Trees sustain livelihoods and mitigate climate change but a predominance of trees outside forests and limited resources make it difficult for many tropical countries to conduct automated nation-wide inventories. Here, we propose an approach to map the carbon stock of each individual overstory tree at the national scale of Rwanda using aerial imagery from 2008 and deep learning. We show that 72% of the mapped trees are located in farmlands and savannas and 17% in plantations, accounting for 48.6% of the national aboveground carbon stocks. Natural forests cover 11% of the total tree count and 51.4% of the national carbon stocks, with an overall carbon stock uncertainty of 16.9%. The mapping of all trees allows partitioning to any landscapes classification and is urgently needed for effective planning and monitoring of restoration activities as well as for optimization of carbon sequestration, biodiversity and economic benefits of trees.

This study aimed at analyzing the spatio-temporal patterns of urban growth and its effects on Rwampara wetland, located in the City of Kigali, Rwanda. First, the study was based on the application of remote sensing technology, where 4 Landsat images (1987, 1999, 2009 & 2018) were classified using maximum likelihood classification algorithm. This helped in analyzing the Land Use and Land Cover (LULC) trends in the study area. Secondly, it used the existing LULC data for the years 1990, 2000, 2010 and 2018 in order to investigate the overall changes in LULC in Kigali City. Finally, semi-structured interviews were used to collect data from local people and decision-makers about their past and future management strategies of Rwampara wetland. In this regard, 30 local communities (mainly natives from the study area), 15 local government authorities at sector and district levels as well as 4 senior government authorities in the central administration were interviewed. The findings revealed that over the past 4 decades, urban growth in Kigali City has rapidly increased at the expense of resource degradation in Rwampara wetland. Specifically, there has been an increase of about 77% of the built-up area over the last 31 years (1987-2018) which has led to the decrease of the wetland surface area from 24 ha in 1987 to only 7.7 ha in 2018. The results also revealed that demographic factors (i.e. a high population growth rate and high population densities) were mainly responsible for urban growth and degradation of wetland resources in the area under investigation. Keywords: urban growth, wetland, wetland resources, wetland degradation, wetland management

Autonomous large-scale high-resolution land cover (HRLC) mapping remains a major challenge in remote sensing due to the scarcity of reliable training data and resolution mismatches between available labels and input of massive of emerging imagery. Existing global land cover products often suffer from coarse spatial resolution and label noise, limiting their utility for fine-scale urban analysis and environmental monitoring. This paper presents C2FNet, a novel Coarse-to-Fine Network designed to generate HRLC maps from noisy, coarse-resolution labels using a weak supervision strategy of the cross-probability. The 2FNet consists of three key modules: (1) Edge Resolution Refinement Backbones (ERRB), which preserve spatial detail via multiscale feature extraction through parallel convolutional branches; (2) Unsupervised Dynamic Shuffle and Diagonal Annotation (UDSDA), which enhances training reliability by identifying confident regions through spatial-consistency analysis and confidence estimation; and (3) a Contrasting Self-Supervised Loss (C2F-Loss) that integrates cross-entropy and cosine similarity terms to mitigate supervision noise and resolution gaps. Evaluations of three benchmark data sets that encompass diverse urban and rural landscapes show that C2FNet achieves state of the art performance, with 80. 01% overall accuracy and a Cohen’s kappa score of 0.7567, outperforming state-of-the-art (SoA) models with weak supervision. Dataset and code: Available here.

Autonomous feature extraction from high-resolution (HR) Remote Sensing (RS) imagery presents complex spatial–spectral heterogeneity and diverse land-cover patterns, challenging conventional single-architecture models that lack multiscale feature diversity and cross-domain generalization. Convolutional neural networks (CNNs) effectively capture localized hierarchical structures, whereas transformers model long-range contextual dependencies via self-attention. However, using these architectures independently restricts their complementary potential. This study proposes FWDNNet (Fused Weights Deep Neural Tokenization Networks), a fusion framework introducing a CNN-to-token conversion mechanism that transforms convolutional feature maps from heterogeneous encoders into transformer-compatible token sequences. Beyond standard concatenation, FWDNNet integrates a TensorDot-based fusion module to learn high-order cross-architectural interactions, while a shared decoding pathway maintains multiscale feature hierarchies and computational efficiency. Experiments across three diverse datasets show that FWDNNet achieves 58.2 ms inference latency with minimal cross-domain performance degradation (2.3–4.2% mIoU drop). Compared to conventional fusion strategies, FWDNNet attains 21% faster inference and 1.7–2.7% higher mIoU over baseline methods. See Materials at https://github.com/BoazGithub/FWDNNet for reproducibility.

Kigali fait partie des villes d’Afrique de l’Est ayant connu une évolution spatiale et physique rapide à partir des années 2010. L’artificialisation des terres progresse au détriment des espaces boisés à cause de la densification du bâti et des routes bitumées ou pavées. Ces dynamiques urbaines contribuent au développement des îlots de chaleur. La littérature scientifique positionne les îlots de chaleur comme un enjeu environnemental et sociétal important à cause des impacts sur la santé humaine. Très peu d’études sont cependant menées à l’échelle des villes africaines afin d’identifier et d’étudier ce phénomène dans les politiques et l’aménagement urbains. Le premier objectif de cet article est d’analyser, avec un recul temporel de trois décennies, l’extension urbaine de la ville de Kigali et ses conséquences sur l’évolution des températures de surface. Le deuxième objectif est d’identifier les quartiers les plus concernés par les îlots de chaleur. Nous analysons l’évolution de la tâche urbaine à partir des informations extraites d’images satellitaires (Landsat) de 1984, 2010 et 2021. Des données sur les températures de surface ont été dérivées de ces images. Des données de températures journalières de l’air sur la période 1984-2018 ont également été utilisées afin d’explorer les tendances des conditions thermiques et des îlots de chaleur urbains. Le travail repose également sur des observations de terrain et l’analyse de documents d’archives (photographies et textes sur l’histoire récente archivés à la bibliothèque publique de la ville de Kigali). Les résultats révèlent que l’évolution spatio-temporelle des surfaces chaudes suit la même trajectoire que l’extension de la tâche urbaine, du centre historique vers les limites est et sud de la ville. Même si Kigali est toujours considérée comme une ville verte, grâce à la présence de nombreux espaces boisés et de quelques reliques forestières, certains quartiers, notamment le centre-ville et Remera, très minéralisés, sont presque dépourvus de végétation et apparaissent comme des îlots de chaleur. Les températures journalières minimales de l’air au centre-ville sont toujours plus élevées que celles observées à l’aéroport international de Kigali, situé en périphérie de la ville. Ces différences peuvent atteindre 1,5 °C au cours des années exceptionnelles, comme en 1998, confirmant le développement d’un îlot de chaleur urbain à Kigali. Cet îlot de chaleur pourrait s’amplifier dans le contexte du réchauffement climatique actuel. L’identification et le suivi des surfaces chaudes peuvent être un outil des politiques d’aménagement urbain et de santé publique, notamment par la création d’îlots de fraîcheur et la sensibilisation des populations exposées.

This study presents the development of a Python-based flood-susceptibility risk-mapping tool, implemented in Jupyter Notebook, applied to Rwanda. A Flood Susceptibility Index (FSI) was developed by integrating 20 causal factors associated with flood occurrences, including topographic, hydrological, geological, and anthropogenic variables. Logistic regression, and Variance Inflation Factor were implemented in Python using libraries such as Numpy, Arcpy, traceback, scipy, Pandas, Seaborn, and statsmodel to assign weights to each factor, and to address multicollinearity. The model was validated against flood extent data derived from Sentinel-1 satellite imagery for the major historical flood event that occurred from 2014 to 2024, ensuring spatial consistency and predictive reliability. To project future flood susceptibility for 2030, precipitation data from the Institut Pierre Simon Laplace Coupled Model, version 5A, Medium Resolution (IPSL-CM5A-MR) climate model under the Representative Concentration Pathway 8.5 (RCP 8.5) scenario were utilized. The resulting FSI was classified into five susceptibility levels, from very low to very high, and visualized using Python’s geospatial and plotting tools within Jupyter Notebook in ArcGIS Pro 3.5. It indicates that areas with high amounts of rainfall, and proximity to wetlands and rivers reveal the highest flood risk. The automated and reproducible approach offered by Python enhances transparency and scalability, providing a decision-support tool for disaster risk reduction and climate adaptation planning in Rwanda.