Leveraging AI and Precision Agriculture to Restore Irrigation Infrastructure Damaged by El Niño and La Niña Events: A Case Study of Mwogo Marshland, Rwanda
Bibliographic record
Abstract
The Mwogo Irrigation Scheme, launched in 2007 in Huye District, Southern Rwanda, was developed into a climate-resilient agricultural hub through a €20 million investment by Welthungerhilfe and its development partners—including EKN, BMZ, VcA, and the Canadian Embassy—until 2014. This extensive effort transformed the marshland into one of the country’s flagship rice production zones, supporting over 2,393 smallholder farmers organized into five cooperatives. However, the devastating effects of the 2024–2025 El Niño, coupled with the forecasted La Niña, have severely compromised this legacy. Key infrastructure, such as the Gatindingoma Dam, Kabakobwa Intake, and Ntaruka canal section, along with water wells for safe community drinking water, collapsed due to extensive flooding, erosion, and siltation. Over 300 hectares of rice land were left without irrigation, leading to a drop in yields from 5.0 to 2.9 tons/ha and threatening local food security and rural livelihoods. This study employs a combination of field evidence, Normalized Difference Vegetation Index (NDVI)-based crop stress analysis, and Random Forest machine learning algorithms to estimate yield loss and identify hotspots of agricultural damage. Furthermore, it proposes a comprehensive recovery strategy centered on smart irrigation systems, Internet of Things (IoT)-enabled monitoring, AI-based early warning systems, and climate-indexed insurance products. The findings call for urgent re-engagement by Welthungerhilfe and its development partners—both former and prospective—to reinvest in the Mwogo Marshland and ensure its transformation into a resilient, technology-enabled agricultural zone.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".