Advanced search
1 file | 464.52 KB Add to list

Exploring the effect of occurrence-bias-adjustment assumptions on hydrological impact modeling

Jorn Van de Velde (UGent) , Matthias Demuzere (UGent) , Bernard De Baets (UGent) and Niko Verhoest (UGent)
(2021) WATER. 13(11).
Author
Organization
Project
Abstract
Bias adjustment of climate model simulations is a common step in the climate impact assessment modeling chain. For precipitation intensity, multiple bias-adjusting methods have been developed, but less so for precipitation occurrence. Intensity-bias-adjusting methods such as ‘Quantile Delta Mapping’ can adjust too many wet days, but not too many dry days. Some occurrence-bias-adjusting methods have been developed to resolve this by the addition of the ability to adjust too dry simulations. Earlier research has shown this to be important when adjusting on a continental scale, when both types of biases can occur. However, the newer occurrence-bias-adjusting methods have their weakness: they might retain a bias in the number of dry days when adjusting data in a region that only has too many wet days. Yet, if this bias is small enough, it is more practical and economical to apply the newer methods when data in the larger region are adjusted. In this study, we consider two recently introduced occurrence-bias-adjusting methods, Singularity Stochastic Removal and Triangular Distribution Adjustment, and compare them in a region with only wet-day biases. This bias adjustment is performed for precipitation intensity and precipitation occurrence, while the evaluation is performed on precipitation intensity, precipitation occurrence and discharge, which combines the former two variables. Despite theoretical weaknesses, we show that both Singularity Stochastic Removal and Triangular Distribution Adjustment perform well. Thus, the methods can be applied for both too wet and too dry simulations, although Triangular Distribution Adjustment may be preferred as it was designed with a broad application in mind.</jats:p>
Keywords
climate change impact, bias adjustment, occurrence-bias-adjustment, hydrological impact, CLIMATE SIMULATIONS, EURO-CORDEX, PRECIPITATION, RAINFALL, DESIGN, SCALES, UCCLE

Downloads

  • KERMIT-A1-627.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 464.52 KB

Citation

Please use this url to cite or link to this publication:

MLA
Van de Velde, Jorn, et al. “Exploring the Effect of Occurrence-Bias-Adjustment Assumptions on Hydrological Impact Modeling.” WATER, vol. 13, no. 11, 2021, doi:10.3390/w13111573.
APA
Van de Velde, J., Demuzere, M., De Baets, B., & Verhoest, N. (2021). Exploring the effect of occurrence-bias-adjustment assumptions on hydrological impact modeling. WATER, 13(11). https://doi.org/10.3390/w13111573
Chicago author-date
Van de Velde, Jorn, Matthias Demuzere, Bernard De Baets, and Niko Verhoest. 2021. “Exploring the Effect of Occurrence-Bias-Adjustment Assumptions on Hydrological Impact Modeling.” WATER 13 (11). https://doi.org/10.3390/w13111573.
Chicago author-date (all authors)
Van de Velde, Jorn, Matthias Demuzere, Bernard De Baets, and Niko Verhoest. 2021. “Exploring the Effect of Occurrence-Bias-Adjustment Assumptions on Hydrological Impact Modeling.” WATER 13 (11). doi:10.3390/w13111573.
Vancouver
1.
Van de Velde J, Demuzere M, De Baets B, Verhoest N. Exploring the effect of occurrence-bias-adjustment assumptions on hydrological impact modeling. WATER. 2021;13(11).
IEEE
[1]
J. Van de Velde, M. Demuzere, B. De Baets, and N. Verhoest, “Exploring the effect of occurrence-bias-adjustment assumptions on hydrological impact modeling,” WATER, vol. 13, no. 11, 2021.
@article{8710878,
  abstract     = {{Bias adjustment of climate model simulations is a common step in the climate impact assessment modeling chain. For precipitation intensity, multiple bias-adjusting methods have been developed, but less so for precipitation occurrence. Intensity-bias-adjusting methods such as ‘Quantile Delta Mapping’ can adjust too many wet days, but not too many dry days. Some occurrence-bias-adjusting methods have been developed to resolve this by the addition of the ability to adjust too dry simulations. Earlier research has shown this to be important when adjusting on a continental scale, when both types of biases can occur. However, the newer occurrence-bias-adjusting methods have their weakness: they might retain a bias in the number of dry days when adjusting data in a region that only has too many wet days. Yet, if this bias is small enough, it is more practical and economical to apply the newer methods when data in the larger region are adjusted. In this study, we consider two recently introduced occurrence-bias-adjusting methods, Singularity Stochastic Removal and Triangular Distribution Adjustment, and compare them in a region with only wet-day biases. This bias adjustment is performed for precipitation intensity and precipitation occurrence, while the evaluation is performed on precipitation intensity, precipitation occurrence and discharge, which combines the former two variables. Despite theoretical weaknesses, we show that both Singularity Stochastic Removal and Triangular Distribution Adjustment perform well. Thus, the methods can be applied for both too wet and too dry simulations, although Triangular Distribution Adjustment may be preferred as it was designed with a broad application in mind.</jats:p>}},
  articleno    = {{1573}},
  author       = {{Van de Velde, Jorn and Demuzere, Matthias and De Baets, Bernard and Verhoest, Niko}},
  issn         = {{2073-4441}},
  journal      = {{WATER}},
  keywords     = {{climate change impact,bias adjustment,occurrence-bias-adjustment,hydrological impact,CLIMATE SIMULATIONS,EURO-CORDEX,PRECIPITATION,RAINFALL,DESIGN,SCALES,UCCLE}},
  language     = {{eng}},
  number       = {{11}},
  pages        = {{17}},
  title        = {{Exploring the effect of occurrence-bias-adjustment assumptions on hydrological impact modeling}},
  url          = {{http://doi.org/10.3390/w13111573}},
  volume       = {{13}},
  year         = {{2021}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: