The convolutional neural network for Pacific decadal oscillation forecast

doi:10.1088/1748-9326/ad8be2

IOCAS-IR > 海洋环流与波动重点实验室

	The convolutional neural network for Pacific decadal oscillation forecast
	Skanupong, Nutta 1,2,3; Xu, Yongsheng1,2,3,4,5 ; Yu, Lejiang 6; Wan, Zhang 7; Wang, Shuo8
	2024-12-01
发表期刊	ENVIRONMENTAL RESEARCH LETTERS
ISSN	1748-9326
卷号	19 期号:12 页码:11
通讯作者	Xu, Yongsheng([email protected])
摘要	The Pacific decadal oscillation (PDO) is often described as a long-lived El Nino-like pattern of Pacific climate variability, and it has widespread climate and ecosystem impacts. PDO forecasts can provide useful information for policymakers on how to handle PDO impacts. Nevertheless, due to the long duration of the PDO cycles and their complex formation mechanisms, it remains a challenge to predict long lead time PDO. In this paper, we propose a transfer-learning-enhanced convolutional neural network (CNN) to tackle complex ocean dynamic forecasting and predict PDO events with up to a one-year lead time. Our method first trains the CNN on historical simulations from Coupled Model Intercomparison Project 6 (CMIP6), covering the period from 1850 to 1972. This prior knowledge is then refined by further training the model with observational data from 1854 to 1972, ensuring robust performance on unseen data. Additionally, k-fold cross-validation is also employed to evaluate the model's performance across diverse subsets of data, enhancing its reliability. Throughout the testing phase from 1983 to 2022, the CNN model consistently outperforms existing dynamical forecast systems, exhibiting superior correlation skills in predicting annual mean PDO indices and PDO phases, including displaying resilience to seasonal variations. The transferred CNN is thus a powerful method to predict PDO events and is potentially valuable for a wide range of applications. This work directly supports the objectives of the World Climate Research Programme Grand Challenge on Climate Prediction.
关键词	convolutional neural network (CNN) machine learning (ML) pacific decadal oscillation (PDO) sea surface temperature (SST)
DOI	10.1088/1748-9326/ad8be2
收录类别	SCI
语种	英语
资助项目	Strategic Priority Research Program of the Chinese Academy of Sciences[LSKJ202201406-2]; Laoshan Laboratory science and technology innovation projects[U22A20587]; NSFC-Shandong Joint Fund Key Project[41906027]; National Natural Science Foundation of China
WOS研究方向	Environmental Sciences & Ecology ; Meteorology & Atmospheric Sciences
WOS类目	Environmental Sciences ; Meteorology & Atmospheric Sciences
WOS记录号	WOS:001350947800001
出版者	IOP Publishing Ltd
WOS关键词	SEA-SURFACE TEMPERATURE ; NORTH PACIFIC ; MODEL ; VARIABILITY
引用统计
文献类型	期刊论文
条目标识符	http://ir.qdio.ac.cn/handle/337002/199418
专题	海洋环流与波动重点实验室
通讯作者	Xu, Yongsheng
作者单位	1.Chinese Acad Sci, Inst Oceanol, Key Lab Ocean Observat & Forecasting, Qingdao, Peoples R China 2.Chinese Acad Sci, Inst Oceanol, Key Lab Ocean Circulat & Waves, Qingdao, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China 4.Laoshan Lab, Lab Ocean & Climate Dynam, Qingdao, Peoples R China 5.Chinese Acad Sci, Ctr Ocean Megasci, Qingdao, Peoples R China 6.Polar Res Inst China, MNR Key Lab Polar Sci, Shanghai, Peoples R China 7.Univ Manchester, Dept Comp Sci, Manchester, England 8.Univ Birmingham, Sch Comp Sci, Birmingham, England
推荐引用方式 GB/T 7714	Skanupong, Nutta,Xu, Yongsheng,Yu, Lejiang,et al. The convolutional neural network for Pacific decadal oscillation forecast[J]. ENVIRONMENTAL RESEARCH LETTERS,2024,19(12):11.
APA	Skanupong, Nutta,Xu, Yongsheng,Yu, Lejiang,Wan, Zhang,&Wang, Shuo.(2024).The convolutional neural network for Pacific decadal oscillation forecast.ENVIRONMENTAL RESEARCH LETTERS,19(12),11.
MLA	Skanupong, Nutta,et al."The convolutional neural network for Pacific decadal oscillation forecast".ENVIRONMENTAL RESEARCH LETTERS 19.12(2024):11.