Mangrove Forest Change in Cacheu, Guinea-Bissau from 2020 to 2025

123,000 hectares of mangrove forest declined by 13% to 107,000 hectares in Cacheu province, Guinea-Bissau, between 2020 and 2025. This dataset was created by Leo Williams using Sentinel-2 satellite imagery and a machine learning classification tool in Google Earth Engine, achieving an overall accuracy above 95%. The mapping reveals opposing regional dynamics of loss at forest margins in the east and canopy densification in the west.

Use Cases

• Quantifying mangrove deforestation and recovery rates based on the reported 13% net loss and regional dynamics.
• Training or validating machine learning models for land cover classification based on the high-accuracy Sentinel-2 classifications.
• Analyzing spatial patterns of environmental change based on the mapped canopy density classes and non-mangrove land cover.
• Assessing the impact of historical drivers like rice paddies on mangrove extent based on the stability analysis mentioned.

Strengths

• High reported classification accuracy above 95% for both time steps.
• Clear temporal coverage from 2020 to 2025 for change detection.
• Specific geographic focus on Cacheu province in Guinea-Bissau.
• Methodology details include the use of Sentinel-2 imagery and a defined machine learning tool (GEM).

Limitations

• Validation relied exclusively on remote sensing interpretation with no field-based ground truthing.
• The description metadata is limited; actual data structure, file formats, and size are unknown.
• Reported high accuracies may reflect the classifier's internal consistency rather than true land cover correspondence.

Provenance

Source

Borealis Harvested Dataverse

Collection Method

Remote sensing analysis using Sentinel 2 imagery classified via the Global Mangrove Mapping Methodology (GEM) in Google Earth Engine.

Time Range

2020 to 2025

Freshness

Last updated 2026-05-02 04:11:13.

Geography

Cacheu province, Northern Guinea-Bissau, West Africa