Automatic calibration of conceptual rainfall-runoff models: Sensitivity to calibration data

Patrice O. Yapo, Hoshin Vijai Gupta, Soroosh Sorooshian

Research output: Contribution to journalArticle

343 Citations (Scopus)

Abstract

The identification of hydrologic models requires that appropriate data be selected for model calibration. In the research presented here, the shuffled complex evolution (SCE-UA) global optimization method was used to calibrate the NWSRFS-SMA conceptual rainfall-runoff flood forecasting model of the US National Weather Service, using a 40-year record of historical data. Based on 344 calibration runs using different lengths of data from different sections of the historical record, we conclude that approximately 8 years of data are required to obtain calibrations that are relatively insensitive to the period selected. Further, the reduction in parameter uncertainty is maximal when the wettest data periods on record are used. A residual analysis is used to compare the performance of the daily root mean square (DRMS) and heteroscedastic maximum likelihood error (HMLE) objective functions. The results suggest that the factor currently limiting model performance is the unavailability of strategies that explicitly account for model error during calibration.

Original languageEnglish (US)
Pages (from-to)23-48
Number of pages26
JournalJournal of Hydrology
Volume181
Issue number1-4
StatePublished - 1996

Fingerprint

hydrologic models
calibration
runoff
rain
rainfall
parameter uncertainty
system optimization
flood forecasting
historical record
weather
limiting factor

ASJC Scopus subject areas

  • Soil Science
  • Earth-Surface Processes

Cite this

Automatic calibration of conceptual rainfall-runoff models : Sensitivity to calibration data. / Yapo, Patrice O.; Gupta, Hoshin Vijai; Sorooshian, Soroosh.

In: Journal of Hydrology, Vol. 181, No. 1-4, 1996, p. 23-48.

Research output: Contribution to journalArticle

@article{512958b8e32f4eea91442c97be608421,
title = "Automatic calibration of conceptual rainfall-runoff models: Sensitivity to calibration data",
abstract = "The identification of hydrologic models requires that appropriate data be selected for model calibration. In the research presented here, the shuffled complex evolution (SCE-UA) global optimization method was used to calibrate the NWSRFS-SMA conceptual rainfall-runoff flood forecasting model of the US National Weather Service, using a 40-year record of historical data. Based on 344 calibration runs using different lengths of data from different sections of the historical record, we conclude that approximately 8 years of data are required to obtain calibrations that are relatively insensitive to the period selected. Further, the reduction in parameter uncertainty is maximal when the wettest data periods on record are used. A residual analysis is used to compare the performance of the daily root mean square (DRMS) and heteroscedastic maximum likelihood error (HMLE) objective functions. The results suggest that the factor currently limiting model performance is the unavailability of strategies that explicitly account for model error during calibration.",
author = "Yapo, {Patrice O.} and Gupta, {Hoshin Vijai} and Soroosh Sorooshian",
year = "1996",
language = "English (US)",
volume = "181",
pages = "23--48",
journal = "Journal of Hydrology",
issn = "0022-1694",
publisher = "Elsevier",
number = "1-4",

}

TY - JOUR

T1 - Automatic calibration of conceptual rainfall-runoff models

T2 - Sensitivity to calibration data

AU - Yapo, Patrice O.

AU - Gupta, Hoshin Vijai

AU - Sorooshian, Soroosh

PY - 1996

Y1 - 1996

N2 - The identification of hydrologic models requires that appropriate data be selected for model calibration. In the research presented here, the shuffled complex evolution (SCE-UA) global optimization method was used to calibrate the NWSRFS-SMA conceptual rainfall-runoff flood forecasting model of the US National Weather Service, using a 40-year record of historical data. Based on 344 calibration runs using different lengths of data from different sections of the historical record, we conclude that approximately 8 years of data are required to obtain calibrations that are relatively insensitive to the period selected. Further, the reduction in parameter uncertainty is maximal when the wettest data periods on record are used. A residual analysis is used to compare the performance of the daily root mean square (DRMS) and heteroscedastic maximum likelihood error (HMLE) objective functions. The results suggest that the factor currently limiting model performance is the unavailability of strategies that explicitly account for model error during calibration.

AB - The identification of hydrologic models requires that appropriate data be selected for model calibration. In the research presented here, the shuffled complex evolution (SCE-UA) global optimization method was used to calibrate the NWSRFS-SMA conceptual rainfall-runoff flood forecasting model of the US National Weather Service, using a 40-year record of historical data. Based on 344 calibration runs using different lengths of data from different sections of the historical record, we conclude that approximately 8 years of data are required to obtain calibrations that are relatively insensitive to the period selected. Further, the reduction in parameter uncertainty is maximal when the wettest data periods on record are used. A residual analysis is used to compare the performance of the daily root mean square (DRMS) and heteroscedastic maximum likelihood error (HMLE) objective functions. The results suggest that the factor currently limiting model performance is the unavailability of strategies that explicitly account for model error during calibration.

UR - http://www.scopus.com/inward/record.url?scp=0030162090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030162090&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0030162090

VL - 181

SP - 23

EP - 48

JO - Journal of Hydrology

JF - Journal of Hydrology

SN - 0022-1694

IS - 1-4

ER -