Advertisement

Journal of Meteorological Research

, Volume 33, Issue 5, pp 797–809 | Cite as

Forecasting Different Types of Convective Weather: A Deep Learning Approach

  • Kanghui ZhouEmail author
  • Yongguang Zheng
  • Bo Li
  • Wansheng Dong
  • Xiaoling Zhang
Regular Article

Abstract

A deep learning objective forecasting solution for severe convective weather (SCW) including short-duration heavy rain (HR), hail, convective gusts (CG), and thunderstorms based on numerical weather prediction (NWP) data was developed. We first established the training datasets as follows. Five years of severe weather observations were utilized to label the NCEP final (FNL) analysis data. A large number of labeled samples for each type of weather were then selected for model training. The local temperature, pressure, humidity, and winds from 1000 to 200 hPa, as well as dozens of convective physical parameters, were taken as predictors in our model. A six-layer convolutional neural network (CNN) model was then built and trained to obtain optimal model weights. After that, the trained model was used to predict SCW based on the Global Forecast System (GFS) forecast data as input. The performances of the CNN model and other traditional methods were compared. The results show that the deep learning algorithm had a higher classification accuracy on HR and hail than support vector machine, random forests, and other traditional machine learning algorithms. The objective forecasts by use of the deep learning algorithm also showed better forecasting skills than the subjective forecasts by the forecasters. The threat scores (TSs) of thunderstorm, HR, hail, and CG were increased by 16.1%, 33.2%, 178%, and 55.7%, respectively. The deep learning forecast model is currently used in the National Meteorological Center of China to provide guidance for the operational SCW forecasting over China.

Key words

deep learning convolutional neural network convective weather forecasting 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes

Acknowledgments

The authors thank the editor, one anonymous reviewer, and Dr. David J. Gagne at NCAR for their constructive comments that have greatly improved the content and presentation of this article. One of the authors, Bo Li, acknowledges partial support from the US NSF grants AGS-1602845 and DMS-1830312.

References

  1. Beijbom, O., P. J. Edmunds, D. I. Kline, et al., 2012: Automated annotation of coral reef survey images. Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Providence, RI, USA, 1170–1177, doi:  https://doi.org/10.1109/CVPR.2012.6247798.Google Scholar
  2. Bengio, Y., 2009: Learning Deep Architectures for AI. Foundations and Trends® in Machine Learning, Vol. 2, No. 1, 1–127, now Publishers Inc., Hanover, MA, USA, doi:  https://doi.org/10.1561/2200000006.Google Scholar
  3. Buda, M., A. Maki, and M. A. Mazurowski, 2018: A systematic study of the class imbalance problem in convolutional neural networks. Neural Netw., 106, 249–259, doi:  https://doi.org/10.1016/j.neunet.2018.07.011.CrossRefGoogle Scholar
  4. Chaudhuri, S., 2010: Convective energies in forecasting severe thunderstorms with one hidden layer neural net and variable learning rate back propagation algorithm. Asia-Pacific J. Atmos. Sci., 46, 173–183, doi:  https://doi.org/10.1007/s13143-010-0016-1.CrossRefGoogle Scholar
  5. Cintineo, J. L., M. J. Pavolonis, J. M. Sieglaff, et al., 2014: An empirical model for assessing the severe weather potential of developing convection. Wea. Forecasting, 29, 639–653, doi:  https://doi.org/10.1175/WAF-D-13-00113.1.CrossRefGoogle Scholar
  6. Ciregan, D., U. Meier, and J. Schmidhuber, 2012: Multi-column deep neural networks for image classification. Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Providence, RI, USA, 3642–3649, doi:  https://doi.org/10.1109/CVPR.2012.6248110.Google Scholar
  7. Doswell III, C. A., 2001: Severe Convective Storms. American Meteorological Society, Boston, MA, USA, 561 pp, doi:  https://doi.org/10.1007/978-1-935704-06-5.CrossRefGoogle Scholar
  8. Doswell III, C. A., H. E. Brooks, and R. A. Maddox, 1996: Flash flood forecasting: An ingredients-based methodology. Wea. Forecasting, 11, 560–581, doi:  https://doi.org/10.1175/1520-0434(1996)011<0560:FFFAIB>2.0.CO;2.CrossRefGoogle Scholar
  9. Gagne II, D. J., A. McGovern, and M. Xue, 2014: Machine learning enhancement of storm-scale ensemble probabilistic quantitative precipitation forecasts. Wea. Forecasting, 29, 1024–1043, doi:  https://doi.org/10.1175/WAF-D-13-00108.1.CrossRefGoogle Scholar
  10. Gagne II, D. J., A. McGovern, J. Brotzge, et al., 2015: Day-ahead hail prediction integrating machine learning with storm-scale numerical weather models. Proceedings of the 27th Conference on Innovative Applications of Artificial Intelligence, AAAI, Austin, TX, USA, 3954–3960.Google Scholar
  11. Gagne II, D. J., A. McGovern, S. E. Haupt, et al., 2017: Storm-based probabilistic hail forecasting with machine learning applied to convection-allowing ensembles. Wea. Forecasting, 32, 1819–1840, doi:  https://doi.org/10.1175/WAF-D-17-0010.1.CrossRefGoogle Scholar
  12. Gardner, M. W., and S. R. Dorling, 1998: Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmos. Environ., 32, 2627–2636, doi:  https://doi.org/10.1016/S1352-2310(97)00447-0.CrossRefGoogle Scholar
  13. Gope, S., S. Sarkar, P. Mitra, et al., 2016: Early prediction of extreme rainfall events: A deep learning approach. Proceedings of the 16th Industrial Conference on Data Mining, Springer, New York, NY, USA, 154–167, doi:  https://doi.org/10.1007/978-3-319-41561-1_12.Google Scholar
  14. Grzymala-Busse, J. W., L. K. Goodwin, W. J. Grzymala-Busse, et al., 2004: An approach to imbalanced data sets based on changing rule strength. Rough-Neural Computing: Techniques for Computing with Words, S. K. Pal, L. Polkowski, and A. Skowron, Eds., Springer, Berlin Heidelberg, 543–553, doi:  https://doi.org/10.1007/978-3-642-18859-6_21.CrossRefGoogle Scholar
  15. Han, L., J. Z. Sun, W. Zhang, et al., 2017: A machine learning nowcasting method based on real-time reanalysis data. J. Geophys. Res. Atmos., 122, 4038–4051, doi:  https://doi.org/10.1002/2016JD025783.CrossRefGoogle Scholar
  16. Herman, G. R., and R. S. Schumacher, 2018: Money doesn’t grow on trees, but forecasts do: Forecasting extreme precipitation with random forests. Mon. Wea. Rev., 146, 1571–1600, doi:  https://doi.org/10.1175/MWR-D-17-0250.1.CrossRefGoogle Scholar
  17. Kingma, D. P., and J. Ba, 2015: Adam: A method for stochastic optimization. Proceedings of the 3rd International Conference on Learning Representations, IEEE, San Diego, USA, 3156–3165.Google Scholar
  18. Klein, B., L. Wolf, and Y. Afek, 2015: A dynamic convolutional layer for short range weather prediction. Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Boston, MA, USA, 4840–4848, doi:  https://doi.org/10.1109/CVPR.2015.7299117.Google Scholar
  19. Klein, W. H., B. M. Lewis, and I. Enger, 1959: Objective prediction of five-day mean temperatures during winter. J. Meteor., 16, 672–682, doi:  https://doi.org/10.1175/1520-0469(1959)016<0672:OPOFDM>2.0.CO;2.CrossRefGoogle Scholar
  20. Krawczyk, B., 2016: Learning from imbalanced data: Open challenges and future directions. Prog. Artif. Intell., 5, 221–232, doi:  https://doi.org/10.1007/s13748-016-0094-0.CrossRefGoogle Scholar
  21. Krizhevsky, A., I. Sutskever, and G. E. Hinton, 2012: ImageNet classification with deep convolutional neural networks. Proceedings of the 25th International Conference on Neural Information Processing Systems, ACM, Lake Tahoe, USA, 1097–1105.Google Scholar
  22. Kubat, M., R. C. Holte, and S. Matwin, 1998: Machine learning for the detection of oil spills in satellite radar images. Mach. Learn., 30, 195–215, doi:  https://doi.org/10.1023/A:1007452223027.CrossRefGoogle Scholar
  23. Lagerquist, R., A. McGovern, and T. Smith, 2017: Machine learning for real-time prediction of damaging straight-line convective wind. Wea. Forecasting, 32, 2175–2193, doi:  https://doi.org/10.1175/WAF-D-17-0038.1.CrossRefGoogle Scholar
  24. Lakshmanan, V., G. Stumpf, and A. Witt, 2005: A neural network for detecting and diagnosing tornadic circulations using the mesocyclone detection and near storm environment algorithms. Proceedings of AI Applications with a Nowcasting Flavor (Joint between the Fourth Conference on Artificial Intelligence and the 21st International Conference on Interactive Information and Processing Systems (IIPS) for Meteorology, Oceanography, and Hydrology), Amer. Meteor. Soc., San Diego, CA, USA, J5.2.Google Scholar
  25. LeCun, Y., and Y. Bengio, 1995: Convolutional networks for images, speech, and time-series. The Handbook of Brain Theory and Neural Networks, M. A. Arbib, Ed., MIT Press, Cambridge, MA, USA, 10 pp.Google Scholar
  26. Mac Namee, B., P. Cunningham, S. Byrne, et al., 2002: The problem of bias in training data in regression problems in medical decision support. Artif. Intell. Med., 24, 51–70, doi:  https://doi.org/10.1016/S0933-3657(01)00092-6.CrossRefGoogle Scholar
  27. Manzato, A., 2005: The use of sounding-derived indices for a neural network short-term thunderstorm forecast. Wea. Forecasting, 20, 896–917, doi:  https://doi.org/10.1175/WAF898.1.CrossRefGoogle Scholar
  28. Manzato, A., 2007: Sounding-derived indices for neural network based short-term thunderstorm and rainfall forecasts. Atmos. Res., 83, 349–365, doi:  https://doi.org/10.1016/j.atmosres.2005.10.021.CrossRefGoogle Scholar
  29. Manzato, A., 2013: Hail in northeast Italy: A neural network ensemble forecast using sounding-derived indices. Wea. Forecasting, 28, 3–28, doi:  https://doi.org/10.1175/WAF-D-12-00034.1.CrossRefGoogle Scholar
  30. Marzban, C., and G. J. Stumpf, 1996: A neural network for tornado prediction based on Doppler radar-derived attributes. J. Appl. Meteor., 35, 617–626, doi:  https://doi.org/10.1175/1520-0450(1996)035<0617:ANNFTP>2.0.CO;2.CrossRefGoogle Scholar
  31. Marzban, C., and G. J. Stumpf, 1998: A neural network for damaging wind prediction. Wea. Forecasting, 13, 151–163, doi:  https://doi.org/10.1175/1520-0434(1998)013<0151:ANNFDW>2.0.CO;2.CrossRefGoogle Scholar
  32. Matsugu, M., K. Mori, Y. Mitari, et al., 2003: Subject independent facial expression recognition with robust face detection using a convolutional neural network. Neural Netw., 16, 555–559, doi:  https://doi.org/10.1016/S0893-6080(03)00115-1.CrossRefGoogle Scholar
  33. Meng, Z. Y., D. C. Yan, and Y. J. Zhang, 2013: General features of squall lines in east China. Mon. Wea. Rev., 141, 1629–1647, doi:  https://doi.org/10.1175/MWR-D-12-00208.1.CrossRefGoogle Scholar
  34. Pedregosa, F., G. Varoquaux, A. Gramfort, et al., 2011: Scikitlearn: Machine learning in Python. J. Mach. Learn. Res., 12, 2825–2830.Google Scholar
  35. Perol, T., M. Gharbi, and M. Denolle, 2018: Convolutional neural network for earthquake detection and location. Sci. Adv., 4, e1700578, doi:  https://doi.org/10.1126/sciadv.1700578.CrossRefGoogle Scholar
  36. Ray, P. S., 1986: Mesoscale Meteorology and Forecasting. American Meteorological Society, Boston, USA, 793 pp, doi:  https://doi.org/10.1007/978-1-935704-20-1_2.CrossRefGoogle Scholar
  37. Sanders, J., and E. Kandrot, 2010: CUDA by Example: An Introduction to General-Purpose GPU Programming. Addison-Wesley Educational Publishers Inc., Upper Saddle River, NJ, USA, 312 pp.Google Scholar
  38. Schmidhuber, J., 2015: Deep learning in neural networks: An overview. Neural Netw., 61, 85–117, doi:  https://doi.org/10.1016/j.neunet.2014.09.003.CrossRefGoogle Scholar
  39. Sheridan, P., 2018: Current gust forecasting techniques, developments and challenges. Adv. Sci. Res., 15, 159–172, doi:  https://doi.org/10.5194/asr-15-159-2018.CrossRefGoogle Scholar
  40. Shi, X. J., Z. R. Chen, H. Wang, et al., 2015: Convolutional LSTM network: A machine learning approach for precipitation now-casting. Proceedings of the 28th International Conference on Neural Information Processing Systems, ACM, Montreal, Canada, 802–810.Google Scholar
  41. Simonyan, K., and A. Zisserman, 2015: Very deep convolutional networks for large-scale image recognition. Proceedings of the 3rd International Conference on Learning Representations, IEEE, San Diego, USA, 313–318.Google Scholar
  42. Stensrud, D. J., M. Xue, L. J. Wicker, et al., 2009: Convective-scale warn-on-forecast system: A vision for 2020. Bull. Amer. Meteor. Soc., 90, 1487–1500, doi:  https://doi.org/10.1175/2009BAMS2795.1.CrossRefGoogle Scholar
  43. Sun, J. S., J. H. Dai, L. F. He, et al., 2014: The Basic Principle and Technical Method of Strong Convective Weather Forecast. China Meteorological Press, Beijing, 1–21. (in Chinese)Google Scholar
  44. Tian, F. Y., Y. G. Zheng, T. Zhang, et al., 2015: Statistical characteristics of environmental parameters for warm season short-duration heavy rainfall over central and eastern China. J. Meteor. Res., 29, 370–384, doi:  https://doi.org/10.1007/s13351-014-4119-y.CrossRefGoogle Scholar
  45. Wang, Y. B., M. S. Long, J. M. Wang, et al., 2017: PredRNN: Recurrent neural networks for predictive learning using spatiotemporal LSTMs. Proceedings of the 31st Conference on Neural Information Processing Systems, Long Beach, CA, USA, 879–888.Google Scholar
  46. Xia, R. D., D.-L. Zhang, and B. L. Wang, 2015: A 6-yr cloud-to-ground lightning climatology and its relationship to rainfall over central and eastern China. J. Appl. Meteor. Climatol., 14, 2443–2460, doi:  https://doi.org/10.1175/JAMC-D-15-0029.1.CrossRefGoogle Scholar
  47. Yang, X. L., J. H. Sun, and W. L. Li, 2015: An analysis of cloud-to-ground lightning in China during 2010–13. Wea. Forecasting, 30, 1537–1550, doi:  https://doi.org/10.1175/WAF-D-14-00132.1.CrossRefGoogle Scholar
  48. Yang, X. L., J. H. Sun, and Y. G. Zheng, 2017: A 5-yr climatology of severe convective wind events over China. Wea. Forecasting, 32, 1289–1299, doi:  https://doi.org/10.1175/WAF-D-16-0101.1.CrossRefGoogle Scholar
  49. Yu, X. D., 2011: Ingredients based forecasting methodology. Meteor. Mon., 37, 913–918. (in Chinese)Google Scholar
  50. Zhang, W., L. Han, J. H. Sun, et al., 2017: Application of multichannel 3D-cube successive convolution network for convective storm nowcasting. IEEE Conference on Computer Vision and Pattern Recognition, IEEE, Honolulu, USA, 118–127.Google Scholar
  51. Zhang, X. L., S. Y. Tao, and J. H. Sun, 2010: Ingredients-based heavy rainfall forecasting. Chinese J. Atmos. Sci., 34, 754–766, doi:  https://doi.org/10.3878/j.issn.1006-9895.2010.04.08. (in Chinese)Google Scholar
  52. Zheng, Y. G., Y. J. Lin, W. J. Zhu, et al., 2013: Operational system of severe convective weather comprehensive monitoring. Meteor. Mon., 39, 234–240. (in Chinese)Google Scholar

Copyright information

© The Chinese Meteorological Society and Springer-Verlag Berlin Heidelberg 2019

Authors and Affiliations

  • Kanghui Zhou
    • 1
    • 2
    • 3
    Email author
  • Yongguang Zheng
    • 3
  • Bo Li
    • 4
  • Wansheng Dong
    • 1
  • Xiaoling Zhang
    • 3
  1. 1.Chinese Academy of Meteorological SciencesChina Meteorological AdministrationBeijingChina
  2. 2.University of Chinese Academy of SciencesBeijingChina
  3. 3.National Meteorological CenterChina Meteorological AdministrationBeijingChina
  4. 4.University of Illinois at Urbana-ChampaignChampaignUSA

Personalised recommendations