Prediction of sea surface <italic style="font-style: italic">p</italic>CO<sub>2</sub> in the South China Sea using Spatiotemporal Convolutional LSTM model

Shuang LI; Yu GAO; Jiannan GAO; Yaqi ZHAO; Peng HAO; Jinbao SONG; Chengcheng YU

doi:10.1007/s00343-025-4257-3

Your Location：

Home >

Browse articles >

Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model

Physics | Updated：2026-03-26

- Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model
- Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model
- 海洋湖沼学报(英文) 2026年44卷第1期页码：19-35
- Affiliations：
  
  1.Ocean College, Zhejiang University, Zhoushan 316021, China
  2.Marine Science and Technology College, Zhejiang Ocean University, Zhoushan 316004, China
- Author bio：
  
  haopeng@zju.edu.cn
- Funds：
  
  the National Key Research and Development Program of China(2023YFC3008202);the National Natural Science Foundation of China(42406019);the Scientific Research Fund of Zhejiang Provincial Education Department(Y202353066)
- DOI：10.1007/s00343-025-4257-3
  中图分类号：
- 收稿：2024-09-24，
  
  录用：2025-02-02，
  
  网络首发：2025-03-14，
  
  纸质出版：2026-01-01
- Accepted：
Scan QR Code
Shuang LI, Yu GAO, Jiannan GAO, 等. Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model[J]. 海洋湖沼学报(英文), 2026,44(1):19-35.

LI Shuang,GAO Yu,GAO Jiannan,et al.Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model[J].Journal of Oceanology and Limnology,2026,44(01):19-35.
Shuang LI, Yu GAO, Jiannan GAO, 等. Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model[J]. 海洋湖沼学报(英文), 2026,44(1):19-35. DOI： 10.1007/s00343-025-4257-3.

LI Shuang,GAO Yu,GAO Jiannan,et al.Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model[J].Journal of Oceanology and Limnology,2026,44(01):19-35. DOI： 10.1007/s00343-025-4257-3.

摘要

Abstract

The prediction of sea surface partial pressure of carbon dioxide (

) in the South China Sea is crucial for understanding the region’s contribution to the global carbon budget and its interactions with climate change. We applied the Spatiotemporal Convolutional Long Short-Term Memory (ST-ConvLSTM) model

integrating key environmental factors including sea surface temperature (SST)

sea surface salinity (SSS)

and chlorophyll

(Chl

)

to predict and analyze sea surface

in the South China Sea. The model demonstrated high accuracy in short-term predictions (1 month)

with a mean absolute error (MAE) of 0.394

a root mean square error (RMSE) of 0.659

and a coefficient of determination (

) of 0.998. For long-term predictions (12 months)

the model maintained its predictive capability

with an MAE of 0.667

RMSE of 1.255

and

of 0.994. Feature importance analysis revealed that sea surface

and SST were the main drivers of the model’s predictions

whereas Chl

and SSS had relatively minor impacts. The model’s generalization ability was further validated in the northwest Pacific Ocean and tropical Pacific Ocean

where it successfully captured the spatiotemporal variation in

with small prediction errors. The ST-ConvLSTM model provides an efficient and accurate tool for forecasting and analyzing sea surface

in the South China Sea

offering new insights into global carbon cycling and climate change. This study demonstrates the potential of deep learning in marine science and provides a significant technical support for global changes and marine ecosystem research.

关键词

Keywords

references

Borges A V , Vanderborght J P , Schiettecatte L S et al . 2004 . Variability of the gas transfer velocity of CO 2 in a macrotidal estuary (the Scheldt) . Estuaries , 27 ( 4 ): 593 - 603 , https://doi.org/10.1007/bf02907647 https://doi.org/10.1007/bf02907647 .

Chen S L , Hu C M , Barnes B B et al . 2019 . A machine learning approach to estimate surface ocean p CO 2 from satellite measurements . Remote Sensing of Environment , 228 : 203 - 226 , https://doi.org/10.1016/j.rse.2019.04.019 https://doi.org/10.1016/j.rse.2019.04.019 .

Cosca C E , Feely R A , Boutin J et al . 2003 . Seasonal and interannual CO 2 fluxes for the central and eastern equatorial Pacific Ocean as determined from f CO 2 -SST relationships . Journal of Geophysical Research: Oceans , 108 ( C8 ): 3278 , https://doi.org/10.1029/2000jc000677 https://doi.org/10.1029/2000jc000677 .

Dai M H , Lu Z M , Zhai W D et al . 2009 . Diurnal variations of surface seawater p CO 2 in contrasting coastal environments . Limnology and Oceanography , 54 ( 3 ): 735 - 745 , https://doi.org/10.4319/lo.2009.54.3.0735 https://doi.org/10.4319/lo.2009.54.3.0735 .

Durand G , van den Broeke M R , Le Cozannet G et al . 2022 . Sea-level rise: from global perspectives to local services. Frontiers in Marine Science , 8 : 709595 , https://doi.org/10.3389/fmars.2021.709595 https://doi.org/10.3389/fmars.2021.709595 .

Ekwurzel B , Boneham J , Dalton M W et al . 2017 . The rise in global atmospheric CO 2 , surface temperature, and sea level from emissions traced to major carbon producers . Climatic Change , 144 ( 4 ): 579 - 590 , https://doi.org/10.1007/s10584-017-1978-0 https://doi.org/10.1007/s10584-017-1978-0 .

Falkowski P , Scholes R J , Boyle E et al . 2000 . The global carbon cycle: a test of our knowledge of earth as a system . Science , 290 ( 5490 ): 291 - 296 , https://doi.org/10.1126/science.290.5490.291 https://doi.org/10.1126/science.290.5490.291 .

Fassbender A J , Rodgers K B , Palevsky H I et al . 2018 . Seasonal asymmetry in the evolution of surface ocean p CO 2 and pH thermodynamic drivers and the influence on sea-air CO 2 flux . Global Biogeochemical Cycles , 32 ( 10 ): 1476 - 1497 , https://doi.org/10.1029/2017gb005855 https://doi.org/10.1029/2017gb005855 .

Fu Z Y , Hu L S , Chen Z D et al . 2020 . Estimating spatial and temporal variation in ocean surface p CO 2 in the Gulf of Mexico using remote sensing and machine learning techniques. Science of the Total Environment , 745 : 140965 , https://doi.org/10.1016/j.scitotenv.2020.140965 https://doi.org/10.1016/j.scitotenv.2020.140965 .

Gao Y , Song J B , Li S et al . 2025 . Parameterization of Langmuir circulation under geostrophic effects using the data-driven approach. Progress in Oceanography , 231 : 103403 , https://doi.org/10.1016/j.pocean.2024.103403 https://doi.org/10.1016/j.pocean.2024.103403 .

Gregor L , Kok S , Monteiro P M S . 2017 . Empirical methods for the estimation of Southern Ocean CO 2 : support vector and random forest regression . Biogeosciences , 14 ( 23 ): 5551 - 5569 , https://doi.org/10.5194/bg-14-5551-2017 https://doi.org/10.5194/bg-14-5551-2017 .

Guemas V , Doblas-Reyes F J , Andreu-Burillo I et al . 2013 . Retrospective prediction of the global warming slowdown in the past decade . Nature Climate Change , 3 ( 7 ): 649 - 653 , https://doi.org/10.1038/nclimate1863 https://doi.org/10.1038/nclimate1863 .

Hansen J E , Sato M , Simons L et al . 2023 . Global warming in the pipeline . Oxford Open Climate Change , 3 ( 1 ): kgad 008 , https://doi.org/10.1093/oxfclm/kgad008 https://doi.org/10.1093/oxfclm/kgad008 .

Hao P , Li S , Gao Y . 2023a . Significant wave height prediction based on deep learning in the South China Sea. Frontiers in Marine Science , 9 : 1113788 , https://doi.org/10.3389/fmars.2022.1113788 https://doi.org/10.3389/fmars.2022.1113788 .

Hao P , Li S , Song J B et al . 2023b . Prediction of sea surface temperature in the South China Sea based on deep learning . Remote Sensing , 15 ( 6 ): 1656 , https://doi.org/10.3390/rs15061656 https://doi.org/10.3390/rs15061656 .

Hao P , Zhao Y Q , Li S et al . 2024 . Deep learning approaches in predicting tropical cyclone tracks: an analysis focused on the northwest Pacific region. Ocean Modelling , 192 : 102444 , https://doi.org/10.1016/j.ocemod.2024.102444 https://doi.org/10.1016/j.ocemod.2024.102444 .

Hochreiter S , Schmidhuber J . 1997 . Long short-term memory . Neural Computation , 9 ( 8 ): 1735 - 1780 , https://doi.org/10.1162/neco.1997.9.8.1735 https://doi.org/10.1162/neco.1997.9.8.1735 .

Joshi A P , Kumar V , Warrior H V . 2022 . Modeling the sea-surface p CO 2 of the central Bay of Bengal region using machine learning algorithms. Ocean Modelling , 178 : 102094 , https://doi.org/10.1016/j.ocemod.2022.102094 https://doi.org/10.1016/j.ocemod.2022.102094 .

Krishna K V , Shanmugam P , Nagamani P V . 2020 . A multiparametric nonlinear regression approach for the estimation of global surface ocean p CO 2 using s atellite oceanographic data . IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 13 : 6220 - 6235 , https://doi.org/10.1109/jstars.2020.3026363 https://doi.org/10.1109/jstars.2020.3026363 .

Kuttippurath J , Peter R , Singh A et al . 2022 . The increasing atmospheric CO 2 over India: comparison to global trends . iScience , 25 ( 8 ): 104863 , https://doi.org/10.1016/j.isci.2022.104863 https://doi.org/10.1016/j.isci.2022.104863 .

Le Quéré C , Raupach M R , Canadell J G et al . 2009 . Trends in the sources and sinks of carbon dioxide . Nature Geoscience , 2 ( 12 ): 831 - 836 , https://doi.org/10.1038/ngeo689 https://doi.org/10.1038/ngeo689 .

Le Quéré C , Takahashi T , Buitenhuis E T et al . 2010 . Impact of climate change and variability on the global oceanic sink of CO 2 . Global Biogeochemical Cycles , 24 ( 4 ): GB 4007 , https://doi.org/10.1029/2009gb003599 https://doi.org/10.1029/2009gb003599 .

Li Q , Guo X H , Zhai W D et al . 2020 . Partial pressure of CO 2 and air-sea CO 2 fluxes in the South China Sea: synthesis of an 18-year dataset. Progress in Oceanography , 182 : 102272 , https://doi.org/10.1016/j.pocean.2020.102272 https://doi.org/10.1016/j.pocean.2020.102272 .

Lohrenz S E , Cai W J , Chen F Z et al . 2010 . Seasonal variability in air-sea fluxes of CO 2 in a river-influenced coastal margin . Journal of Geophysical Research: Oceans , 115 ( C10 ): C 10034 , https://doi.org/10.1029/2009jc005608 https://doi.org/10.1029/2009jc005608 .

Ma Y , Li T J , Xia H Y et al . 2022 . Effects of water masses and circulation on the surface water partial pressure of carbon dioxide in summer in eastern Beibu Gulf, China . Journal of Marine Science and Engineering , 11 ( 1 ): 46 , https://doi.org/10.3390/jmse11010046 https://doi.org/10.3390/jmse11010046 .

Peter R , Kuttippurath J , Chakraborty K et al . 2023 . A high concentration CO 2 pool over the Indo-Pacific Warm Pool . Scientific Reports , 13 ( 1 ): 4314 , https://doi.org/10.1038/s41598-023-31468-0 https://doi.org/10.1038/s41598-023-31468-0 .

Reichstein M , Camps-Valls G , Stevens B et al . 2019 . Deep learning and process understanding for data-driven Earth system science . Nature , 566 ( 7743 ): 195 - 204 , https://doi.org/10.1038/s41586-019-0912-1 https://doi.org/10.1038/s41586-019-0912-1 .

Ripple W J , Wolf C , Gregg J W et al . 2023 . The 2023 state of the climate report: entering uncharted territory . BioScience , 73 ( 12 ): 841 - 850 , https://doi.org/10.1093/biosci/biad080 https://doi.org/10.1093/biosci/biad080 .

Rutgersson A , Smedman A , Sahlée E . 2011 . Oceanic convective mixing and the impact on air-sea gas transfer velocity . Geophysical Research Letters , 38 ( 2 ): L 02602 , https://doi.org/10.1029/2010gl045581 https://doi.org/10.1029/2010gl045581 .

Sabine C L , Feely R A , Gruber N et al . 2004 . The oceanic sink for anthropogenic CO 2 . Science , 305 ( 5682 ): 367 - 371 , https://doi.org/10.1126/science.1097403 https://doi.org/10.1126/science.1097403 .

Shi X J , Chen Z R , Wang H et al . 2015 . Convolutional LSTM network: a machine learning approach for precipitation nowcasting . In: Proceedings of the 29th International Conference on Neural Information Processing Systems . MIT Press , Montreal, Canada . p. 802 - 910 .

Sunanda N , Kuttippurath J , Chakraborty A et al . 2023 . Stressors of primary productivity in the north Indian Ocean revealed by satellite, reanalysis and CMIP6 data. Progress in Oceanography , 219 : 103164 , https://doi.org/10.1016/j.pocean.2023.103164 https://doi.org/10.1016/j.pocean.2023.103164 .

Takahashi T , Sutherland S C , Wanninkhof R et al . 2009 . Climatological mean and decadal change in surface ocean p CO 2 , and net sea-air CO 2 flux over the global oceans . Deep Sea Research Part II: Topical Studies in Oceanography , 56 ( 8-10 ): 554 - 577 , https://doi.org/10.1016/j.dsr2.2008.12.009 https://doi.org/10.1016/j.dsr2.2008.12.009 .

Thunell R C , Qingmin M , Calvert S E et al . 1992 . Glacial-Holocene biogenic sedimentation patterns in the South China Sea: productivity variations and surface water p CO 2 . Paleoceanography , 7 ( 2 ): 143 - 162 , https://doi.org/10.1029/92pa00278 https://doi.org/10.1029/92pa00278 .

Tseng C M , Wong G T F , Lin I I et al . 2005 . A unique seasonal pattern in phytoplankton biomass in low-latitude waters in the South China Sea . Geophysical Research Letters , 32 ( 8 ): L 08608 , https://doi.org/10.1029/2004gl022111 https://doi.org/10.1029/2004gl022111 .

Tu Z B , Le C F , Bai Y et al . 2021 . Increase in CO 2 uptake capacity in the Arctic Chukchi Sea during summer revealed by satellite-based estimation . Geophysical Research Letters , 48 ( 15 ): e2021 GL 093844 , https://doi.org/10.1029/2021gl093844 https://doi.org/10.1029/2021gl093844 .

Wang G Z , Shen S S P , Chen Y et al . 2020 . Feasibility of reconstructing the basin-scale sea surface partial pressure of carbon dioxide from sparse in situ observations over the South China Sea . Earth System Science Data , 13 ( 3 ): 1403 - 1417 , https://doi.org/10.5194/essd-13-1403-2021 https://doi.org/10.5194/essd-13-1403-2021 .

Wanninkhof R , Pickers P A , Omar A M et al . 2019 . A surface ocean CO 2 reference network, SOCONET and associated marine boundary layer CO 2 measurements. Frontiers in Marine Science , 6 : 400 , https://doi.org/10.3389/fmars.2019.00400 https://doi.org/10.3389/fmars.2019.00400 .

Zhai W D , Dai M H , Chen B S et al . 2013 . Seasonal variations of sea-air CO 2 fluxes in the largest tropical marginal sea (South China Sea) based on multiple-year underway measurements . Biogeosciences , 10 ( 11 ): 7775 - 7791 , https://doi.org/10.5194/bg-10-7775-2013 https://doi.org/10.5194/bg-10-7775-2013 .

Zhang M Y , Zhu X M , Ji X L et al . 2023a . Controlling factor analysis of oceanic surface p CO 2 in the South China Sea using a three-dimensional high-resolution biogeochemical model. Frontiers in Marine Science , 10 : 1155979 , https://doi.org/10.3389/fmars.2023.1155979 https://doi.org/10.3389/fmars.2023.1155979 .

Zhang S Q , Bai Y , He X Q et al . 2023b . The carbon sink of the Coral Sea, the world's second largest marginal sea, weakened during 2006-2018. Science of the Total Environment , 872 : 162219 , https://doi.org/10.1016/j.scitotenv.2023.162219 https://doi.org/10.1016/j.scitotenv.2023.162219 .

Zhong G R , Li X G , Song J M et al . 2022 . Reconstruction of global surface ocean p CO 2 using region-specific predictors based on a stepwise FFNN regression algorithm . Biogeosciences , 19 ( 3 ): 845 - 859 , https://doi.org/10.5194/bg-19-845-2022 https://doi.org/10.5194/bg-19-845-2022 .

浏览量

Downloads

CSCD

文章被引用时，请邮件提醒。

Submit

工具集

关联资源

A physics-enhanced deep-learning model for estimating turbid shallow water depth from SAR images

Comparative analysis of mesoscale eddies based on Eulerian and Lagrangian frameworks

Deep neural network based on adversarial training for short-term high-resolution precipitation nowcasting from radar echo images

Distribution, source, and sink of microplastics in the nearshore-estuarine systems of the South China Sea basin

Prediction of sea surface pCO2 in the South China Sea using Spatiotemporal Convolutional LSTM model

Prediction of sea surface pCO2 in the South China Sea using Spatiotemporal Convolutional LSTM model

DOI：10.1007/s00343-025-4257-3

摘要

Abstract

关键词

Keywords

references

Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model

Prediction of sea surface pCO₂ in the South China Sea using Spatiotemporal Convolutional LSTM model