• Breiman, L., 2001: Random forests. Mach. Learn., 45, 532, https://doi.org/10.1023/A:1010933404324.

  • Chambolle, A., and T. Pock, 2011: A first-order primal-dual algorithm for convex problems with applications to imaging. J. Math. Imaging Vis., 40, 120145, https://doi.org/10.1007/s10851-010-0251-1.

    • Search Google Scholar
    • Export Citation
  • Chawla, N. V., K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, 2002: SMOTE: Synthetic minority over-sampling technique. J. Artif. Intell. Res., 16, 321357, https://doi.org/10.1613/jair.953.

    • Search Google Scholar
    • Export Citation
  • Dixon, M., and G. Wiener, 1993: TITAN: Thunderstorm identification, tracking, analysis, and nowcasting—A radar-based methodology. J. Atmos. Oceanic Technol., 10, 785797, https://doi.org/10.1175/1520-0426(1993)010<0785:TTITAA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • George, J. J., 2014: Weather Forecasting for Aeronautics. Academic Press, 684 pp.

  • Germann, U., and I. Zawadzki, 2002: Scale dependence of the predictability of precipitation from continental radar images. Part I: Description of the methodology. Mon. Wea. Rev., 130, 28592873, https://doi.org/10.1175/1520-0493(2002)130<2859:SDOTPO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Guo, X., Y. Yin, C. Dong, G. Yang, and G. Zhou, 2008: On the class imbalance problem. 2008 Fourth Int. Conf. on Natural Computation, Jinan, China, IEEE, 192201, https://doi.org/10.1109/ICNC.2008.871.

    • Crossref
    • Export Citation
  • Hanssen-Bauer, I., and P. Nordli, 1998: Annual and seasonal temperature variations in Norway 1876-1997. DNMI Rep. 25, 98 pp.

  • Hastie, T., R. Tibshirani, and J. Friedman, 2009: The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed. Springer Science & Business Media, 763 pp.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hwang, Y., A. J. Clark, V. Lakshmanan, and S. E. Koch, 2015: Improved nowcasts by blending extrapolation and model forecasts. Wea. Forecasting, 30, 12011217, https://doi.org/10.1175/WAF-D-15-0057.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Inness, P. M., and S. Dorling, 2012: Operational Weather Forecasting. John Wiley & Sons, 248 pp.

    • Crossref
    • Export Citation
  • Japkowicz, N., and S. Stephen, 2002: The class imbalance problem: A systematic study. Intell. Data Anal., 6, 429449, https://doi.org/10.3233/IDA-2002-6504.

    • Search Google Scholar
    • Export Citation
  • Li, L., W. Schmid, and J. Joss, 1995: Nowcasting of motion and growth of precipitation with radar over a complex orography. J. Appl. Meteor., 34, 12861300, https://doi.org/10.1175/1520-0450(1995)034<1286:NOMAGO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Liu, Z., G. Gilbert, J. M. Cepeda, A. O. K. Lysdahl, L. Piciullo, H. Hefre, and S. Lacasse, 2020: Modelling of shallow landslides with machine learning algorithms. Geosci. Front., https://doi.org/10.1016/j.gsf.2020.04.014, in press.

    • Search Google Scholar
    • Export Citation
  • Longadge, R., and S. Dongre, 2013: Class imbalance problem in data mining review. arXiv preprint arXiv:1305.1707.

  • Mandapaka, P. V., U. Germann, L. Panziera, and A. Hering, 2012: Can Lagrangian extrapolation of radar fields be used for precipitation nowcasting over complex alpine orography? Wea. Forecasting, 27, 2849, https://doi.org/10.1175/WAF-D-11-00050.1.

    • Search Google Scholar
    • Export Citation
  • Mao, Y., 2020: Random forest nowcasts. Accessed 19 May 2020, https://doi.org/10.17632/smxkyhtdvj.3.

    • Crossref
    • Export Citation
  • MathWorks, 2019: Predict responses using ensemble of bagged decision trees-matlab-mathworks nordic. Accessed 1 December 2019, https://se.mathworks.com/help/stats/treebagger.predict.html.

  • Mosavi, A., P. Ozturk, and K. Chau, 2018: Flood prediction using machine learning models: Literature review. Water, 10, 1536, https://doi.org/10.3390/w10111536.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Müller, M., and Coauthors, 2017: AROME-MetCoOp: A nordic convective-scale operational weather prediction model. Wea. Forecasting, 32, 609627, https://doi.org/10.1175/WAF-D-16-0099.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Norwegian Meteorological Institute, 2011: Free access to weather-and climate data from Norwegian meteorological institute from historical data to real time observations. eKlima, accessed 1 May 2018, https://www.met.no/en/free-meteorological-data/Download-services.

  • Nurmi, P., 2003: Recommendations on the verification of local weather forecasts. ECMWF Tech. Memo. 430, 19 pp., https://www.ecmwf.int/en/elibrary/11401-recommendations-verification-local-weather-forecasts.

  • Reyniers, M., 2008: Quantitative precipitation forecasts based on radar observations: Principles, algorithms and operational systems. Royal Meteorological Institute of Belgium Publ. Scientifique et Technique 52, 62 pp.

  • Roebber, P. J., 2009: Visualizing multiple measures of forecast quality. Wea. Forecasting, 24, 601608, https://doi.org/10.1175/2008WAF2222159.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Scikit-Learn Developers, 2019: Permutation importance with multicollinear or correlated features. Accessed 1 December 2019, https://scikit-learn.org/stable/auto_examples/inspection/plot_permutation_importance_multicollinear.html.

  • Shi, X., Z. Chen, H. Wang, D.-Y. Yeung, W.-K. Wong, and W.-c. Woo, 2015: Convolutional LSTM network: A machine learning approach for precipitation nowcasting. NIPS’15: Proc. 28th Int. Conf. on Neural Information Processing Systems, Cambridge, MA, NIPS, 802810, https://dl.acm.org/doi/10.5555/2969239.2969329.

  • Testik, F. Y., and M. Gebremichael, 2010: Rainfall: State of the Science. Wiley Online Library, 287 pp.

    • Crossref
    • Export Citation
  • Wang, Y., and Coauthors, 2017: Guidelines for nowcasting techniques. WMO Publ. 1198, 82 pp., https://library.wmo.int/doc_num.php?explnum_id=3795.

  • Wilks, D. S., 2011: Statistical Methods in the Atmospheric Sciences. 3rd ed. International Geophysics Series, Vol. 100, Academic Press, 704 pp.

    • Search Google Scholar
    • Export Citation
  • Zou, H., S. Wu, J. Shan, and X. Yi, 2019: A method of radar echo extrapolation based on TREC and Barnes filter. J. Atmos. Oceanic Technol., 36, 17131727, https://doi.org/10.1175/JTECH-D-18-0194.1.

    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 108 133 1
Full Text Views 20 45 17
PDF Downloads 32 57 14

Improving Radar-Based Precipitation Nowcasts with Machine Learning Using an Approach Based on Random Forest

View More View Less
  • 1 Geophysical Institute, University of Bergen, Bergen, Norway
© Get Permissions
Restricted access

Abstract

A binary classification model is trained by random forest using data from 41 stations in Norway to predict the precipitation in a given hour. The predictors consist of results from radar nowcasts and numerical weather predictions. The results demonstrate that the random forest model can improve the precipitation predictions by the radar nowcasts and the numerical weather predictions. This study clarifies whether certain potential factors related to model training can influence the predictive skill of the random forest method. The results indicate that enforcing a balanced prediction by resampling the training datasets or lowering the threshold probability for classification cannot improve the predictive skill of the random forest model. The study reveals that the predictive skill of the random forest model shows seasonality, but is only weakly influenced by the geographic diversity of the training dataset. Finally, the study shows that the most important predictor is the precipitation predictions by the radar nowcasts followed by the precipitation predictions by the numerical weather predictions. Although meteorological variables other than precipitation are weaker predictors, the results suggest that they can help to reduce the false alarm ratio and to increase the success ratio of the precipitation prediction.

Significance Statement

Machine learning can be useful in improving weather forecasts with relatively inexpensive computational efforts. Specifically, this study has demonstrated that radar nowcasts can be improved by integrating the information from radar and numerical weather prediction using the random forest method. The random forest method’s performance shows seasonality but is only weakly influenced by the geographic diversity of the training dataset. Also, there is no need to use specific strategies to address the imbalance of the precipitation and no precipitation frequency from the observations during model training. However, future study is needed to identify better predictor choices to further improve the random forest method.

Current affiliation: Geophysical Institute, University of Bergen, Bergen, Norway.

© 2020 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: Yiwen Mao, yiwen.mao@uib.no

Abstract

A binary classification model is trained by random forest using data from 41 stations in Norway to predict the precipitation in a given hour. The predictors consist of results from radar nowcasts and numerical weather predictions. The results demonstrate that the random forest model can improve the precipitation predictions by the radar nowcasts and the numerical weather predictions. This study clarifies whether certain potential factors related to model training can influence the predictive skill of the random forest method. The results indicate that enforcing a balanced prediction by resampling the training datasets or lowering the threshold probability for classification cannot improve the predictive skill of the random forest model. The study reveals that the predictive skill of the random forest model shows seasonality, but is only weakly influenced by the geographic diversity of the training dataset. Finally, the study shows that the most important predictor is the precipitation predictions by the radar nowcasts followed by the precipitation predictions by the numerical weather predictions. Although meteorological variables other than precipitation are weaker predictors, the results suggest that they can help to reduce the false alarm ratio and to increase the success ratio of the precipitation prediction.

Significance Statement

Machine learning can be useful in improving weather forecasts with relatively inexpensive computational efforts. Specifically, this study has demonstrated that radar nowcasts can be improved by integrating the information from radar and numerical weather prediction using the random forest method. The random forest method’s performance shows seasonality but is only weakly influenced by the geographic diversity of the training dataset. Also, there is no need to use specific strategies to address the imbalance of the precipitation and no precipitation frequency from the observations during model training. However, future study is needed to identify better predictor choices to further improve the random forest method.

Current affiliation: Geophysical Institute, University of Bergen, Bergen, Norway.

© 2020 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: Yiwen Mao, yiwen.mao@uib.no
Save