07 Dec 2024

The destruction of natural resources in Africa has escalated at an alarming rate, but without the necessary measurement and monitoring to draw the necessary conclusions. Africa had the largest annual rate of net forest loss in 2010–2020, at 3.9 million hectares. This rate has increased in each of the three decades since 1990. 

Utilisation of Digital Earth Africa maps and Digital Earth Africa Sandbox, are giving geospatial scientists from the African Regional Institute for Geospatial Information Science and Technology (AFRIGIST), the tools to call attention to this regressive evolution as it specifically affects forested land in Benin. AFRIGIST is a key implementing partner of Digital Earth Africa, helping drive uptake and awareness of Digital Earth Africa's accessible, analysis-ready tools and services. 

Research undertaken by a group of academics at AFRIGIST, with contribution by Digital Earth Africa, entitled, Remote Sensing-based Analysis of Phenology Metrics in the Ouémé-Boukou Reserve Forest (OBRF), Benin Republic examined the phenology metrics of the OBRF using Earth Observation data. Phenology, the study of periodic phenomena in plant growth and development influenced by seasons and living environments, is a critical tool for understanding the complex interactions between plants and their surroundings.

The study utilised Sentinel 1 data to analyse phenology statistics, while a cropland mask extracted from Digital Earth Africa's cropland extent map was employed to estimate the location of croplands within the forest before analysing the metrics. The presence and amount of vegetation in the forest was measured using the Radar Vegetation Index (RVI) calculated using XR_Phenology under the Jupiter notebook of the Sandbox. RVI was plotted over time between January 2019 and January 2021, generating per-pixel-based phenology statistics related to key parameters such as the start of the season (SOS), peak of the season (POS), end of the season (EOS), value at the start of the season (vSOS), value at peak of the season (vPOS), value at end of the season (vEOS), length of the season (LOS), amplitude of season (AOS), rate of greening (ROG), and rate of senescence (ROS).

The application of the Digital Earth Africa cropland mask revealed that a significant portion of the forest is being unlawfully used for agricultural activities. Forests in Benin, as in the rest of Africa, are affected by agricultural expansion which includes over-grazing and clearing fires. The effects on biodiversity, erosion and even poverty are well-known. The World Bank reports that in Africa, forest-related income can lift 11% of rural households out of extreme poverty and forests can sustainably enhance food security.

NDVI image of the study area

The RVI analysis showed a decline in vegetation phenology in May 2019, coinciding with the end of the dry season and the beginning of the rainy season. The growing season from May 2019 to January 2020 exhibited a bimodal pattern, with two peaks observed in August and December 2019. This indicates that any afforestation activity (in other words, re-establishing trees) would be more suitable within that period. From January 2020 to January 2021, the same trend was observed with a slight shift of the peak in December.

geomedian image

A recent report by the Food and Agriculture Organisation (FAO) on World Forest Resource Assessment reveals a regressive evolution of forests, with a negative growth rate of -0.19, -0.13, and -0.12 for the decades of 1990-2000, 2000-2010, and 2010-2020, respectively. This troubling trend is particularly evident in the major part of Africa's tropical forest, necessitating the establishment of a monitoring system to provide essential information for the conservation and management policies of African forests. 

The research findings of the Ouémé-Boukou Reserve Forest emphasise the importance of active remote sensing in providing valuable information for the monitoring of forest reserves, with the potential to aid conservation efforts and inform sustainable management practices. 

The research findings, led by Joseph Oloukoi, were presented at the 14th AARSE International Conference held in Abidjan, in November 2024. Contributors to the research include Momodou Soumah, Hubert O. Yadjemi, and Sena Ghislain C. Adimou of AFRIGIST, and Edward Boamah, technical manager at Digital Earth Africa. Going forward, this research will be published in paper format to be submitted for inclusion in the official AARSE conference publication.