Scholarly record
ENSO EVENTS RECOGNITION AND PREDICTION USING MACHINE LEARNING
Abstract
The El Nino-Southern Oscillation (ENSO) plays a key role in the interannual variability of the global climate. Its prediction and recognition has been a big challenge to the atmospheric and ocean science modeling community despite significant advances in understanding, improved models, and enhanced observational networks. The present study proposes a new approach by using the support vector machines (SVM) classification model of machine learning algorithm to identify and possible predict the future ENSO events. The daily sea surface temperature data in the Pacific Ocean from NOAA National Centers for Environmental Information (NCEI) is used as the training set, and each daily data in the training set is assigned an ENSO index as the primary indicator of ENSO events. The Principal component analysis (PCA) is used to decompose the training set in a set of successive orthogonal components (eigen maps) that explain a maximum amount of the variance and bring out the strong patterns. The test results show the great match between the true ENSO index and the predicted ENSO index as shown in the attached plot. The proposed method can also be used to other research areas such as cyclone and tornado pattern search.
Publication Impact Profile
Publication details
References12
K. E. Trenberth and J. M. Caron, “The Southern Oscillation revisited: Sea level pressures, surface temperatures, and precipitation,” J. Clim., vol. 13, pp. 4358–4365, 2000.
K. Wolter, “The Southern Oscillation in surface circulation and climate over the tropical Atlantic, Eastern Pacific, and Indian Oceans as captured by cluster analysis,” J. Climate Appl. Meteor., vol. 26, pp. 540–558, 1987.
K. Wolter and M.S. Timlin, “Monitoring ENSO in COADS with a seasonally adjusted principal component index,” in Proc. of the 17th Climate Diagnostics Workshop, Norman, OK, NOAA/NMC/CAC, NSSL, Oklahoma Clim. Survey, CIMMS and the School of Meteor., Univ. of Oklahoma, 1993, pp. 52–57.
N.E. Graham and J. Michaelson and T.P. Barnett, “An investigation of the El Niño-Southern Oscillation cycle with statistical models: II. Model results,” J Geophys Res., vol. 92, pp. 14271–14289, 1997.
M.A. Cane and S.E. Zebiak and S.C. Dolan, “Experimental forecasts of El Nino,” Nature, vol. 321, pp. 827–832, 1959.
A. L. Samuel, “The COLA anomaly coupled model: Ensemble ENSO prediction,” Mon Weather Rev, vol. 131, pp. 2324–2341, 2003.
A. Kumar and A. G. Barnston and M. P. Hoerling, “Seasonal predictions, probabilistic verifications, and ensemble size,” J. Climate, vol. 14, pp. 1671–1676, 2001.
A. Wu and W. W. Hsieh and B. Tang, “Neural network forecasts of the tropical Pacific sea surface temperatures,” Neural Networks, vol. 19, pp. 145–154, 2006.
Q. Y. Feng and R. Vasile and M. Segond and A. Gozolchiani and Y. Wang and M. Abel and S. Havlin and A. Bunde and H.A.I. Dijkstra, “ClimateLearn: A machine-learning approach for climate prediction using network measures,” J. Climate, vol. 14, pp. 1671–1676, 2001.
M. R. Hush, “Machine learning for quantum physics,” Science, vol. 355, 2017.
V. N. Vapnik, “The Nature of Statistical Learning Theory,” Springer-Verlag, New York, Inc., New York, 1995.
C. M. Bishop, “Pattern recognition and machine learning,” Springer, New York, 2006.
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.
