This paper investigates the completeness and inequalities of global urban building data in OpenStreetMap (OSM). Using a machine-learning model, the authors infer the completeness of OSM building stock data for 13,189 urban agglomerations. Results show significant spatial biases, with high completeness in some areas (16% of the urban population exceeding 80% completeness) and very low completeness in others (48% of the urban population below 20%). While inequalities have receded somewhat due to humanitarian mapping efforts, complex patterns remain, varying across development index groups, population sizes, and geographic regions. The paper provides recommendations for data producers and analysts to address these biases.
Publisher
Nature Communications
Published On
Jul 06, 2023
Authors
Benjamin Herfort, Sven Lautenbach, João Porto de Albuquerque, Jennings Anderson, Alexander Zipf
Tags
OpenStreetMap
urban completeness
building data
spatial bias
humanitarian mapping
data inequalities
machine learning
Related Publications
Explore these studies to deepen your understanding of the subject.
