ENSO predictive analytics based on layered rendering

Abstract

El Niño-Southern Oscillation (ENSO) is a periodic climate phenomenon in the equatorial Pacific that significantly influences global climate patterns. Accurate prediction and monitoring of ENSO events are essential for meteorological agencies and governmental institutions. This study introduces a content-guided attention module within a convolutional neural network to improve prediction accuracy. This module models inter-channel relationships and enhances information interaction by integrating channel and spatial attention weights. These advancements substantially improve prediction accuracy and help overcome the spring prediction barrier in ENSO forecasting. The research emphasizes global feature modeling and proposes a novel content-guided ENSO prediction model. It also includes an ocean data generation model utilizing global attention. Furthermore, a layered rendering technique is employed to invert ocean data, facilitating detailed analysis and contributing to the development of an ocean synthetic dataset.

Keywords

ENSO prediction / Deep learning / Climate modeling / Content-guided attention / Ocean data generation

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: 167239

[3] Bao Q, Wu XF, Li JX, Wang L, He B, Wang XC et al (2018) Outlook for El Niño and the Indian Ocean Dipole in autumn-winter 2018–2019. Chin Sci Bull 64(1):73–78 (in Chinese with English abstract)

[4] Boucharel J, Jin FF. A simple theory for the modulation of tropical instability waves by ENSO and the annual cycle. Tellus A-Dyn Meteorol Oceanogr, 2020, 72(1): 1-14

[5] Broni-Bedaiko C, Katsriku FA, Unemi T, Atsumi M, Abdulai JD, Shinomiya N et al (2019) El Niño-Southern Oscillation forecasting using complex networks analysis of LSTM neural networks. Artif Life Robot 24:445–451

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

[7] Chen DK, Lian T, Fu CB, Cane MA, Tang YM, Murtugudde R, et al. Strong influence of westerly wind bursts on El Niño diversity. Nat Geosci, 2015, 8(5): 339-345

[8] Davis RE, Talley LD, Roemmich D, Owens WB, Rudnick DL, Toole J, et al. 100 years of progress in ocean observing systems. Meteorol Monogr, 2019, 59: 1-46

[9] Fang XH, Mu M. Both air-sea components are crucial for El Niño forecast from boreal spring. Sci Rep, 2018, 8(1): 10501

[10] Feng LC, Zhang RH, Wang ZG, Chen XR. Processes leading to second-year cooling of the 2010–12 La Niña event, diagnosed using GODAS. Adv Atmos Sci, 2015, 32: 424-438

[11] Galanti E, Tziperman E, Harrison M, Rosati A, Sirkes Z. A study of ENSO prediction using a hybrid coupled model and the adjoint method for data assimilation. Mon Weather Rev, 2003, 131(11): 2748-2764

[12] Glantz MH (2001) Currents of change: impacts of El Niño and La Niña on climate and society. Cambridge University Press

[13] Goodfellow IJ, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S et al (2014) Generative adversarial nets. In: 28th Conference on Neural Information Processing Systems (NIPS), Montreal, pp 2672–2680

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

[15] Jiang J, Wang T, Chi X, Hao H, Wang Y, Chen Y et al (2016) SC-ESAP: a parallel application platform for earth system model. In: International Conference on Computational Science 2016, San Diego, pp 1612–1623

[16] Kug JS, Ham YG. Are there two types of La Niña?. Geophys Res Lett, 2011, 38: L16704

[17] Liu YC, Shao ZR, Hoffmann N (2021) Global attention mechanism: retain information to enhance channel-spatial interactions. Preprint at arXiv:2112.05561

[18] Luo JJ, Masson S, Behera S, Shingu S, Yamagata T. Seasonal climate predictability in a coupled OAGCM using a different approach for ensemble forecasts. J Clim, 2005, 18(21): 4474-4497

[19] 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

[20] Mahesh A, Evans M, Jain G, Castillo M, Lima A, Lunghino B et al (2019) Forecasting El Niño with convolutional and recurrent neural networks. In: 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, pp 8–14

[21] McLeod AI, Li WK. Diagnostic checking ARMA time series models using squared-residual autocorrelations. J Time Ser Anal, 1983, 4(4): 269-273

[22] Messié M, Chavez F. Global modes of sea surface temperature variability in relation to regional climate indices. J Clim, 2011, 24(16): 4314-4331

[23] Mu M, Xu H, Duan WS. A kind of initial errors related to “spring predictability barrier” for El Niño events in Zebiak-Cane model. Geophys Res Lett, 2007, 34(3): L03709

[24] Neng S. A multi-statistical analysis of the southern oscillation (SO) and its relation to the mean monthly atmospheric circulation at 500 hPa in the northern hemisphere. Adv Atmos Sci, 1988, 5(3): 345-359

[25] Petrova D, Ballester J, Koopman SJ, Rodó X. Multiyear statistical prediction of ENSO enhanced by the tropical Pacific observing system. J Clim, 2020, 33(1): 163-174

[26] Ren HL, Wu YJ, Bao Q, Ma JH, Liu CZ, Wan JH, et al. The China multi-model ensemble prediction system and its application to flood-season prediction in 2018. J Meteorol Res, 2019, 33(3): 540-552

[27] Ren HL, Zheng F, Luo JJ, Wang R, Liu MH, Zhang WJ, et al. A review of research on tropical air-sea interaction, ENSO dynamics, and ENSO prediction in China. J Meteorol Res, 2020, 34(1): 43-62

[28] Toms BA, Barnes EA, Hurrell JW. Assessing decadal predictability in an earth-system model using explainable neural networks. Geophys Res Lett, 2021, 48(12): e2021GL093842

[29] Wen N, Liu Z, Li L. Direct ENSO impact on East Asian summer precipitation in the developing summer. Clim Dyn, 2019, 52: 6799-6815

[30] Xin XG, Wu TW, Zhang J (2012) Introduction of CMIP5 experiments carried out by BCC climate system model. Adv Clim Change Res 8:378–382 (in Chinese)

[31] Zhai PM, Yu R, Guo YJ, Li QX, Ren XJ, Wang YQ et al (2016) The strong El Niño in 2015/2016 and its dominant impacts on global and China’s climate. Acta Meteorol Sin 74(3):309–321 (in Chinese with English abstract)

[32] Zhang Q, Wang H, Dong JY, Zhong GQ, Sun X. Prediction of sea surface temperature using long short-term memory. IEEE Geosci Remote Sens Lett, 2017, 14(10): 1745-1749

[33] Zhang RH. A hybrid coupled model for the pacific ocean-atmosphere system. Part I: Description and basic performance. Adv Atmos Sci, 2015, 32: 301-318

[34] 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: 3901-3915

[35] Zhou L, Zhang RH. A hybrid neural network model for ENSO prediction in combination with principal oscillation pattern analyses. Adv Atmos Sci, 2022, 39(6): 889-902

[36] Zhou L, Zhang RH. A self-attention-based neural network for three-dimensional multivariate modeling and its skillful ENSO predictions. Sci Adv, 2023, 9(10): eadf2827

Funding

National Major Science and Technology Projects of China(2022ZD0117201)