Uncovering urban circadian pulses based on an animated cartogram: the example of Bogotá

ABSTRACT

This paper describes a new kind of dynamic map showing the differing location of ‘day’ and ‘night’ populations and how they balance out over 24 h. To this end, we created a smoothed animated cartogram, with visual effectiveness as the main criterion. The area of study is the Bogotá metropolitan area in Colombia, where an urban mobility survey (EODH) was carried out in 2019. This paper emphasizes the method developed and the discussions which arose over the course of our work. We used R language to process the data and create the animations in order to make them reproducible, following an open-science approach. We mainly used the cartogramR and spatstat packages. The R script is appended.

RÉSUMÉ

Cet article propose un nouveau mode de représentation cartographique dynamique donnant à voir la localisation différenciée de la « population de jour » et de la « population de nuit » et restituant les rééquilibrages sous-jacents sur 24 heures. Pour ce faire, nous avons mis au point un cartogramme animé lissé, en visant l’efficacité visuelle pour le lecteur. La zone d’étude choisie est l’aire métropolitaine de Bogotá en Colombie où une enquête de mobilité urbaine (EODH) a été mise en œuvre en 2019. Cet article met l’accent sur la méthodologie développée et les discussions qui ont jalonné la démarche. Dans une optique de reproductibilité et de science ouverte, l’ensemble des traitements a été réalisé au moyen du langage R et repose principalement sur deux packages cartogramR et spatstat. Le code est fourni en annexe.

KEYWORDS:

• Circadian urban pulses
• micro data
• animated cartogram
• R script
• Bogotá

MOTS-CLEFS:

• Pulsation urbaine circadienne
• données individuelles
• cartogramme animé
• script R
• Bogotá

Disclosure statement

No potential conflict of interest was reported by the author(s).

Declaration of interest statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 Link to the Mobiliscope Project headed by Julie Vallée (UMR Géographie-cité): https://mobiliscope.cnrs.fr/en 

2 Thanks to the involvement of a team headed by Guillaume Leroux (the “Housing, spatial inequalities, and trajectories” and “Mobility, trajectories, and territories” research units).

3 Link to the Manhattan Population Explorer: http://manpopex.us/

4 Although this is not suited for visualizing stock values.

5 Original mobility surveys data: https://www.simur.gov.co/encuestas-de-movilidad

6 https://cran.r-project.org/web/packages/cartogramR/index.html

7 https://cran.r-project.org/web/packages/spatstat/index.html

8 https://modural.hypotheses.org/le-projet

9 A Bus Rapid Transit (or BRT) is a public transport system based on a trunk bus network with dedicated lanes and controlled-access stations. Feeder buses connect the trunk lanes to neighboring districts.

10 In Spanish, Sistema Integrado de Transporte Público (SITP).

11 90 min were needed to generate the 96 cartograms on a regular PC.

12 Another advantage is that the Gastner et al. (2018) algorithm and its implementation in the cartogramR package can easily handle 0 values. As described in Cornillon, Demoraes 2021, the default mp parameter is set to 0.2. If a spatial unit has a zero value, it will be replaced by mp times the smallest positive value in any spatial unit. It is also possible to distort spatial units using absolute error. Absolute error refers to the maximum difference between target surface areas (areas strictly proportional to values) and effective surface areas (output areas that the algorithm was able to compute). For instance, if we set a maximum 1,000,000 m² absolute error, we accept that the surface area of each spatial unit may differ to that extent from target area. To do so allows the algorithm to reach the final result more quickly. In our case, we did not have any 0 values and we did not have to change the default options of the algorithm.

13 Gastner et al. (2018) introduced a fast flow-based algorithm whose equations are easier to solve than using the Gastner et al. (2004) method.

14 Especially with G. Le Roux (Ined) and J. Vallée (CNRS).

15 See GitHub link in appendix.

16 We have retained the Spanish acronym, which stands for “Unidad Territorial de Análisis de Movilidad”. Indeed, this acronym is widely used not only for Bogotá but also for other Latin-American metropolises.

17 https://www.dane.gov.co/index.php/estadisticas-por-tema/demografia-y-poblacion/censo-nacional-de-poblacion-y-vivenda-2018

18 Computing the arrivals instead of the departures gives the same result.

19 4 cartograms per hour multiplied by 24 h.

Additional information

Funding

This work was supported by the French National Research Agency under Grant [ANR-MODURAL 19-CE22-00169-01].

Notes on contributors

Hugo Thomas

Hugo Thomas is a civil and transport engineer who graduated from the École Nationale des Ponts et Chaussées (Paris). He joined the ANR-funded Modural research program in 2021 to carry out research on sustainable mobility in Latin-American cities, looking specifically at the environmental impact of transport in Bogotá and Lima. He is currently working for a bi-disciplinary PhD on the levers for more sustainable mobility in Bogotá and Lima, within the frame of an international cotutelle agreement. This PhD has a Geography component, co-supervised at Université Rennes 2 (France) by Vincent Gouëset and Florent Demoraes, and an engineering component, supervised at the Environment Engineering Department of the Universidad de Los Andes (Colombia) by Luis Angel Guzmán. https://orcid.org/0000-0002-7762-4976; https://www.linkedin.com/in/ht128/.

Florent Demoraes

Florent Demoraes holds a PhD and HDR in Geography. He is a full professor at Université Rennes 2 and a member of the CNRS 6590 Research Unit (Espaces et Sociétés) which he directed from 2016 to 2019 (Rennes site). He is currently on a two-year secondment to the French Institute of Andean Studies (IFEA) in Bogotá to conjointly steer with Vincent Gouëset the ANR-funded Modural research program. He works in Latin American metropolises on (1) the relationship individuals have with the city according to their place of residence, their position in the social hierarchy, and their biographical stage; (2) sustainable mobility practices and social inequalities; (3) transformations to social divisions of space; (4) innovative mapping technique. https://orcid.org/0000-0001-6113-9960 / https://perso.univ-rennes2.fr/florent.demoraes.