, Volume 23, Issue 3, pp 375–396 | Cite as

A hybrid CNN-LSTM model for typhoon formation forecasting

  • Rui Chen
  • Xiang WangEmail author
  • Weimin Zhang
  • Xiaoyu Zhu
  • Aiping Li
  • Chao Yang


A typhoon is an extreme weather event that can cause huge loss of life and economic damage in coastal areas and beyond. As a consequence, the search for more accurate predictive models of typhoon formation; and, intensity have become imperative as meteorologists, governments, and other agencies seek to mitigate the impact of these catastrophic events. While work in this field has progressed diligently, this paper argues, that the existing models are deficient. Traditional numerical forecast models based on fluid mechanics have difficulty in predicting the intensity of typhoons. Forecasts based on statistics and machine learning fail to take into account the spatial and temporal relationships among typhoon formation variables leading to weaknesses in the predictive power of this model. Therefore, we propose a hybrid model, which we argue, can produce a more realist and accurate account of typhoon ‘behavior’ as it focuses on both the spatio-temporal correlations of atmospheric and oceanographic variables. Our CNN-LSTM model introduces 3D convolutional neural networks (3DCNN) and 2D convolutional neural networks (2DCNN) as a method to better understand the spatial relationships of the features of typhoon formation. We also use LSTM to examine the temporal sequence of relations in typhoon progression. Extensive experiments based on three datasets show that our hybrid CNN-LSTM model is superior to existing methods, including numerical forecast models used by many official organizations; and, statistical forecast and machine learning based methods.


Typhoon Forecasting Spatio-temporal CNN LSTM 



This research is partially supported by the the National Key Research and Development Program of China (No. 2018YFC1406206) and National Natural Science Foundation of China (61802424, 61502516, 61472433, 41675097). The final amendation about the English expression is under the help of David Hughes.


  1. 1.
    Peng MS, et al (2012) Developing versus nondeveloping disturbances for tropical cyclone formation. Part I: North Atlantic. Mon Weather Rev 140.4:1047–1066CrossRefGoogle Scholar
  2. 2.
    Fu B, et al (2012) Developing versus nondeveloping disturbances for tropical cyclone formation. Part II: Western North Pacific. Mon Weather Rev 140(4):1067–1080CrossRefGoogle Scholar
  3. 3.
    Hennon CC (2017) Forecasting tropical cyclogenesis over the atlantic basin using Large-Scale data. Monthly weather review. 1–14Google Scholar
  4. 4.
    Kerns BW, Chen SS (2013) Cloud Clusters and tropical cyclogenesis: Developing and nondeveloping systems and their Large-Scale environment. Mon Weather Rev 141 (1):192–210CrossRefGoogle Scholar
  5. 5.
    Xingjian S, et al (2015) Convolutional LSTM network: A machine learning approach for precipitation nowcasting. 802–810Google Scholar
  6. 6.
    Yang Y, et al (2018) A CFCC-LSTM model for sea surface temperature prediction. IEEE Geosci Remote Sens Lett 15(2):207–211CrossRefGoogle Scholar
  7. 7.
    Donahue J, et al (2015) Long-term recurrent convolutional networks for visual recognition and description. 2625–2634Google Scholar
  8. 8.
    Zhang L, et al (2017) Learning Spatiotemporal Features Using 3DCNN and Convolutional LSTM for Gesture Recognition. 3120–3128Google Scholar
  9. 9.
    Qu Q, Liu S, Zhu F, Jensen CS (2016) Efficient online summarization of Large-Scale dynamic networks. IEEE Trans Knowl Data Eng 28(12):3231–3245CrossRefGoogle Scholar
  10. 10.
    Liu S, Qu Q (2016) Dynamic collective routing using crowdsourcing data. Transp Res B Methodol 93:450–469CrossRefGoogle Scholar
  11. 11.
    Liu S, Qu Q, Wang S (2018) Heterogeneous anomaly detection in social diffusion with discriminative feature discovery. Inf Sci 439-440:1–18CrossRefGoogle Scholar
  12. 12.
    Liu S, Qu Q, Wang S (2015) Rationality analytics from trajectories. TKDD 10(1):10:1–10:22CrossRefGoogle Scholar
  13. 13.
    CMA Verification on Forecasts of Tropical Cyclones over Western North Pacific (2012 ∼ 2016)Google Scholar
  14. 14.
    Chen P, Yu H, Chan JCL (2011) A Western North Pacific tropical cyclone intensity prediction scheme[J]. Acta meteorologica sinica 25(5):611–624CrossRefGoogle Scholar
  15. 15.
    Franklin JL (2008) National Hurricane center forecast verification report. National Hurricane center Rep. (2012 ∼ 2016)Google Scholar
  16. 16.
    Song J, Wang Y, Chen P (2011) A statistical prediction scheme of tropical cyclone intensity over the western North Pacific based on the partial least square regression [J]. Acta Meteorol Sinica 69(5):745–756Google Scholar
  17. 17.
    Leiming M (2014) Research progress on China typhoon numerical prediction models and associated major techniques. Progress in Geophys.(in Chinese) 29(3):1013–1022DOIGoogle Scholar
  18. 18.
    Camargo SJ, et al (2007) Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis. J Climate 20(19):4819–4834CrossRefGoogle Scholar
  19. 19.
    Murakami H, Sugi M (2010) Effect of model resolution on tropical cyclone climate projections. Sola 6:73–76CrossRefGoogle Scholar
  20. 20.
    Zhang M, et al (2016) A genesis potential index for Western North Pacific tropical cyclones by oceanic parameters. J Geophys Res Oceans 121(9):7176–7191CrossRefGoogle Scholar
  21. 21.
    Chand SS, Walsh KJE (2011) Forecasting Tropical cyclone formation in the Fiji region: A Probit Regression Approach Using Bayesian Fitting. Weather Forecast 26(2):150–165CrossRefGoogle Scholar
  22. 22.
    Su Y, et al (2018) A New Data Mining Model for Hurricane Intensity Prediction. 98–105Google Scholar
  23. 23.
    Wijnands JS, et al (2016) Variable selection for tropical cyclogenesis predictive modeling. Mon Weather Rev 144(12):4605–4619CrossRefGoogle Scholar
  24. 24.
    Zhang W, et al (2015) Discriminating developing versus nondeveloping tropical disturbances in the Western North Pacific through decision tree analysis. Weather Forecast 30(2):446–454CrossRefGoogle Scholar
  25. 25.
    Zhang W, et al (2013) The application of decision tree to intensity change classification of tropical cyclones in western North Pacific. Geophys Res Lett 40(9):1883–1887CrossRefGoogle Scholar
  26. 26.
    LeCun Y, et al (1989) Backpropagation applied to handwritten zip code recognition. Neural Comput 1(4):541–551CrossRefGoogle Scholar
  27. 27.
    LeCun Y, et al (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324CrossRefGoogle Scholar
  28. 28.
    Hochreiter S, Schmidhuber J (1997) Long Short-term memory. Neural Comput 9(8):1735–1780CrossRefGoogle Scholar
  29. 29.
    Kalchbrenner N, et al (2015) Grid long short-term memory. arXiv:1507.01526
  30. 30.
    Li Y, et al (2017) Leveraging LSTM for rapid intensifications prediction of tropical cyclones. IV-4/W2, 101–105Google Scholar
  31. 31.
    Qiu M, et al (2017) A Short-Term Rainfall Prediction Model using Multi-Task Convolutional Neural Networks. 395–404Google Scholar
  32. 32.
    Ji S, et al (2013) 3D convolutional neural networks for human action recognition. IEEE Trans Pattern Anal Mach Intell 35(1):221–231CrossRefGoogle Scholar
  33. 33.
    Li Y, et al (2016) Large-scale gesture recognition with a fusion of rgb-d data based on the c3d model. 25–30Google Scholar
  34. 34.
    Tran D, et al (2015) Learning spatiotemporal features with 3d convolutional networks. 4489–4497Google Scholar
  35. 35.
    Zhou Z (2016) Machine Learning, Tsinghua University pressGoogle Scholar
  36. 36.
    Schumacher AB, et al (2009) Objective estimation of the 24-h probability of tropical cyclone formation. Weather Forecast 24(2):456–471CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  • Rui Chen
    • 1
  • Xiang Wang
    • 1
    Email author
  • Weimin Zhang
    • 2
  • Xiaoyu Zhu
    • 3
  • Aiping Li
    • 3
  • Chao Yang
    • 3
  1. 1.College of Meteorology and Oceanology, College of ComputerNational University of Defense TechnologyChangshaChina
  2. 2.College of MeteorologyNational University of Defense Technology, Laboratory of Software Engineering for Complex SystemsChangshaChina
  3. 3.College of ComputerNational University of Defense TechnologyChangshaChina

Personalised recommendations