Forest Typology 2020: Global 10-Meter Resolution Forest and Land-Use Probability Maps

Forest Typology (ForTy) v1 provides global 10-meter spatial resolution multi-class probability maps for the baseline year 2020. The dataset was developed by Maxim Neumann using a deep learning pipeline trained on over 1.7 million globally distributed samples. It categorizes land cover into six classes aligned with FAO and EUDR definitions, including Primary Forest and Planted Forest.

Use Cases

• Monitoring deforestation and forest degradation based on the high-resolution classification of primary and regenerating forests.
• Assessing compliance with regulations like the EU Deforestation-free Products Regulation (EUDR) based on the aligned forest type definitions.
• Modeling carbon stocks and biodiversity by distinguishing between natural and planted forest types.
• Analyzing agricultural tree cover and agroforestry systems based on the Tree Crops and Agroforestry probability band.

Strengths

• High spatial resolution of 10 meters per pixel for detailed land cover analysis.
• Trained on a large sample of over 1.7 million globally distributed data points.
• Global coverage of land areas between 65 degrees S and 84 degrees N latitude.
• Provides probability scores for each forest class, allowing for uncertainty assessment.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• The 'Other Land' class is not stored as a band and must be derived programmatically, adding a processing step.
• Data freshness should be verified as the baseline year is 2020.

Provenance

Source

Maxim Neumann via figshare.

Collection Method

Developed using a three-stage cascaded deep learning pipeline trained on globally distributed samples.

Time Range

Baseline year 2020.

Freshness

Last updated 2026-05-22 23:22:31; baseline temporal scope is 2020.

Geography

Global land areas between 65 degrees S and 84 degrees N latitude.