The paper addresses the limitations of fragmented and delayed hydrological information systems in supporting timely disaster risk mitigation. The paper introduces the RoWaterAPI, a framework that integrates near real-time hydrological measurements with geospatial analytics to improve awareness during flood-related events. The methodology utilizes open-source technologies, including Django, Kafka, and PostGIS, to support scalable data ingestion and hazard mapping. Initial baseline evaluation under a simulated bursty workload indicates an end-to-end latency of ≈1–3 s and a peak throughput of ≈6000–8500 messages/s. This performance supports real-time alerts for data variations, bridging advanced geoprocessing with user-centered design for public and institutional stakeholders. Ultimately, RoWaterAPI provides a transferable implementation model that can be adapted to any national context facing similar constraints in data fragmentation and operational accessibility.
Cristian et al. (Thu,) studied this question.