Neural network forecasts of the tropical Pacific sea surface temperatures
Aiming Wu, William W. Hsieh and Benyang Tang
A neural network (NN) model had been used by our group to predict the
sea surface temperature anomalies (SSTA) in the Nino3.4 region in the equatorial Pacific (Tang et
al, 2000). The
NN model has been extended to forecast the SSTA
over the entire tropical Pacific, and in the latest version 3.3
(introduced in Dec. 2005), the model used is a Bayesian NN (Wu et
Click here to see details about this model.
Using sea level pressure and sea surface temperature data up to the
end of October, 2006,
forecasts were made with the NN model. Ensemble-averaged forecasts for
the SSTA in the Nino3.4 region at
various lead times are shown in Fig.1, and the forecasted
SSTA fields over the tropical Pacific are displayed in Fig.2,
showing the El Niño peaking in early 2007.
Figure 1. The SSTA (in degree Celsius) in the Nino3.4
area (170W-120W, 5S-5N) predicted by the ensemble-averaged nonlinear
model at 3, 6, 9 and 12 months of lead time (circles), with
observations denoted by the solid line. Tick marks along the abscissa
indicate the January of the given years.
(The postscript file of Fig.1 is also available).
Figure 2. SSTA (in ºC) predicted by the ensemble-averaged
nonlinear model at 3, 6, 9 and 12 months of lead time, corresponding to
the four consecutive seasons starting with
DJF (December, 2006 - February, 2007).
The zero contour is shown as a thick black curve.
(The postscript file of Fig.2 is also available).
Data in tabular format for the Nino3.4 SSTA (in ºC) at 3, 6, 9 and
12 months of lead time:
Tang, B., W.W. Hsieh, A.H. Monahan and F.T. Tangang, 2000. Skill
comparisons between neural networks and canonical correlation analysis
in predicting the equatorial Pacific sea surface temperatures.
J.Climate, 13: 287-293.
Wu, A., W.W. Hsieh and B. Tang, 2006. Neural network forecasts of the
tropical Pacific sea surface temperatures. Neural Networks. 19: 145-154.
Back to [UBC Climate
Prediction Group Home Page]