Forecasting regional sea surface temperatures in the tropical Pacific by neural network models, with wind stress and sea level pressure as predictors

Fredolin Tangang @ Tajudin Mahmud, William W. Hsieh, Benyang Tang

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

We constructed two types of neural network models for forecasting the sea surface temperature anomaly (SSTA) over several standard equatorial Pacific regions (Niño 3, 3.4, 3.5, 4, P2, P4, and P5). The first type used the sea level pressure (SLP) as predictors, while the second one used the wind stress. By ensemble averaging over 20 forecasts with random initial weights, the resulting forecasts were much less noisy than those in our earlier models. The models performed best in the western-central equatorial regions and less well in the eastern boundary regions. At longer leads of 9 - 12 months, the cross-validated skills (1952 - 1993) for the models using the tropical Pacific SLP as predictors were statistically higher than those using the wind stress. Overall, the models using the tropical SLP showed usable cross-validated skills up to 12-month lead. The true out-of-sample forecast performances during the 1982 - 1993 period for the Niño 3.5 SSTA at lead times of 9, 12, and 15 months attained correlation skills of 0.78, 0.80, and 0.75, respectively.

Original languageEnglish
Pages (from-to)7511-7522
Number of pages12
JournalJournal of Geophysical Research C: Oceans
Volume103
Issue number3334
Publication statusPublished - 1998

Fingerprint

Wind stress
sea surface temperature
Sea level
sea level pressure
sea level
wind stress
forecasting
neural networks
surface temperature
Neural networks
predictions
temperature anomaly
Temperature
anomalies
equatorial regions
forecast

ASJC Scopus subject areas

  • Oceanography
  • Astronomy and Astrophysics
  • Atmospheric Science
  • Space and Planetary Science
  • Earth and Planetary Sciences (miscellaneous)
  • Geophysics
  • Geochemistry and Petrology

Cite this

@article{42a179cd0252458ebcc66a560e499bd5,
title = "Forecasting regional sea surface temperatures in the tropical Pacific by neural network models, with wind stress and sea level pressure as predictors",
abstract = "We constructed two types of neural network models for forecasting the sea surface temperature anomaly (SSTA) over several standard equatorial Pacific regions (Ni{\~n}o 3, 3.4, 3.5, 4, P2, P4, and P5). The first type used the sea level pressure (SLP) as predictors, while the second one used the wind stress. By ensemble averaging over 20 forecasts with random initial weights, the resulting forecasts were much less noisy than those in our earlier models. The models performed best in the western-central equatorial regions and less well in the eastern boundary regions. At longer leads of 9 - 12 months, the cross-validated skills (1952 - 1993) for the models using the tropical Pacific SLP as predictors were statistically higher than those using the wind stress. Overall, the models using the tropical SLP showed usable cross-validated skills up to 12-month lead. The true out-of-sample forecast performances during the 1982 - 1993 period for the Ni{\~n}o 3.5 SSTA at lead times of 9, 12, and 15 months attained correlation skills of 0.78, 0.80, and 0.75, respectively.",
author = "{Tangang @ Tajudin Mahmud}, Fredolin and Hsieh, {William W.} and Benyang Tang",
year = "1998",
language = "English",
volume = "103",
pages = "7511--7522",
journal = "Journal of Geophysical Research",
issn = "2169-9380",
publisher = "American Geophysical Union",
number = "3334",

}

TY - JOUR

T1 - Forecasting regional sea surface temperatures in the tropical Pacific by neural network models, with wind stress and sea level pressure as predictors

AU - Tangang @ Tajudin Mahmud, Fredolin

AU - Hsieh, William W.

AU - Tang, Benyang

PY - 1998

Y1 - 1998

N2 - We constructed two types of neural network models for forecasting the sea surface temperature anomaly (SSTA) over several standard equatorial Pacific regions (Niño 3, 3.4, 3.5, 4, P2, P4, and P5). The first type used the sea level pressure (SLP) as predictors, while the second one used the wind stress. By ensemble averaging over 20 forecasts with random initial weights, the resulting forecasts were much less noisy than those in our earlier models. The models performed best in the western-central equatorial regions and less well in the eastern boundary regions. At longer leads of 9 - 12 months, the cross-validated skills (1952 - 1993) for the models using the tropical Pacific SLP as predictors were statistically higher than those using the wind stress. Overall, the models using the tropical SLP showed usable cross-validated skills up to 12-month lead. The true out-of-sample forecast performances during the 1982 - 1993 period for the Niño 3.5 SSTA at lead times of 9, 12, and 15 months attained correlation skills of 0.78, 0.80, and 0.75, respectively.

AB - We constructed two types of neural network models for forecasting the sea surface temperature anomaly (SSTA) over several standard equatorial Pacific regions (Niño 3, 3.4, 3.5, 4, P2, P4, and P5). The first type used the sea level pressure (SLP) as predictors, while the second one used the wind stress. By ensemble averaging over 20 forecasts with random initial weights, the resulting forecasts were much less noisy than those in our earlier models. The models performed best in the western-central equatorial regions and less well in the eastern boundary regions. At longer leads of 9 - 12 months, the cross-validated skills (1952 - 1993) for the models using the tropical Pacific SLP as predictors were statistically higher than those using the wind stress. Overall, the models using the tropical SLP showed usable cross-validated skills up to 12-month lead. The true out-of-sample forecast performances during the 1982 - 1993 period for the Niño 3.5 SSTA at lead times of 9, 12, and 15 months attained correlation skills of 0.78, 0.80, and 0.75, respectively.

UR - http://www.scopus.com/inward/record.url?scp=0032522032&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032522032&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0032522032

VL - 103

SP - 7511

EP - 7522

JO - Journal of Geophysical Research

JF - Journal of Geophysical Research

SN - 2169-9380

IS - 3334

ER -