Le Parc national de la Salonga (PNS) et ses ressources biologiques revêtent une importance considérable tant sur le plan environnemental qu'économique pour les communautés riveraines qui dépendent de ses ressources pour leur survie. Cette étude vise à quantifier l'impact des activités humaines sur les forêts primaires de la partie sud du PNS en utilisant la classification d'images Landsat des années 2002, 2007, 2014 et 2020, combinée aux outils d'analyse de l'écologie du paysage. Les résultats ont révélé que le paysage reste prédominé par les forêts primaires entre 2002 et 2020, bien que leur superficie ait constamment diminué au profit des forêts secondaires et des zones agricoles. En effet, la superficie des forêts primaires est passée de 12 636,82 km² en 2002 à 11 266 km² en 2020. La perte totale des forêts primaires au cours de cette période s'élève à 10,4 %, soit un taux annuel de déforestation de 0,6 %. Le nombre total de fragments de la forêt primaire a également légèrement diminué, passant de 525 142 en 2002 à 524 584 en 2020. De plus, les résultats ont également révélé que la proportion de forêts primaires augmente à mesure que l'on s'éloigne des villages, indiquant un impact significatif des zones villageoises sur la dégradation forestière. L'anthropisation de la partie sud du PNS se manifeste clairement par la réduction des surfaces de forêts primaires, principalement due aux activités agricoles. Ces résultats soulignent la nécessité d'un suivi régulier des activités humaines au sein du PNS, ainsi que le développement d'une stratégie de gestion intégrée pour préserver cette zone d'une importance écologique cruciale.
This study maps and quantifies the green infrastructure of the cities of Bujumbura, Kinshasa and Lubumbashi. A remote sensing approach, in conjunction with landscape ecology metrics, characterized this analysis, which was based upon three Landsat images acquired in 2000, 2013 and 2022 for each city. Spatial pattern indices show that green infrastructure was suppressed in the cities of Bujumbura and Kinshasa, in contrast to the city of Lubumbashi, which was characterized by a fragmentation of its green areas. Additionally, the values of the stability, aggregation and fractal dimension metrics suggest that the city of Bujumbura underwent rather intense dynamics and a reduction in the continuity of its green areas, while the city of Kinshasa shows weaker dynamics and tendances of patch aggregation in the study period. The city of Lubumbashi, in its turn, exhibits strong dynamics and an aggregation of its green infrastructure, in a context of a high level of anthropization. The evolution of the Normalized Difference Vegetation Index shows a sawtooth evolution of tall vegetation patches for the city of Bujumbura, against a gradual decrease in Kinshasa and Lubumbashi. It is advocated that the urban growth of these cities should be planned carefully in order to integrate the presence of sufficient green infrastructure.
The Luilu sector, located in the Katangan Copperbelt (southeastern DR Congo), has expe-rienced intensified extractive activities over several decades. While this dynamic contrib-utes to national economic growth, it generates environmental and social impacts that re-main insufficiently documented, particularly regarding landscape structure and resili-ence. This study examines changes in land cover and spatial organization between 1990 and 2024 under increasing anthropogenic pressures. The methodology combines a mul-ti-temporal cartographic approach with landscape ecology indices (composition, frag-mentation, connectivity, spatial complexity) applied to satellite imagery. The analysis re-veals a significant decline in natural formations, notably Miombo woodlands (from 50.83% to 38.89%), correlated with the rapid expansion of agricultural (from 4.25% to 13.41%) and urban areas (from 0.64% to 5.05%), primarily driven by shifting cultivation and urbanization. Ecological indices indicate growing instability, increased fragmenta-tion, and reduced ecosystem resilience. The largest patch index shows a reduction in for-est dominance (29.62% to 24.34%), while fractal analysis highlights rapid, disorganized, and spatially complex urban expansion. These dynamics reflect a rapid, unplanned landscape reconfiguration, undermining ecological connectivity and reinforcing regional socio-environmental vulnerability. The study underscores the need to integrate landscape dynamics into local land management policies in order to balance mining development, food security, and ecosystem conservation.
Forests of Lualaba Province (DR Congo) form a compositionally complex mosaic of dry dense forest, gallery forest, and Miombo woodland. Yet, categorical land-cover maps impose discrete boundaries on these inherently continuous vegetation gradients, systematically discarding subpixel compositional information critical for forest monitoring and carbon accounting. The magnitude of this information loss at the landscape scale, however, remains largely unquantified. In this study, we train a Multi-Output Neural Network (MONN) using Sentinel-2 spectral and textural predictors (2025) to estimate the proportional cover of three forest types across the province. Model performance is benchmarked against a normalised Random Forest (RF) using spatial block cross-validation. Categorical information loss is quantified pixel-wise using two complementary metrics, dominant class proportion and Shannon compositional entropy, alongside a derived interpretive quantity, categorical information loss. The MONN slightly outperformed RF (R² = 0.648 vs 0.630; RMSE = 0.224 vs 0.229), yet the results reveal a fundamentally heterogeneous landscape structure. The mean dominant-class proportion was only 56.2%, indicating that categorical maps discard, on average, 43.8% of compositional information per pixel. Only 7.9% of forested pixels exceeded the 75% dominance threshold, while Shannon entropy reached 74.1% of its theoretical maximum, indicating that forest types coexist in near-equal proportions across most pixels. This renders categorical attribution structurally inadequate for most of the forested landscape. Across 92.1% of forested pixels, no single forest type achieved clear dominance. These results show that compositional mixing is the dominant structural condition of the landscape, and that compositional mapping is essential for representing tropical forest structure in heterogeneous drylands. By formally quantifying categorical information loss at the landscape scale, this study shows that continuous compositional mapping converts this structural ambiguity into a spatially explicit ecological signal, with direct implications for monitoring vegetation dynamics and biodiversity, highlighting a structural source of error in carbon stock estimation in tropical dry forests.
The Luilu sector, located in the Katangan Copperbelt (southeastern DR Congo), has experienced intensified extractive activities over several decades. While this dynamic contributes to national economic growth, it generates environmental and social impacts that remain insufficiently documented, particularly regarding landscape structure and resilience. This study examines changes in land use and spatial organization between 1990 and 2024 under increasing anthropogenic pressures. The methodology combines a multi-temporal cartographic approach with landscape ecology indices (composition, fragmentation, connectivity, spatial complexity) applied to satellite imagery. The analysis reveals a significant decline in natural formations, notably Miombo woodlands (from 50.83% to 38.89%), correlated with the rapid expansion of agricultural (from 4.25% to 13.41%) and urban areas (from 0.64% to 5.05%), primarily driven by shifting cultivation and urbanization. Ecological indices indicate growing instability, increased fragmentation, and reduced ecosystem resilience. The largest patch index shows a reduction in forest dominance (29.62% to 24.34%), while fractal analysis highlights rapid, disorganized, and spatially complex urban expansion. These dynamics reflect a rapid, unplanned landscape reconfiguration, undermining ecological connectivity and reinforcing regional socio-environmental vulnerability. The study underscores the need to integrate landscape dynamics into local land management policies in order to balance mining development, food security, and ecosystem conservation.