Annual Cropland and Landcover Maps for Africa from Planet Imagery

High-resolution annual maps for selected African countries developed by Clark University's Agricultural Impacts Research Group. The maps include field boundaries, cultivated frequency, and multi-class land cover, generated by applying machine learning approaches to Planet satellite imagery. The data is hosted on AWS Open Data and is licensed under the Planet NICFI participant agreement.

Use Cases

• Monitor agricultural expansion and contraction based on annual cropland maps.
• Analyze land-use change patterns based on multi-class land cover classifications.
• Estimate cultivation intensity and field sizes based on field boundary and cultivated frequency maps.
• Train or validate machine learning models for remote sensing based on labeled satellite imagery products.

Strengths

• High-resolution imagery from Planet satellites.
• Includes multiple derived products: field boundaries, cultivated frequency, and land cover.
• Annual time series allows for change detection.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Clark University's Agricultural Impacts Research Group

Collection Method

Machine learning approaches applied to Planet satellite imagery.

Time Range

Annual, but specific years are unknown.

Freshness

Annual maps, but specific publication years are unknown.

Geography

Selected African countries.