LandScan

LandScan
Name	LandScan
Caption	Global population distribution model example
Established	1998
Developer	Oak Ridge National Laboratory
Country	United States
Discipline	Population distribution modeling

LandScan is a high-resolution global population distribution dataset produced to estimate ambient population counts by allocating census counts to grid cells. It supports analyses for disaster response, public health, transportation, humanitarian relief, and national security by providing gridded population estimates tied to geographic features and ancillary datasets. The dataset is maintained and distributed by research groups associated with Oak Ridge National Laboratory and used by agencies such as United States Department of Defense, United States Agency for International Development, United Nations Office for the Coordination of Humanitarian Affairs, and academic institutions.

Overview

LandScan is a global raster product that represents estimated population counts at approximately 1 km spatial resolution, produced as annual datasets. The model outputs ambient population — people present during a 24-hour period — rather than nighttime residential population, supporting operational needs of organizations like Federal Emergency Management Agency, World Health Organization, International Committee of the Red Cross, and United Nations High Commissioner for Refugees. The dataset covers sovereign states, territories, and subnational units used by entities such as World Bank, European Commission, and national statistical offices for situational awareness and planning.

Methodology and Data Sources

The LandScan methodology is a multi-variable dasymetric modeling approach that integrates census counts with ancillary geospatial layers. Primary census inputs derive from national statistical agencies such as United States Census Bureau, Office for National Statistics (United Kingdom), and National Institute of Statistics and Geography (Mexico) where available. Ancillary layers used in the allocation algorithm include land cover from sources like Moderate Resolution Imaging Spectroradiometer, built-up area products such as Global Human Settlement Layer, road networks from OpenStreetMap and Global Road Data, nighttime lights from Defense Meteorological Satellite Program's Operational Linescan System, and elevation data from Shuttle Radar Topography Mission. Additional inputs can include points of interest, Transportation Research Board datasets, and infrastructure inventories compiled by organizations like Esri or Humanitarian OpenStreetMap Team. The allocation process applies weights derived from travel time, urban/rural classification, proximity to roads and built environment, and land cover to disaggregate census counts into grid cells.

Applications and Use Cases

LandScan is applied across sectors by stakeholders including NASA, Centers for Disease Control and Prevention, United Nations Children's Fund, and private firms. Typical use cases include disaster risk analysis for events such as the 2004 Indian Ocean earthquake and tsunami, pandemic modeling like for COVID-19 pandemic, humanitarian response mapping used by Médecins Sans Frontières, evacuation planning for hazards catalogued by United States Geological Survey, service accessibility assessments with ties to International Telecommunication Union infrastructure planning, and exposure estimation for climate impacts associated with Intergovernmental Panel on Climate Change scenarios. Planners in metropolitan areas including New York City, Tokyo, London, and Mumbai utilize LandScan-derived layers together with transportation models and census tract data for resilience and infrastructure projects.

Accuracy, Limitations, and Validation

Accuracy assessments compare LandScan outputs with alternate gridded population datasets produced by groups like WorldPop, Global Human Settlement Layer, and national census products from Statistics Canada or Instituto Nacional de Estadística y Censos (Argentina). Validation studies by universities such as University of Oxford, University of California, Berkeley, and Massachusetts Institute of Technology evaluate error using high-resolution building footprints from Microsoft and household survey clusters from Demographic and Health Surveys. Limitations include dependency on the spatial and temporal resolution of input censuses (varying by United Nations member state), potential misallocation in rapidly urbanizing areas like Lagos or informal settlements such as in Dhaka, and sensitivity to ancillary data quality (e.g., cloud-contaminated MODIS scenes or outdated road maps). Users must consider uncertainty when applying LandScan for critical operational decisions and often complement it with local surveys or remote sensing-derived building inventories from providers like Maxar Technologies.

Licensing and Access

Access to LandScan is managed by organizations tied to its developer and is subject to licensing terms used by agencies such as Oak Ridge National Laboratory and partners in the U.S. Department of Energy. Distribution pathways include institutional agreements with entities like National Geospatial-Intelligence Agency and data portals used by United Nations agencies and academic consortia. Commercial users and government customers often obtain datasets under specific license agreements, while some research collaborations secure access through memoranda with institutions such as University of Tennessee. Alternative open products with different licensing models include WorldPop and the Global Human Settlement Layer produced by the European Commission.

History and Development

LandScan originated in the late 1990s at Oak Ridge National Laboratory with initial funding and collaboration involving U.S. Department of Energy and defense research partners. Early project milestones included prototype releases in 1998 and operational annual updates introduced in the 2000s, aligning with international efforts such as Global Earth Observation System of Systems and partnerships with United States Agency for International Development. Over time the project incorporated new data streams—nighttime lights from Defense Meteorological Satellite Program, terrain from SRTM, and global remote sensing products from NASA missions—while engaging with academic groups at Carnegie Mellon University and University of Oxford for methodological research. Continued development reflects evolving needs of actors like World Health Organization and United Nations Office for the Coordination of Humanitarian Affairs for timely population distribution information.

Category:Geographic information systems