Flood management in climate change circumstance — a case study in central Vietnam

Do Hoai Nam, Keiko Udo, Akira Mano

    Research output: Chapter in Book/Report/Conference proceedingChapter


    Climate change is likely to affect on most aspects of the society. Increase in temperature has been resulting in change of the hydrological cycle. Extreme precipitation events that usually cause severe floods are expected to increase in most tropical regions. Accurate and timely prediction of runoff is crucial in flood management in order to respond to climate change. Recently, numerical weather prediction (NWP) has demonstrated its usefulness in flood forecasting, particularly the extension of forecast lead time. This study proposes a shortterm flood forecast model for the upper Thu Bon River (3,150 km2) in Central Vietnam, by coupling the global NWP with the super tank model. Given the intrinsic errors of the NWP, flood forecasts based on its direct outputs were not what we had expected. Thus, model output statistic (MOS) approach has been proposed to increase forecast skill through the improved rainfall prediction, by using artificial neural network. As a result, MOS-driven flood forecast has demonstrated potential inputs in development of an early flood warning system that has been considered as substantial benefits in developing countries where weather observation is scarce and access to high resolution NWP is limited.

    Original languageEnglish
    Title of host publicationAdvances in Geosciences
    Subtitle of host publicationVolume 23: Hydrological Science (HS)
    PublisherWorld Scientific Publishing Co.
    Number of pages16
    ISBN (Electronic)9789814355339
    ISBN (Print)9814355321, 9789814355322
    Publication statusPublished - 2011 Jan 1

    ASJC Scopus subject areas

    • Earth and Planetary Sciences(all)
    • Engineering(all)
    • Environmental Science(all)


    Dive into the research topics of 'Flood management in climate change circumstance — a case study in central Vietnam'. Together they form a unique fingerprint.

    Cite this