El Niño southern-oscillation prediction using southern oscillation index and Niño3 as onset indicators

Application of artificial neural networks

Mahad S. Baawain, Mohamed H. Nour, Ahmed G. El-Din, Mohamed Gamal El-Din

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

El Niño southern-oscillation (ENSO) is known to be the strongest climatic variation on seasonal to inter-annual time scales. It causes severe droughts, floods, fires, and hurricanes leading to economical disasters. This study explores the use of relatively simple inputs in developing artificial neural network (ANN) models for predicting the onset of ENSO by forecasting some of its indicators. Two indicators, southern oscillation index (SOI) and Niño3, were used one at a time to model the ENSO occurrence using monthly averaged data. Both models performed well in forecasting and predicting ENSO occurrence up to 12 months in advance. Correlation coefficient values of more than 0.8 and 0.9 (one month lead time), and above 0.7 and 0.8 (12 month lead time) were obtained for SOI and Niño3, respectively. Both models apply the feed forward multilayer perceptron network trained with error back-propagation algorithm. The final models were compared with each other and found to be highly consistent with 75% agreement in their forecasting ability.

Original languageEnglish
Pages (from-to)113-121
Number of pages9
JournalJournal of Environmental Engineering and Science
Volume4
Issue number2
DOIs
Publication statusPublished - Mar 2005

Fingerprint

Southern Oscillation
artificial neural network
Neural networks
prediction
Backpropagation algorithms
Drought
Hurricanes
Multilayer neural networks
back propagation
Disasters
Fires
hurricane
index
indicator
disaster
drought
timescale

Keywords

  • Artificial neural networks
  • Cllimate anomalies
  • El Niño
  • ENSO
  • La Niña
  • Niño3
  • Southern oscillation index (SOI)
  • Teleconnections

ASJC Scopus subject areas

  • Environmental Chemistry
  • Environmental Engineering

Cite this

El Niño southern-oscillation prediction using southern oscillation index and Niño3 as onset indicators : Application of artificial neural networks. / Baawain, Mahad S.; Nour, Mohamed H.; El-Din, Ahmed G.; El-Din, Mohamed Gamal.

In: Journal of Environmental Engineering and Science, Vol. 4, No. 2, 03.2005, p. 113-121.

Research output: Contribution to journalArticle

@article{c21acba2e94c4d3aa7c2e1a5939fa356,
title = "El Ni{\~n}o southern-oscillation prediction using southern oscillation index and Ni{\~n}o3 as onset indicators: Application of artificial neural networks",
abstract = "El Ni{\~n}o southern-oscillation (ENSO) is known to be the strongest climatic variation on seasonal to inter-annual time scales. It causes severe droughts, floods, fires, and hurricanes leading to economical disasters. This study explores the use of relatively simple inputs in developing artificial neural network (ANN) models for predicting the onset of ENSO by forecasting some of its indicators. Two indicators, southern oscillation index (SOI) and Ni{\~n}o3, were used one at a time to model the ENSO occurrence using monthly averaged data. Both models performed well in forecasting and predicting ENSO occurrence up to 12 months in advance. Correlation coefficient values of more than 0.8 and 0.9 (one month lead time), and above 0.7 and 0.8 (12 month lead time) were obtained for SOI and Ni{\~n}o3, respectively. Both models apply the feed forward multilayer perceptron network trained with error back-propagation algorithm. The final models were compared with each other and found to be highly consistent with 75{\%} agreement in their forecasting ability.",
keywords = "Artificial neural networks, Cllimate anomalies, El Ni{\~n}o, ENSO, La Ni{\~n}a, Ni{\~n}o3, Southern oscillation index (SOI), Teleconnections",
author = "Baawain, {Mahad S.} and Nour, {Mohamed H.} and El-Din, {Ahmed G.} and El-Din, {Mohamed Gamal}",
year = "2005",
month = "3",
doi = "10.1139/s04-047",
language = "English",
volume = "4",
pages = "113--121",
journal = "Journal of Environmental Engineering and Science",
issn = "1496-2551",
publisher = "National Research Council",
number = "2",

}

TY - JOUR

T1 - El Niño southern-oscillation prediction using southern oscillation index and Niño3 as onset indicators

T2 - Application of artificial neural networks

AU - Baawain, Mahad S.

AU - Nour, Mohamed H.

AU - El-Din, Ahmed G.

AU - El-Din, Mohamed Gamal

PY - 2005/3

Y1 - 2005/3

N2 - El Niño southern-oscillation (ENSO) is known to be the strongest climatic variation on seasonal to inter-annual time scales. It causes severe droughts, floods, fires, and hurricanes leading to economical disasters. This study explores the use of relatively simple inputs in developing artificial neural network (ANN) models for predicting the onset of ENSO by forecasting some of its indicators. Two indicators, southern oscillation index (SOI) and Niño3, were used one at a time to model the ENSO occurrence using monthly averaged data. Both models performed well in forecasting and predicting ENSO occurrence up to 12 months in advance. Correlation coefficient values of more than 0.8 and 0.9 (one month lead time), and above 0.7 and 0.8 (12 month lead time) were obtained for SOI and Niño3, respectively. Both models apply the feed forward multilayer perceptron network trained with error back-propagation algorithm. The final models were compared with each other and found to be highly consistent with 75% agreement in their forecasting ability.

AB - El Niño southern-oscillation (ENSO) is known to be the strongest climatic variation on seasonal to inter-annual time scales. It causes severe droughts, floods, fires, and hurricanes leading to economical disasters. This study explores the use of relatively simple inputs in developing artificial neural network (ANN) models for predicting the onset of ENSO by forecasting some of its indicators. Two indicators, southern oscillation index (SOI) and Niño3, were used one at a time to model the ENSO occurrence using monthly averaged data. Both models performed well in forecasting and predicting ENSO occurrence up to 12 months in advance. Correlation coefficient values of more than 0.8 and 0.9 (one month lead time), and above 0.7 and 0.8 (12 month lead time) were obtained for SOI and Niño3, respectively. Both models apply the feed forward multilayer perceptron network trained with error back-propagation algorithm. The final models were compared with each other and found to be highly consistent with 75% agreement in their forecasting ability.

KW - Artificial neural networks

KW - Cllimate anomalies

KW - El Niño

KW - ENSO

KW - La Niña

KW - Niño3

KW - Southern oscillation index (SOI)

KW - Teleconnections

UR - http://www.scopus.com/inward/record.url?scp=20544448454&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=20544448454&partnerID=8YFLogxK

U2 - 10.1139/s04-047

DO - 10.1139/s04-047

M3 - Article

VL - 4

SP - 113

EP - 121

JO - Journal of Environmental Engineering and Science

JF - Journal of Environmental Engineering and Science

SN - 1496-2551

IS - 2

ER -