Innovative approaches to mapping forest change and direct drivers in Central Africa using the SEPAL platform

The project team finally met in person in July 2024 in Kinshasa to present the project results at the Congo Basin Forest Partnership (CBFP) 20th Meeting of Parties.
©FAO
In 2020, during the uncertainties of the COVID-19 pandemic, a bold idea emerged: estimate deforestation trends and associated drivers of deforestation across six Central African countries using FAO’s SEPAL (System for Earth Observation Data Access, Processing and Analysis for Land Monitoring) platform. What began as an ambitious concept soon evolved into a multi-year effort to generate data, uncover complex drivers of forest change, and build capacity to support national forest monitoring and reporting.
Project inception in times of COVID-19
Launched during the height of the pandemic, the project was marked by remote collaboration, logistical hurdles, and tight timelines. It aimed to assess deforestation and its drivers across more than 4 million square kilometers, an area larger than the European Union.
Early on, the team recognized the need for strong leadership to steer the project. Aurélie Shapiro, then transitioning from the World Wide Fund for Nature, joined FAO’s Forest Monitoring and Data Platforms team as project manager. Her arrival energized the effort, helping guide six countries through the complexities of forest monitoring under global pandemic restrictions, which limited meetings, workshops and trainings to online interactions.
The project was organized into four interconnected work packages:
- Assessing deforestation and degradation using SEPAL for satellite time series analysis. The methodology evolved over time from wall-to-wall processing to sample-based analysis. Over 40 analysts collaborating on the project processed and visually interpreted data in a few months, generating the first regional map of deforestation and degradation for the 2015-2020 period.
- Analyzing direct and indirect drivers of deforestation through visual interpretation, enabling nuanced understanding of multiple, overlapping drivers at regional, national, and subnational levels.
- Conducting socio-economic surveys across five of the six Central African countries in the study captured local perceptions and realities, validating remote sensing findings and enriching the analysis with ground-level insights.
- Knowledge management and dissemination through reports, live dashboards, and interactive maps were developed to ensure the results reached key stakeholders. All outputs aimed to support transparency, usability, and national ownership of data.
The success of this project relied upon strong institutional relationships, at the national level, with FAO offices, but also through strong collaboration between national governments, which was fostered by the Central Africa Forest Initiative (CAFI), the project resource partner.
Delivering results at scale: Leveraging SEPAL, collaboration and resilience
The project's accomplishments include:
- Comprehensive maps and data: Forest masks, fragmentation maps and visualizations of deforestation and degradation trends, classified by year, change type and forest type.
- Statistical validation: Over 63,000 systematically sampled points were validated through visual interpretation to ensure accuracy.
- Granular reporting: Results were delivered at both national and subnational levels, supporting targeted policy action.
- Project extensions: With additional funding, the project developed new datasets, refined methods, and expanded its analytical reach.
- Transparency: All data are publicly available and accessible, encouraging uptake and common baselines.
- Emissions estimates: Using FAO and the Intergovernmental Panel on Climate Change (IPCC) tier 1 and 2 data and FAO’s Environmental eXternalities ACcounting (Ex-ACT) tool, the team calculated forest emissions related to deforestation and degradation.
Countries like Cameroon and Congo are now exploring the integration of these results into updated forest reference emission levels, illustrating the project’s potential to inform official reporting, climate financing and action.
Scaling impact from Central to West Africa
Given the value of the socio-economic surveys and the replicable methodology the team is implementing the approach in West Africa. This includes sharing tested templates and methods to broaden the project's geographic and policy impact.
In support of open data principles, the team is working to upload datasets to FAO’s Data Catalog developing country dashboard for tailored access; and publishing peer-reviewed articles to contribute to global research. Additional Open Foris platforms such as Open Foris Arena and Open Foris Earth Map are also being applied in West Africa.
Despite its complexity, the project stands as a testament to resilience, innovation, and interagency collaboration between six countries, along with international organizations such as the Central African Forest Observatory (OFAC) and the Joint Research Centre (JRC) of the European Commission.
As Aurélie summarized, “What started with a Skype call and a seemingly overwhelming task has now become a benchmark in regional forest monitoring. There’s plenty more to be done, including disseminating our research and data through various channels and peer-review publications, and ensuring data availability via FAO’s platforms and solutions. So, stay tuned for more!”
SEPAL Drivers website (available in English, French and Spanish)
Story Map in three languages summarizing methods and updated results
National and subnational deforestation and degradation area estimates in interactive tableau dashboards and Google Drive.
Emissions calculations in Google Sheets and dashboards
Explore the visually interpreted data in ArcGIS Online and the FAO Data Catalog
The publication "A global methodology for the assessment of direct drivers associated with deforestation and degradation" will be available soon.
Report

Drivers Dashboard
Explore the direct drivers and the type and date of disturbance and impact on various forest types