Using Landsat and Machine Learning to Map Urban Change

Social scientists at the U.S. Geological Survey (USGS) Fort Collins Science Center – in collaboration with the USGS National Land Imaging Program – conduct Earth observation user case studies using qualitative research methods. Using standard scientific methods, they are better able to understand the variety of Earth observation users, including how they use and value Earth observation data. This graphic illustration guides you through an Earth observation user case study and provides the in-depth user experience of Descartes Labs – one example of an Earth observation user. Descartes Labs is a New Mexico-based company that developed a geospatial data refinery for building and running machine learning models on Earth observation data. This case study shows how they used Landsat imagery to map urban growth and heating. Landsat is a joint USGS/NASA Program that provides the longest continuous space-based record of Earth’s land surface. Every day, Landsat satellites provide essential information to help land managers and policy makers make decisions about resources and the environment. In the context of rapid urbanization, the Landsat program is well-suited to monitor urban growth and its impacts to social and ecological systems.