A new scientific project is offering an unprecedented look at how humans have reshaped the Earth. Using artificial intelligence and satellite data, researchers say they can now pinpoint almost every structure built by people worldwide.
Others are reading now
The scale of the effort sets it apart from previous mapping attempts.
According to UNILAD, researchers at the Technical University of Munich have created a high-resolution global map that tracks nearly every physical structure on Earth, from major cities to remote rural areas. The project processed data on more than 2.75 billion buildings.
The map, known as the GlobalBuildingAtlas, was built using over 800,000 satellite images captured in 2019 and analysed with machine-learning tools.
A global first
Scientists involved in the project said the atlas contains roughly one billion more buildings than any comparable dataset. It also offers around 30 times more detail than existing 3D building models.
For the first time, researchers can observe on a global scale how human construction has transformed landscapes over time. The system captures not only building footprints but also height, volume and spacing.
Also read
“With 3D models, we see not only the footprint but also the volume of each building, enabling far more precise insights into living conditions,” said Professor Xiaoxiang Zhu, lead author of the study and head of the dataset project.
Beyond cities
The AI model estimates the height and volume of every structure on Earth, including buildings in rural and sparsely populated regions. It also measures distances between neighbouring properties.
Researchers said this makes the tool valuable for studying housing density, wealth inequality and environmental impact across both urban and remote areas.
Because the model relies on existing satellite imagery, the team did not need to deploy costly aircraft for ground scanning, reducing both expense and complexity, reports UNILAD.
New indicators
According to UNILAD, Zhu said traditional two-dimensional maps often fail to capture social and economic realities. The project introduces what researchers call a new global indicator: building volume per capita.
Also read
“We introduce a new global indicator: building volume per capita, the total building mass relative to population – a measure of housing and infrastructure that reveals social and economic disparities,” he said.
The data also reflects how wealthier areas tend to have more space and larger buildings per person than poorer regions.
Policy potential
The team hopes governments and planners will use the atlas to guide urban development and respond more effectively to natural disasters such as floods and earthquakes.
Since its launch, nearly 300,000 people have visited the project’s website to explore the interactive database, according to the researchers.
Sources: UNILAD, Technical University of Munich