Editorial Type:
Article
Article Type:
Research Article

A Multivariable Convolutional Neural Network for Forecasting Synoptic-Scale Sea Surface Temperature Anomalies in the South China Sea

Yonglan Miao aSchool of Marine Science and Technology, Tianjin University, Tianjin, China

Search for other papers by Yonglan Miao in
Current site
Google Scholar
PubMed Close
,
Cuicui Zhang aSchool of Marine Science and Technology, Tianjin University, Tianjin, China

Search for other papers by Cuicui Zhang in
Current site
Google Scholar
PubMed Close
,
Xuefeng Zhang aSchool of Marine Science and Technology, Tianjin University, Tianjin, China

Search for other papers by Xuefeng Zhang in
Current site
Google Scholar
PubMed Close
, and
Lianxin Zhang bKey Laboratory of Marine Environmental Information Technology, National Marine Data and Information Service, Ministry of Natural Resources, Tianjin, China

Search for other papers by Lianxin Zhang in
Current site
Google Scholar
PubMed Close
Online Publication:
01 Jun 2023
Print Publication:
01 Jun 2023
DOI:
https://doi.org/10.1175/WAF-D-22-0094.1
Page(s):
849–863
Restricted access

Abstract

The sea surface temperature anomaly (SSTA) plays a key role in climate change and extreme weather processes. Usually, SSTA forecast methods consist of numerical and conventional statistical models, and the former can be seriously influenced by the uncertainty of physical parameterization schemes, the nonlinearity of ocean dynamic processes, and the nonrobustness of numerical discretization algorithms. Recently, deep learning has been explored to address forecast issues in the field of oceanography. However, existing deep learning models for ocean forecasting are mainly site specific, which were designed for forecasting on a single point or for an independent variable. Moreover, few special deep learning networks have been developed to deal with SSTA field forecasts under typhoon conditions. In this study, a multivariable convolutional neural network (MCNN) is proposed, which can be applied for synoptic-scale SSTA forecasting in the South China Sea. In addition to the SSTA itself, the surface wind speed and the surface current velocity are regarded as input variables for the prediction networks, effectively reflecting the influences of both local atmospheric dynamic forcing and nonlocal oceanic thermal advection. Experimental results demonstrate that MCNN exhibits better performance than a single-variable convolutional neural network (SCNN), especially for the SSTA forecast during the typhoon passage. While forecast results deteriorate rapidly in the SCNN during the passage of a typhoon, forecast errors in the MCNN can be effectively restrained to slowly increase over the forecast time due to the introduction of the surface wind speed in this network.

© 2023 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Xuefeng Zhang, xuefeng.zhang@tju.edu.cn

Abstract

The sea surface temperature anomaly (SSTA) plays a key role in climate change and extreme weather processes. Usually, SSTA forecast methods consist of numerical and conventional statistical models, and the former can be seriously influenced by the uncertainty of physical parameterization schemes, the nonlinearity of ocean dynamic processes, and the nonrobustness of numerical discretization algorithms. Recently, deep learning has been explored to address forecast issues in the field of oceanography. However, existing deep learning models for ocean forecasting are mainly site specific, which were designed for forecasting on a single point or for an independent variable. Moreover, few special deep learning networks have been developed to deal with SSTA field forecasts under typhoon conditions. In this study, a multivariable convolutional neural network (MCNN) is proposed, which can be applied for synoptic-scale SSTA forecasting in the South China Sea. In addition to the SSTA itself, the surface wind speed and the surface current velocity are regarded as input variables for the prediction networks, effectively reflecting the influences of both local atmospheric dynamic forcing and nonlocal oceanic thermal advection. Experimental results demonstrate that MCNN exhibits better performance than a single-variable convolutional neural network (SCNN), especially for the SSTA forecast during the typhoon passage. While forecast results deteriorate rapidly in the SCNN during the passage of a typhoon, forecast errors in the MCNN can be effectively restrained to slowly increase over the forecast time due to the introduction of the surface wind speed in this network.

© 2023 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Xuefeng Zhang, xuefeng.zhang@tju.edu.cn
Save
Cover Weather and Forecasting
Weather and Forecasting

  • Albawi, S., T. A. Mohammed, and S. Al-Zawi, 2017: Understanding of a convolutional neural network. 2017 Int. Conf. on Engineering and Technology (ICET), Antalya, Turkey, Institute of Electrical and Electronics Engineers, 1–6, https://doi.org/10.1109/ICEngTechnol.2017.8308186.

  • Barnett, T. P., and R. W. Preisendorfer, 1987: Origins and levels of monthly and seasonal forecast skill for United States surface air temperatures determined by canonical correlation analysis. Mon. Wea. Rev., 115, 1825–1850, https://doi.org/10.1175/1520-0493(1987)115<1825:OALOMA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation

  • Barnston, A. G., and T. M. Smith, 1996: Specification and prediction of global surface temperature and precipitation from global SST using CCA. J. Climate, 9, 2660–2697, https://doi.org/10.1175/1520-0442(1996)009<2660:SAPOGS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation

  • Chao, S.-Y., P.-T. Shaw, and J. Wang, 1995: Wind relaxation as possible cause of the South China Sea warm current. J. Oceanogr., 51, 111–132, https://doi.org/10.1007/BF02235940.

    • Search Google Scholar
    • Export Citation

  • Choy, C.-W., M.-C. Wu, and T.-C. Lee, 2020: Assessment of the damages and direct economic loss in Hong Kong due to Super Typhoon Mangkhut in 2018. Trop. Cyclone Res. Rev., 9, 193–205, https://doi.org/10.1016/j.tcrr.2020.11.001.

    • Search Google Scholar
    • Export Citation

  • Clevert, D. A., U. Thomas, and H. Sepp, 2016: Fast and accurate deep network learning by exponential linear units (ELUs). arXiv, 1511.07289v5, https://doi.org/10.48550/arXiv.1511.07289.

  • Collins, D. C., C. J. C. Reason, and F. Tangang, 2004: Predictability of Indian Ocean sea surface temperature using canonical correlation analysis. Climate Dyn., 22, 481–497, https://doi.org/10.1007/s00382-004-0390-4.

    • Search Google Scholar
    • Export Citation

  • Dare, R. A., and J. L. McBride, 2011: Sea surface temperature response to tropical cyclones. Mon. Wea. Rev., 139, 3798–3808, https://doi.org/10.1175/MWR-D-10-05019.1.

    • Search Google Scholar
    • Export Citation

  • Feng, W. K., and C. W. Bao, 1982: Topographic and geomorphological characteristics of South China Sea. Mar. Geol. Res., 4, 80–93.

  • Gao, X., and S. Mathur, 2021: Predictability of U.S. regional extreme precipitation occurrence based on large-scale meteorological patterns (LSMPs). J. Climate, 34, 7181–7198, https://doi.org/10.1175/JCLI-D-21-0137.1.

    • Search Google Scholar
    • Export Citation

  • Goodfellow, I., Y. Bengio, and A. Courville, 2016: Deep Learning. MIT Press, 800 pp.

  • Graham, N. E., J. Michaelson, and T. P. Barnett, 1987: An investigation of the El Niño–Southern Oscillation cycle with statistical models: 1. Predictor field characteristics. J. Geophys. Res., 92, 14 251–14 270, https://doi.org/10.1029/JC092iC13p14251.

    • Search Google Scholar
    • Export Citation

  • Ham, Y.-G., J.-H. Kim, and J.-J. Luo, 2019: Deep learning for multi-year ENSO forecasts. Nature, 573, 568–572, https://doi.org/10.1038/s41586-019-1559-7.

    • Search Google Scholar
    • Export Citation

  • Helber, R. W., J. F. Shriver, C. N. Barron, and O. M. Smedstad, 2010: Evaluating the impact of the number of satellite altimeters used in an assimilative ocean prediction system. J. Atmos. Oceanic Technol., 27, 528–546, https://doi.org/10.1175/2009JTECHO683.1.

    • Search Google Scholar
    • Export Citation

  • Hu, W., R. Wu, and Y. Liu, 2014: Relation of the South China Sea precipitation variability to tropical Indo-Pacific SST anomalies during spring-to-summer transition. J. Climate, 27, 5451–5467, https://doi.org/10.1175/JCLI-D-14-00089.1.

    • Search Google Scholar
    • Export Citation

  • Ioffe, S., and C. Szegedy, 2015: Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv, 1502.03167v3, https://doi.org/10.48550/arXiv.1502.03167.

  • Krizhevsky, A., I. Sutskever, and G. E. Hinton, 2012: ImageNet classification with deep convolutional neural networks. Commun. ACM, 25, 1097–1105.

    • Search Google Scholar
    • Export Citation

  • LeCun, Y., Y. Bengio, and G. Hinton, 2015: Deep learning. Nature, 521, 436–444, https://doi.org/10.1038/nature14539.

  • Li, Q., L. Q. Wang, W. M. Ma, and M. Jiang, 2014: Forecasting performance analysis of ECMWF and GFS model for a case of extra-tropical cyclone. Mar. Forecasts, 31, 22–28.

    • Search Google Scholar
    • Export Citation

  • Li, Y.-H., N. Wang, J. Shi, X. Hou, and J. Liu, 2018: Adaptive batch normalization for practical domain adaptation. Pattern Recognit., 80, 109–117, https://doi.org/10.1016/j.patcog.2018.03.005.

    • Search Google Scholar
    • Export Citation

  • Li, Y., C. Fan, Y. Li, Q. Wu, and Y. Ming, 2018: Improving deep neural network with multiple parametric exponential linear units. Neurocomputing, 301, 11–24, https://doi.org/10.1016/j.neucom.2018.01.084.

    • Search Google Scholar
    • Export Citation

  • Lin, A. L., and R. H. Zhang, 2009: The impact of atmospheric wind at low level on sea surface temperature over the South China Sea and its relationship to monsoon. Mark. Sci., 33, 95–100.

    • Search Google Scholar
    • Export Citation

  • Liu, J., B. Jin, L. Wang, and L. Xu, 2020: Sea surface height prediction with deep learning based on attention mechanism. IEEE Geosci. Remote Sens. Lett., 19, 1–5, https://doi.org/10.1109/LGRS.2020.3039062.

    • Search Google Scholar
    • Export Citation

  • Liu, Y., 2014: Performance verification of medium-range forecasting by T639, ECMWF and Japan models from September to November 2013. Meteor. Mon., 40, 247–252.

    • Search Google Scholar
    • Export Citation

  • Ma, C. H., 2004: Preliminary study on the relationship among the main features of landform, the distribution of bottom sediment and fish distribution. Trans. Oceanol. Limnol., 1, 44–51.

    • Search Google Scholar
    • Export Citation

  • McGovern, A., R. Lagerquist, D. J. Gagne, G. E. Jergensen, K. L. Elmore, C. R. Homeyer, and T. Smith, 2019: Making the black box more transparent: Understanding the physical implications of machine learning. Bull. Amer. Meteor. Soc., 100, 2175–2199, https://doi.org/10.1175/BAMS-D-18-0195.1.

    • Search Google Scholar
    • Export Citation

  • Nair, V., and G. E. Hinton, 2010: Rectified linear units improve restricted Boltzmann machines. Proc. 27th Int. Conf. on Machine Learning (ICML-10), Haifa, Israel, Association for Computing Machinery, 807–814, https://dl.acm.org/doi/10.5555/3104322.3104425.

  • Oquab, M., L. Bottou, I. Laptev, and J. Sivic, 2014: Learning and transferring mid-level image representations using convolutional neural networks. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Columbus, OH, Institute of Electrical and Electronics Engineers, 1717–1724, https://doi.org/10.1109/CVPR.2014.222.

  • Patil, K., and M. C. Deo, 2018: Basin-scale prediction of sea surface temperature with artificial neural networks. J. Atmos. Oceanic Technol., 35, 1441–1455, https://doi.org/10.1175/JTECH-D-17-0217.1.

    • Search Google Scholar
    • Export Citation

  • Peng, S.-Q., and L. Xie, 2006: Effect of determining initial conditions by four-dimensional variational data assimilation on storm surge forecasting. Ocean Modell., 14, 1–18, https://doi.org/10.1016/j.ocemod.2006.03.005.

    • Search Google Scholar
    • Export Citation

  • Preacher, K. J., P. J. Curran, and D. J. Bauer, 2006: Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. J. Educ. Behav. Stat., 31, 437–448, https://doi.org/10.3102/10769986031004437.

    • Search Google Scholar
    • Export Citation

  • Qian, Y. F., Q. Q. Wang, and P. Chu, 1999: Numerical experiments of effects of ocean bottom topography on ocean currents, sea surface heights and temperatures in the South China Sea. J. Trop. Meteor., 15, 289–296.

    • Search Google Scholar
    • Export Citation

  • Repelli, C. A., and P. Nobre, 2004: Statistical prediction of sea-surface temperature over the tropical Atlantic. Int. J. Climatol., 24, 45–55, https://doi.org/10.1002/joc.982.

    • Search Google Scholar
    • Export Citation

  • Sadeghi, M., A. A. Asanjan, M. Faridzad, P. Nguyen, K. Hsu, S. Sorooshian, and D. Braithwaite, 2019: PERSIANN-CNN: Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks–convolutional neural networks. J. Hydrometeor., 20, 2273–2289, https://doi.org/10.1175/JHM-D-19-0110.1.

    • Search Google Scholar
    • Export Citation

  • Selvaraju, R., M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, 2017: Grad-CAM: Visual explanations from deep networks via gradient-based localization. 2017 IEEE Int. Conf. on Computer Vision, Venice, Italy, Institute of Electrical and Electronics Engineers, 618–626, https://doi.org/10.1109/ICCV.2017.74.

  • Shao, Q., W. Li, G. J. Han, G. Hou, S. Liu, Y. Gong, and P. Qu, 2021: A deep learning model for forecasting sea surface height anomalies and temperatures in the South China Sea. J. Geophys. Res. Oceans, 126, e2021JC017515, https://doi.org/10.1029/2021JC017515.

    • Search Google Scholar
    • Export Citation

  • Sun, Y., X. Yao, X. Bi, X. Huang, X. Zhao, and B. Qiao, 2021: Time-series graph network for sea surface temperature prediction. Big Data Res., 25, 100237, https://doi.org/10.1016/j.bdr.2021.100237.

    • Search Google Scholar
    • Export Citation

  • Tangang, F. T., W. W. Hsieh, and B. Y. Tang, 1998: Forecasting regional sea surface temperatures in the tropical Pacific by neural network models, with wind stress and sea level pressure as predictors. J. Geophys. Res., 103, 7511–7522, https://doi.org/10.1029/97JC03414.

    • Search Google Scholar
    • Export Citation

  • Wang, G., J. Li, C. Wang, and Y. Yan, 2012: Interactions among the winter monsoon, ocean eddy and ocean thermal front in the South China Sea. J. Geophys. Res., 117, C08002, https://doi.org/10.1029/2012JC008007.

    • Search Google Scholar
    • Export Citation

  • Wang, W. Q., D. X. Wang, and P. Shi, 2001: Annual and interannual variations of large-scale dynamic field in South China Sea. J. Trop. Oceanogr., 20, 61–68.

    • Search Google Scholar
    • Export Citation

  • Wu, R., G. Huang, Z. C. Du, and K. Hu, 2014: Cross-season relation of the South China Sea precipitation variability between winter and summer. Climate Dyn., 43, 193–207, https://doi.org/10.1007/s00382-013-1820-y.

    • Search Google Scholar
    • Export Citation

  • Yang, D. Q., R. X. Hao, J. X. He, and Z. J. He, 2019: Research progress of statistical forecasting methods in ocean prediction. Mar. Info., 34, 1–9, https://doi.org/10.19661/j.cnki.mi.2019.02.001.

    • Search Google Scholar
    • Export Citation

  • Yang, J., D. Wu, and X. Lin, 2008: On the dynamics of the South China Sea warm current. J. Geophys. Res., 113, C08003, https://doi.org/10.1029/2007JC004427.

    • Search Google Scholar
    • Export Citation

  • Ye, L., 1994: On the mechanism of South China Sea warm current and Kuroshio branch in winter—Preliminary results of 3-D baroclinic experiments. Terr. Atmos. Oceanic Sci., 5, 597–610, https://doi.org/10.3319/TAO.1994.5.4.597(O).

    • Search Google Scholar
    • Export Citation

  • Ying, X., Z. Zheng, J. Ni, and K. Zhao, 2022: Numerical simulation study on the dynamic impact of typhoon “Mangkhut” storm surge on the sea area near the Hong Kong-Zhuhai-Macao Bridge. Phys. Chem. Earth, 128, 103269, https://doi.org/10.1016/j.pce.2022.103269.

    • Search Google Scholar
    • Export Citation

  • Yu, L., X. Jin, and R. A. Weller, 2007: Annual, seasonal, and interannual variability of air–sea heat fluxes in the Indian Ocean. J. Climate, 20, 3190–3209, https://doi.org/10.1175/JCLI4163.1.

    • Search Google Scholar
    • Export Citation

  • Zhao, Y., D. Yang, W. Li, C. Liu, X. Deng, R. Hao, and Z. He, 2020: Medium- to long-term forecasts of sea surface height anomalies using a spatiotemporal empirical orthogonal function method. J. Atmos. Oceanic Technol., 37, 2225–2237, https://doi.org/10.1175/JTECH-D-20-0029.1.

    • Search Google Scholar
    • Export Citation

  • Zheng, G., X. Li, R.-H. Zhang, and B. Liu, 2020: Purely satellite data-driven deep learning forecast of complicated tropical instability waves. Sci. Adv., 6, eaba1482, https://doi.org/10.1126/sciadv.aba1482.

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 274 274 17
Full Text Views 218 218 0
PDF Downloads 174 174 2