Sentinel-2 10m Land Use/Land Cover Time Series

Global imagery layer optimized for analysis. Sentinel-2 10m land use/land cover time series of the world. Produced by Impact Observatory and Esri.

This layer displays a global map of land use/land cover (LULC) derived from ESA Sentinel-2 imagery at 10m resolution. Each year is generated with Impact Observatory’s deep learning AI land classification model, trained using billions of human-labeled image pixels from the National Geographic Society. The global maps are produced by applying this model to the Sentinel-2 Level-2A image collection on Microsoft’s Planetary Computer, processing over 400,000 Earth observations per year.

The algorithm generates LULC predictions for nine classes, described in detail below.  

The year 2017 has a land cover class assigned for every pixel, but its class is based upon fewer images than the other years. The years 2018-2024 are based upon a more complete set of imagery. For this reason, the year 2017 may have less accurate land cover class assignments than the years 2018-2024.
 

Key Properties 

  Variable mapped: Land use/land cover in 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024
Source Data Coordinate System: Universal Transverse Mercator (UTM) WGS84
Service Coordinate System: Web Mercator Auxiliary Sphere WGS84 (EPSG:3857)
Extent: Global
Source imagery: Sentinel-2 L2A
Cell Size: 10-meters
Type: Thematic
Attribution: Esri, Impact Observatory Analysis: Optimized for analysis Class Definitions: 
  ValueNameDescription1WaterAreas where water was predominantly present throughout the year; may not cover areas with sporadic or ephemeral water; contains little to no sparse vegetation, no rock outcrop nor built up features like docks; examples: rivers, ponds, lakes, oceans, flooded salt plains.2TreesAny significant clustering of tall (~15 feet or higher) dense vegetation, typically with a closed or dense canopy; examples: wooded vegetation,  clusters of dense tall vegetation within savannas, plantations, swamp or mangroves (dense/tall vegetation with ephemeral water or canopy too thick to detect water underneath).4Flooded vegetationAreas of any type of vegetation with obvious intermixing of water throughout a majority of the year; seasonally flooded area that is a mix of grass/shrub/trees/bare ground; examples: flooded mangroves, emergent vegetation, rice paddies and other heavily irrigated and inundated agriculture.5CropsHuman planted/plotted cereals, grasses, and crops not at tree height; examples: corn, wheat, soy, fallow plots of structured land.7Built AreaHuman made structures; major road and rail networks; large homogenous impervious surfaces including parking structures, office buildings and residential housing; examples: houses, dense villages / towns / cities, paved roads, asphalt.8Bare groundAreas of rock or soil with very sparse to no vegetation for the entire year; large areas of sand and deserts with no to little vegetation; examples: exposed rock or soil, desert and sand dunes, dry salt flats/pans, dried lake beds, mines.9Snow/IceLarge homogenous areas of permanent snow or ice, typically only in mountain areas or highest latitudes; examples: glaciers, permanent snowpack, snow fields.10CloudsNo land cover information due to persistent cloud cover.11RangelandOpen areas covered in homogenous grasses with little to no taller vegetation; wild cereals and grasses with no obvious human plotting (i.e., not a plotted field); examples: natural meadows and fields with sparse to no tree cover, open savanna with few to no trees, parks/golf courses/lawns, pastures. Mix of small clusters of plants or single plants dispersed on a landscape that shows exposed soil or rock; scrub-filled clearings within dense forests that are clearly not taller than trees; examples: moderate to sparse cover of bushes, shrubs and tufts of grass, savannas with very sparse grasses, trees or other plants.

NOTE: Land use focus does not provide the spatial detail of a land cover map. As such, for the built area classification, yards, parks, and groves will appear as built area rather than trees or rangeland classes.


Usage Information and Best Practices

Processing Templates
  • This layer includes a number of preconfigured processing templates (raster function templates) to provide on-the-fly data rendering and class isolation for visualization and analysis. 
  • Each processing template includes labels and descriptions to characterize the intended usage. This may include for visualization, for analysis, or for both visualization and analysis.

 

Visualization

  • The default rendering on this layer displays all classes.
  • There are a number of on-the-fly renderings/processing templates designed specifically for data visualization.
  • By default, the most recent year is displayed. To discover and isolate specific years for visualization in Map Viewer, try using the Image Collection Explorer.

 

Analysis

  • In order to leverage the optimization for analysis, the capability must be enabled by your ArcGIS organization administrator. More information on enabling this feature can be found in the ‘Regional data hosting’ section of this help doc.
  • Optimized for analysis means this layer does not have size constraints for analysis and it is recommended for multisource analysis with other layers optimized for analysis. See this group for a complete list of imagery layers optimized for analysis.
  • Prior to running analysis, users should always provide some form of data selection with either a layer filter (e.g. for a specific date range, cloud cover percent, mission, etc.) or by selecting specific images. To discover and isolate specific images for analysis in Map Viewer, try using the Image Collection Explorer.
  • Zonal Statistics is a common tool used for understanding the composition of a specified area by reporting the total estimates for each of the classes. 

 

General

  • If you are new to Sentinel-2 LULC, the Sentinel-2 Land Cover Explorer provides a good introductory user experience for working with this imagery layer. For more information, see this Quick Start Guide.
  • Global land use/land cover maps provide information on conservation planning, food security, and hydrologic modeling, among other things. This dataset can be used to visualize land use/land cover anywhere on Earth.

 

Classification Process

These maps include Version 003 of the global Sentinel-2 land use/land cover data product. It is produced by a deep learning model trained using over five billion hand-labeled Sentinel-2 pixels, sampled from over 20,000 sites distributed across all major biomes of the world.

The underlying deep learning model uses 6-bands of Sentinel-2 L2A surface reflectance data: visible blue, green, red, near infrared, and two shortwave infrared bands. To create the final map, the model is run on multiple dates of imagery throughout the year, and the outputs are composited into a final representative map for each year.

The input Sentinel-2 L2A data was accessed via Microsoft’s Planetary Computer and scaled using Microsoft Azure Batch.

 
Citation
Karra, Kontgis, et al. “Global land use/land cover with Sentinel-2 and deep learning.” IGARSS 2021-2021 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2021.

Acknowledgements

Training data for this project makes use of the National Geographic Society Dynamic World training dataset, produced for the Dynamic World Project by National Geographic Society in partnership with Google and the World Resources Institute.

Jeux de données disponibles au téléchargement

Informations supplémentaires

Champ Valeur
Dernière mise à jour mars 24, 2026, 21:07 (TU)
Créé mars 24, 2026, 21:07 (TU)
Domaine / Sujet
Domaine ou sujet du jeu de données catalogué.
Titre
Titre du jeu de données.
Sentinel-2 10m Land Use/Land Cover Time Series
Description
Une description du jeu de données.

Global imagery layer optimized for analysis. Sentinel-2 10m land use/land cover time series of the world. Produced by Impact Observatory and Esri.

This layer displays a global map of land use/land cover (LULC) derived from ESA Sentinel-2 imagery at 10m resolution. Each year is generated with Impact Observatory’s deep learning AI land classification model, trained using billions of human-labeled image pixels from the National Geographic Society. The global maps are produced by applying this model to the Sentinel-2 Level-2A image collection on Microsoft’s Planetary Computer, processing over 400,000 Earth observations per year.

The algorithm generates LULC predictions for nine classes, described in detail below.  

The year 2017 has a land cover class assigned for every pixel, but its class is based upon fewer images than the other years. The years 2018-2024 are based upon a more complete set of imagery. For this reason, the year 2017 may have less accurate land cover class assignments than the years 2018-2024.
 

Key Properties 

  Variable mapped: Land use/land cover in 2017, 2018, 2019, 2020, 2021, 2022, 2023, 2024
Source Data Coordinate System: Universal Transverse Mercator (UTM) WGS84
Service Coordinate System: Web Mercator Auxiliary Sphere WGS84 (EPSG:3857)
Extent: Global
Source imagery: Sentinel-2 L2A
Cell Size: 10-meters
Type: Thematic
Attribution: Esri, Impact Observatory Analysis: Optimized for analysis Class Definitions: 
  ValueNameDescription1WaterAreas where water was predominantly present throughout the year; may not cover areas with sporadic or ephemeral water; contains little to no sparse vegetation, no rock outcrop nor built up features like docks; examples: rivers, ponds, lakes, oceans, flooded salt plains.2TreesAny significant clustering of tall (~15 feet or higher) dense vegetation, typically with a closed or dense canopy; examples: wooded vegetation,  clusters of dense tall vegetation within savannas, plantations, swamp or mangroves (dense/tall vegetation with ephemeral water or canopy too thick to detect water underneath).4Flooded vegetationAreas of any type of vegetation with obvious intermixing of water throughout a majority of the year; seasonally flooded area that is a mix of grass/shrub/trees/bare ground; examples: flooded mangroves, emergent vegetation, rice paddies and other heavily irrigated and inundated agriculture.5CropsHuman planted/plotted cereals, grasses, and crops not at tree height; examples: corn, wheat, soy, fallow plots of structured land.7Built AreaHuman made structures; major road and rail networks; large homogenous impervious surfaces including parking structures, office buildings and residential housing; examples: houses, dense villages / towns / cities, paved roads, asphalt.8Bare groundAreas of rock or soil with very sparse to no vegetation for the entire year; large areas of sand and deserts with no to little vegetation; examples: exposed rock or soil, desert and sand dunes, dry salt flats/pans, dried lake beds, mines.9Snow/IceLarge homogenous areas of permanent snow or ice, typically only in mountain areas or highest latitudes; examples: glaciers, permanent snowpack, snow fields.10CloudsNo land cover information due to persistent cloud cover.11RangelandOpen areas covered in homogenous grasses with little to no taller vegetation; wild cereals and grasses with no obvious human plotting (i.e., not a plotted field); examples: natural meadows and fields with sparse to no tree cover, open savanna with few to no trees, parks/golf courses/lawns, pastures. Mix of small clusters of plants or single plants dispersed on a landscape that shows exposed soil or rock; scrub-filled clearings within dense forests that are clearly not taller than trees; examples: moderate to sparse cover of bushes, shrubs and tufts of grass, savannas with very sparse grasses, trees or other plants.

NOTE: Land use focus does not provide the spatial detail of a land cover map. As such, for the built area classification, yards, parks, and groves will appear as built area rather than trees or rangeland classes.


Usage Information and Best Practices

Processing Templates
  • This layer includes a number of preconfigured processing templates (raster function templates) to provide on-the-fly data rendering and class isolation for visualization and analysis. 
  • Each processing template includes labels and descriptions to characterize the intended usage. This may include for visualization, for analysis, or for both visualization and analysis.

 

Visualization

  • The default rendering on this layer displays all classes.
  • There are a number of on-the-fly renderings/processing templates designed specifically for data visualization.
  • By default, the most recent year is displayed. To discover and isolate specific years for visualization in Map Viewer, try using the Image Collection Explorer.

 

Analysis

  • In order to leverage the optimization for analysis, the capability must be enabled by your ArcGIS organization administrator. More information on enabling this feature can be found in the ‘Regional data hosting’ section of this help doc.
  • Optimized for analysis means this layer does not have size constraints for analysis and it is recommended for multisource analysis with other layers optimized for analysis. See this group for a complete list of imagery layers optimized for analysis.
  • Prior to running analysis, users should always provide some form of data selection with either a layer filter (e.g. for a specific date range, cloud cover percent, mission, etc.) or by selecting specific images. To discover and isolate specific images for analysis in Map Viewer, try using the Image Collection Explorer.
  • Zonal Statistics is a common tool used for understanding the composition of a specified area by reporting the total estimates for each of the classes. 

 

General

  • If you are new to Sentinel-2 LULC, the Sentinel-2 Land Cover Explorer provides a good introductory user experience for working with this imagery layer. For more information, see this Quick Start Guide.
  • Global land use/land cover maps provide information on conservation planning, food security, and hydrologic modeling, among other things. This dataset can be used to visualize land use/land cover anywhere on Earth.

 

Classification Process

These maps include Version 003 of the global Sentinel-2 land use/land cover data product. It is produced by a deep learning model trained using over five billion hand-labeled Sentinel-2 pixels, sampled from over 20,000 sites distributed across all major biomes of the world.

The underlying deep learning model uses 6-bands of Sentinel-2 L2A surface reflectance data: visible blue, green, red, near infrared, and two shortwave infrared bands. To create the final map, the model is run on multiple dates of imagery throughout the year, and the outputs are composited into a final representative map for each year.

The input Sentinel-2 L2A data was accessed via Microsoft’s Planetary Computer and scaled using Microsoft Azure Batch.

 
Citation
Karra, Kontgis, et al. “Global land use/land cover with Sentinel-2 and deep learning.” IGARSS 2021-2021 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2021.

Acknowledgements

Training data for this project makes use of the National Geographic Society Dynamic World training dataset, produced for the Dynamic World Project by National Geographic Society in partnership with Google and the World Resources Institute.
Étiquettes / Mots-clés
Mots-clés/étiquettes catégorisant le jeu de données.
Format (CSV, XLS, TXT, PDF, etc.)
Format de fichier du jeu de données.
Taille du jeu de données
Taille du jeu de données en mégaoctets.
0.09
Identifiant des métadonnées
Identifiant des métadonnées — peut être utilisé comme identifiant unique pour l’entrée du catalogue.
Date de publication
Date de publication du jeu de données.
2022-10-18
Période temporelle des données (date de début)
Date de début des données dans le jeu de données.
Période temporelle des données (date de fin)
Date de fin des données temporelles du jeu de données.
Zone géospatiale couverte
Région spatiale ou lieu nommé couvert par le jeu de données.
Champ Valeur
Catégorie d’accès
Type d’accès accordé pour le jeu de données (ouvert, restreint, service, etc.).
public (affichage de la valeur pour English)
Licence
Licence utilisée pour accéder au jeu de données.
Creative Commons Attribution
Restrictions d’utilisation
Restrictions d’utilisation des données.
Emplacement
Emplacement du jeu de données.
https://climate.esri.ca/datasets/esri::sentinel-2-10m-land-use-land-cover-time-series-1 (affichage de la valeur pour English)
Service de données
Service de données pour accéder à un jeu de données.
Propriétaire
Propriétaire du jeu de données.
esri_imagery (affichage de la valeur pour English)
Point de contact
Qui contacter concernant l’accès ?
Esri (affichage de la valeur pour English)
Courriel du point de contact
Courriel à utiliser pour l’accès ?
Éditeur
Éditeur du jeu de données.
Esri (affichage de la valeur pour English)
Courriel de l’éditeur
Adresse courriel de l’éditeur.
Auteur
Auteur du jeu de données.
esri_imagery
Courriel de l'auteur
Adresse courriel de l’auteur.
Date d’accès
Date à laquelle les données et métadonnées ont été consultées.
2026-01-24
Champ Valeur
Identifiant
Identifiant unique du jeu de données.
cfcb7609de5f478eb7666240902d4d3d (affichage de la valeur pour English)
Langue
Langue(s) du jeu de données
English
Lien vers la description du jeu de données
Une URL vers un document externe décrivant le jeu de données.
https://climate.esri.ca/datasets/esri::sentinel-2-10m-land-use-land-cover-time-series-1 (affichage de la valeur pour English)
Identifiant persistant
Les données sont identifiées par un identifiant persistant.
Identifiant globalement unique
Les données sont identifiées par un identifiant persistant et globalement unique.
Contient des données sur des individus
Le jeu de données contient-il des données sur des individus ?
Contient des données identifiables sur des individus
Le jeu de données contient-il des données identifiables sur des individus ?
Contient des données autochtones
Le jeu de données contient-il des données sur des communautés autochtones ?
Type de portail
Type de plateforme du portail source.
Champ Valeur
Version
Version du jeu de données
None
Source
Source du jeu de données.
None
Notes de version
Notes de version à propos du jeu de données.
Est une version d’un autre jeu de données
Lien vers le jeu de données dont celui-ci est une version.
Autres versions
Lien vers les jeux de données qui en sont des versions.
Texte de provenance
Texte de provenance des données.
Esri (affichage de la valeur pour English)
URL de provenance
URL de provenance des données.
Résolution temporelle
Décrit la granularité des données temporelles dans le jeu de données.
Résolution géospatiale en mètres
Décrit la granularité (en mètres) des données géospatiales dans le jeu de données.
Résolution géospatiale (par régions)
Décrit la granularité (par régions) des données géospatiales dans le jeu de données.
Champ Valeur
Permission de la communauté autochtone
Qui détient la permission de la communauté autochtone. Qui contacter pour l’accès à un jeu de données concernant des communautés autochtones.
Permission communautaire
Permission communautaire (qui a accordé la permission).
Les communautés autochtones concernées par le jeu de données
Communautés autochtones dont sont issues les données.
Champ Valeur
Nombre de lignes de données
Pour un jeu de données tabulaire, nombre total de lignes.
8
Nombre de colonnes de données
Pour un jeu de données tabulaire, nombre total de colonnes uniques.
19
Nombre de cellules de données
Pour un jeu de données tabulaire, nombre total de cellules contenant des données.
152
Nombre de relations de données
Pour un jeu de données RDF, nombre total de triplets.
Nombre d’entités
Pour un jeu de données RDF, nombre total d’entités.
Nombre de propriétés de données
Pour un jeu de données RDF, nombre total de propriétés uniques utilisées par les triplets.
Qualité des données
Décrit la qualité des données du jeu de données.
Métrique de qualité des données
Une métrique utilisée pour mesurer la qualité des données, telle que les valeurs manquantes ou les formats invalides.

0 Commentaires

Please login or register to comment.