Land Cover 15-Class

Impact Observatory — version 1.4

Description

This dataset classifies land cover at 10 m spatial resolution, with global coverage for custom time periods (> 3 months) from 2017 onwards. Pixels are assigned one of 15 land cover classes.

Dataset ID
e7a24340
Category
Land use & land cover
Type
Raster
CRS
EPSG:3857
Spatial coverage
Global
Spatial resolution
10 m
Temporal coverage
2017+
Temporal resolution
User-defined

Usage notes

Data are delivered as composites of predictions made on Sentinel-2 observations over a date range. Custom date ranges are permitted, although annual composites are recommended to ensure sufficient Sentinel-2 observations for reliable predictions.

Variables

land_cover_class

TypeInteger

Description

Index of the predicted land cover class. Predictions are made using a deep learning segmentation model trained to predict > 5 billion manually annotated land cover classifications from Sentinel-2 imagery.

Usage notes

Overall accuracy is 76% compared to a majority consensus. Lowest accuracy is for Flooded Vegetation (57%) and Rangeland (58%). Highest accuracy is for Crops (90%) and Water (87%).

IndexNameDescription
11Water Channel ExtentFull extent of water body under normal circumstances (i.e. outside of flooding or other exceptional events). Includes seasonal water bodies.
12Variable WaterIntermittent water flow or standing water due to fluctuation in cover, weather events, or human activity. May also occur due to turbidity, algal blooms, ice cover, pollution, or glare.
13Persistent WaterPermanent water flow or standing water.
30Snow/IceHomogenous areas of permanent snow or ice. Typically only found in mountainous and high latitude regions.
40Bare GroundRock, soil, or sand with little to no vegetation.
61Sparse RangelandVegetation mixes of grasses and/or sparse, short (< 5 m), woody scrub, with or without bare ground. May contain small isolated trees.
62Dense RangelandVegetation dominated by dense, short (< 5 m), woody scrub and little to no grass or bare ground. May contain small isolated trees. Can include lawns or fields when adjacent to built areas.
80Flooded VegetationAny vegetation mixed with water throughout most of the year, including seasonally flooded areas. Thick mangrove, swamp, or seasonal wetland may fall under next-best-fit classes if water is not observed during the given time period.
101Sparse TreesMainly tree canopy cover (> 5 m height), but with mixing of other vegetation or bare ground due to thin leaf cover or dispersed tree cover. Can indicate gain/loss of leaves or dying/growing trees.
102Dense TreesDense tree canopy cover (> 5 m height) with little to no other vegetation or bare ground.
141Inactive CroplandFallow or otherwise inactive fields, sometimes interspersed with small infrastructure.
142Active CroplandActively growing cereals, grasses, irrigated pastures, or other human managed crops.
172Low Density BuiltArtificial impervious surfaces, buildings, and structures mixed with vegetation that cannot be disentangled at 10 m resolution. Typically represents low-density residential/suburban areas or path networks.
174High Density BuiltArtificial impervious surfaces, buildings, and structures with little to no mixed vegetation or bare ground.
240CloudsNo land cover information due to persistent cloud cover.

Pricing

A data request returns a temporal composite of the requested date range. Data acquisition costs are per hectare:

VolumePrice
> 0 ha$0.02 / ha

Resources

Provider

Impact Observatory

www.impactobservatory.com

Impact Observatory is a nature data analytics company based in Washington DC. Impact Observatory uses AI-based approaches to build land cover products that inform environmental analysis for governments, industries, and markets.