Using network theory and machine learning to predict El Niño
Peter D. Nooteboom1,3,Qing Yi Feng1,3,Cristóbal López2,Emilio Hernández-García2,and Henk A. Dijkstra1,3Peter D. Nooteboom et al. Peter D. Nooteboom1,3,Qing Yi Feng1,3,Cristóbal López2,Emilio Hernández-García2,and Henk A. Dijkstra1,3
1Institute for Marine and Atmospheric Research Utrecht (IMAU), Department of Physics, Utrecht University, Utrecht, the Netherlands
2Instituto de Física Interdisciplinar y Sistemas Complejos (IFISC, CSIC-UIB), University of the Balearic Islands, Balearic Islands, Spain
3Centre for Complex Systems Studies, Utrecht University, Utrecht, the Netherlands
1Institute for Marine and Atmospheric Research Utrecht (IMAU), Department of Physics, Utrecht University, Utrecht, the Netherlands
2Instituto de Física Interdisciplinar y Sistemas Complejos (IFISC, CSIC-UIB), University of the Balearic Islands, Balearic Islands, Spain
3Centre for Complex Systems Studies, Utrecht University, Utrecht, the Netherlands
Correspondence: Peter D. Nooteboom (p.d.nooteboom@uu.nl)
Received: 07 Mar 2018 – Discussion started: 13 Mar 2018 – Revised: 22 Jun 2018 – Accepted: 26 Jun 2018 – Published: 23 Jul 2018
Abstract. The skill of current predictions of the warm phase of the El Niño Southern Oscillation (ENSO) reduces significantly beyond a lag time of 6 months. In this paper, we aim to increase this prediction skill at lag times of up to 1 year. The new method combines a classical autoregressive integrated moving average technique with a modern machine learning approach (through an artificial neural network). The attributes in such a neural network are derived from knowledge of physical processes and topological properties of climate networks, and they are tested using a Zebiak–Cane-type model and observations. For predictions up to 6 months ahead, the results of the hybrid model give a slightly better skill than the CFSv2 ensemble prediction by the National Centers for Environmental Prediction (NCEP). Interestingly, results for a 12-month lead time prediction have a similar skill as the shorter lead time predictions.
The prediction of the El Niño phenomenon, an increased sea surface temperature in the eastern Pacific, fascinates people for a long time. El Niño is associated with natural disasters, such as droughts and floods. Current methods can make a reliable prediction of this phenomenon up to 6 months ahead. However, this article presents a method which combines network theory and machine learning which predicts El Niño up to 1 year ahead.
The prediction of the El Niño phenomenon, an increased sea surface temperature in the eastern...
