special issue. This has led us to withdraw move our special issue to a another journal, and we are now very glad to reopen our cfp on Environment and Planning B.

Chart of the day

Wed, 17 Mar 2021 00:00:00 +0000

Just came across this chart on Twitter, and it reminded of of this quote

“It would be nice if all of the data which sociologists require could be enumerated because then we could run them through IBM machines and draw charts as the economists do. However, not everything that can be counted counts, and not everything that counts can be counted.” (William Bruce Cameron)

This is on page 13 of his 1963 book “Informal Sociology: A Casual Introduction to Sociological Thinking”.

Calculating isochrones and accessibility with r5r

Sun, 07 Mar 2021 00:00:00 +0000

Some of you might remember that few months ago we released r5r, an R package for rapid realistic routing on multimodal transport networks. I’m glad to share we have a new paper paper introducing and demonstrating the r5r package, published open-access on the Findings journal.

Pereira, R. H. M., Saraiva, M., Herszenhut, D., Braga, C. K. V., & Conway, M. W. (2021). r5r: Rapid Realistic Routing on Multimodal Transport Networks with R⁵ in R. Findings, 21262. https://doi.org/10.32866/001c.21262

The paper talks about the routing algorithms behind R5, and demonstrates how r5r can be easily used to estimate travel time matrices, isochrones and accessibility on multimodal transport networks 🚴‍♀️ 🚶‍♀️ 🚅 🚌 🚗 🚡 The paper includes an .Rmarkdown file with code to reproduce the study.

For a quick overview, you may also check the package vignettes:

Urban Picture

Thu, 04 Mar 2021 00:00:00 +0000

Beautiful photo of Tromsø in Norway, by David Jensen one. Tromsø is one of the best places in the world to watch the northern lights. It’s also the only place I know where it’s cheaper to buy salmon than cucumber.

credit: David Jensen

Spatial, income and racial inequalities in access to COVID-19 healthcare

Wed, 24 Feb 2021 00:00:00 +0000

I’m very glad to share our new paper analyzing spatial, income and racial inequalities in access to COVID-19 healthcare using the new balanced floating catchment area approach (BFCA). I have summarized the key findings and contributions of the study in the Twitter thread, but please feel free to read the full paper. Manuscript PDF and code available here.

Pereira, R. H., Braga, C. K. V., Servo, L. M., Serra, B., Amaral, P., Gouveia, N., & Paez, A. (2021). Geographic access to COVID-19 healthcare in Brazil using a balanced float catchment area approach. Social Science & Medicine, v273, 113773.

Abstract:

The rapid spread of COVID-19 across the world has raised concerns about the responsiveness of cities and healthcare systems during pandemics. Recent studies try to model how the number of COVID-19 infections will likely grow and impact the demand for hospitalization services at national and regional levels. However, less attention has been paid to the geographic access to COVID-19 healthcare services and to hospitals’ response capacity at the local level, particularly in urban areas in the Global South. This paper shows how transport accessibility analysis can provide actionable information to help improve healthcare coverage and responsiveness. It analyzes accessibility to COVID-19 healthcare at high spatial resolution in the 20 largest cities of Brazil. Using network-distance metrics, we estimate the vulnerable population living in areas with poor access to healthcare facilities that could either screen or hospitalize COVID-19 patients. We then use a new balanced floating catchment area (BFCA) indicator to estimate spatial, income, and racial inequalities in access to hospitals with intensive care unit (ICU) beds and mechanical ventilators while taking into account congestion effects. Based on this analysis, we identify substantial social and spatial inequalities in access to health services during the pandemic. The availability of ICU equipment varies considerably between cities, and it is substantially lower among black and poor communities. The study maps territorial inequalities in healthcare access and reflects on different policy lessons that can be learned for other countries based on the Brazilian case.

As a quick teaser, the image below depicts the city of Rio de Janeiro. It shows A1) The number of ICU beds per person in the catchment area of each hospital, A2) BFCA accessibility level with competition, A3) the combined spatial distribution of population ∩ accessibility. The bivariate choropleth map (A3) helps draw attention to places that deserve more (or less) attention:

bright pink = areas with large population underserved by health services
bright cyan = areas with small population with high access to health services
dark purple = areas with small population with the lowest access to health

The political economy of car dependence

Sat, 20 Feb 2021 00:00:00 +0000

image credit: Michael Kerbow

Last year, Giulio Mattioli (TU Dortmund University) and colleagues published a great literature review paper on the political economy of car dependence. The study has over 320 references (!) and yet the authors do a great job in 1) systematizing the reasons why it is so hard to reduce car ownership and use, and 2) casting a critical eye over sustainable transport research and discourse.

Mattioli, G., Roberts, C., Steinberger, J. K., & Brown, A. (2020). The political economy of car dependence: A systems of provision approach. Energy Research & Social Science, v66, 101486.

I believe this paper will soon be recognized as a classic on the topic, so you should probably read the whole study (it’s open access). If you don’t have time, you may read this funny Twitter thread with some key points of the study, or this one-slide summary.

Now, if if the reason why you don’t have time to read any of the above is because you ~~procrastinate~~ are on Twitter, you should probably follow the authors there: Giulio Mattioli, Cameron Roberts and Julia Steinberger.

btw, Giulio is one of my favorite people on Twitter, and you only need to read a few of his papers to realize he’s one of the best scholars of our generation.

Music for the weekend

Sat, 13 Feb 2021 00:00:00 +0000

Soundtrack for a carnival season at home.

Wilson Simonal:

Spatial Analytics + Data webinar series

Wed, 03 Feb 2021 00:00:00 +0000

Rachel Franklin ( Twitter) and Levi Wolf ( Twitter) have been organizing Spatial Analytics + Data, an incredible webinar series hosted at The Alan Turing Institute and the Newcastle and Bristol Universities.

I’ll be presenting next week (Feb 9th at 4pm UTC / GMT) about ‘Transport and Inequalities in Access to Opportunities’. I’ll be talking about some of the recent developments of the Access to Opportunities Project, both in terms of empirical applications analyzing access to COVID-19 healthcare, methodological innovations, and computational tools we’re developing at the project.

In case you cannot attend the webinar, it will be recorded and made available here.

UPDATE [10/Feb/2021]. Here is the recording of my presentation.

This is the program with the next speakers ( registration here):

And in case you missed the first part of the #SAD2020 webinar series, all presentations with an amazing lineup of speakers are available on their YouTube channel.

The joys of academic writing

Mon, 01 Feb 2021 00:00:00 +0000

Oh, the joys of academic writing! Thanks Ahmed El-Geneidy for sharing this on Twitter.

Assorted Links

Wed, 27 Jan 2021 00:00:00 +0000

An interactive map to visualize over 100 years of neighborhood changes in the UK, via Ahmad Barclay
The Transit Costs Project, a growing and incredible a global database of public transport construction costs. The project is led by Eric Goldwyn, Alon Levy and Elif Ensari at the NYU Marron Institute
paper: Life expectancy and mortality in 363 cities of Latin America, by Usama Bilal and the SALURBAL project team
In case you’re looking for a research question for a Masters or PhD research project, you might appreciate this new report. 100 Social Sciences and Humanities priority research questions for transport and mobility in Horizon Europe. The report was written by an incredible team with some of the most prominent scholars in the field
The Online Spatial & Urban Seminar has announced the new program with the next speakers up to April 2021
Coordinate precision, by XKCD

Thanks future me

Thu, 14 Jan 2021 00:00:00 +0000

As a researcher whose career is built upon a solid base of daily procrastination, I endorse this message. Thanks future me.

video by Griffin.

r5r: Rapid Realistic Routing with R5 in R

Thu, 07 Jan 2021 00:00:00 +0000

I’m glad to announce r5r, an R package for rapid realistic routing on multimodal transport networks (walk, bike, public transport and car). The package has been developed by my team and I as part of the Access to Opportunities Project because ~~it’s fun~~ we wanted to make transport modeling and accessibility analysis easier and faster.

r5r provides a simple and friendly interface to R⁵, a powerful routing engine developed by Conveyal. The r5r package is a simple way to run R⁵ locally, what allows users to generate detailed routing analysis or calculate travel time matrices using seamless parallel computing.

The package has very low data requirements, which means you can use it to analyze pretty much any city, region or country. It only requires data on street networks from OpenStreetMap. It does allow for multimodal transport routing if public transport data is provided in the standard GTFS format. The package is also easily scalable allowing for fast computation at either the city or country level.

I’ll give a quick overview of the package in this post. For a detailed demonstration of r5r, see in this intro Vignette.

Installation

To use r5r, you need to have Java SE Development Kit version >= 11.0.8 installed on your computer. No worries, you don’t have to pay for it.

# install package from CRAN
install.packages("r5r")

# allocate some RAM memory to Java, 2GB in this case
options(java.parameters = "-Xmx2G")

# load package
library(r5r)

Demonstration on sample data

This quick demo shows how you can use r5r in three simple steps. The package includes a small sample data set of the public transport and Open Street Map networks of São Paulo (Brazil), which I’ll be using here.

In this example, we will be calculating the travel time by public transport between every possible pair of origins and destinations in the sample data.

1) Build a routable transport network

You only need to point to the local directory with your data. The function will automatically download and store locally an R5 Jar file (Jar file is downloaded only once), and builds a multimodal transport network given an OSM street network in .pbf format (mandatory) and one or more public transport networks in GTFS.zip format (optional).

path <- system.file("extdata/spo", package = "r5r")
r5r_core <- setup_r5(data_path = path, verbose = FALSE)

2) Set trip parameters

# Load origin/destination points
points <- read.csv(system.file("extdata/spo/spo_hexgrid.csv", package = "r5r"))

# Set transport mode
mode <- c("transit")

# Set trip departure time
departure_datetime <- as.POSIXct("13-05-2019 10:00:00", format = "%d-%m-%Y %H:%M:%S")
time_window <- 60 # minutes
percentiles <- 50  # 50th percentile

We all know travel times might vary at different times of the day. To account for this variation, what we do in this example is to compute several travel time matrices with multiple departure times within a 60-minute window and retrieve the median travel time within that window.

3) Compute travel time matrix

Now we only need to compute the travel times, which is blazing fast thanks to R⁵.

ttm <- travel_time_matrix(r5r_core = r5r_core,
                          origins = points,
                          destinations = points,
                          mode = mode,
                          departure_datetime = departure_datetime,
                          time_window = time_window,
                          percentiles = percentiles)

The output is a data.table that brings time estimates (in minutes), and it looks like this:

head(ttm)
>             fromId            toId travel_time
> 1: 89a8100c603ffff 89a8100c603ffff           0
> 2: 89a8100c603ffff 89a8100c617ffff          13
> 3: 89a8100c603ffff 89a8100c60fffff           6
> 4: 89a8100c603ffff 89a8100c607ffff          11
> 5: 89a8100c603ffff 89a8100c6abffff          20

In this example, we’ve calculated the median travel time within a one hour-window for over 104 thousand origin/destination pairs, what took 38 seconds in my laptop. When using a single departure time, it takes less than 4 seconds.

r5r Allows for other routing parameters and it also brings the detailed_itineraries() function, which returns detailed information on transport mode, travel time, walk distance, geometry etc for each trip section of multiple alternative routes between origin/destination pairs. More details can be found on the package webpage.

ps. The package has been published on CRAN for a few months and we already have over 1600 downloads, but only now I managed to have some time to share about the package here on my website.

New study on social and racial inequalities and COVID-19 in Brazil

Tue, 15 Dec 2020 00:00:00 +0000

We have a new study preprint analyzing the “Social and racial inequalities in COVID-19 risk of hospitalisation and death across Sao Paulo state, Brazil”.

In this paper, we combine epidemiological, serological, household survey, NPI & mobility data to 1) Estimate the social and racial inequalities in the risk of hospitalization and death by COVID-19, and 2) Show how vulnerability to COVID is shaped by preexisting social and health inequalities.

Our findings show that vulnerability to COVID-19 is strongly shaped by preexisting health inequalities. We find black and low-income communities (big overlap here) are disproportionately more likely to be hospitalized and to die from COVID. This is largely due to structural social and health inequities that lead these groups to have more comorbidities, and to live under socioeconomic conditions that limit their access to health services and their capacity to adhere to social distancing. One of the key takeaways from the study is that disadvantaged groups (alongside healthcare workers) should be prioritized in COVID-19 vaccination campaigns. This would be optimal both to prevent and slow down community transmission AND to help reduce health inequities.

You may find a more detailed summary the main findings and figures of the paper in these Twitter threads in English and in Portuguese. For more info on the context, data, methods, findings and discussion, check out the preprint itself.

New study preprint analyzing the "Social and racial inequalities in #COVID19 risk of hospitalisation and death across Sao Paulo state, Brazil" https://t.co/V7HDPZ9xvC Another study by the @CaddeProject team, this one co-led by @sabrinalyli @CarlosPrete1 and I.
🧶with main results pic.twitter.com/hfFIgW9JCs
— Rafael H.M. Pereira 🚡 Urban Demographics (@UrbanDemog) December 14, 2020

Presentation at the Geographic Data Science Lab

Fri, 11 Dec 2020 00:00:00 +0000

The team at the Geographic Data Science Lab at the University of Liverpool kindly invited me to present at their Brown Bag Seminar. The seminar was recorded and you can watch it in the video below. I talked about the Access to Opportunities Project and about some of the data and computational packages we’re developing along the way and making publicly available.

By the way, you should probably follow the Geographic Data Science Lab on Twitter. The other seminars they have organized are available here.

Comparing objective and subjective accounts of accessibility

Tue, 08 Dec 2020 00:00:00 +0000

Jean Ryan and I have a new preprint “What are we missing when we measure accessibility? Comparing calculated and self-reported accounts”. The manuscript is currently under review, but one can already read/cite it from SocArXiv.

TL;DR: In this study we examine the ways in which “objective” and “subjective” accounts of access differ and can complement one another. The study includes a comparison of (1) so called ‘objective’ indicators of accessibility to key activities by various modes of transport; and (2) individuals’ own perceptions of their capability to access valuable out-of-home activities and the modal options available to them. We show that conventional transport models, by overlooking the heterogeneity in people’s perceptions of their mobility conditions, overestimate accessibility levels & underestimate accessibility inequalities. We also show how perceived accounts of access can be incorporated into conventional models to improve accessibility analysis.

Ryan, J., & Pereira, R. H. M. (2020). What are we missing when we measure accessibility? Comparing calculated and self-reported accounts. SocArXiv. https://doi.org/10.31235/osf.io/ja34d

Abstract:

Accessibility is increasingly recognised as a key purpose of transport policies. Most of the common practices found both in academic studies and policy planning draw on relatively simple accessibility measures taken as ‘objective’ indicators that only focus on the interaction between land use and transport. Relatively little attention has been paid to heterogeneity in individual characteristics and in self-reported perceptions of accessibility (‘subjective’ indicators), and the corresponding differences with respect to available modal options. This study includes a comparison of (1) ‘objective’ indicators of accessibility to key activities by various modes of transport; and (2) individuals’ own perceptions of their capability to access valuable out-of-home activities and the modal options available to them. This study examines the key differences between the two representations of accessibility. The calculated measure was developed using door-to-door travel times to supermarkets and healthcare centres using OpenTripPlanner. The self-reported measure was based on a dedicated capability-oriented travel survey of people aged 65-79 in Sweden’s large metropolitan regions: Stockholm, Gothenburg and Malmö. The data were analysed using descriptive statistics and binary and multinomial logistic regressions. The results of this study allow us to gain a greater insight into the ways in which the two accounts differ and can complement one another. We find that conventional transport methods, by overlooking the heterogeneity in people’s perceptions of their accessibility, tend to overestimate accessibility levels and underestimate accessibility inequalities. This study shows how perceived accounts of accessibility can be incorporated into conventional accessibility models and improve accessibility analyses.

Transport Access Manual

Mon, 07 Dec 2020 00:00:00 +0000

Just a few days ago we published the Transport Access Manual: A Guide for Measuring Connection between People and Places. It presents loads of information on concepts, methods and tools to help planning professionals rethink cities and mobility systems.

This is a remarkable and open access publication on transport accessibility, led by David Levison and David King with contributions from 37 coauthors.

Open position at TSU Oxford

Fri, 04 Dec 2020 00:00:00 +0000

The Transport Studies Unit (TSU) and the School of Geography and the Environment at Oxford University are recruiting for a permanent post in transport geographies and mobilities. This is a wonderful opportunity to work alongside a brilliant team at one of the best universities in the world.

The history of humankind is the history of urbanization

Wed, 11 Nov 2020 00:00:00 +0000

The history of humankind is the history of urbanization. Few images speak to this phrase so well as these maps showing how the spatial distributional of population densities in France changed between the late 1800s and nowadays.

These maps were created by Magali Talandier, and they are part of her research on the long-term changes in the regional development of France. This paper might be of particular interest.

New study: situating Transportation Equity

Mon, 19 Oct 2020 00:00:00 +0000

Alex Karner and I have new study defining what transportation equity is, and giving a literature review on how this topic relates to broader concerns over transport and mobility justice and inclusive urban planning. The study is as an entry for the International Encyclopedia of Transportation, edited by Roger Vickerman, forthcoming in 2021. For now, we’re sharing the preprint of the study, which can download from here.

The manuscript has been accepted/ready for months but only now I got some time to share it. Please help us spread the word.

Pereira, R. H. M. & Karner, A. (forthcoming 2021) Transportation Equity. In R. Vickerman (Ed.), International Encyclopedia of Transportation (2021). 1st Edition, Elsevier.

Abstract:

Transportation equity is a way to frame distributive justice concerns in relation to how social, economic, and government institutions shape the distribution of transportation benefits and burdens in society. It focuses on the evaluative standards used to judge the differential impacts of policies and plans, asking who benefits from and is burdened by them and to what extent. Questions of transportation equity involve both sufficientarian and egalitarian concerns with both absolute levels of wellbeing, transport-related poverty and social exclusion as well as with relative levels of transport-related inequalities. Ultimately, the study of transport equity explores the multiple channels through which transport and land use policies can create conditions for more inclusive cities and transport systems that allow different people to flourish, to satisfy their basic needs and lead a meaningful life. Transportation equity issues broadly encompass how policy decisions shape societal levels of environmental externalities and what groups are more or less exposed to them, as well as how those decisions affect the lives of different groups in terms of their ability to access life-enhancing opportunities such as employment, healthcare and education. Equity is a crucial part of a broader concern with transport and mobility justice. The call for transport justice goes beyond distributive concerns, and yet justice cannot be achieved without equity.

Roads colored by orientation

Sun, 18 Oct 2020 00:00:00 +0000

The figure below shows the roads in Brasília colored by orientation. Beautiful rainbow patterns but ~~a horrible~~ not exactly a great experience for pedestrians though. These maps were created using a really neat interactive tool developed by Andrei Kashcha & @mitsuchi using data from OpenStreetMap. You can try it on other cities here or check the source code. Hat tip @mapmakerdavid and @stevevance.

If you’re interested on the topic of street network configuration and urban morphology, you should have a look at Geoff Boeing‘s work, in case you haven’t already. In this paper, in particular, Geoff analyzes street network orientation and entropy in 100 cities around the world.

Detail:

Quote of the day: data

Thu, 15 Oct 2020 00:00:00 +0000

Hot take - Data is not the new oil. Data is the new glitter:

- Lures humans in with its shininess
- Very easy to accumulate
- Found in places you least likely expect to find it
- Almost impossible to get rid of
- Everyone insists on using it w/o thinking through the consequences
— Becky Yoose (@yo_bj) October 14, 2020

Social and natural lottery of life

Sun, 20 Sep 2020 00:00:00 +0000

Science communication is one of the greatest things on the internet. While a lot of SciCom videos focus on hard science stuff such as black wholes, gravitational waves or gene editing, this video below is probably the most impactful video you’ll watch. This one, by Derek Muller over Veritasium, is about how our life outcomes are shaped by our own choices and efforts but also fundamentally by luck, what John Rawls call the social and natural lottery.

The video covers a mix of things from social psychology, collective behavior, social inequalities and moral philosophy, and how these things impact how you see yourself and others in the world. It should help you ~~not to become a jerk~~ to be kind to yourself and to others.

The glacial speed of academic journals

Fri, 18 Sep 2020 00:00:00 +0000

Paper submitted / Paper accepted

Academic achievement unlocked

Mon, 07 Sep 2020 00:00:00 +0000

A couple months ago I unlocked this academic achievement you see in the picture below. BTW, Mendeley’s response is brilliant. Just two months later, I received today an email inviting me to review my own paper submitted a few weeks ago. At this pace they will get me retired soon lol.

Next level (professor level): reviewer number 2 suggesting me to read my own work. I’m afraid this one might take ages to happen though.

Global population density spikes

Sat, 22 Aug 2020 00:00:00 +0000

It’s been a while since Alasdair Rae started sharing some of his amazing work rendering 3D images of population densities. Some of his most beautiful maps came out in this piece by Visual Capitalist (ht Todd Gardner).

Alasdair Rae has been kind enough to share a few instructions on how to create maps like these.

Socioeconomic bias in testing and diagnostics of COVID-19

Sat, 01 Aug 2020 00:00:00 +0000

We have a new paper looking at the demographic, epidemiological and clinical characteristics of the first 3 months of COVID-19 in Brazil, a country with one of the fastest-growing COVID-19 epidemics in the world. The study brings some really interesting results that help us (1) understand how Brazil has ~~screwed things up~~ got to this point, and (2) put things in an international perspective.

There is one finding I would like to highlight, though. In this study we find an important socioeconomic bias in access to COVID-19 diagnosis. Looking at the metropolitan region of São Paulo, we find that reported cases of severe acute respiratory infection (SARI) are much more likely to be diagnosed with COVID-19 for patients who live in high-income census tracts. Patients living in low-income urban peripheries, on the other hand, are much more likely to be reported with SARI with unknown cause. This can be explained in part by: (1) the high proportion of early cases in returning travelers with higher income and better access to private laboratories for diagnostics; and (2) the more limited access to freely available diagnostic screening. This finding suggests that the true burden of the epidemic in lower-income neighborhoods an vulnerable population groups is probably underestimated.

This study is another great collective effort from a group of researchers from Brazil and the UK under the CADDE project. Thanks William de Souza, Nuno Faria and Ester Sabino for leading the team on this one!

de Souza, W. M., et al (2020). Epidemiological and clinical characteristics of the COVID-19 epidemic in Brazil. Nature Human Behaviour, 1–9. https://doi.org/10.1038/s41562-020-0928-4

Rafa Pereira

Abstract

The first case of COVID-19 was detected in Brazil on 25 February 2020. We report and contextualize epidemiological, demographic and clinical findings for COVID-19 cases during the first 3 months of the epidemic. By 31 May 2020, 514,200 COVID-19 cases, including 29,314 deaths, had been reported in 75.3% (4,196 of 5,570) of municipalities across all five administrative regions of Brazil. The R0 value for Brazil was estimated at 3.1 (95% Bayesian credible interval = 2.4–5.5), with a higher median but overlapping credible intervals compared with some other seriously affected countries. A positive association between higher per-capita income and COVID-19 diagnosis was identified. Furthermore, the severe acute respiratory infection cases with unknown aetiology were associated with lower per-capita income. Co-circulation of six respiratory viruses was detected but at very low levels. These findings provide a comprehensive description of the ongoing COVID-19 epidemic in Brazil and may help to guide subsequent measures to control virus transmission.

An animated 3D hologram of population density

Fri, 31 Jul 2020 00:00:00 +0000

Some of you might have read the title of this post and thought about all the recent advances in computational models and data visualizations we have seen over the past few years. Right?

Well, this is a 3D hologram created in 1978 (!) showing how the population density of the US changed between 1790-1970. It was created by Geoffrey Dutton at the Harvard Lab for Computer Graphics and Spatial Analysis. This is a gem found by James Cheshire, who wrote a really good post on Joy Division, Population Surfaces and Pioneering Electronic Cartography.

ps. My jaw also dropped when I first saw this gif below. What an incredible effort it must have been to create this back then, something many of us take for granted today.

Combining mobility, genomic and epidemiological data to look at the transmission of COVID-19

Thu, 23 Jul 2020 00:00:00 +0000

We have a new paper out combining mobility, genomic and epidemiological data to look at the transmission of COVID-19 in Brazil. I’m biased, but this is a groundbreaking study that was only possible due to the collaboration of great team with over 70 researchers from Brazil and the UK under the CADDE project. See the abstract and link to the ungated version of the paper below. Thanks Darlan Candido, Nuno Faria and Ester Sabino for leading the team.

Candido, D. S., et al (2020). Evolution and epidemic spread of SARS-CoV-2 in Brazil. Science. https://doi.org/10.1126/science.abd2161

Rafa Pereira

Abstract

Brazil currently has one of the fastest growing SARS-CoV-2 epidemics in the world. Owing to limited available data, assessments of the impact of non-pharmaceutical interventions (NPIs) on virus spread remain challenging. Using a mobility-driven transmission model, we show that NPIs reduced the reproduction number from >3 to 1–1.6 in São Paulo and Rio de Janeiro. Sequencing of 427 new genomes and analysis of a geographically representative genomic dataset identified >100 international virus introductions in Brazil. We estimate that most (76%) of the Brazilian strains fell in three clades that were introduced from Europe between 22 February11 March 2020. During the early epidemic phase, we found that SARS-CoV-2 spread mostly locally and within-state borders. After this period, despite sharp decreases in air travel, we estimated multiple exportations from large urban centers that coincided with a 25% increase in average travelled distances in national flights. This study sheds new light on the epidemic transmission and evolutionary trajectories of SARS-CoV-2 lineages in Brazil, and provide evidence that current interventions remain insufficient to keep virus transmission under control in the country.

Mars landscape color palette

Sat, 04 Jul 2020 00:00:00 +0000

I came across this gorgeous image of Mars landscape and I was compelled to create a color palette. This image was captured by HiRISE (NASA) and processed by Seán Doran. I created this color scale with support from Chroma.js Color Palette Helper.

library(ggplot2)
library(hexbin)
library(cowplot)
library(chroma)

# color palettes
chroma::show_col(interp_colors(25, colors=c("#3a5251",  "#d5b087"), interp="bezier"))
chroma::show_col(interp_colors(25, colors=c("#19323b", "#d5b087"), interp="bezier"))

# plot
green_ish <- ggplot(data.frame(x = rnorm(100000), y = rnorm(100000)), aes(x = x, y = y)) +
              geom_hex() + coord_fixed() + theme_void() +
              scale_fill_gradient(space = "Lab", 
                                  low = "#273e3d",
                                  high = "#d5b087")

blue_ish <- ggplot(data.frame(x = rnorm(100000), y = rnorm(100000)), aes(x = x, y = y)) +
              geom_hex() + coord_fixed() + theme_void() +
              scale_fill_gradient(space = "Lab", 
                                  low = "#19323b",
                                  high = "#d5b087") 

cowplot::plot_grid(green_ish, blue_ish)

Rafa Pereira

10th Anniversary of Urban Demographics!

Fri, 26 Jun 2020 00:00:00 +0000

It feels like I created the Urban Demographics blog a long time ago. My motivation was to create a personal repository of interesting and funny things I find on the web related to my research interests. Ten years, a marriage, a PhD, three cities and a dog later, my ~~procrastination~~ activity on the blog and on Twitter is still quite rewarding, more than I could expect really. It has helped me learn about the work of many incredible researchers from across the globe. Perhaps more importantly, the blog has helped me get to know some really nice people, a few of whom have become co-authors and friends. I like to think the blog has also helped others as well.

By the stats of the blog, I’m glad to see a few people still find it useful too. Here are just some quick stats that show a summary of the blog over the past year.

82 posts, an average of ~1.6 posts per week
24,113 (21,476 + 2,637) visits, an average of ~66 visits per day
11,810 followers on Twitter
3,475 likes/followers on Facebook
682 RSS feed subscribers

The 5 most popular posts:

and 5 of my favorite posts:

Where do readers come from? (156 countries | 2874 Cities)

United States (30.2%)
Brazil (21.1%)
United Kingdom (05.6%)
Germany (03.4%)
Canada (03.0%)
France (02.5%)

Here’s simple density map showing where our visitors come from:

Density of Sessions, data from Google Analytics

Soundtrack for the weekend

Sun, 14 Jun 2020 00:00:00 +0000

Pipelinefunk, by Armin Küpper.

2020 Lectures on Urban Economics

Thu, 11 Jun 2020 00:00:00 +0000

The 2020 Lectures on Urban Economics, organized by the Urban Economics Association, is entirely online and open to the public. The series has some amazing speakers, including Edward Glaeser (Harvard), Gilles Duranton (Wharton), Vernon Henderson (LSE), Dave Donaldson (MIT), Veronica Guerrieri (Chicago Booth) as well as Leah Boustan, Esteban Rossi-Hansberg and Stephen Redding from Princeton.

The series started today with a presentation by Edward Glaeser giving an introductory overview of a few core concepts and models of spatial equilibrium. In case you missed his presentation, you can watch it on YouTube. I might have to watch it again now that I can set the speed to 0.75The man talks quite fast.

You might also like to join the Online Spatial & Urban Seminar, co-organized by Clare Balboni, Fabian Eckert, Caitlin Gorback, and Chris Severen.

Assorted Links

Tue, 02 Jun 2020 00:00:00 +0000

Racial wealth inequality in the USA in charts.
In the USA, more than 1,000 people were killed by the police in 2019. In the state of Rio de Janeiro alone, the police killed 1,810 people last year, an average of five per day. In both cases, injustices disproportionately kill young black men.
The expansion and decline of slavery across the US between 1790 to 1870, by Bill Rankin.

credit: Bill Rankin

Nice piece celebrating the work of Du Bois and his data visualizations to portray black America in the late 1800 and early 1900s.
Interactive map of racial lynchings in the United States from 1835 to 1964. This is based on extensive data collected by Monroe and Florence work. via Keir Clarke.

credit: Monroe and Florence Work

Top favorited tweets

Thu, 28 May 2020 00:00:00 +0000

Here is a 2-minute ~~procrastination~~ exercise in . This a very short script that allows you to download the data on all previous 3K tweets you favorited and find out the tweeps you like the most. This post was inspired by Fernando Barbalho, who shared an original version of this code on his Github.

The first thing to do is to load a few libraries and download the tweets favorited by a Twitter account:

library(rtweet)
library(magrittr)
library(ggplot2)
library(dplyr)

# download the data
df_favorites <- rtweet::get_favorites("@UrbanDemog", n=3000)

Count the number of favorited tweets by user:

users_favorite <-
  df_favorites %>%
  group_by(screen_name, user_id) %>%
  summarise(quant_fav = n())

Now you can filter the top ‘n’ users using the super handy dplyr::top_n function and plot the results on chart like this:

# filter top 15 users
users_favorite %>%
  ungroup() %>%
  top_n(15,quant_fav) %>%
  mutate(screen_name = reorder(screen_name,quant_fav)) %>%

# plot
ggplot() +
    geom_col(aes(x = screen_name, y = quant_fav)) +
    theme_light() +
    theme(panel.grid.major.y = element_blank(),
          panel.grid.minor = element_blank(),
          panel.border = element_blank()) +
    coord_flip() +
    labs(title = "Top 15 Twitter accounts I like more frequently",
         y = "Number of tweets favorited by @UrbanDemog",
         x = "@ user names")

Webinar series: Researching Transport Inequalities in Global South Cities

Thu, 14 May 2020 00:00:00 +0000

If your calendar is not filled with webinars already, this post is for you. Karen Lucas (Leeds and Manchester), Daniel Oviedo (UCL) and I are putting together a webinar series on “Researching Transport Inequalities in Global South Cities”. This series brings together an international community of established and emerging researchers to share their recent work and experiences and to reflect on the key challenges and opportunities we face under six themes:

Access to opportunities - May 21st
Informality - June 4th
Transitions - June 18th
Health - July 2nd
Gender - July 16th
Age and disabilities - July 30th

Please see the full program below, help us spread the word and register here. The webinar series is organized under the International Network for Transport and Accessibility in Low Income Communities (INTALInC) and INTALInC LAC, the Latin American branch of the network.

Urban Picture

Wed, 13 May 2020 00:00:00 +0000

President Vargas Avenue in the city center of Rio de Janeiro during quarantine days.

credit: Lucas Magno

Webinar: Access to COVID-19 Healthcare in Brazil

Sun, 10 May 2020 00:00:00 +0000

Early April this year, a few colleagues and I published a study in Portuguese looking at Urban mobility and access to the healthcare system by patients with suspected and severe cases of COVID-19 in the 20 largest cities of Brazil. I shared a post with a summary of the study in English a few weeks ago, here

I will be presenting this study online tomorrow as part of a webinar series on Transport, Urban Mobility and COVID-19 in Brazil. See full program below. This webinar series has been organized by Silvana Zioni, Rafel Siqueira, Thiago Guimaraes and Renato Anelli at the Federal University of ABC (UFABC). Thank you for the invitation! My presentation will start at 9pm (GMT) and you can watch it below. If you would like to participate in the Zoom session, you can find info about the transmission link here.

ps. Sorry for short notice.

The size of Particulate Matter

Tue, 05 May 2020 00:00:00 +0000

Particulate matter is an air pollutant that is very damaging to health and extremely common in urban areas. Sotirios Papathanasiou have created this beautiful and yet scary image comparing the size of Particulate Matter (PM2.5 and PM10) against a red blood cell, and pathogens including SARS-CoV-2 (the virus that causes COVID-19). Regular medical masks don’t have much protection effect, if any, against Particulate Matter.

credit: Sotirios Papathanasiou

Oxford in lockdown

Thu, 30 Apr 2020 00:00:00 +0000

A glimpse of Oxford completely empty during the lockdown. Such a wonderful city with stunning architecture. How I miss this place.

We cannot exactly say that Oxford has a unique architecture, though. This is because Oxford was shamelessly copied by ‘the other place’. Matheus Nunes, a good friend of mine from the other place shared this other video showing their lockdown. Is it beautiful? Ok-ish. Do I miss it? Not strongly. Would I live there again? I might visit them, shortly.

Credit:

Oxford video by Sean Hattersley
Cambridge video by Synergy Film

Global trends toward urban street-network sprawl

Tue, 28 Apr 2020 00:00:00 +0000

The world is sprawling out. This one of the findings of a fascinating new paper that measures urban sprawl based on the local connectivity of street networks using. The authors combine OpenStreetMap data with satellite data to analyze the extent to which human settlements have sprawled since 1975 at a global, national, and city levels. The supplemantary material includes Python code used in the paper, and the data outputs + interactive website. Thanks Sabrina Lai for the pointer.

credit: Barrington-Leigh & Millard-Ball, McGill University

Barrington-Leigh, C., & Millard-Ball, A. (2020). Global trends toward urban street-network sprawl. Proceedings of the National Academy of Sciences. 117 (4) 1941-1950 https://doi.org/10.1073/pnas.1905232116

Abstract

We present a global time series of street-network sprawl—that is, sprawl as measured through the local connectivity of the street network. Using high-resolution data from OpenStreetMap and a satellite-derived time series of urbanization, we compute and validate changes over time in multidimensional street connectivity measures based on graph-theoretic and geographic concepts. We report on global, national, and city-level trends since 1975 in the street-network disconnectedness index (SNDi), based on every mapped node and edge in the world. Streets in new developments in 90% of the 134 most populous countries have become less connected since 1975, while just 29% show an improving trend since 2000. The same period saw a near doubling in the relative frequency of a street-network type characterized by high circuity, typical of gated communities. We identify persistence in street-network sprawl, indicative of path-dependent processes. Specifically, cities and countries with low connectivity in recent years also had relatively low preexisting connectivity in our earliest time period. We discuss implications for policy intervention in road building in new and expanding cities as a top priority for sustainable urban development.

credit: Barrington-Leigh & Millard-Ball, McGill University

Brasília turns 60 today

Tue, 21 Apr 2020 00:00:00 +0000

The city of Brasilia turns 60 today. Brasilia is well known as one of the boldest experiments of modern urban planning, and one of the most controversial ones as well. As someone deeply involved in studying cities and the social-spatial inequalities reflected in and reinforced by their human mobility systems, I love to hate Brasilia. There are days, though, when I hate to love it. This is why I like to say ‘I was born in Brasilia but it’s not my fault’.

In a celebratory post, I’d like to share these two aerial footages of the city. It is truly ironic how Brasilia can be at the same time so splendid from above and yet so shitty at the human scale.

Happy birthday!

credit: These videos have been created by Bento Viana, one of my favorite artists in town. Bento is a talented and versatile photographer and video maker.

New home: Urban Demographics 2.0

Sat, 18 Apr 2020 00:00:00 +0000

Photo by Toa Heftiba on Unsplash

Very excited to share the new home of the Urban Demographics blog. When I created the blog in June 2010, I decided to use Google’s platform Blogger. I was sure Google was going to conquer the world so that seemed the right choice to make at that time.

For a long time, I was really satisfied with how Blogger makes blogging so simple and easy. It didn’t take too many years, though, to feel it a bit limited. Blogger is not exactly versatile and it doesn’t work really well to share data analysis with code or interactive data visualizations… I know, I know… Urban Demographics 1.0 was also looking a bit old-fashioned as well.

I’ve decided to give it a go and create Urban Demographics 2.0 entirely from within RStudio . This website was created using Blogdown and the Academic theme for Hugo. Thanks Kaue Braga for the push I needed to get this started.

The blog has changed quite a bit from the old version and it will keep changing as I go along breaking things while trying to learn how to manage this. The spirit of Urban Demographics remains the same, though! I will keep using this as a source for structured procrastination, a personal repository to share interesting, beautiful and curious ‘stuff’ related to cities, transport, spatial analysis, demography and R. Oh, and I’ll also use the blog every now and then ~~for self-promotion~~ to share news about my own research.

ps. I haven’t decided whether I’ll try to bring the content from the old blog to this one but I’m certainly keeping an online version of Urban Demographics 1.0 as a frozen blog for the record.

Soundtrack:

Posts | Urban Demographics

Trip to Boston and DC in March

Boston, 1st week of March

Washington, DC, 2nd week of March

CfP Decolonising Research in Transport Geography

Guest editors:

Special issue information:

Accessibility and spatial interaction models

Abstract:

Crime and school access

Abstract:

Fare-free transit and the travel behavior of older adults

Proximity-centred accessibility

Abstract

The social and spatial effects of fare cuts on public transport

Abstract

Exposure to extreme heat in public transit

Geocoding Brazilian data with {geocodebr}

Two key contributions

An invitation to the community

Reuniting spatial interaction and accessibility

Abstract:

Advancing Urban Accessibility for Inclusive Cities

Toronto visit

The AAG is just around the corner

Highlights of 2024

Visiting Professor at Toronto

Finalist in the Academic Jabuti Award

Exposure to extreme heat in public transit

Abstract:

How experienced segregation is shaped by homophily and limited travel

Abstract:

WSTLUR workshop: a crash course on urban accessibility with R

Equity implications of Transit Oriented Development

Measuring transportation equity

Abstract

A summary of my 2023, and happy 2024

Route-based accessibility and its association with transit ridership

The intersection of racial and income inequalities in access to healthcare

The trade-off between time and money in ride-hailing and transit accessibility

Advances in geospatial approaches to transport networks and sustainable mobility

Online workshop: Getting to Services

Free public transit and voter turnout

Abstract:

censobr: Download Data from Brazil’s Population Census

The Causal Quartet

Accessibility within planetary boundaries

Abstract:

Accessibility and space-time differences in when and how people travel

Abstract:

New paper: time interval metric for cumulative opportunity accessibility

Contribution

Context of the problem

The proposed metric

Empirical analysis and tests

Takeaway

Calculating Urban Centrality in R

Installation

Data input

Calculating UCI

Quote of the day

CfP: Environmentally and socially sustainable mobility

Scope

Submission deadline

Guest editors

Related Workshops at GIScience Conference

The impact of travel time inaccuracy and variability on accessibility

Estimating public transport emissions from GTFS data

Open source transport network analysis

Celebrating a milestone of the Access to Opportunities Project

Changing gears

Celebrate

A new special friend

Trading off costs with time in accessibility by ride-hailing and transit

Abstract:

English version of the book on introduction to urban accessibility with R

A quick summary of the book:

Impact of network centrality and income on slowing infection spread

Abstract:

Measuring just accessibility within planetary boundaries