Multi-scale spatial features and temporal attention mechanisms: advancing the accuracy of ENSO prediction

Abstract

The exponential progression in oceanic observational technology has fostered the accumulation of substantial time series data pivotal for predictions in ocean meteorology. Foremost among the phenomena observed is El Niño-Southern Oscillation (ENSO), a critical determinant in the interplay of global ocean atmosphere interactions, with its severe manifestations inducing extreme meteorological conditions. Therefore, precisely predicting ENSO events carries immense gravitas. Historically, predictions hinged primarily on dynamic models and statistical approaches; however, the intricate and multifaceted spatiotemporal dynamics of ENSO events have often impeded the accuracy of these traditional methodologies. A notable lacuna in contemporary research is the insufficient exploration of long-term dependencies within oceanic data and the suboptimal integration of spatial information derived from spatiotemporal data. To address these limitations, this study introduces a forward-thinking ENSO prediction framework synergizing multiscale spatial features with temporal attention mechanisms. This innovation facilitates a more profound exploration of temporal and spatial domains, enhancing the retention of extensive-period data while optimizing the use of spatial information. Preliminary analyses executed on the global ocean data assimilation system dataset attest to the superior efficacy of the proposed method, underscoring a substantial improvement over established methods including SA-convolutional long short-term memory, particularly in facilitating long-term predictions.The source code and datasets are provided. The code is available at https://github.com/tse1998/ENSO-prediction.

Keywords

Deep learning / Multiscale spatiotemporal features / ENSO prediction / Sequence prediction

References

[1] Adams RM, Chen CC, McCarl BA, Weiher RF. The economic consequences of ENSO events for agriculture. Clim Res, 1999, 13(3): 165-172,

[2] Aguilar-Martinez S, Hsieh WW. Forecasts of tropical pacific sea surface temperatures by neural networks and support vector regression. Int J Oceanogr, 2009, 2009: 1-13,

[3] Bahdanau D, Cho K, Bengio Y (2014) Neural machine translation by jointly learning to align and translate. Preprint arXiv:1409.0473

[4] Barnston AG, Tippett MK, L’Heureux ML, Li SH, DeWitt DG. Skill of real-time seasonal ENSO model predictions during 2002–11: is our capability increasing?. Bull Amer Meteorol Soc, 2012, 93(5): 631-651,

[5] Chen D, Cane MA, Kaplan A, Zebiak SE, Huang DJ. Predictability of El Niño over the past 148 years. Nature, 2004, 428(6984): 733-736,

[6] Glantz MH. . Currents of change: impacts of El Niño and La Niña on climate and society, 2001 London Cambridge University Press

[7] Ham YG, Kim JH, Luo JJ. Deep learning for multi-year ENSO forecasts. Nature, 2019, 573(7775): 568-572,

[8] He DD, Lin PF, Liu HL, Ding L, Jiang JR (2019) DLENSO: a deep learning ENSO forecasting model. In: 16th Pacific Rim International Conference on Artificial Intelligence (PRICAI), Cuvu, Yanuca Island, pp 12–23

[9] Jiang JR, Wang TY, Chi XB, Hao HQ, Wang YZ, Chen YR et al (2016) SC-ESAP: a parallel application platform for earth system model. In: 16th Annual International Conference on Computational Science (ICCS), San Diego, pp 1612–1623

[10] Jin EK, Kinter JL, Wang B, Park CK, Kang IS, Kirtman BP, et al.. Current status of ENSO prediction skill in coupled ocean-atmosphere models. Clim Dyn, 2008, 31(6): 647-664,

[11] Lin ZH, Li MM, Zheng ZB, Cheng YY, Yuan C (2020) Self-attention ConvLSTM for spatiotemporal prediction. In: 34th AAAI Conference on Artificial Intelligence, New York, pp 11531–11538

[12] Luo JJ, Masson S, Behera SK, Yamagata T. Extended ENSO predictions using a fully coupled ocean-atmosphere model. J Clim, 2008, 21(1): 84-93,

[13] Luong MT, Pham H, Manning CD (2015) Effective approaches to attention-based neural machine translation. Preprint arXiv:1508.04025

[14] Mu B, Peng C, Yuan SJ, Chen L (2019) ENSO forecasting over multiple time horizons using ConvLSTM network and rolling mechanism. In: 2019 International Joint Conference on Neural Networks (IJCNN), Budapest, pp 1–8

[15] Shi XJ, Chen ZR, Wang H, Yeung DY, Wong WK, Woo WC (2015) Convolutional LSTM network: a machine learning approach for precipitation nowcasting. Preprint arXiv:1506.04214

[16] Szegedy C, Liu W, Jia YQ, Sermanet P, Reed S, Anguelov D et al (2015) Going deeper with convolutions. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, pp 1–9

[17] Zhang RH. A hybrid coupled model for the Pacific Ocean-atmosphere system. Part I: description and basic performance. Adv Atmos Sci, 2015, 32(3): 301-318,

[18] Zheng F, Zhu J. Improved ensemble-mean forecasting of ENSO events by a zero-mean stochastic error model of an intermediate coupled model. Clim Dyn, 2016, 47(12): 3901-3915,

Funding

Key Technologies Research and Development Program(2022ZD0117201)