Propagation of input errors: implications for model simulations and risk analysis

Kerrie M. Tomkins*, Andrew J. Davidson

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

3 Citations (Scopus)

Abstract

Hydrological models are widely used in water resources management in Australia. Unpinning these models is streamflow data, which is commonly used as inputs, for calibration of parameters and, for verification of model performance. One of the lesser investigated issues in modelling and uncertainty analysis is how the choice of error models impacts on simulations, and how this propagates into decision-making where simulations are used to determine the volume, frequency and reliability of flows. We used an analysis of the deviations in gaugings from flow gauges in the Namoi River catchment, to derive empirically-based error models for the data. The error models were used to generate uncertainty in tributary and residual inflows in the Namoi River Integrated Quality and Quantity Model (IQQM). Several scenarios were run, including empirically-derived best-fit, empirically-derived Gaussian and standard Gaussian error models, with reference to a baseline simulation where the data are assumed to be error free. Analysis of end-of-system flows showed that there was no conclusive difference in the effect of the error models; however, this was likely to be due to the addition of random rather than auto-correlated errors, which arise from fitting of rating curves to gaugings. This study highlights the need for further investigation into rating curve uncertainty, error autocorrelation and sampling of error models.

Original languageEnglish
Title of host publicationRisk in Water Resources Management
Subtitle of host publicationproceedings of symposium H03 held during IUGG2011
EditorsG. Bloschl, K. Takeuchi, S. Jain, A. Farnleitner, A. Schumann
Place of PublicationWallingford, Oxfordshire, UK
PublisherInternational Association of Hydrological Sciences
Pages101-106
Number of pages6
Volume347
ISBN (Print)9781907161223
Publication statusPublished - 2011
Externally publishedYes
EventSymposium H03 on Risk in Water Resources Management, Held During the 25th General Assembly of the International Union of Geodesy and Geophysics, IUGG 2011 - Melbourne, VIC, Australia
Duration: 28 Jun 20117 Jul 2011

Other

OtherSymposium H03 on Risk in Water Resources Management, Held During the 25th General Assembly of the International Union of Geodesy and Geophysics, IUGG 2011
CountryAustralia
CityMelbourne, VIC
Period28/06/117/07/11

Keywords

  • Namoi River
  • rating curve
  • river model
  • uncertainty

Fingerprint Dive into the research topics of 'Propagation of input errors: implications for model simulations and risk analysis'. Together they form a unique fingerprint.

Cite this