Climate-smart agriculture (CSA) is one of the innovative approaches for sustainably increasing the agricultural productivity, improving livelihoods and incomes of farmers, while at the same time improving resilience and contributing to climate change mitigation. In spite of the fact that there is neither explicit policy nor practices branded as CSA in Democratic Republic of Congo (DRC), farmers are utilizing an array of farming practices whose attributes meet the CSA criteria. However, the intensity, distribution, efficiency, and dynamics of use as well as the sources of these technologies are not sufficiently documented. Therefore, this review paper provides a comprehensive evidence of CSA-associated farming practices in DRC, public and private efforts to promote CSA practices, and the associated benefits accruing from the practices as deployed by farmers in the DRC. We find evidence of progress among farming communities in the use of practices that can be classified as CSA. Communities using these practices are building on the traditional knowledge systems and adaptation of introduced technologies to suit the local conditions. Reported returns on use of these practices are promising, pointing to their potential continued use into the future. While progressive returns on investment are reported, they are relatively lower than those reported from other areas in sub-Saharan Africa deploying similar approaches. We recommend for strategic support for capacity building at various levels, including public institutions for policy development and guidance, extension and community level to support uptake of technologies and higher education institutions for mainstreaming CSA into curricula and training a generation of CSA sensitive human resources.

Forests offer numerous ecosystem services that enhance human livelihoods. However, without proper regulation, human activities can deteriorate forest ecosystems. This study aims to evaluate the land use and cover dynamics in Mumosho municipality, South Kivu, Democratic Republic of Congo (DRC), and assess the diversity and carbon sequestration capacity of its woody species. We conducted a supervised classification of Landsat images from 1994, 2004, 2013, and 2023 and calculated spatial structure indices to analyze anthropization and deforestation in the area. From 1994 to 2013, the Mumosho landscape experienced significant deforestation, with rates of 3.69% per year (1994–2004) and 0.12% per year (2004–2013), while the anthropization rate increased from 0.45 in 1994 to 1.33 in 2013. However, from 2013 to 2023, forest cover began to recover at a rate of 0.99% per year. The current floristic diversity in the Mumosho forest is notable, with a high Shannon index of 3.59, including 30 woody species from 19 botanical families. Eucalyptus globulus and Grevillea robusta are the most abundant species, with relative frequencies of 25.84% and 17.16%, respectively. Collectively, these woody species contribute an average carbon sequestration capacity of 113.429 ± 91.354 tC/ha. These findings underscore the importance of reinforcing (re)forestation initiatives and sustainable forest management strategies to safeguard ecosystem services and mitigate environmental degradation.

Timely crop monitoring and yield prediction are essential in guiding management decision making. The aim of the study was to estimate the agronomic traits of paddy rice ( Oryza sativa L.) using unmanned aerial vehicle (UAV)‐multispectral imaging. A randomized complete block design field experiment with a split–split plot arrangement was set up in the Ruzizi plain, Democratic Republic of Congo (DRC). Spectral imaging data were collected at rice tillering and panicle initiation stages. Predictive analysis of rice agronomic traits was performed using linear and decision tree‐based machine learning techniques. Paddy rice trait predictions were critically sensitive to the timing of image acquisition but not largely affected by the model. The most accurate predictions were made at rice panicle initiation stage, with R ² values of 0.62, 0.65, and 0.75 for yield, aboveground biomass, and plant nitrogen (N) uptake, respectively. The visible atmospherically resistant index (VARI), modified chlorophyll absorption in reflective index, and ratio vegetation index, along with near infrared and green bands, played a critical role in predicting paddy rice N uptake and yield. The same spectral features associated with crop height and canopy data were essential for predicting paddy rice aboveground biomass. UAV‐multispectral data were able to assess agricultural intensification strategies at field/landscape scale irrespective of soil types, watering regimes, and cultivars. Special consideration should be attributed to VARI, as it enables economical prediction of paddy rice traits. The UAV technologies are therefore reliable tools for monitoring rice production and can be applied in agricultural extension in the DRC.

Background: Soil fertility provides the basis for agricultural productivity and food security, but conventional laboratory analyses remain inaccessible to smallholder farmers in sub-Saharan Africa. In the eastern Democratic Republic of the Congo, farmers rely on locally derived indicators such as weed communities, soil color, crops productivity, and erosion, but their validity remains underexplored. Objective: This study assessed the reliability of farmer-identified soil fertility indicators and examined how they align with laboratory-based measurements, with the goal of integrating local knowledge into practical fertility assessments in Lurhala and Kamisimbi, during Season A 2019 and Season A 2023, corresponding to the main cropping period (September–January) in South Kivu. Methods: On 148 plots, we combined farmer surveys, weed inventories and soil analyses including pH, nitrogen (N), phosphorus (P), potassium (K), and organic matter (OM). We constructed a composite Soil Fertility Index (SFI), computed Bray–Curtis (BC) similarities and trained decision-tree classifiers (0.7:0.3 split). Results: Twenty-eight weed species were recorded; Bidens pilosa , Conyza sumantrensis , and Galinsoga ciliata associated with higher fertility, while Imperata cylindrica and Hyparrhenia rufa marked degraded plots. Laboratory means for high-fertility soils were pH = 6.1 ± 0.6, OM = 3.9 ± 1.5% and K = 1.9 ± 0.5 cmol kg ⁻¹ ; low-fertility soils averaged pH = 4.8 ± 0.2, OM = 3.0 ± 1.2% and K = 0.8 ± 0.3 cmol kg ⁻¹ . Overall SFI = 0.514 ± 0.097 (low = 0.341 ± 0.073; medium = 0.537 ± 0.10; high = 0.526 ± 0.086), clearly indicating general degradation and limited separation of farmers’ medium and high classes. Decision trees ranked soil color (53.8%) and weed species (51.6%) highest indicators; model accuracies ranged from 79.7 to 93.2% (overall 82.4%), area under the ROC curve (AUC): 0.96–0.99, indicating reliable relationships between standard and local knowledge. Phosphorus content was unexpectedly higher in farmer-classified low-fertility plots, highlighting complex nutrient dynamics incorporated into farmers’ holistic assessments. Conclusion: Farmer-derived indicators offer cost-effective, context-specific diagnostics but struggle to resolve subtle fertility gradients. We argue that the way forward lies not in choosing one knowledge system over the other, but in forging a synergistic partnership between them.

Waste management in developing countries faces several challenges, including optimizing basic services and integrating sustainable innovations. However, local empirical studies that could guide this transition are often lacking. This study analyses residents’ perceptions of household waste management in the city of Bukavu, comparing traditional methods and automated systems. A probabilistic survey of 442 households across the city was conducted using mobile data collection (KoboCollect). Data were analyzed quantitatively using the Python programming language in the Google Colab environment. While a structured municipal service exists (used by 82.8% of households), satisfaction is only relative (57.2%). Critical health concerns are odors (86.9%) and disease proliferation (84.8%). Residents prioritize increasing collection frequency (78.5%) and providing public bins (76%). Although 53.6% believe their neighborhood is ready for automation, major concerns regarding job losses (66.1%) and cost (59%) persist. The study provides a three-tiered empirical roadmap: it diagnoses the need to optimize an existing service, delivers a citizen-centered action plan for immediate improvements, and defines the social and informational prerequisites for a sustainable automation transition. Success requires improving basic services while managing the perceived socioeconomic risks of innovation.