Forecasting daily streamflows using NARMAX models: How disturbances may affect model performance

Paolo Bartolini, Elena C. Carcano, Luigi Piroddi, Juan B Valdes

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The realization of a random event (e.g. basin outflows, rainfall depths) may be represented by a deterministic link between forcings (input system series) and outputs. In those circumstances when such a non-linear link (e.g. rainfall-runoff models) is not known or when sufficient physical insight of the process is not available, non linear empirical models may be applied with fairly good chances of success. Moreover, even if knowledge of the process causality produces a deterministic relationship between input and output, randomness still remains in the realization of the event, e.g. streamflow events. NARMAX models, initially developed by Leontaritis and Billings (1985) are an example of non-linear models that may be used in this application. Herein the original and simplified version of NARMAX models (NARX) is introduced in a comparative study in the reproduction of daily streamflows data for two small catchments in the Ligurian region (Italy). NARMAX models, in general, yield an input-output representation of a non linear system where the current output a function of lagged inputs, outputs and noise. The influence of disturbances in rainfall-runoff modeling at daily scale using NARMAX models is evaluated in this paper. The corresponding NARX models are used when the focus is on the deterministic input- output relationship with a crude simplification of the disturbance model.

Original languageEnglish (US)
Title of host publicationWorld Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008
Volume316
DOIs
StatePublished - 2008
EventWorld Environmental and Water Resources Congress 2008: Ahupua'a - Honolulu, HI, United States
Duration: May 12 2008May 16 2008

Other

OtherWorld Environmental and Water Resources Congress 2008: Ahupua'a
CountryUnited States
CityHonolulu, HI
Period5/12/085/16/08

Fingerprint

streamflow
disturbance
Rain
Runoff
rainfall-runoff modeling
rainfall
Catchments
Nonlinear systems
comparative study
outflow
catchment
runoff
basin

Keywords

  • Forecasting
  • Parameters
  • Sensitivity analysis
  • Streamflow

ASJC Scopus subject areas

  • Management, Monitoring, Policy and Law
  • Water Science and Technology
  • Pollution

Cite this

Bartolini, P., Carcano, E. C., Piroddi, L., & Valdes, J. B. (2008). Forecasting daily streamflows using NARMAX models: How disturbances may affect model performance. In World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008 (Vol. 316) https://doi.org/10.1061/40976(316)613

Forecasting daily streamflows using NARMAX models : How disturbances may affect model performance. / Bartolini, Paolo; Carcano, Elena C.; Piroddi, Luigi; Valdes, Juan B.

World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008. Vol. 316 2008.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bartolini, P, Carcano, EC, Piroddi, L & Valdes, JB 2008, Forecasting daily streamflows using NARMAX models: How disturbances may affect model performance. in World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008. vol. 316, World Environmental and Water Resources Congress 2008: Ahupua'a, Honolulu, HI, United States, 5/12/08. https://doi.org/10.1061/40976(316)613
Bartolini P, Carcano EC, Piroddi L, Valdes JB. Forecasting daily streamflows using NARMAX models: How disturbances may affect model performance. In World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008. Vol. 316. 2008 https://doi.org/10.1061/40976(316)613
Bartolini, Paolo ; Carcano, Elena C. ; Piroddi, Luigi ; Valdes, Juan B. / Forecasting daily streamflows using NARMAX models : How disturbances may affect model performance. World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008. Vol. 316 2008.
@inproceedings{02baf963b5af4571ad7a31dad842d852,
title = "Forecasting daily streamflows using NARMAX models: How disturbances may affect model performance",
abstract = "The realization of a random event (e.g. basin outflows, rainfall depths) may be represented by a deterministic link between forcings (input system series) and outputs. In those circumstances when such a non-linear link (e.g. rainfall-runoff models) is not known or when sufficient physical insight of the process is not available, non linear empirical models may be applied with fairly good chances of success. Moreover, even if knowledge of the process causality produces a deterministic relationship between input and output, randomness still remains in the realization of the event, e.g. streamflow events. NARMAX models, initially developed by Leontaritis and Billings (1985) are an example of non-linear models that may be used in this application. Herein the original and simplified version of NARMAX models (NARX) is introduced in a comparative study in the reproduction of daily streamflows data for two small catchments in the Ligurian region (Italy). NARMAX models, in general, yield an input-output representation of a non linear system where the current output a function of lagged inputs, outputs and noise. The influence of disturbances in rainfall-runoff modeling at daily scale using NARMAX models is evaluated in this paper. The corresponding NARX models are used when the focus is on the deterministic input- output relationship with a crude simplification of the disturbance model.",
keywords = "Forecasting, Parameters, Sensitivity analysis, Streamflow",
author = "Paolo Bartolini and Carcano, {Elena C.} and Luigi Piroddi and Valdes, {Juan B}",
year = "2008",
doi = "10.1061/40976(316)613",
language = "English (US)",
isbn = "9780784409763",
volume = "316",
booktitle = "World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008",

}

TY - GEN

T1 - Forecasting daily streamflows using NARMAX models

T2 - How disturbances may affect model performance

AU - Bartolini, Paolo

AU - Carcano, Elena C.

AU - Piroddi, Luigi

AU - Valdes, Juan B

PY - 2008

Y1 - 2008

N2 - The realization of a random event (e.g. basin outflows, rainfall depths) may be represented by a deterministic link between forcings (input system series) and outputs. In those circumstances when such a non-linear link (e.g. rainfall-runoff models) is not known or when sufficient physical insight of the process is not available, non linear empirical models may be applied with fairly good chances of success. Moreover, even if knowledge of the process causality produces a deterministic relationship between input and output, randomness still remains in the realization of the event, e.g. streamflow events. NARMAX models, initially developed by Leontaritis and Billings (1985) are an example of non-linear models that may be used in this application. Herein the original and simplified version of NARMAX models (NARX) is introduced in a comparative study in the reproduction of daily streamflows data for two small catchments in the Ligurian region (Italy). NARMAX models, in general, yield an input-output representation of a non linear system where the current output a function of lagged inputs, outputs and noise. The influence of disturbances in rainfall-runoff modeling at daily scale using NARMAX models is evaluated in this paper. The corresponding NARX models are used when the focus is on the deterministic input- output relationship with a crude simplification of the disturbance model.

AB - The realization of a random event (e.g. basin outflows, rainfall depths) may be represented by a deterministic link between forcings (input system series) and outputs. In those circumstances when such a non-linear link (e.g. rainfall-runoff models) is not known or when sufficient physical insight of the process is not available, non linear empirical models may be applied with fairly good chances of success. Moreover, even if knowledge of the process causality produces a deterministic relationship between input and output, randomness still remains in the realization of the event, e.g. streamflow events. NARMAX models, initially developed by Leontaritis and Billings (1985) are an example of non-linear models that may be used in this application. Herein the original and simplified version of NARMAX models (NARX) is introduced in a comparative study in the reproduction of daily streamflows data for two small catchments in the Ligurian region (Italy). NARMAX models, in general, yield an input-output representation of a non linear system where the current output a function of lagged inputs, outputs and noise. The influence of disturbances in rainfall-runoff modeling at daily scale using NARMAX models is evaluated in this paper. The corresponding NARX models are used when the focus is on the deterministic input- output relationship with a crude simplification of the disturbance model.

KW - Forecasting

KW - Parameters

KW - Sensitivity analysis

KW - Streamflow

UR - http://www.scopus.com/inward/record.url?scp=79251491715&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79251491715&partnerID=8YFLogxK

U2 - 10.1061/40976(316)613

DO - 10.1061/40976(316)613

M3 - Conference contribution

AN - SCOPUS:79251491715

SN - 9780784409763

VL - 316

BT - World Environmental and Water Resources Congress 2008: Ahupua'a - Proceedings of the World Environmental and Water Resources Congress 2008

ER -