RESUMO
Vu Gia-Thu Bon (VGTB) river basin is an area where flash flood and heavy flood events occur frequently, negatively impacting the local community and socio-economic development of Quang Nam Province. In recent years, structural and non-structural solutions have been implemented to mitigate damages due to floods. However, under the impact of climate change, natural disasters continue to happen unpredictably day by day. It is, therefore, necessary to develop a spatial decision support system for real-time flood warnings in the VGTB river basin, which will support in ensuring the area's socio-economic development. The main purpose of this study is to develop an online flood warning system in real-time based on Internet-of-Things (IoT) technologies, GIS, telecommunications, and modeling (Soil and Water Assessment Tool (SWAT) and Hydrologic Engineering Center's River Analysis System (HEC-RAS)) in order to support the local community in the vulnerable downstream areas in the event of heavy rainfall upstream. The structure of the designed system consists of these following components: (1) real-time hydro-meteorological monitoring network, (2) IoT communication infrastructure (Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), wireless networks), (3) database management system (bio-physical, socio-economic, hydro-meteorological, and inundation), (4) simulating and predicting model (SWAT, HEC-RAS), (5) automated simulating and predicting module, (6) flood warning module via short message service (SMS), (7) WebGIS, application for providing and managing hydro-meteorological and inundation data, and (8) users (citizens and government officers). The entire operating processes of the flood warning system (i.e., hydro-meteorological data collecting, transferring, updating, processing, running SWAT and HEC-RAS, visualizing) are automated. A complete flood warning system for the VGTB river basin has been developed as an outcome of this study, which enables the prediction of flood events 5 h in advance and with high accuracy of 80%.