Mapping of forest damage in Mayotte after Cyclone Chido

Remote sensing and artificial intelligence

  • Africa
  • Mayotte

ONF International • 2025

Context

Cyclone Chido: Mayotte’s forests severely affected

Cyclone Chido caused significant damage to Mayotte’s forests, uprooting thousands of trees and weakening the soil. This destruction increases the risk of flooding and threatens the island’s unique biodiversity.

In response to this emergency, the ONF, together with its subsidiary ONF International and its partners, is launching a restoration programme involving reforestation, monitoring of affected areas and sustainable ecosystem management. The objective is clear: to protect the forests, support rural communities and preserve Mayotte’s natural heritage.

Activities carried out

Nine maps were produced to show the damage in each forest in Mayotte, with a clear indication of the level of severity.

The project in figures

7,000 hectares of forest surveyed: using drones and artificial intelligence, windfall areas are mapped and the severity of the damage assessed.

0,25 ha

Precise mapping of Mayotte's state-owned forests, 0.25 ha grid

1 125 ha

Three forests in Mayotte—Combani, Voundze and Dapani—covering a mapped area of 1,125 hectares.

IA

Semi-automatic method → pixel classification to detect felled trees

Project details

At the request of the ONF, ONF International conducted an impact study after Cyclone Chido in Mayotte.

Across nearly 7,000 hectares of state-owned and departmental forests, the team used drones and artificial intelligence (deep learning) to map windfall and classify the severity of the damage. Working closely with the ONF Mayotte, field assessments were used to calibrate and validate the analyses.

©ONF International

Post-cyclone drone image acquisition, Choungui forest, Mayotte

The main phases of the project

1

Drone acquisitions (RGB) as input data, and observation of the cyclone's impact on forests

2

Scientific framing and definition of damage severity classes as a proxy for estimating the observed impact

3

Orientation of field missions to link observed damage to acquisitions and create local descriptive sheets

4

Creation of training labels through unsupervised machine learning on a training area

5

Deep learning model training (Unet) on self-generated labels

6

Calibration of severity thresholds using references from field missions

7

Production of maps (0.25 ha) showing the severity of damage across all forests studied

Expected results

ONF International has mapped the damage caused by Cyclone Chido to the forests of Mayotte.

Accurate maps show the condition of the forests and help to plan their restoration.

Budget
2 X 20 000€
Partners
ONF MAYOTTE
Beneficiaries
ONF MAYOTTE
Donor
ONF

Other projects to discover

TerrAmaz - Amazonian Territories

South America

TerrAmaz - Amazonian Territories

TerrAmaz - Amazonian Territories is a programme that supports the development of Amazonian territories.

See the project
Forland: Monitoring deforestation and encroachment on forests in Ghana

Africa

Forland: Monitoring deforestation and encroachment on forests in Ghana

Development of a deforestation monitoring system for B-BOVID in Ghana, including mapping, risk analysis, a digital platform and training to support RSPO certification.

See the project
Technical assistance to Algeria to promote sustainable forest management and fire prevention

Africa

Technical assistance to Algeria to promote sustainable forest management and fire prevention

Support for Algeria to strengthen sustainable forest management, improve fire prevention, establish an integrated fire management system and develop forestry expertise.

See the project