Biological Sciences Division
Published September 3, 2025 | Version v1
Journal article Metadata-only

Infrastructure deficits and informal settlements in sub-Saharan Africa

Creators

  • 1. University of Chicago

Description

Sustainable development is an imperative worldwide but metrics and data on poverty and quality of life have remained too coarse and abstract to characterize challenges adequately and guide practical progress. Nowhere is this challenge greater than in Africa, where we still know little about the spatial details of development. Here we leverage a comprehensive, high-precision dataset of building footprints to identify infrastructure deficits and infer informal settlements down to the street block level everywhere in sub-Saharan Africa. We identify a general pattern of informality with cities showing, on average, greater access to infrastructure and services than rural and peri-urban areas. We show that such patterns of informality are characterized by consistent statistical distributions reflecting uneven local development. We also show that these physical measures of informality are systematically associated with many indicators of human deprivation, which form a single principal component co-varying predictably with specific changes in street access to buildings. These results demonstrate that the localization of sustainable development is possible down to the street level at a continental scale and provide a general distributed strategy for accelerating progress in infrastructure and service expansion that taps local innovations in systematic, equitable and context-appropriate ways.

Data availability

All block-level data for sub-Saharan Africa, including aggregations to GHSL and Africapolis urban definitions, are available for public download via the Harvard Dataverse repository at https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/DQY54U. The database is also available online at www.millionneighborhoods.africa/download as an interactive map to visualize diverse statistics including block complexity and population density. All primary data sources are available via public sources with the exception of the DHS and Ecopia datasets, which require data usage agreements. To access DHS data, users must request access to 'SURVEY' and 'GPS' data for all countries in sub-Saharan Africa following these instructions: https://dhsprogram.com/data/Access-Instructions.cfm. We provide application processing interface (API) code to download the surveys in the kblock-analysis repository. To access the Ecopia building footprint data, users should email admin@digitizeafrica.ai to obtain credentials for the DigitizeAfrica Platform. Users who have difficulty accessing the above data should contact the authors, who maintain back-ups of the source data. Source data are provided with this paper.

The code for generating the underlying database, including block complexity, population and block geometries, is available on GitHub at https://github.com/mansueto-institute/kblock with DOI-minted source code available via Zenodo at https://doi.org/10.5281/zenodo.12636819 (ref. 52). The code for reproducing the analysis, including all figures and tables, is available on GitHub at https://github.com/mansueto-institute/kblock-analysis with DOI-minted source code available via Zenodo at https://doi.org/10.5281/zenodo.15702173 (ref. 53). A Code Ocean capsule with a fully reproducible example is available at https://doi.org/10.24433/CO.1487090.v1.

Additional details

Identifiers

DOI
10.1038/s41586-025-09465-2
Other
oai:uchicago.tind.io:16183

Related works

Cites
https://doi.org/10.7910/DVN/DQY54U (URL)
http://www.millionneighborhoods.africa/download (URL)
https://doi.org/10.5281/zenodo.12636819 (URL)
https://doi.org/10.5281/zenodo.15702173 (URL)

Funding

Bill & Melinda Gates Foundation
University of Chicago
University of Chicago

UChicago Information

Division(s)
Biological Sciences Division
Department(s)
Ecology and Evolution