Abstract
An ensemble machine learning model for tropical cyclone (TC) track forecasts over the Western North Pacific was developed and evaluated in this study. First, we investigated predictors including TC climatology and persistence factors which were extracted from TC best-track dataset and storm’s surrounding atmospheric conditions which were extracted from ERA-Interim reanalysis. Then, we built a Gradient Boosting Decision Tree (GBDT) nonlinear model for TC track forecasts, in which 30-year data was used. Finally, using tenfold cross-validation method, the GBDT model was compared with a frequently used technique: climatology and persistence (CLIPER) model. The experimental results show that the GBDT model performs well in three forecast times (24 h, 48 h, and 72 h) with relatively small forecast error of 138, 264, and 363.5 km, respectively. The model obtains excellent TC moving direction aspects. However, the model is still insufficient to produce aspects of storm acceleration and deceleration, with mean moving velocity sensitivities all less than 60%. Nevertheless, the model obtains much more robust and accurate TC tracks relative to CLIPER model, where the forecast skills are 17.5%, 26.3%, and 32.1% at three forecast times, respectively. The presented study demonstrates that the GBDT model could provide reliable evidence and guidance for operational TC track forecasts.
Similar content being viewed by others
References
Aberson Sim D (1998) Five-day tropical cyclone track forecasts in the North Atlantic Basin[J]. Weather Forecast 13(4):1005–1015. https://doi.org/10.1175/1520-0434(1998)013<1005:fdtctf>2.0.co;2
Bender MA, Ginis I, Tuleya R et al (2007) The Operational GFDL Coupled Hurricane-Ocean prediction system and a summary of its performance. Monthly Weather Rev. 135(12):3965–3989. https://doi.org/10.1175/2007mwr2032.1
Bessafi M, Lasserrebigorry A, Neumann CJ et al (2002) Statistical prediction of tropical cyclone motion: an analog CLIPER approach. Weather Forecast 17(4):821–831. https://doi.org/10.1175/1520-0434(2002)0172.0.CO;2
Camargo SJ, Robertson AW, Gaffney SJ, Smyth P, Ghil M (2007) Cluster analysis of typhoon tracks Part I: General properties. J Clim 20(14):3635–53. https://doi.org/10.1175/JCLI4188.1
Chan JCL, Shi J, Lam C (1998) Seasonal forecasting of tropical cyclone activity over theWestern North Pacific and the South China Sea. Weather and forecasting 13(4):997–1004. https://doi.org/10.1175/1520-0434(1998)013<0997:SFOTCA>2.0.CO;2
Chen JM, Elsberry RL, Boothe MA et al (2010) A simple statistical-synoptic track prediction technique for Western North Pacific tropical cyclones. Mon Weather Rev 10(127):124–133. https://doi.org/10.1175/1520-0493(1999)1272.0.CO;2
Chu, J-H, et al. (2002) The joint typhoon warning center tropical cyclone best-tracks, 1945–2000. Ref. NRL/MR/7540-02 16
Dee DP, Uppala SM, Simmons AJ et al (2011) The ERA-Interim reanalysis: configuration and performance of the data assimilation system. Q J R Meteorol Soc 137(656):553–597. https://doi.org/10.1002/qj.828
Dietterich TG (2000) Ensemble methods in machine learning. In: International workshop on multiple classifier systems. Springer, Berlin, Heidelberg, pp 1–15. https://doi.org/10.1007/3-540-45014-9_1
Elsberry RL (1995) Recent advancements in dynamical tropical cyclone track predictions. Meteorol Atmos Phys 56(1–2):81–99. https://doi.org/10.1007/BF01022522
Elsberry RL, Dobos PH, Bacon AB (1991) Lagged-average predictions of tropical cyclone tracks. Mon Weather Rev 119(4):1031–1039. https://doi.org/10.1175/1520-0493(1991)119<1031:LAPOTC>2.0.CO;2
Fraedrich K, Raible CC, Sielmann F (2003) Analog ensemble forecasts of tropical cyclone tracks in the Australian Region. Weather Forecast 18(1):3–11. https://doi.org/10.1175/1520-0434(2003)018<0003:AEFOTC>2.0.CO;2
Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Ann Stat 29(5):1189–1232. https://doi.org/10.2307/2699986
Goerss James S (2009) Impact of satellite observations on the tropical cyclone track forecasts of the navy operational global atmospheric prediction system. Monthly Wea Rev 137(1):41–50. https://doi.org/10.1175/2008mwr2601.1
Goerss JS, Jeffries RA (1994) Assimilation of synthetic tropical cyclone observations into the Navy Operational Global Atmospheric Prediction System. Weather Forecast 9(4):557–576. https://doi.org/10.1175/1520-0434(1994)009<0557:AOSTCO>2.0.CO;2
Goerss JS, Sampson CR, Gross JM (2004) A history of western North Pacific tropical cyclone track forecast skill. Weather Forecast 19(3):633–638. https://doi.org/10.1175/1520-0434(2004)019<0633:AHOWNP>2.0.CO;2
Hagen AB, Strahan-Sakoskie D, Luckett C (2012) A Reanalysis of the 1944–53 Atlantic Hurricane Seasons-the first decade of aircraft reconnaissance. J Clim 25(13):4441–4460. https://doi.org/10.1175/JCLI-D-11-00419.1
Hall TM, Jewson S (2007) Statistical modelling of North Atlantic tropical cyclone tracks. Tellus A Dyn Meteorol Oceanogr 59(4):486–498. https://doi.org/10.1111/j.1600-0870.2007.00240.x
Holland GJ (1993) The Global guide to tropical cyclone forecasting. Phys Solid State 35(6):753–756
Huang XY, ** L, Huang Y (2012) A Genetic-Neural Network Model Based on Multidimensional Scaling for Typhoon Intensity. Computational Sciences and Optimization (CSO), Fifth International Joint Conference on. IEEE, 2012. https://doi.org/10.1109/CSO.2012.196
Huang Y, ** L (2013) A prediction scheme with genetic neural network and Isomap algorithm for tropical cyclone intensity change over western North Pacific. Meteorol Atmos Phys 121(3–4):143–152. https://doi.org/10.1007/s00703-013-0263-7
Huang Y, ** L, Zhao H et al (2018) Fuzzy neural network and LLE Algorithm for forecasting precipitation in tropical cyclones: comparisons with interpolation method by ECMWF and stepwise regression method. Nat Hazards 91(1):201–220. https://doi.org/10.1007/s11069-017-3122-x
Ihshaish H, Cortés A, Senar MA (2012) Towards improving numerical weather predictions by evolutionary computing techniques. Procedia Comput Sci 9:1056–1063. https://doi.org/10.1016/j.procs.2012.04.114
** Long, **aoyan H, Shi Xuming (2010) A study on influence of predictor multicollinearity on performance of the stepwise regression prediction equation. J Meteorol Res 24(5):593–601. https://doi.org/10.1002/wea.667
Johnson R, Zhang T (2014) Learning nonlinear functions using regularized Greedy forest. IEEE Trans Pattern Anal Mach Intell 36(5):942–954. https://doi.org/10.1109/tpami.2013.159
Kinney JB, Atwal GS (2014) Equitability, mutual information, and the maximal information coefficient. Proc Nat Acad Sci 111(9):3354–3359. https://doi.org/10.1073/pnas.1309933111
Knapp KR, Kruk MC, Levinson DH et al (2010) The international best track archive for climate stewardship (IBTrACS). Bull Am Meteorol Soc 91(3):363–376. https://doi.org/10.1175/2009BAMS2755.1
Lin KP, Dong Y, Huang XY (2014) GA-ANN Forecast Model for Typhoon Gale in South China Sea Based on MDS. Appl Mech Mater 556–562:5618–5622. https://doi.org/10.4028/www.scientific.net/amm.556-562.5618
Long J, Huang X, Shi X et al (2013) A fuzzy neural network prediction model based on manifold learning to reduce dimensions for typhoon intensity. Ind Electr Appl. https://doi.org/10.1109/ICIEA.2013.6566432
Marshall JFL, Leslie LM, Abbey RF et al (2002) Tropical cyclone track and intensity prediction: the generation and assimilation of high-density, satellite-derived data. Meteorol Atmos Phys 80(1–4):43–57. https://doi.org/10.1007/s007030200013
Nakamura J, Lall U, Kushnir Y, Camargo SJ (2009) Classifying North Atlantic tropical cyclone tracks by mass moments. J Clim 22(20):5481–5494. https://doi.org/10.1175/2009jcli2828.1
Nakamura J, Lall U, Kushnir Y, Rajagopalan B (2015) HITS: Hurricane intensity and track simulator with North Atlantic Ocean applications for risk assessment. J Appl Meteorol Climatol 54(7):1620–1636. https://doi.org/10.1175/JAMC-D-14-0141.1
Nakamura J, Camargo SJ, Sobel AH et al (2017) Western North Pacific tropical cyclone model tracks in present and future climates. J Geophys Res Atmos. https://doi.org/10.1002/2017JD027007
Nakamura J, Camargo SJ, Sobel AH, Henderson N, Emanuel KA, Kumar A, Vidale PL (2017) Western North Pacific tropical cyclone model tracks in present and future climates. J Geophys Res Atmos 122(18):9721–9744. https://doi.org/10.1002/2017JD027007
Neumann CJ (1979) A guide to Atlantic and eastern Pacific models for the prediction of tropical cyclone motion. NOAA Technical Memorandum NWS No. 932. National Hurricane Center, Miami, Florida
Neumann CJ, Hope JR (1972) Performance analysis of the HURRAN tropical cyclone forecast system. Mon Weather Rev 100(4):245–255. https://doi.org/10.1175/1520-0493(1972)1002.3.CO;2
Pradhan R, Aygun RS, Maskey M et al (2018) Tropical Cyclone Intensity Estimation Using a Deep Convolutional Neural Network. IEEE Trans Image Process. https://doi.org/10.1109/TIP.2017.2766358
Roy Chandan, Kovordányi Rita (2012) Tropical cyclone track forecasting techniques–A review. Atmos Res 104:40–69. https://doi.org/10.1016/j.atmosres.2011.09.012
Sievers O, Fraedrich K, Raible CC (2000) Self-adapting analog ensemble predictions of tropical cyclone tracks. Weather Forecast 15(5):623–629. https://doi.org/10.1175/1520-0434(2000)015<0623:saaepo>2.0.co;2
Song HJ, Huh SH, Kim JH, Ho CH, Park SK (2005) Typhoon track prediction by a support vector machine using data reduction methods. In: Hao Y et al (eds) Computational intelligence and security. CIS 2005, vol 3801. Springer, Berlin. https://doi.org/10.1007/11596448_74
Tian J, Li M, Chen F et al (2012) Coevolutionary learning of neural network ensemble for complex classification tasks. Pattern Recognit 45(4):1373–1385. https://doi.org/10.1016/j.patcog.2011.09.012
Zhou C, Yu H, Ding Y, Guo F, Gong XJ (2017) Multi-scale encoding of amino acid sequences for predicting protein interactions using gradient boosting decision tree. PloS ONE. https://doi.org/10.1371/journal.pone.0181426
Acknowledgements
The authors are grateful to three anonymous reviewers for their constructive comments, which helped improve the representation and quality of this work. This research is jointly supported by the National Key Research and Development Program of China (No. 2018YFC1508803), the National Natural Science Foundation of China (Grant Nos. 41671095, 41971199, 41875182), Shanghai Science and Technology Support Program (Grant No. 19DZ1201505), Guangzhou Science and Technology plan projects (No. 201904010162), Sun Yat-sen University “100 Top Talents Program” (No. 74110-18841203).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled ’Western North Pacific Tropical Cyclone Track Forecasts by a Machine Learning Model’.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Tan, J., Chen, S. & Wang, J. Western North Pacific tropical cyclone track forecasts by a machine learning model. Stoch Environ Res Risk Assess 35, 1113–1126 (2021). https://doi.org/10.1007/s00477-020-01930-w
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00477-020-01930-w