Nonlinear Principal Component Analysis: Tropical Indo–Pacific Sea Surface Temperature and Sea Level Pressure

Adam Hugh Monahan Oceanography Unit, Department of Earth and Ocean Sciences, and Crisis Points Group, Peter Wall Institute for Advanced Studies, University of British Columbia, Vancouver, British Columbia, Canada

Search for other papers by Adam Hugh Monahan in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Nonlinear principal component analysis (NLPCA) is a generalization of traditional principal component analysis (PCA) that allows for the detection and characterization of low-dimensional nonlinear structure in multivariate datasets. The authors consider the application of NLPCA to two datasets: tropical Pacific sea surface temperature (SST) and tropical Indo–Pacific sea level pressure (SLP). It is found that for the SST data, the low-dimensional NLPCA approximations characterize the data better than do PCA approximations of the same dimensionality. In particular, the one-dimensional NLPCA approximation characterizes the asymmetry between spatial patterns characteristic of average El Niño and La Niña events, which the 1D PCA approximation cannot. The differences between NLPCA and PCA results are more modest for the SLP data, indicating that the lower-dimensional structures of this dataset are nearly linear.

Corresponding author address: Dr. Adam H. Monahan, Oceanography Unit, Dept. of Earth and Ocean Sciences, University of British Columbia, 6270 University Boulevard, Vancouver, BC V6T1Z4, Canada.

Email: monahan@ocgy.ubc.ca

Abstract

Nonlinear principal component analysis (NLPCA) is a generalization of traditional principal component analysis (PCA) that allows for the detection and characterization of low-dimensional nonlinear structure in multivariate datasets. The authors consider the application of NLPCA to two datasets: tropical Pacific sea surface temperature (SST) and tropical Indo–Pacific sea level pressure (SLP). It is found that for the SST data, the low-dimensional NLPCA approximations characterize the data better than do PCA approximations of the same dimensionality. In particular, the one-dimensional NLPCA approximation characterizes the asymmetry between spatial patterns characteristic of average El Niño and La Niña events, which the 1D PCA approximation cannot. The differences between NLPCA and PCA results are more modest for the SLP data, indicating that the lower-dimensional structures of this dataset are nearly linear.

Corresponding author address: Dr. Adam H. Monahan, Oceanography Unit, Dept. of Earth and Ocean Sciences, University of British Columbia, 6270 University Boulevard, Vancouver, BC V6T1Z4, Canada.

Email: monahan@ocgy.ubc.ca

Save
  • Baldi, P., and K. Hornik, 1989: Neural network and principal component analysis: Learning from examples without local minima. Neural Networks,2, 53–58.

  • Barnston, A. G., 1994: Linear statistical short-term climate predictive skill in the Northern Hemisphere. J. Climate,7, 1513–1564.

  • ——, and R. E. Livezey, 1987: Classification, seasonality, and persistence of low-frequency atmospheric circulation patterns. Mon. Wea. Rev.,115, 1083–1126.

  • Bishop, C. M., 1995: Neural Networks for Pattern Recognition. Clarendon Press, 482 pp.

  • Burgers, G., and D. B. Stephenson, 1999: The “normality” of El Niño. Geophys. Res. Lett.,26, 1027–1030.

  • Cybenko, G., 1989: Approximation by superpositions of a sigmoidal function. Math. Contrib. Signals Syst.,2, 303–314.

  • Finnoff, W., F. Hergert, and H. G. Zimmermann, 1993: Improving model selection by nonconvergent methods. Neural Networks,6, 771–783.

  • Hoerling, M. P., A. Kumar, and M. Zhong, 1997: El Niño, La Niña, and the nonlinearity of their teleconnections. J. Climate,10, 1769–1786.

  • Hsieh, W. W., and B. Tang, 1998: Applying neural network models to prediction and data analysis in meteorology and oceanography. Bull. Amer. Meteor. Soc.,79, 1855–1870.

  • Kramer, M. A., 1991: Nonlinear principal component analysis using autoassociative neural networks. AIChE J.,37, 233–243.

  • Malthouse, E. C., 1998: Limitations of nonlinear PCA as performed with generic neural networks. IEEE Trans. Neural Networks,9, 165–173.

  • Monahan, A. H., 2000a: Nonlinear principal component analysis by neural networks: Theory and application to the Lorenz system. J. Climate,13, 821–835.

  • ——, 2000b: Nonlinear principal component analysis of climate data. Ph.D. dissertation, University of British Columbia, 157 pp. [Available from Dept. of Earth and Ocean Sciences, University of British Columbia, Vancouver, BC V6T 1Z4, Canada.].

  • Penland, C., and P. D. Sardeshmukh, 1995: The optimal growth of tropical sea surface temperature anomalies. J. Climate,8, 1999–2024.

  • Sengupta, S. K., and J. S. Boyle, 1995: Nonlinear principal component analysis of climate data. PCMDI Tech. Rep. 29, 21 pp. [Available from Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Laboratory, University of California, Livermore, CA 94550.].

  • Smith, T. M., R. W. Reynolds, R. E. Livezey, and D. C. Stokes, 1996:Reconstruction of historical sea surface temperatures using empirical orthogonal functions. J. Climate,9, 1403–1420.

  • Tangang, F. T., B. Tang, A. H. Monahan, and W. W. Hsieh, 1998: Forecasting ENSO events: A neural network-extended EOF approach. J. Climate,11, 29–41.

  • von Storch, H., and F. W. Zwiers, 1999: Statistical Analysis in Climate Research. Cambridge University Press, 494 pp.

  • Wang, B., 1995: Interdecadal changes in El Niño onset in the last four decades. J. Climate,8, 267–285.

  • Woodruff, S. D., R. J. Slutz, R. L. Jenne, and P. M. Steurer, 1987: A Comprehensive Ocean–Atmosphere Data Set. Bull. Amer. Meteor. Soc.,68, 1239–1250.

  • Yuval, 2000: Neural network training for prediction of climatological time series, regularized by minimization of the generalized cross-validation function. Mon. Wea. Rev.,128, 1456–1473.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 651 114 16
PDF Downloads 354 60 1