Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning
The Pacific Decadal Oscillation (PDO), the dominant pattern of sea surface temperature anomalies in the North Pacific basin, is an important low-frequency climate phenomenon. Leveraging data spanning from 1871 to 2010, we employed machine learning models to predict the PDO based on variations in sev...
Published in: | Remote Sensing |
---|---|
Main Authors: | , , , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
MDPI AG
2024
|
Subjects: | |
Online Access: | https://doi.org/10.3390/rs16132261 https://doaj.org/article/5f8004b058ac46f09a6f5879f38b265c |
id |
ftdoajarticles:oai:doaj.org/article:5f8004b058ac46f09a6f5879f38b265c |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:5f8004b058ac46f09a6f5879f38b265c 2024-09-09T19:26:13+00:00 Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning Zhixiong Yao Dongfeng Xu Jun Wang Jian Ren Zhenlong Yu Chenghao Yang Mingquan Xu Huiqun Wang Xiaoxiao Tan 2024-06-01T00:00:00Z https://doi.org/10.3390/rs16132261 https://doaj.org/article/5f8004b058ac46f09a6f5879f38b265c EN eng MDPI AG https://www.mdpi.com/2072-4292/16/13/2261 https://doaj.org/toc/2072-4292 doi:10.3390/rs16132261 2072-4292 https://doaj.org/article/5f8004b058ac46f09a6f5879f38b265c Remote Sensing, Vol 16, Iss 13, p 2261 (2024) PDO prediction machine learning model explanation Science Q article 2024 ftdoajarticles https://doi.org/10.3390/rs16132261 2024-08-05T17:48:57Z The Pacific Decadal Oscillation (PDO), the dominant pattern of sea surface temperature anomalies in the North Pacific basin, is an important low-frequency climate phenomenon. Leveraging data spanning from 1871 to 2010, we employed machine learning models to predict the PDO based on variations in several climatic indices: the Niño3.4, North Pacific index (NPI), sea surface height (SSH), and thermocline depth over the Kuroshio–Oyashio Extension (KOE) region (SSH_KOE and Ther_KOE), as well as the Arctic Oscillation (AO) and Atlantic Multi-decadal Oscillation (AMO). A comparative analysis of the temporal and spatial performance of six machine learning models was conducted, revealing that the Gated Recurrent Unit model demonstrated superior predictive capabilities compared to its counterparts, through the temporal and spatial analysis. To better understand the inner workings of the machine learning models, SHapley Additive exPlanations (SHAP) was adopted to present the drivers behind the model’s predictions and dynamics for modeling the PDO. Our findings indicated that the Niño3.4, North Pacific index, and SSH_KOE were the three most pivotal features in predicting the PDO. Furthermore, our analysis also revealed that the Niño3.4, AMO, and Ther_KOE indices were positively associated with the PDO, whereas the NPI, SSH_KOE, and AO indices exhibited negative correlations. Article in Journal/Newspaper Arctic Directory of Open Access Journals: DOAJ Articles Arctic Pacific Oyashio ENVELOPE(157.000,157.000,50.000,50.000) Remote Sensing 16 13 2261 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
PDO prediction machine learning model explanation Science Q |
spellingShingle |
PDO prediction machine learning model explanation Science Q Zhixiong Yao Dongfeng Xu Jun Wang Jian Ren Zhenlong Yu Chenghao Yang Mingquan Xu Huiqun Wang Xiaoxiao Tan Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning |
topic_facet |
PDO prediction machine learning model explanation Science Q |
description |
The Pacific Decadal Oscillation (PDO), the dominant pattern of sea surface temperature anomalies in the North Pacific basin, is an important low-frequency climate phenomenon. Leveraging data spanning from 1871 to 2010, we employed machine learning models to predict the PDO based on variations in several climatic indices: the Niño3.4, North Pacific index (NPI), sea surface height (SSH), and thermocline depth over the Kuroshio–Oyashio Extension (KOE) region (SSH_KOE and Ther_KOE), as well as the Arctic Oscillation (AO) and Atlantic Multi-decadal Oscillation (AMO). A comparative analysis of the temporal and spatial performance of six machine learning models was conducted, revealing that the Gated Recurrent Unit model demonstrated superior predictive capabilities compared to its counterparts, through the temporal and spatial analysis. To better understand the inner workings of the machine learning models, SHapley Additive exPlanations (SHAP) was adopted to present the drivers behind the model’s predictions and dynamics for modeling the PDO. Our findings indicated that the Niño3.4, North Pacific index, and SSH_KOE were the three most pivotal features in predicting the PDO. Furthermore, our analysis also revealed that the Niño3.4, AMO, and Ther_KOE indices were positively associated with the PDO, whereas the NPI, SSH_KOE, and AO indices exhibited negative correlations. |
format |
Article in Journal/Newspaper |
author |
Zhixiong Yao Dongfeng Xu Jun Wang Jian Ren Zhenlong Yu Chenghao Yang Mingquan Xu Huiqun Wang Xiaoxiao Tan |
author_facet |
Zhixiong Yao Dongfeng Xu Jun Wang Jian Ren Zhenlong Yu Chenghao Yang Mingquan Xu Huiqun Wang Xiaoxiao Tan |
author_sort |
Zhixiong Yao |
title |
Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning |
title_short |
Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning |
title_full |
Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning |
title_fullStr |
Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning |
title_full_unstemmed |
Predicting and Understanding the Pacific Decadal Oscillation Using Machine Learning |
title_sort |
predicting and understanding the pacific decadal oscillation using machine learning |
publisher |
MDPI AG |
publishDate |
2024 |
url |
https://doi.org/10.3390/rs16132261 https://doaj.org/article/5f8004b058ac46f09a6f5879f38b265c |
long_lat |
ENVELOPE(157.000,157.000,50.000,50.000) |
geographic |
Arctic Pacific Oyashio |
geographic_facet |
Arctic Pacific Oyashio |
genre |
Arctic |
genre_facet |
Arctic |
op_source |
Remote Sensing, Vol 16, Iss 13, p 2261 (2024) |
op_relation |
https://www.mdpi.com/2072-4292/16/13/2261 https://doaj.org/toc/2072-4292 doi:10.3390/rs16132261 2072-4292 https://doaj.org/article/5f8004b058ac46f09a6f5879f38b265c |
op_doi |
https://doi.org/10.3390/rs16132261 |
container_title |
Remote Sensing |
container_volume |
16 |
container_issue |
13 |
container_start_page |
2261 |
_version_ |
1809895872130973696 |