Caractérisation de la structure de la canopée à l'échelle de la République Démocratique du Congo.

Abstract

Tropical forests play a crucial role in regulating the global climate and maintaining biodiversity. They act as major carbon sinks but are increasingly threatened by deforestation, climate change, and anthropogenic disturbances. Within these forests, canopy structure—and particularly tree crown surface area—determines light capture efficiency, stand productivity, and ecosystem resilience. While the influence of topography on biomass and tree height has been well documented, its effect on crown surface remains poorly understood, especially in Central Africa. Advances in remote sensing, notably the use of LiDAR data and very high-resolution aerial imagery, now make it possible to finely characterise canopy structure at large scales. The specific issue addressed in this study is the relationship between selected topographic variables (altitude, slope, Topographic Position Index, Topographic Ruggedness Index, and slope aspect) and crown surface in the tropical forests of the Democratic Republic of Congo. Based on a dataset of 8,779 manually delineated crowns, the results show that, when considered individually, TPI, slope, and aspect significantly and non-linearly influence crown surfaces, whereas altitude and TRI appear to exert only marginal effects. These findings reveal that crown expansion is favoured in intermediate slopes and topographic positions, while aspect reflects adaptations to local microclimatic conditions. More broadly, this study provides a basis for training automated crown delineation models, enabling large-scale analyses and improving the modelling of Central African forests role in carbon sequestration and global climate regulation.