Review of application of deep learning in Indian Ocean Dipole prediction

doi:10.3969/j.issn.1001-909X.2024.03.004

Abstract

Abstract:

The Indian Ocean Dipole (IOD) is a pivotal climate phenomenon in the Indian Ocean region, exerting a significant impact on the climate change of the surrounding areas and the global climate system. Accurate prediction of IOD is essential for comprehending the dynamics of the global climate, yet traditional forecasting methods are limited in capturing its complexity and nonlinearity, constraining predictive capabilities. This paper begins by outlining the relevant theories of IOD and evaluates the strengths and weaknesses of traditional forecasting methods. It then provides a comprehensive analysis of the application and development of deep learning in the field of IOD prediction, highlights the advantages of deep learning models over traditional methods in terms of automatic feature extraction, nonlinear relationship modeling, and large data processing capabilities. Additionally, the paper discusses the challenges faced by deep learning models in IOD forecasting: including data scarcity, overfitting, and model interpretability issues, and proposes future research directions to promote innovation and progress in the application of deep learning technology in the field of climate prediction.

Key words: global climate change, neural network, CNN, LSTM, ConvLSTM, climate prediction, climate variability, data processing

CLC Number:

P732

ZHENG Mengke, FANG Wei, ZHANG Xiaozhi. Review of application of deep learning in Indian Ocean Dipole prediction[J]. Journal of Marine Sciences, 2024, 42(3): 51-63.

Figures/Tables 7

References 65

[1]	肖莺, 张祖强, 何金海. 印度洋偶极子研究进展综述[J]. 热带气象学报, 2009, 25(5):621-627.
	XIAO Y, ZHANG Z Q, HE J H. Progresses in the studies on Indian Ocean Dipoles[J]. Journal of Tropical Meteorology, 2009, 25(5): 621-627.
[2]	黄怡陶, 张文君, 薛奥运. ENSO对印度洋偶极子非对称性的影响及机制研究[J]. 气象科学, 2023, 43(1):1-14.
	HUANG Y T, ZHANG W J, XUE A Y. Influence of ENSO on Indian Ocean Dipole skewness and its physical mechanism[J]. Journal of the Meteorological Sciences, 2023, 43(1): 1-14.
[3]	BAHIYAH A, WIRASATRIYA A, MARDIANSYAH W, et al. Massive SST-front anomaly in the tip of Sumatra waters triggered by extreme positive IOD 2019 event[J]. International Journal of Remote Sensing, 2023: 1-18.
[4]	姜继兰, 刘屹岷, 李建平, 等. 印度洋偶极子研究进展回顾[J]. 地球科学进展, 2021, 36(6):579-591. DOI
	JIANG J L, LIU Y M, LI J P, et al. Indian Ocean Dipole: A review and perspective[J]. Advances in Earth Science, 2021, 36(6): 579-591.
[5]	MAGEE A D, KIEM A S. Using indicators of ENSO, IOD, and SAM to improve lead time and accuracy of tropical cyclone outlooks for Australia[J]. Journal of Applied Meteorology and Climatology, 2020, 59(11): 1901-1917.
[6]	ROY I, MLIWA M, TROCCOLI A. Important drivers of East African monsoon variability and improving rainy season onset prediction[J]. Natural Hazards, 2024, 120: 429-445.
[7]	LIGUORI G, MCGREGOR S, SINGH M, et al. Revisiting ENSO and IOD contributions to Australian precipitation[J]. Geophysical Research Letters, 2022, 49(1): e2021GL094295.
[8]	KURNIADI A, WELLER E, MIN S K, et al. Independent ENSO and IOD impacts on rainfall extremes over Indonesia[J]. International Journal of Climatology, 2021, 41(6): 3640-3656.
[9]	NGUYEN-LE D, NGO-DUC T, MATSUMOTO J. The tele-connection of the two types of ENSO and Indian Ocean Dipole on Southeast Asian autumn rainfall anomalies[J]. Climate Dynamics, 2024, 62(6): 1-23.
[10]	SAJI N H, GOSWAMI B N, VINAYACHANDRAN P N, et al. A dipole mode in the tropical Indian Ocean[J]. Nature, 1999, 401(6751): 360-363.
[11]	CAI W J, SANTOSO A, WANG G J, et al. Increased frequency of extreme Indian Ocean Dipole events due to greenhouse warming[J]. Nature, 2014, 510(7504): 254-258.
[12]	DU Y, ZHANG Y H, ZHANG L Y, et al. Thermo-cline warming induced extreme Indian Ocean Dipole in 2019[J]. Geophysical Research Letters, 2020, 47(18):e2020GL090079.
[13]	ABRAM N J, GAGAN M K, COLE J E, et al. Recent intensification of tropical climate variability in the Indian Ocean[J]. Nature Geoscience, 2008, 1(12): 849-853.
[14]	ABRAM N J, WRIGHT N M, ELLIS B, et al. Coupling of Indo-Pacific climate variability over the last millennium[J]. Nature, 2020, 579(7799): 385-392.
[15]	LING F H, LUO J J, LI Y, et al. Multi-task machine learning improves multi-seasonal prediction of the Indian Ocean Dipole[J]. Nature Communications, 2022, 13(1): 7681.
[16]	MCKENNA S, SANTOSO A, GUPTA A S, et al. Indian Ocean Dipole in CMIP5 and CMIP6: Characteristics, biases, and links to ENSO[J]. Scientific Reports, 2020, 10(1): 11500.
[17]	ZHANG Y, ZHOU W, WANG X, et al. IOD,ENSO, and seasonal precipitation variation over Eastern China[J]. Atmospheric Research, 2022, 270: 106042.
[18]	HUANG K, HUANG B H, WANG D X, et al. Diversity of strong negative Indian Ocean Dipole events since 1980: Characteristics and causes[J]. Climate Dynamics, 2024, 62(3): 2017-2040.
[19]	LESTARI R K, KOH T Y. Statistical evidence for asy-mmetry in ENSO-IOD interactions[J]. Atmosphere-Ocean, 2016, 54(5): 498-504.
[20]	董笛, 何金海. 印度洋偶极子与两类El Niño的关联及其可能机理[J]. 热带气象学报, 2015, 31(2):182-192.
	DONG D, HE J H. The linkage between Indian Ocean Dipole and two types of El Niño event and its possible mechanisms[J]. Journal of Tropical Meteorology, 2015, 31(2): 182-192.
[21]	LI Z G, CAI W J, LIN X P. Dynamics of changing impacts of tropical Indo-Pacific variability on Indian and Australian rainfall[J]. Scientific Reports, 2016, 6: 31767. DOI PMID
[22]	POWER K, AXELSSON J, WANGDI N, et al. Regional and local impacts of the ENSO and IOD events of 2015 and 2016 on the Indian summer monsoon—A Bhutan case study[J]. Atmosphere, 2021, 12(8): 954.
[23]	SEMENOV S M, POPOV I O, YASYUKEVICH V V. Statis-tical model for assessing the formation of climate-related hazards based on climate monitoring data[J]. Russian Meteorology and Hydrology, 2020, 45(5): 339-344.
[24]	PEREVEDENTSEV Y P, SHANTALINSKII K M, GURY-ANOV V V, et al. Empirical statistical model of climatic changes in the Volga region[J]. IOP Conference Series: Earth and Environmental Science, 2018, 211: 012016.
[25]	WANG Y F, XU Y P, SONG S, et al. Assessing the impacts of climatic and anthropogenic factors on water level variation in the Taihu Plain based on non-stationary statistical models[J]. Environmental Science and Pollution Research International, 2020, 27(18): 22829-22842. DOI PMID
[26]	方玥炜, 唐佑民, 李俊德, 等. 几种统计模型对热带印度洋海温异常的预报[J]. 海洋学研究, 2018, 36(1):1-15. DOI
	FANG Y W, TANG Y M, LI J D, et al. Several statistical models to predict tropical Indian Ocean sea surface tempe-rature anomaly[J]. Journal of Marine Sciences, 2018, 36(1): 1-15.
[27]	LI X W, BORDBAR M H, LATIF M, et al. Monthly to seasonal prediction of tropical Atlantic sea surface tempe-rature with statistical models constructed from observations and data from the Kiel Climate Model[J]. Climate Dyna-mics, 2020, 54(3): 1829-1850.
[28]	YANG L C, FRANZKE C L E, FU Z T. Evaluation of the ability of regional climate models and a statistical model to represent the spatial characteristics of extreme precipitation[J]. International Journal of Climatology, 2020, 40(15): 6612-6628.
[29]	FERREIRA N C R, MIRANDA J H, COOKE R. Climate change and extreme events on drainage systems: Numerical simulation of soil water in corn crops in Illinois (USA)[J]. International Journal of Biometeorology, 2021, 65(7): 1001-1013. DOI PMID
[30]	FANG X W, LI Z, CHENG C, et al. Response of freezing/thawing indexes to the wetting trend under warming climate conditions over the Qinghai-Tibetan Plateau during 1961-2010: A numerical simulation[J]. Advances in Atmospheric Sciences, 2023, 40(2): 211-222.
[31]	晏红明, 杨辉, 李崇银. 赤道印度洋海温偶极子的气候影响及数值模拟研究[J]. 海洋学报, 2007, 29(5):31-39.
	YAN H M, YANG H, LI C Y. Numerical simulations on the climate impacts of temperature dipole in the equatorial Indian Ocean[J]. Acta Oceanologica Sinica, 2007, 29(5): 31-39.
[32]	胡帅, 吴波, 周天军. 近期气候预测系统IAP-DecPreS对印度洋偶极子的回报技巧:全场同化和异常场同化的比较[J]. 大气科学, 2019, 43(4):831-845.
	HU S, WU B, ZHOU T J. Predictive skill of the near-term climate prediction system IAP-DecPreS for the Indian Ocean Dipole: A comparison of full-field and anomaly initializations[J]. Chinese Journal of Atmospheric Sciences, 2019, 43(4): 831-845.
[33]	FANG M, LI X. Paleoclimate data assimilation: Its moti-vation, progress and prospects[J]. Science China Earth Sciences, 2016, 59(9): 1817-1826.
[34]	SHAHABADI M B, BÉLAIR S, BILODEAU B, et al. Impact of weak coupling between land and atmosphere data assimilation systems on environment and climate change Canada’s global deterministic prediction system[J]. Weather and Forecasting, 2019, 34(6): 1741-1758.
[35]	HU G N, DANCE S L, BANNISTER R N, et al. Progress, challenges, and future steps in data assimilation for convection-permitting numerical weather prediction: Report on the virtual meeting held on 10 and 12 November 2021[J]. Atmospheric Science Letters, 2023, 24(1): e1130.
[36]	DANIEL D, AYENEW T, FLETCHER C G, et al. Nume-rical groundwater flow modelling under changing climate in Abaya-Chamo lakes basin, Rift Valley, Southern Ethiopia[J]. Modeling Earth Systems and Environment, 2022, 8(3): 3985-3999.
[37]	SHINDE P P, SHAH S. A review of machine learning and deep learning applications[C]//2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA), August 16-18, 2018, Pune, India. IEEE, 2018: 1-6.
[38]	HAM Y G, KIM J H, LUO J J. Deep learning for multi-year ENSO forecasts[J]. Nature, 2019, 573(7775): 568-572.
[39]	SUN M, CHEN L, LI T, et al. CNN-based ENSO forecasts with a focus on SSTA zonal pattern and physical interpre-tation[J]. Geophysical Research Letters, 2023, 50(20):e2023GL105175.
[40]	LIU J, TANG Y M, WU Y L, et al. Forecasting the Indian Ocean Dipole with deep learning techniques[J]. Geophysical Research Letters, 2021, 48(20): e2021GL094407.
[41]	ZHANG Y, GU Z H, VAN GRIENSVEN THÉ J, et al. The discharge forecasting of multiple monitoring station for Humber River by Hybrid LSTM models[J]. Water, 2022, 14(11): 1794.
[42]	NIGAM A, SRIVASTAVA S. Hybrid deep learning models for traffic stream variables prediction during rainfall[J]. Multimodal Transportation, 2023, 2(1): 100052.
[43]	CAO X Q, GUO Y N, LIU B N, et al. ENSO prediction based on long short-term memory (LSTM)[J]. IOP Confe-rence Series: Materials Science and Engineering, 2020, 799: 012035.
[44]	HUANG A, VEGA-WESTHOFF B, SRIVER R L. Analyz-ing El Niño-Southern Oscillation predictability using long-short-term-memory models[J]. Earth and Space Science, 2019, 6(2): 212-221.
[45]	KHEYRURI Y, SHARAFATI A, NESHAT A. Predicting agricultural drought using meteorological and ENSO parameters in different regions of Iran based on the LSTM model[J]. Stochastic Environmental Research and Risk Assessment, 2023, 37(9): 3599-3613.
[46]	HAQ D Z, RINI NOVITASARI D C, HAMID A, et al. Long short-term memory algorithm for rainfall prediction based on El- Niño and IOD data[J]. Procedia Computer Science, 2021, 179: 829-837.
[47]	谢文鸿, 徐广珺, 董昌明. 基于ConvLSTM机器学习的风暴潮漫滩预报研究[J]. 大气科学学报, 2022, 45(5):674-687.
	XIE W H, XU G J, DONG C M. Research on storm surge floodplain prediction based on ConvLSTM machine learning[J]. Transactions of Atmospheric Sciences, 2022, 45(5): 674-687.
[48]	GENG H T, HU Z Y, WANG T L. ConvLSTM based temperature forecast modification model for north China[J]. Journal of Tropical Meteorology, 2022, 28(4): 405-412.
[49]	MOISHIN M, DEO R C, PRASAD R, et al. Designing deep-based learning flood forecast model with ConvLSTM hybrid algorithm[J]. IEEE Access, 2021, 9: 50982-50993.
[50]	LI C, FENG Y, SUN T Y, et al. Long term Indian Ocean Dipole (IOD) index prediction used deep learning by ConvLSTM[J]. Remote Sensing, 2022, 14(3): 523.
[51]	RATNAM J V, DIJKSTRA H A, BEHERA S K. A machine learning based prediction system for the Indian Ocean Dipole[J]. Scientific Reports, 2020, 10(1): 284.
[52]	CHEN P, SUN B, WANG H J, et al. Improving the CFSv2 prediction of the Indian Ocean Dipole based on a physical-empirical model and a deep-learning approach[J]. International Journal of Climatology, 2022, 42(16): 9200-9214.
[53]	WANG G G, CHENG H L, ZHANG Y M, et al. ENSO analysis and prediction using deep learning: A review[J]. Neurocomputing, 2023, 520: 216-229.
[54]	LUCIA S, KARG B. A deep learning-based approach to robust nonlinear model predictive control[J]. IFAC PapersOnLine, 2018, 51(20): 511-516.
[55]	HABIBI O, CHEMMAKHA M, LAZAAR M. Effect of features extraction and selection on the evaluation of machine learning models[J]. IFAC-PapersOnLine, 2022, 55(12): 462-467.
[56]	CHARILAOU P, BATTAT R. Machine learning models and over-fitting considerations[J]. World Journal of Gastroen-terology, 2022, 28(5): 605-607.
[57]	化盈盈, 张岱墀, 葛仕明. 深度学习模型可解释性的研究进展[J]. 信息安全学报, 2020, 5(3):1-12.
	HUA Y Y, ZHANG D C, GE S M. Research progress in the interpretability of deep learning models[J]. Journal of Cyber Security, 2020, 5(3): 1-12.
[58]	DING W P, ABDEL-BASSET M, HAWASH H, et al. Explainability of artificial intelligence methods, applications and challenges: A comprehensive survey[J]. Information Sciences, 2022, 615: 238-292.
[59]	MARIN I, KUZMANIC SKELIN A, GRUJIC T. Empirical evaluation of the effect of optimization and regularization techniques on the generalization performance of deep convolutional neural network[J]. Applied Sciences, 2020, 10(21): 7817.
[60]	SHORTEN C, KHOSHGOFTAAR T M. A survey on image data augmentation for deep learning[J]. Journal of Big Data, 2019, 6(1): 60.
[61]	ZHUANG F Z, QI Z Y, DUAN K Y, et al. A compre-hensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43-76.
[62]	MATVEEV S A, OSELEDETS I V, PONOMAREV E S, et al. Overview of visualization methods for artificial neural networks[J]. Computational Mathematics and Mathematical Physics, 2021, 61(5): 887-899.
[63]	曾春艳, 严康, 王志锋, 等. 深度学习模型可解释性研究综述[J]. 计算机工程与应用, 2021, 57(8):1-9. DOI
	ZENG C Y, YAN K, WANG Z F, et al. Survey of interpretability research on deep learning models[J]. Computer Engineering and Applications, 2021, 57(8): 1-9. DOI
[64]	SAMO M, MASE J M M, FIGUEREDO G. Deep learning with attention mechanisms for road weather detection[J]. Sensors, 2023, 23(2): 798.
[65]	MOHAMMED A, KORA R. A comprehensive review on ensemble deep learning: Opportunities and challenges[J]. Journal of King Saud University-Computer and Information Sciences, 2023, 35(2): 757-774.

方法	优点	局限性
统计模型	简易且计算效率高	对历史数据的过度依赖使其对未知的气候变化处理困难;无法捕捉复杂的非线性关系;对观测误差敏感
数值模式	可处理复杂的物理机制和相互作用,有助于理解IOD的物理机制;可提供详细的时空分辨率;可进行长期预测	对初始条件和参数化方案敏感;计算复杂度高,计算资源需求大

方法	优点	局限性
统计模型	简易且计算效率高	对历史数据的过度依赖使其对未知的气候变化处理困难;无法捕捉复杂的非线性关系;对观测误差敏感
数值模式	可处理复杂的物理机制和相互作用,有助于理解IOD的物理机制;可提供详细的时空分辨率;可进行长期预测	对初始条件和参数化方案敏感;计算复杂度高,计算资源需求大

项目	优势	局限性
数据处理	能够有效地处理和分析大规模的气候数据集,从历史数据中发现不易被传统方法察觉的复杂模式和关联。	需要大量计算资源的同时还对大数据集有较高的依赖性,对数据质量和完整性比较敏感。
拟合能力	其卓越的非线性拟合能力,能够建立和模拟数据中复杂的非线性关系,提高预测的准确度。	在数据量有限或模型过于复杂时,深度学习模型可能会过度拟合训练数据,从而降低其在新数据上的泛化能力。
自动特征学习	能够自动从原始数据中提取有用的特征,减少了对传统特征提取方法的依赖,提高了效率。	无
可解释性	无	决策过程的不透明性,使得其预测难以被直观理解和验证,从而可能限制其在关键领域(如灾害预警)的应用。

项目	优势	局限性
数据处理	能够有效地处理和分析大规模的气候数据集,从历史数据中发现不易被传统方法察觉的复杂模式和关联。	需要大量计算资源的同时还对大数据集有较高的依赖性,对数据质量和完整性比较敏感。
拟合能力	其卓越的非线性拟合能力,能够建立和模拟数据中复杂的非线性关系,提高预测的准确度。	在数据量有限或模型过于复杂时,深度学习模型可能会过度拟合训练数据,从而降低其在新数据上的泛化能力。
自动特征学习	能够自动从原始数据中提取有用的特征,减少了对传统特征提取方法的依赖,提高了效率。	无
可解释性	无	决策过程的不透明性,使得其预测难以被直观理解和验证,从而可能限制其在关键领域(如灾害预警)的应用。