Description
This dataset classifies the land cover class / habitat type present on the land surface at 20 m or 25 m spatial resolution, with national coverage across 6 tropical countries with annual temporal resolution and varying temporal coverage. Pixels are assigned one of 8+ land cover classes, depending on country.
Usage notes
Spatial coverage is available for 6 countries, with spatial resolution and temporal coverage differing among them (see Variables section for details). Use of this dataset for EUDR compliance must be made through the ICE CoT system. This dataset will no longer be updated, with updates instead being made to the Land Cover 10 m dataset from the same provider.
Licence
Commercial
Version
1.0
Release date
Coming soon
CRS
EPSG:6933
Update schedule
None
Provider cost
TBC
Description
Land cover index provides the index of the predicted land cover class for a pixel. Predictions are made using machine learning models trained to predict expert annotated polygons from multiple satellite data sources.
Usage notes
Annotation polygons and model predictions are designed to achieve high accuracy over challenging land cover transitions (e.g. tree crops and natural forest). The dataset employs a minimum of 8 land cover classes, with additional classes (e.g. dense forest, open forest) also being employed for some countries.
Variable category
Land cover class
Variable type
Integer