article · Scientific African
Opening spillways at major dams like the Weija Dam in Accra, Ghana, can cause sudden and destructive downstream flooding. To address this risk, an early dam spillage warning system was developed by combining geospatial artificial intelligence, local data, the Internet of Things, and geographic information systems. The framework was informed by an evaluation of the existing Weija Dam Emergency Preparedness Plan. A Lagged Feedback Random Forest model replaces conventional rule-based approaches, delivering improved predictive performance, while spatial analysis identifies vulnerable downstream communities and evacuation routes. The resulting system turns predictions into operational intelligence for urban authorities, disaster management bodies, and local residents. It provides a collaborative digital platform intended to strengthen smart city infrastructure, streamline emergency coordination, issue targeted flood alerts, and support community-level risk communication across African urban centres.
Sudden flooding caused by opening dam spillways poses serious threats to nearby urban settlements. By combining machine learning with spatial mapping, early warning systems can give municipal authorities and residents timely, actionable intelligence. This helps disaster response teams plan safer evacuation paths, distribute targeted warnings, and protect vulnerable downstream infrastructure in rapidly expanding African cities.
The framework enables a platform-based early warning service for municipal authorities, water resource managers, and disaster management organisations. Potential applications include integrated risk communication and municipal emergency planning tools. Given that the predictive model was developed and tested using data from the Weija Dam, the technology appears to be applied and tested in an operational case study, requiring further platform integration and sensor connectivity before broad commercial or municipal deployment.
AI-generated from the published abstract. Always read the original work before citing.
Early warning systems are essential in reducing the effects of dam spillage, particularly during events like the opening of spillways at major dams such as the Weija Dam, which can lead to sudden flooding in downstream communities. This study presents a GeoAI-enabled Early Dam Spillage Warning System in support of the resilience of smart cities and urban water governance. The focus of this study is the Weija Dam in Accra, Ghana. A critical assessment and review of the existing Weija Dam Emergency Preparedness Plan informed the development of the GeoAI-enabled Early Dam Spill Warning System. A Lagged Feedback Random Forest model was employed to improve the traditional rule-based method, whilst spatial analysis was used to identify evacuation routes and at-risk downstream communities. The proposed system demonstrates improved predictive performance. This translates the model’s performance into operational intelligence for city authorities, residents, and the disaster management organisation. The results indicate that integrating local data, GeoAI, IoT, and Geographic Information System (GIS) enhanced the preparedness of smart cities and the resilience of urban infrastructure, and supports community-level risk communication regarding dam spill risks in African cities. The GeoAI-enabled early warning system can function as a collaborative platform-based core infrastructure for smart cities, bringing together various stakeholders for planning, targeted alerts, evacuation, and coordinated emergency response.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1016/j.sciaf.2026.e03591
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
