Spring Predictability Barrier Phenomenon in ENSO Prediction Model Based on LSTM Deep Learning Algorithm

被引：0

作者：

Zhou P. ^{[1
]}

Huang Y. ^{[1
]}

Hu B. ^{[1
,2
]}

Wei J. ^{[1
,3
]}

机构：

[1] Key Laboratory of Tropical Atmosphere-Ocean System (MOE), School of Atmospheric Sciences, Sun Yat-Sen University, Guangzhou

[2] HSBC Business School, Peking University, Shenzhen

[3] School of Marine Sciences, Guangxi University, Nanning

来源：

Beijing Daxue Xuebao (Ziran Kexue Ban)/Acta Scientiarum Naturalium Universitatis Pekinensis | 2021年 / 57卷 / 06期

关键词：

ENSO; LSTM; Niño3.4; index; Prediction error; SPB;

D O I：

10.13209/j.0479-8023.2021.114

中图分类号：

学科分类号：

摘要：

A LSTM (long-short term memory) model is applied to the prediction of the Niño3.4 index, and the spring prediction barrier (SPB) issue has been further investigated in the LSTM model. The results show that the model can predict the trend of the Niño3.4 index well, yet revealing different performance in different El Niño events. For the 1997/1998 El Niño and 2015/2016 El Niño, which are strong EP El Niño events, the model performes well on the prediction of Niño3.4 index trend and peaks, and anomaly correlation coefficient (ACC) reaches more than 0.93. But for the weak CP El Niño events, e.g. the 1991/1992 El Niño and 2002/2003 El Niño, it shows relatively poor performance on the prediction of the peak. In the growing period, the maximum season growth rate of prediction error are in AMJ quarter, which indicates obvious SPB phenomenon. However, in the decaying period, the maximum have similar distribution in the same type of events: for the weak CP El Niño events, the maximum are in AMJ quarter, indicating obvious SPB phenomenon; for strong EP El Niño events, the maximum are in other quarter, indicating that there is no SPB phenomenon. The differences in the performance among individuals may be related to the development characteristics of the event itself (such as event type and intensity). © 2021 Peking University.

引用

页码：1071 / 1078

页数：7

共 20 条

[1]

Oldenborgh G J V, Balmaseda M A, Ferranti L, Et al., Did the ECMWF seasonal forecast model outperform statistical ENSO forecast models over the last 15 years?, Journal of Climate, 18, 16, pp. 3240-3249, (2005)

[2]

Dijkstra H A, Petersik P, Hernandez-Garcia E, Et al., The application of machine learning techniques to improve El Niño prediction skill, Frontiers in Physics, 7, (2019)

[3]

Barnston A G, Tippett M K, L'Heureux M L, Et al., Skill of real-time seasonal ENSO model predictions during 2002-11: is our capability increasing?, Bulletin of the American Meteorological Society, 93, 5, pp. 631-651, (2012)

[4]

Mcphaden M J, Zebiak S E, Glantz M H., ENSO as an integrating concept in earth science, Science, 314, pp. 1740-1745, (2006)

[5]

Ham Y G, Kim J H, Luo J J., Deep learning for multi-year ENSO forecasts, Nature, 573, pp. 568-572, (2019)

[6]

Webster P J, Yang S., Monsoon and ENSO: selectively interactive system, Quarterly Journal of the Royal Meteorological Society, 118, pp. 877-926, (1992)

[7]

Webster P J., The annual cycle and the predictability of the tropical coupled ocean-atmosphere system, Meteorology and Atmospheric Physics, 56, 1, pp. 33-55, (1995)

[8]

Torrence C, Webster P J., The annual cycle of persistence in the El Niño/Southern Oscillation, Quarterly Journal of the Royal Meteorological Society, 124, pp. 1985-2004, (1998)

[9]

Wang B, Fang Z., Chaotic oscillations of tropical climate: a dynamic system theory for ENSO, Journal of the Atmospheric Sciences, 53, 19, pp. 2786-2802, (1996)

[10]

Mu M, Duan W S., A new approach to studying ENSO predictability: conditional nonlinear optimal perturbation, Chinese Science Bulletin, 48, 10, pp. 1045-1047, (2003)

← 1 2 →