A comparison of nonlinear flood forecasting methods

F. Laio; A. Porporato; R. Revelli; L. Ridolfi

doi:10.1029/2002WR001551

A comparison of nonlinear flood forecasting methods

F. Laio, A. Porporato, R. Revelli, L. Ridolfi

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

54 Scopus citations

Abstract

Two nonlinear models, nonlinear prediction (NLP) and artificial neural networks (ANN), are compared for multivariate flood forecasting. For NLP the calibration of the locally linear model is quite simple, while for ANN the validation and identification of the model can be cumbersome, mainly because of overfitting. Very good results are obtained with the two methods: NLP performs slightly better at short forecast times while the situation is reversed for longer times.

Original language	English (US)
Pages (from-to)	TNN21-TNN24
Journal	Water Resources Research
Volume	39
Issue number	5
DOIs	https://doi.org/10.1029/2002WR001551
State	Published - May 2003
Externally published	Yes

All Science Journal Classification (ASJC) codes

Water Science and Technology

Keywords

Artificial neural networks
Flood forecasting
Nonlinear prediction
Overfitting

Access to Document

10.1029/2002WR001551

Cite this

@article{e514ca214ffe4a91a708cd4d45e21b87,

title = "A comparison of nonlinear flood forecasting methods",

abstract = "Two nonlinear models, nonlinear prediction (NLP) and artificial neural networks (ANN), are compared for multivariate flood forecasting. For NLP the calibration of the locally linear model is quite simple, while for ANN the validation and identification of the model can be cumbersome, mainly because of overfitting. Very good results are obtained with the two methods: NLP performs slightly better at short forecast times while the situation is reversed for longer times.",

keywords = "Artificial neural networks, Flood forecasting, Nonlinear prediction, Overfitting",

author = "F. Laio and A. Porporato and R. Revelli and L. Ridolfi",

year = "2003",

month = may,

doi = "10.1029/2002WR001551",

language = "English (US)",

volume = "39",

pages = "TNN21--TNN24",

journal = "Water Resources Research",

issn = "0043-1397",

publisher = "American Geophysical Union",

number = "5",

}

TY - JOUR

T1 - A comparison of nonlinear flood forecasting methods

AU - Laio, F.

AU - Porporato, A.

AU - Revelli, R.

AU - Ridolfi, L.

PY - 2003/5

Y1 - 2003/5

N2 - Two nonlinear models, nonlinear prediction (NLP) and artificial neural networks (ANN), are compared for multivariate flood forecasting. For NLP the calibration of the locally linear model is quite simple, while for ANN the validation and identification of the model can be cumbersome, mainly because of overfitting. Very good results are obtained with the two methods: NLP performs slightly better at short forecast times while the situation is reversed for longer times.

AB - Two nonlinear models, nonlinear prediction (NLP) and artificial neural networks (ANN), are compared for multivariate flood forecasting. For NLP the calibration of the locally linear model is quite simple, while for ANN the validation and identification of the model can be cumbersome, mainly because of overfitting. Very good results are obtained with the two methods: NLP performs slightly better at short forecast times while the situation is reversed for longer times.

KW - Artificial neural networks

KW - Flood forecasting

KW - Nonlinear prediction

KW - Overfitting

UR - http://www.scopus.com/inward/record.url?scp=1542646171&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1542646171&partnerID=8YFLogxK

U2 - 10.1029/2002WR001551

DO - 10.1029/2002WR001551

M3 - Article

AN - SCOPUS:1542646171

SN - 0043-1397

VL - 39

SP - TNN21-TNN24

JO - Water Resources Research

JF - Water Resources Research

IS - 5

ER -

A comparison of nonlinear flood forecasting methods

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this