Journal metrics

Journal metrics

  • IF value: 4.256 IF 4.256
  • IF 5-year value: 4.819 IF 5-year 4.819
  • CiteScore value: 4.10 CiteScore 4.10
  • SNIP value: 1.412 SNIP 1.412
  • SJR value: 2.023 SJR 2.023
  • IPP value: 3.97 IPP 3.97
  • h5-index value: 58 h5-index 58
  • Scimago H index value: 99 Scimago H index 99
Volume 22, issue 9 | Copyright
Hydrol. Earth Syst. Sci., 22, 5021-5039, 2018
https://doi.org/10.5194/hess-22-5021-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Research article 28 Sep 2018

Research article | 28 Sep 2018

Parameter uncertainty analysis for an operational hydrological model using residual-based and limits of acceptability approaches

Aynom T. Teweldebrhan1, John F. Burkhart1,2, and Thomas V. Schuler1 Aynom T. Teweldebrhan et al.
  • 1Department of Geosciences, University of Oslo, Oslo, Norway
  • 2Statkraft, Oslo, Norway

Abstract. Parameter uncertainty estimation is one of the major challenges in hydrological modeling. Here we present parameter uncertainty analysis of a recently released distributed conceptual hydrological model applied in the Nea catchment, Norway. Two variants of the generalized likelihood uncertainty estimation (GLUE) methodologies, one based on the residuals and the other on the limits of acceptability, were employed. Streamflow and remote sensing snow cover data were used in conditioning model parameters and in model validation. When using the GLUE limit of acceptability (GLUE LOA) approach, a streamflow observation error of 25% was assumed. Neither the original limits nor relaxing the limits up to a physically meaningful value yielded a behavioral model capable of predicting streamflow within the limits in 100% of the observations. As an alternative to relaxing the limits, the requirement for the percentage of model predictions falling within the original limits was relaxed. An empirical approach was introduced to define the degree of relaxation. The result shows that snow- and water-balance-related parameters induce relatively higher streamflow uncertainty than catchment response parameters. Comparable results were obtained from behavioral models selected using the two GLUE methodologies.

Download & links
Publications Copernicus
Download
Citation
Share