Forecasting the equatorial Pacific sea surface temperatures by neural network models

Fredolin Tangang @ Tajudin Mahmud, W. W. Hsieh, B. Tang

Research output: Contribution to journalArticle

64 Citations (Scopus)

Abstract

We used neural network models to seasonally forecast the tropical Pacific sea surface temperature anomalies (SSTA) in the Niño 3.4 region (6°S-6°N, 120°W-170°W). The inputs to the neural networks (i.e., the predictors) were the first seven wind stress empirical orthogonal function (EOF) modes of the tropical Pacific (20°S-20°N, 120°E-70°W) for four seasons and the Niño 3.4 SSTA itself for the final season. The period of 1952-1981 was used for training the neural network models, and the period 1982-1992 for forecast validation. At 6-month lead time, neural networks attained forecast skills comparable to the other E1 Niño-Southern Oscillation (ENSO) models. Our results suggested that neural network models were viable for ENSO forecasting even at longer lead times of 9 to 12 months. We hypothesized that at these longer leads, the underlying relationship between the wind stress and Niño 3.4 SSTA became increasingly nonlinear. The neural network results were interpreted in light of current theories, e.g., the role of the "off-equatorial" Rossby waves in triggering the onset of an ENSO event and the delayed-oscillator theory in the development and termination of an ENSO event.

Original languageEnglish
Pages (from-to)135-147
Number of pages13
JournalClimate Dynamics
Volume13
Issue number2
Publication statusPublished - Feb 1997
Externally publishedYes

Fingerprint

Southern Oscillation
sea surface temperature
temperature anomaly
wind stress
equatorial wave
Rossby wave
forecast

ASJC Scopus subject areas

  • Atmospheric Science

Cite this

Forecasting the equatorial Pacific sea surface temperatures by neural network models. / Tangang @ Tajudin Mahmud, Fredolin; Hsieh, W. W.; Tang, B.

In: Climate Dynamics, Vol. 13, No. 2, 02.1997, p. 135-147.

Research output: Contribution to journalArticle

@article{6193f095e7894b5bafea424ab9454a7a,
title = "Forecasting the equatorial Pacific sea surface temperatures by neural network models",
abstract = "We used neural network models to seasonally forecast the tropical Pacific sea surface temperature anomalies (SSTA) in the Ni{\~n}o 3.4 region (6°S-6°N, 120°W-170°W). The inputs to the neural networks (i.e., the predictors) were the first seven wind stress empirical orthogonal function (EOF) modes of the tropical Pacific (20°S-20°N, 120°E-70°W) for four seasons and the Ni{\~n}o 3.4 SSTA itself for the final season. The period of 1952-1981 was used for training the neural network models, and the period 1982-1992 for forecast validation. At 6-month lead time, neural networks attained forecast skills comparable to the other E1 Ni{\~n}o-Southern Oscillation (ENSO) models. Our results suggested that neural network models were viable for ENSO forecasting even at longer lead times of 9 to 12 months. We hypothesized that at these longer leads, the underlying relationship between the wind stress and Ni{\~n}o 3.4 SSTA became increasingly nonlinear. The neural network results were interpreted in light of current theories, e.g., the role of the {"}off-equatorial{"} Rossby waves in triggering the onset of an ENSO event and the delayed-oscillator theory in the development and termination of an ENSO event.",
author = "{Tangang @ Tajudin Mahmud}, Fredolin and Hsieh, {W. W.} and B. Tang",
year = "1997",
month = "2",
language = "English",
volume = "13",
pages = "135--147",
journal = "Climate Dynamics",
issn = "0930-7575",
publisher = "Springer Verlag",
number = "2",

}

TY - JOUR

T1 - Forecasting the equatorial Pacific sea surface temperatures by neural network models

AU - Tangang @ Tajudin Mahmud, Fredolin

AU - Hsieh, W. W.

AU - Tang, B.

PY - 1997/2

Y1 - 1997/2

N2 - We used neural network models to seasonally forecast the tropical Pacific sea surface temperature anomalies (SSTA) in the Niño 3.4 region (6°S-6°N, 120°W-170°W). The inputs to the neural networks (i.e., the predictors) were the first seven wind stress empirical orthogonal function (EOF) modes of the tropical Pacific (20°S-20°N, 120°E-70°W) for four seasons and the Niño 3.4 SSTA itself for the final season. The period of 1952-1981 was used for training the neural network models, and the period 1982-1992 for forecast validation. At 6-month lead time, neural networks attained forecast skills comparable to the other E1 Niño-Southern Oscillation (ENSO) models. Our results suggested that neural network models were viable for ENSO forecasting even at longer lead times of 9 to 12 months. We hypothesized that at these longer leads, the underlying relationship between the wind stress and Niño 3.4 SSTA became increasingly nonlinear. The neural network results were interpreted in light of current theories, e.g., the role of the "off-equatorial" Rossby waves in triggering the onset of an ENSO event and the delayed-oscillator theory in the development and termination of an ENSO event.

AB - We used neural network models to seasonally forecast the tropical Pacific sea surface temperature anomalies (SSTA) in the Niño 3.4 region (6°S-6°N, 120°W-170°W). The inputs to the neural networks (i.e., the predictors) were the first seven wind stress empirical orthogonal function (EOF) modes of the tropical Pacific (20°S-20°N, 120°E-70°W) for four seasons and the Niño 3.4 SSTA itself for the final season. The period of 1952-1981 was used for training the neural network models, and the period 1982-1992 for forecast validation. At 6-month lead time, neural networks attained forecast skills comparable to the other E1 Niño-Southern Oscillation (ENSO) models. Our results suggested that neural network models were viable for ENSO forecasting even at longer lead times of 9 to 12 months. We hypothesized that at these longer leads, the underlying relationship between the wind stress and Niño 3.4 SSTA became increasingly nonlinear. The neural network results were interpreted in light of current theories, e.g., the role of the "off-equatorial" Rossby waves in triggering the onset of an ENSO event and the delayed-oscillator theory in the development and termination of an ENSO event.

UR - http://www.scopus.com/inward/record.url?scp=0001275437&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001275437&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0001275437

VL - 13

SP - 135

EP - 147

JO - Climate Dynamics

JF - Climate Dynamics

SN - 0930-7575

IS - 2

ER -