Global Urban Footprint

Global Urban Footprint
Name	Global Urban Footprint
Caption	High-resolution mapping of built-up areas
Type	Remote sensing dataset
Creator	German Aerospace Center (DLR)
Released	2016
Format	Raster
Resolution	12–84 m
Coverage	Global

Contents

Global Urban Footprint is a high-resolution global dataset mapping built-up areas derived from satellite remote sensing. Developed principally by the German Aerospace Center (DLR) with contributions from partners, the dataset has been used by researchers and institutions including European Space Agency, NASA, United Nations Environment Programme, World Bank, and Esri to analyze urbanization patterns. It has informed work by scholars at Massachusetts Institute of Technology, University of Oxford, Stanford University, University College London, and University of Tokyo as well as planning agencies such as UN-Habitat, OECD, and national statistical offices in Germany, United States, India, Brazil, and China.

Overview

The Global Urban Footprint provides a consistent, wall-to-wall mapping of built-up areas based on high-resolution synthetic aperture radar and optical imagery collected by missions like TanDEM-X, TerraSAR-X, Landsat, and Sentinel-1. It was produced in collaboration with institutions including DLR, University of Bonn, Fraunhofer Society, Max Planck Society, and private contractors such as Airbus Defence and Space and Planet Labs. The product complements earlier global land maps produced by GlobCover, MODIS Land Cover, Global Human Settlement Layer, and initiatives led by European Commission research programs and feeds into assessments by Intergovernmental Panel on Climate Change and Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services.

Production of the dataset combined machine learning classifiers, object-based image analysis, and manual quality assessment drawing on data from TanDEM-X interferometry, TerraSAR-X backscatter, and multispectral stacks from Landsat 8 and Sentinel-2. Algorithms were trained using reference samples from mapping projects such as OpenStreetMap, CORINE Land Cover, and national cadasters like United States Geological Survey and India Remote Sensing Centre. Processing employed high-performance computing centers such as German Climate Computing Center (DKRZ), supercomputers at Lawrence Berkeley National Laboratory, and cloud platforms offered by Amazon Web Services, Google Earth Engine, and Microsoft Azure. Validation used ground truth from field campaigns coordinated with agencies like United Nations Development Programme and academic surveys from Harvard University, Columbia University, and University of California, Berkeley.

Mapped built-up density reveals rapid expansion in metropolitan regions like Tokyo, Delhi, São Paulo, Shanghai, and Lagos and shows contrasting dynamics in regions such as Sub-Saharan Africa, Southeast Asia, Eastern Europe, North America, and Australia. The footprint highlights peri-urban growth corridors connecting megacities such as Beijing–Tianjin, Mexico City–Guadalajara, and Cairo–Alexandria and identifies smaller urban agglomerations in island states like Philippines and Indonesia. Comparative studies using the dataset with datasets like WorldPop, Gridded Population of the World, and Global Rural-Urban Mapping Project illustrate relationships between built-up extent and urban hierarchy in regions covered by programmes such as Belt and Road Initiative and analyses of post-industrial change in cities like Detroit and Manchester.

Analyses linking built-up maps to environmental datasets from Copernicus, Global Forest Watch, International Energy Agency, and United Nations Framework Convention on Climate Change inform assessments of land cover conversion, urban heat island effects in cities such as London, Paris, New York City, and Sydney, and flood exposure in deltas like Ganges Delta, Nile Delta, and Mekong Delta. Socioeconomic research using the product interacts with census agencies such as United States Census Bureau, Office for National Statistics (UK), Statistics Canada, and National Bureau of Statistics of China to study housing, transportation networks like Trans-Siberian Railway or Pan-American Highway, and infrastructure investment from institutions such as the Asian Development Bank and the Inter-American Development Bank.

The dataset supports urban planning at municipal agencies in Berlin, Delhi, São Paulo, and Johannesburg, disaster risk planning by International Federation of Red Cross and Red Crescent Societies, public health studies with World Health Organization, and climate adaptation projects funded by the Green Climate Fund and Global Environment Facility. Researchers combine it with socioeconomic layers from World Resources Institute, mobility data from Uber, Google, and Apple mobility reports, and cadastral maps held by agencies like Land Registry (UK) and National Land Commission (Kenya). It is also used in conservation planning coordinated with IUCN, heritage assessments with UNESCO, and infrastructure modeling for transport authorities such as Transport for London.

Limitations arise from sensor resolution differences across platforms like TanDEM-X, Landsat, and Sentinel-2, temporal mismatches with census epochs of UN Population Division and national bureaus, and classification errors common to automated mapping efforts employed by groups including Google Earth Engine teams and academic labs at ETH Zurich and Imperial College London. Urban form heterogeneity in informal settlements in cities such as Kampala, Port-au-Prince, and Dhaka poses commission and omission errors relative to cadastral ground truth from municipal authorities and NGOs like Habitat for Humanity. Users should consider uncertainty in slope and elevation data from Shuttle Radar Topography Mission when applying the dataset to hydrological or risk models developed by NOAA and European Centre for Medium-Range Weather Forecasts.

Category:Remote sensing datasets