Published on Sustainable Cities

The Global Urban Footprint: A map of nearly every human settlement on Earth

This page in:

Urbanization is increasingly central to the global development process, but until recently, basic spatial information on the world’s urban areas has been unavailable, inconsistent, or unreliable. The lack of consistent data on the world’s cities makes it hard to understand the overall impact of urbanization.  However, innovations in geospatial mapping are now helping to provide one major piece of the puzzle: maps of practically all built-up areas around the world are available thanks to new uses of satellite data. 
Scientists at the German Aerospace Center (DLR) have succeeded in using a newly developed method to map the world’s built spaces at an unprecedented spatial resolution,  resulting in the ‘Global Urban Footprint’ (GUF ), a global map of human settlements at a spatial resolution of 12 meters per grid cell (aggregated to 75m for public use).
The German radar satellites TerraSAR X and TanDEM X acquired over 180,000 images between 2010 and 2013, which were processed, together with additional data such as digital terrain models, to produce the Global Urban Footprint. In total, the researchers processed over 20 million datasets with a combined volume of more than 320 terabytes.
The result is a map which, besides possessing a strange beauty that has been likened to Chinese ink drawings, documents practically all of humanity’s physical presence on the surface of the earth: Greater Tokyo, home to over 35 million people; Delhi, set amidst a constellation of hundreds of tiny villages; tendrils of linear settlement along roads, canals and rivers spreading out from Ho Chi Minh City.
Despite the name, the Global Urban Footprint does not just identify ‘urban’ areas. At a resolution of 12 meters, it in fact has the potential to locate even the smallest rural settlements, from remote outposts in the vast plains of Mongolia to small industrial complexes in the Sahara.
A key innovation of GUF is that it uses radar data, as opposed to optical imagery. Relying on radar data has proved to be an extremely effective approach to mapping human settlement as it detects vertical structures characteristic of human habitations with great precision.
The two radar satellites, orbiting at an altitude of over 500 km, covered the entire surface of Earth within two years. This was possible as the sibling satellites are able to 'see' through clouds and can even record data at night. This is a decisive advantage of radar technology compared to optical satellites, as optical technology involves laborious piecing together of scenes undisturbed by clouds.
The high degree of precision means that the resulting images may show objects as small as chimneys, pylons or even road signs. On the other hand, what looks like human-made objects might actually be large individual trees in the African landscape or rocky outcrops in the desert. To filter out these errors, the research team undertook a rigorous quality enhancement process, with support from the World Bank and the Swiss State Secretariat for Economic Affairs (SECO). This involved comparing the radar-based data to reference data such as Open Street Map and maps of impervious surfaces, as well as an automated elimination of wrongly classified areas. What remains is human-made objects that make up human settlements. However, GUF does not just replicate what might already be represented in one of the reference layers; it also finds settlements that are not found in any other similar data set. For example, the large area of linear settlements along ridges north of Nairobi, Kenya, which are too small or dispersed to be picked up by most such maps, are clearly visible in GUF.
Linear settlements north of Nairobi, Kenya, which are too small or dispersed to be picked up by most maps, are clearly visible in the Global Urban Footprints map (right). Left: Google Earth, © Google 2010, Image Landsat; Right: DLR
Linear settlements north of Nairobi, Kenya, which are too small or dispersed to be picked up by most maps, are clearly visible in the Global Urban Footprints map (right). Left: Google Earth, © Google 2010, Image Landsat; Right: DLR
This new global data set is invaluable to development organizations like the World Bank, governments around the world, and researchers at universities and think tanks. It allows us to quantitatively and qualitatively analyze and compare settlement patterns, at the metropolitan, national, and global scales. This can act as a valuable input into the design of urban and regional development projects, and can also inform our understanding of how urbanization is related to economic growth, poverty reduction, and carbon emissions.
But the applications of this data go beyond urban development. Among other uses, accurate data on built-up areas can vastly improve the accuracy of population distribution mapping. Knowing where people and assets are located is crucial for climate change mitigation and disaster risk management. It can help design projects that more effectively connect settlements with services through transportation and other infrastructure networks.
  In light of the potential development applications of GUF, DLR will soon release the data set to be used free of charge at full spatial resolution for any scientific use, and at 75m resolution for any non-profit use. Currently available by request from DLR, by the middle of 2016, the data set will also be accessible on the World Bank’s PUMA platform, as well as the European Space Agency’s Urban Thematic Exploitation Platform. Together with other new data sets derived from optical Earth observation data or showing the population distribution, the Global Urban Footprint map provides a deeper understanding of the wide-ranging impacts of urbanization. In this context, the World Bank and DLR will continue their fruitful cooperation in order to further utilize Earth Observation data and techniques for sustainable urban and rural development.


Thomas Esch

Leader of the research team “Urban Areas and Land Management” and project manager at the German Remote Sensing Data Center

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000