Design streamflow estimation for ungauged catchments in Eastern NSW : identification of important predictor variables

James Pirozzi, Ataur Rahman

    Research output: Chapter in Book / Conference PaperConference Paperpeer-review

    Abstract

    Peak flow estimation is frequently required in water engineering applications. For small to medium sized ungauged catchments this is generally obtained using regional flood estimation technique. The most commonly adopted regional flood estimation technique in Australia has been the Probabilistic Rational Method (PRM) which was recommended in Australian Rainfall and Runoff in 1987. Availability of additional data and advancement in regional flood estimation methods since 1987 warrants replacement of the PRM by more statistically and hydrologically meaningful regional flood frequency analysis method. This paper presents a study in that a simple form of Quantile Regression Technique is tested using a data from 56 catchments in Eastern NSW. It has been found that a prediction equation based on only two predictor variables (catchment area and rainfall intensity) is capable of providing quite accurate peak flow estimation for Eastern NSW. The particular advantage of this model as compared to the existing PRM is that this is based on a sounder statistical basis and this does not involve use of runoff coefficient maps.
    Original languageEnglish
    Title of host publicationProceedings of Ozwater '10, Australia's National Water Conference and Exhibition, Held in Brisbane, Queensland, 8-10 March 2010
    PublisherAustralian Water Association
    Number of pages8
    ISBN (Print)9781921335099
    Publication statusPublished - 2010
    EventOzwater Convention and Exhibition -
    Duration: 8 Mar 2010 → …

    Conference

    ConferenceOzwater Convention and Exhibition
    Period8/03/10 → …

    Fingerprint

    Dive into the research topics of 'Design streamflow estimation for ungauged catchments in Eastern NSW : identification of important predictor variables'. Together they form a unique fingerprint.

    Cite this