Skip to main content

Geospatial Data Sets

Oak Ridge National Laboratory collects, curates, and delivers unique geospatial datasets that offer a full picture of human activities, the natural and built environment, and the interactions among them. The following datasets and platforms provide federal agencies and industry with information to better understand our increasingly connected and digitally integrated infrastructure, anticipate population changes at global and local scales, and respond quickly to disasters that threaten human and economic security.

Screenshot of the EAGLE-I platform

EAGLE-I —The Environment for Analysis of Geo-Located Energy Information, or EAGLE-I, is an interactive geographic information system that provides capabilities for monitoring energy infrastructure, reporting outages, and coordinating emergency response and recovery efforts.

ORNL manages and maintains the tool, which was developed by the Department of Energy as the first-ever technology to track power outages nationwide. It is the authoritative federal source for historical and real-time situational awareness for the nation’s energy infrastructure, which includes electricity, natural gas, and petroleum.

EAGLE-I delivers near-real-time power outage data, as well as analysis and visualization capabilities—which now include the URBAN-NET modeling tool for critical infrastructure interdependency analysis. The tool can home in on impacted areas at the county level, overlay weather and satellite data, locate area hospitals, and identify other key points of critical infrastructure. With that information readily available, DOE, FEMA, and other first responders can efficiently plan, react, respond, recover, and restore critical energy systems after a disaster.

World map showing color-coded population density
Within LandScan, human populations are represented by the yellow and orange colors. The darker the orange, the higher the population density.

LandScan—LandScan datasets (Global, USA, and HD) provide globally consistent and high-resolution data on where populations are on the planet. This data provides significantly greater clarity than the residential locations as reported in official censuses, by incorporating and allocating working populations, shopping populations, and students in schools. To date, LandScan data has been used around the world to estimate populations at risk and to plan, aid, and direct disaster recovery efforts in the wake of earthquakes, tsunamis, hurricanes, and other such events.

The full archive of LandScan’s three distinct datasets—LandScan Global, LandScan HD, and LandScan USA—is available under a Creative Commons Open-Source license at

Homeland Infrastructure Foundation Level Data (HIFLD)—ORNL helped provided many of the nation’s authoritative geospatial datasets for critical infrastructure analysis and protection, preparedness planning, and crisis management. Led by the Department of Homeland Security, HIFLD hosts more than 400 open data layers for operational, situational, and strategic awareness of critical infrastructures. Among the layers developed by ORNL are:

  • Schools (public/private, K-12, colleges, and universities)
  • Childcare facilities
  • Prisons/correctional facilities
  • Rail systems
  • Solid waste facilities
  • Mobile home parks
  • Hospitals
  • Foundation energy data (substations, distribution/transmission lines, etc.)
  • Major sports venues
  • National shelters
  • Nursing homes/assisted living facilities
  • Law enforcement facilities
  • Convention centers
Logo for the Outage Data Initiative Nationwide

ODIN—The Outage Data Initiative Nationwide program, led by Oak Ridge National Laboratory and the U.S. Department of Energy's Office of Electricity, seeks to establish a comprehensive digital reporting standard for power outage data to enable utilities and others to exchange data freely with designated stakeholders at all levels. ODIN will remove barriers to data sharing — improving restoration, reliability, risk mitigation, and emergency response. (Factsheet)

PlanetSense—The PlanetSense platform gathers open-source data to support geospatial intelligence, national security, and humanitarian missions. PlanetSense seamlessly integrates archived data with dynamic location intelligence information, runs sophisticated data analytics algorithms and visualization tools, and generates geospatial intelligence in real time. With the PlanetSense platform, ORNL maintains and delivers more than 250 million global points of interest data to support national security missions.

Population Density Tables—ORNL’s Population Density Tables report ambient building occupancy estimates of people/1000 ft2 for day, night, and episodic events at regional, national, and subnational levels across the planet. The PDT database has been used for emergency preparedness and response, population mapping, and environmental and socioeconomic applications.

UrbanPop—ORNL’s UrbanPop dataset simulates realistic full populations with detailed demographic, cultural, and economic characteristics, as well as residential and workplace information. This unique collection of demographically and spatially detailed population datasets has been used in several studies, from determining the effect of hurricane Sandy on residential populations within New York to effects of electricity outages on economically vulnerable populations in the Knoxville Metro Area.

Satellite image with AI-detected building outlines showing intact and destroyed structures following a hurricane.
This image from Sept. 30, 2022, shows how the Federal Emergency Management Agency used ORNL's USA Structures data along with new satellite images to identify structures that were destroyed in Lee County, during Hurricane Ian. Credit: ORNL, U.S. Dept. of Energy

USA Structures—USA Structures is the first-ever dataset of all building structures 450 ft2 and larger in the U.S. and its territories—more than 125 million structures total. Researchers at ORNL combined expertise in geospatial science, high-performance computing, and artificial intelligence to process images, compile data, detect patterns, and extract building features. To date, our researchers have processed 1.1 petabytes of imagery—stitching together and describing what is roughly the equivalent 1 billion digital photographs.

This publicly available dataset provides a foundation for understanding location-based information, and helps improve data sharing across our Federal, Local, State, Tribal and Territorial partners. Accurate and detailed maps are critical in emergency response and recovery efforts in communities of any size, speeding recovery efforts such as damage assessments and helping communities in need begin to rebuild in a matter of days rather than weeks or months.

WorldSTAMP—ORNL's WorldSTAMP platform integrates and harmonizes more than 30 authoritative sources of global data—including datasets from the World Bank, United Nations, and World Health Organization—into a spatiotemporal data cube supporing analytics across 200+ countries, covering 50+ years, and providing 90,000+ attributes. ORNL applies advanced analytics to identify trends, patterns, anomalies, and changes in national landscapes and populations in support of geospatial intelligence missions.