Systematic analysis of uncertainty in 2D flood inundation models

Thomas Willis; Nigel Wright; Andrew Sleigh

doi:10.1016/j.envsoft.2019.104520

Systematic analysis of uncertainty in 2D flood inundation models

Thomas Willis^*, Nigel Wright, Andrew Sleigh

^*Corresponding author for this work

Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

Assessing uncertainty is a critical part of understanding and developing flood inundation models for use in risk assessment. Typically, uncertainties are investigated by comparing the effects of an ensemble of key model inputs, such as friction values and hydrographic uncertainties, on model outputs. In this study, an approach is adopted that also consider the uncertainty associated with the computational models. Using the LISFLOOD-FP code, which contains multiple methods for solving floodplain flow, two test cases with different hydraulic characteristics are used in a systematic uncertainty analysis. An ensemble of inputs including cell size, hydrological uncertainty, and representation of buildings are assessed for impact on modelling results. Results show the numerical complexity is a significant source of uncertainty in complex flow regimes, but this reduces in typical fluvial flood events. The method of assessing the modelling output is also found to be important in determining the overall influence of parameters.

Original language	English
Article number	104520
Journal	Environmental Modelling and Software
Volume	122
DOIs	https://doi.org/10.1016/j.envsoft.2019.104520
Publication status	Published - Dec 2019

Bibliographical note

Funding Information:
Thomas Willis was funded by the Flood Risk Management Research Consortium which was supported by grant number EP/F20511/1 from the EPSRC and the DEFRA/EA Joint Research Programme on Flood and Coastal Defence. The authors wish to thank the anonymous referees and Dr Mark Smith who provided comments and feedbacks that helped improve the quality of the manuscript. The authors also wish to thank for valuable insight Professor Paul Bates and Dr Jeffery Neal for use of and insights into the LISFLOOD-FP code, and Professor Rob Lamb for use and insight into the Mexborough dataset. Appendix A

Publisher Copyright:
© 2019

Keywords

Flood inundation modelling
Hydraulics
Numerical modelling
Uncertainty analysis

ASJC Scopus subject areas

Software
Environmental Engineering
Ecological Modelling

Access to Document

10.1016/j.envsoft.2019.104520

Cite this

@article{b5c5a28ad27a4289b27c25beabcf3964,

title = "Systematic analysis of uncertainty in 2D flood inundation models",

abstract = "Assessing uncertainty is a critical part of understanding and developing flood inundation models for use in risk assessment. Typically, uncertainties are investigated by comparing the effects of an ensemble of key model inputs, such as friction values and hydrographic uncertainties, on model outputs. In this study, an approach is adopted that also consider the uncertainty associated with the computational models. Using the LISFLOOD-FP code, which contains multiple methods for solving floodplain flow, two test cases with different hydraulic characteristics are used in a systematic uncertainty analysis. An ensemble of inputs including cell size, hydrological uncertainty, and representation of buildings are assessed for impact on modelling results. Results show the numerical complexity is a significant source of uncertainty in complex flow regimes, but this reduces in typical fluvial flood events. The method of assessing the modelling output is also found to be important in determining the overall influence of parameters.",

keywords = "Flood inundation modelling, Hydraulics, Numerical modelling, Uncertainty analysis",

author = "Thomas Willis and Nigel Wright and Andrew Sleigh",

note = "Funding Information: Thomas Willis was funded by the Flood Risk Management Research Consortium which was supported by grant number EP/F20511/1 from the EPSRC and the DEFRA/EA Joint Research Programme on Flood and Coastal Defence. The authors wish to thank the anonymous referees and Dr Mark Smith who provided comments and feedbacks that helped improve the quality of the manuscript. The authors also wish to thank for valuable insight Professor Paul Bates and Dr Jeffery Neal for use of and insights into the LISFLOOD-FP code, and Professor Rob Lamb for use and insight into the Mexborough dataset. Appendix A Publisher Copyright: {\textcopyright} 2019",

year = "2019",

month = dec,

doi = "10.1016/j.envsoft.2019.104520",

language = "English",

volume = "122",

journal = "Environmental Modelling and Software",

issn = "1364-8152",

publisher = "Elsevier",

}

TY - JOUR

T1 - Systematic analysis of uncertainty in 2D flood inundation models

AU - Willis, Thomas

AU - Wright, Nigel

AU - Sleigh, Andrew

N1 - Funding Information: Thomas Willis was funded by the Flood Risk Management Research Consortium which was supported by grant number EP/F20511/1 from the EPSRC and the DEFRA/EA Joint Research Programme on Flood and Coastal Defence. The authors wish to thank the anonymous referees and Dr Mark Smith who provided comments and feedbacks that helped improve the quality of the manuscript. The authors also wish to thank for valuable insight Professor Paul Bates and Dr Jeffery Neal for use of and insights into the LISFLOOD-FP code, and Professor Rob Lamb for use and insight into the Mexborough dataset. Appendix A Publisher Copyright: © 2019

PY - 2019/12

Y1 - 2019/12

N2 - Assessing uncertainty is a critical part of understanding and developing flood inundation models for use in risk assessment. Typically, uncertainties are investigated by comparing the effects of an ensemble of key model inputs, such as friction values and hydrographic uncertainties, on model outputs. In this study, an approach is adopted that also consider the uncertainty associated with the computational models. Using the LISFLOOD-FP code, which contains multiple methods for solving floodplain flow, two test cases with different hydraulic characteristics are used in a systematic uncertainty analysis. An ensemble of inputs including cell size, hydrological uncertainty, and representation of buildings are assessed for impact on modelling results. Results show the numerical complexity is a significant source of uncertainty in complex flow regimes, but this reduces in typical fluvial flood events. The method of assessing the modelling output is also found to be important in determining the overall influence of parameters.

AB - Assessing uncertainty is a critical part of understanding and developing flood inundation models for use in risk assessment. Typically, uncertainties are investigated by comparing the effects of an ensemble of key model inputs, such as friction values and hydrographic uncertainties, on model outputs. In this study, an approach is adopted that also consider the uncertainty associated with the computational models. Using the LISFLOOD-FP code, which contains multiple methods for solving floodplain flow, two test cases with different hydraulic characteristics are used in a systematic uncertainty analysis. An ensemble of inputs including cell size, hydrological uncertainty, and representation of buildings are assessed for impact on modelling results. Results show the numerical complexity is a significant source of uncertainty in complex flow regimes, but this reduces in typical fluvial flood events. The method of assessing the modelling output is also found to be important in determining the overall influence of parameters.

KW - Flood inundation modelling

KW - Hydraulics

KW - Numerical modelling

KW - Uncertainty analysis

UR - http://www.scopus.com/inward/record.url?scp=85072935476&partnerID=8YFLogxK

U2 - 10.1016/j.envsoft.2019.104520

DO - 10.1016/j.envsoft.2019.104520

M3 - Article

AN - SCOPUS:85072935476

SN - 1364-8152

VL - 122

JO - Environmental Modelling and Software

JF - Environmental Modelling and Software

M1 - 104520

ER -

Systematic analysis of uncertainty in 2D flood inundation models

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this