Seasonal forecasting of tropical cyclone activity in the Australian and the South Pacific Ocean regions

J.S. Wijnands; G. Qian; K.L. Shelton; R.J.B. Fawcett; J.C.L. Chan; Y. Kuleshov

Mathematics of Climate and Weather Forecasting (2015)

  • Volume: 1, Issue: 1
  • ISSN: 2353-6438

Abstract

top
The Australian Bureau of Meteorology (Bureau) issues operational tropical cyclone (TC) seasonal forecasts for the Australian region (AR) and the South Pacific Ocean (SPO) and subregions therein. The forecasts are issued in October, ahead of the Southern Hemisphere TC season (November to April). Improvement of operational TC seasonal forecasts can lead to more accurate warnings for coastal communities to prepare for TC hazards. This study investigates the use of support vector regression (SVR) models, exploring new explanatory variables and non-linear relationships between them, the use of model averaging, and lastly the integration of forecast intervals based on a bias-corrected and accelerated non-parametric bootstrap. Hindcasting analyses show that the SVR model outperforms several benchmark methods. Analysis of the generated models shows that the Dipole Mode Index, 5VAR index and the Southern Oscillation Index are the most frequently selected as explanatory variables for TC seasonal forecasting in all regions. The usage of ENSOrelated covariates implies that definitions of regions and subregions may have to be updated to achieve optimal forecasting performance. Overall, the new SVR methodology is an improvement over the current linear discriminant analysis models and has the potential to increase accuracy of TC seasonal forecasts in the AR and SPO.

How to cite

top

J.S. Wijnands, et al. "Seasonal forecasting of tropical cyclone activity in the Australian and the South Pacific Ocean regions." Mathematics of Climate and Weather Forecasting 1.1 (2015): null. <http://eudml.org/doc/276509>.

@article{J2015,
abstract = {The Australian Bureau of Meteorology (Bureau) issues operational tropical cyclone (TC) seasonal forecasts for the Australian region (AR) and the South Pacific Ocean (SPO) and subregions therein. The forecasts are issued in October, ahead of the Southern Hemisphere TC season (November to April). Improvement of operational TC seasonal forecasts can lead to more accurate warnings for coastal communities to prepare for TC hazards. This study investigates the use of support vector regression (SVR) models, exploring new explanatory variables and non-linear relationships between them, the use of model averaging, and lastly the integration of forecast intervals based on a bias-corrected and accelerated non-parametric bootstrap. Hindcasting analyses show that the SVR model outperforms several benchmark methods. Analysis of the generated models shows that the Dipole Mode Index, 5VAR index and the Southern Oscillation Index are the most frequently selected as explanatory variables for TC seasonal forecasting in all regions. The usage of ENSOrelated covariates implies that definitions of regions and subregions may have to be updated to achieve optimal forecasting performance. Overall, the new SVR methodology is an improvement over the current linear discriminant analysis models and has the potential to increase accuracy of TC seasonal forecasts in the AR and SPO.},
author = {J.S. Wijnands, G. Qian, K.L. Shelton, R.J.B. Fawcett, J.C.L. Chan, Y. Kuleshov},
journal = {Mathematics of Climate and Weather Forecasting},
keywords = {tropical cyclones; seasonal forecasting; support vector regression; forecast interval; Australian region; South Pacific Ocean},
language = {eng},
number = {1},
pages = {null},
title = {Seasonal forecasting of tropical cyclone activity in the Australian and the South Pacific Ocean regions},
url = {http://eudml.org/doc/276509},
volume = {1},
year = {2015},
}

TY - JOUR
AU - J.S. Wijnands
AU - G. Qian
AU - K.L. Shelton
AU - R.J.B. Fawcett
AU - J.C.L. Chan
AU - Y. Kuleshov
TI - Seasonal forecasting of tropical cyclone activity in the Australian and the South Pacific Ocean regions
JO - Mathematics of Climate and Weather Forecasting
PY - 2015
VL - 1
IS - 1
SP - null
AB - The Australian Bureau of Meteorology (Bureau) issues operational tropical cyclone (TC) seasonal forecasts for the Australian region (AR) and the South Pacific Ocean (SPO) and subregions therein. The forecasts are issued in October, ahead of the Southern Hemisphere TC season (November to April). Improvement of operational TC seasonal forecasts can lead to more accurate warnings for coastal communities to prepare for TC hazards. This study investigates the use of support vector regression (SVR) models, exploring new explanatory variables and non-linear relationships between them, the use of model averaging, and lastly the integration of forecast intervals based on a bias-corrected and accelerated non-parametric bootstrap. Hindcasting analyses show that the SVR model outperforms several benchmark methods. Analysis of the generated models shows that the Dipole Mode Index, 5VAR index and the Southern Oscillation Index are the most frequently selected as explanatory variables for TC seasonal forecasting in all regions. The usage of ENSOrelated covariates implies that definitions of regions and subregions may have to be updated to achieve optimal forecasting performance. Overall, the new SVR methodology is an improvement over the current linear discriminant analysis models and has the potential to increase accuracy of TC seasonal forecasts in the AR and SPO.
LA - eng
KW - tropical cyclones; seasonal forecasting; support vector regression; forecast interval; Australian region; South Pacific Ocean
UR - http://eudml.org/doc/276509
ER -

References

top
  1. [1] R. E. Basher and X. Zheng. “Tropical cyclones in the southwest Pacific: Spatial patterns and relationships to Southern Oscillation and sea surface temperature.” In: Journal of Climate 8.5 (1995), pp. 1249–1260. doi: 10.1175/1520-0442(1995) 008<1249:TCITSP>2.0.CO;2. [Crossref] 
  2. [2] A.I. Belousov, S.A.Verzakov, and J. von Frese. “A flexible classification approach with optimal generalisation performance: support vector machines.” In: Chemometrics and Intelligent Laboratory Systems 64.1 (2002), pp. 15–25. doi: 10.1016/S0169- 7439(02)00046-1. [Crossref] 
  3. [3] S. J. Camargo, K. A. Emanuel, and A. H. Sobel. “Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis.” In: Journal of Climate 20.19 (2007), pp. 4819–4834. doi: 10.1175/JCLI4282.1. [Crossref] 
  4. [4] S. J. Camargo and A. H. Sobel. “Western North Pacific tropical cyclone intensity and ENSO.” In: Journal of Climate 18.15 (2005), pp. 2996–3006. doi: 10.1175/JCLI3457.1. [Crossref] 
  5. [5] S. J. Camargo, A. G. Barnston, P. J. Klotzbach, and C.W. Landsea. “Seasonal tropical cyclone forecasts.” In: WMO Bulletin 56.4 (2007), pp. 297–309.  
  6. [6] J. Camp, M. Roberts, C. MacLachlan, E. Wallace, L. Hermanson, A. Brookshaw, A. Arribas, and A. A. Scaife. “Seasonal forecasting of tropical storms using the Met Office GloSea5 seasonal forecast system.” In: Quarterly Journal of the Royal Meteorological Society 141.691 (2015), pp. 2206–2219. doi: 10.1002/qj.2516. [Crossref] 
  7. [7] S. S. Chand, K. J. Tory, J. L.McBride, M. C. Wheeler, R. A. Dare, and K. J. E.Walsh. “The different impact of positive-neutral and negative-neutral ENSO regimes on Australian tropical cyclones.” In: Journal of Climate 26.20 (2013), pp. 8008–8016. doi: 10.1175/JCLI-D-12-00769.1. [Crossref] 
  8. [8] C.-C. Chang and C.-J. Lin. “LIBSVM: A library for support vector machines.” In: ACM Transactions on Intelligent Systems and Technology 2.3 (2011), 27:1–27:27. doi: 10.1145/1961189.1961199. [Crossref] 
  9. [9] K. Davis, X. Zeng, and E. A. Ritchie. “A new statistical model to predict seasonal North Atlantic hurricane activity.” In: Weather and Forecasting 141 (691 2015), pp. 2206–2219. doi: 10.1175/WAF-D-14-00156.1. [Crossref] 
  10. [10] T. G. Dietterich. “Ensemble methods in machine learning.” In: Multiple classifier systems. Vol. 1857. Lecture Notes in Computer Science. Springer Berlin Heidelberg, 2000, pp. 1–15. isbn: 978-3-540-67704-8. doi: 10.1007/3-540-45014-9_1.  
  11. [11] A. J. Dowdy. “Long-term changes in Australian tropical cyclone numbers.” In: Atmospheric Science Letters 15.4 (2014), pp. 292–298. doi: 10.1002/asl2.502. [Crossref] 
  12. [12] A. J. Dowdy, L. Qi, D. Jones, H. Ramsay, R. Fawcett, and Y. Kuleshov. “Tropical cyclone climatology of the South Pacific Ocean and its relationship to El Niño-Southern Oscillation.” In: Journal of Climate 25.18 (2012), pp. 6108–6122. doi: 10. 1175/JCLI-D-11-00647.1. [Crossref] 
  13. [13] W. Drosdowsky and L. E. Chambers. Near global sea surface temperature anomalies as predictors of Australian seasonal rainfall. Tech. rep. 65. Melbourne: Bureau of Meteorology Research Centre, 1998.  
  14. [14] B. Efron. “Better bootstrap confidence intervals.” In: Journal of the American Statistical Association 82.397 (1987), pp. 171– 185. doi: 10.1080/01621459.1987.10478410. [Crossref] Zbl0622.62039
  15. [15] K. Emanuel. “Increasing destructiveness of tropical cyclones over the past 30 years.” In: Nature 436.7051 (2005), pp. 686– 688. doi: 10.1038/nature03906. [Crossref] 
  16. [16] J. L. Evans and R. J. Allan. “El Niño/Southern Oscillation modification to the structure of the monsoon and tropical cyclone activity in the Australasian region.” In: International Journal of Climatology 12.6 (1992), pp. 611–623. doi: 10.1002/joc. 3370120607. [Crossref] 
  17. [17] W. M. Frank and G. S. Young. “The interannual variability of tropical cyclones.” In:MonthlyWeather Review 135.10 (2007), pp. 3587–3598. doi: 10.1175/MWR3435.1. [Crossref] 
  18. [18] A. Z.-C. Goh and J. C. L. Chan. “An improved statistical scheme for the prediction of tropical cyclones making landfall in South China.” In: Weather and Forecasting 25.2 (2010), pp. 587–593. doi: 10.1175/2009WAF2222305.1. [Crossref] 
  19. [19] T. Hastie, R. Tibshirani, and J. Friedman. The elements of statistical learning. 2nd ed. Springer, 2009. isbn: 978-0-387-84857- 0.  Zbl1273.62005
  20. [20] G. J. Holland. “On the quality of the Australian tropical cyclone data base.” In: Australian Meteorological Magazine 29 (1981), pp. 169–181.  
  21. [21] N.-Y. Kang and J. B. Elsner. “Consensus on climate trends inwestern North Pacific tropical cyclones.” In: Journal of Climate 25.21 (2012), pp. 7564–7573. doi: 10.1175/JCLI-D-11-00735.1. [Crossref] 
  22. [22] A. Karatzoglou, A. Smola, K. Hornik, and A. Zeileis. “kernlab - An S4 package for kernel methods in R.” In: Journal of Statistical Software 11.9 (2004), pp. 1–20. url: http://www.jstatsoft.org/v11/i09.  
  23. [23] Y. Kuleshov, L. Qi, R. Fawcett, and D. Jones. “Improving preparedness to natural hazards: Tropical cyclone seasonal prediction for the Southern Hemisphere.” In: Advances in Geosciences, Volume 12: Ocean Science (OS). Ed. by J. Gan. Vol. 12. Singapore: World Scientific Publishing, 2009, pp. 127–143. doi: 10.1142/9789812836168_0010.  
  24. [24] Y. Kuleshov, R. Fawcett, L. Qi, B. Trewin, D. Jones, J. McBride, and H. Ramsay. “Trends in tropical cyclones in the South Indian Ocean and the South Pacific Ocean.” In: Journal of Geophysical Research 115.D01101 (2010). doi: 10.1029/ 2009JD012372. [Crossref] 
  25. [25] Y. Kuleshov, Y.Wang, J. Apajee, R. Fawcett, and D. Jones. “Prospects for improving the operational seasonal prediction of tropical cyclone activity in the Southern Hemisphere.” In: Atmospheric and Climate Sciences 2.3 (2012), pp. 298–306. doi: 10.4236/acs.2012.23027. [Crossref] 
  26. [26] Y. Kuleshov, C. Spillman, Y.Wang, A. Charles, R. deWit, K. Shelton, D. Jones, H. Hendon, C. Ganter, A.Watkins, J. Apajee, and A. Griesser. “Seasonal prediction of climate extremes for the Pacific: Tropical cyclones and extreme ocean temperatures.” In: Journal of Marine Science and Technology 20.6 (2012), pp. 675–683. doi: 10.6119/JMST-012-0628-1. [Crossref] 
  27. [27] S. Kullback and R. A. Leibler. “On information and sufficiency.” In: The Annals of Mathematical Statistics 22.1 (1951), pp. 79–86. doi: 10.1214/aoms/1177729694. [Crossref] Zbl0042.38403
  28. [28] M. A. Lander. “An exploratory analysis of the relationship between tropical storm formation in the western North Pacific and ENSO.” In: Monthly Weather Review 122.4 (1994), pp. 636–651. doi: 10.1175/1520-0493(1994)122<0636:AEAOTR>2.0. CO;2. [Crossref] 
  29. [29] J.-H. Lee and C.-J. Lin. Automatic model selection for support vector machines. Tech. rep. Department of Computer Science and Information Engineering, National Taiwan University, 2000.  
  30. [30] C.-J. Lin and R. C. Weng. Simple probabilistic predictions for support vector regression. Tech. rep. National Taiwan University, 2004.  
  31. [31] J. Malilay. “Tropical cyclones.” In: The public health consequences of disasters. Ed. by E. K. Noji. Oxford University Press, 1996. Chap. 10, pp. 207–227. isbn: 9780199747689.  
  32. [32] F. Molteni, R. Buizza, T. N. Palmer, and T. Petroliagis. “The ECMWF ensemble prediction system: Methodology and validation.” In: Quarterly Journal of the Royal Meteorological Society 122.529 (1996), pp. 73–119. doi: 10.1002/qj.49712252905. [Crossref] 
  33. [33] M. Momma and K. P. Bennett. “A pattern search method for model selection of support vector regression.” In: 2002 SIAM International Conference on Data Mining. SIAM, 2002, pp. 261–274. doi: 10.1137/1.9781611972726.16.  
  34. [34] N. Nicholls. “A possible method for predicting seasonal tropical cyclone activity in the Australian region.” In: Monthly Weather Review 107.9 (1979), pp. 1221–1224. doi: 10.1175/1520-0493(1979)107<1221:APMFPS>2.0.CO;2. [Crossref] 
  35. [35] N. Nicholls. “The Southern Oscillation, sea-surface-temperature, and interannual fluctuations in Australian tropical cyclone activity.” In: Journal of Climatology 4.6 (1984), pp. 661–670. doi: 10.1002/joc.3370040609. [Crossref] 
  36. [36] N. Nicholls. “Recent performance of a method for forecasting Australian seasonal tropical cyclone activity.” In: Australian Meteorological Magazine 40.2 (1992), pp. 105–110.  
  37. [37] J. C. Platt. “Fast training of support vector machines using sequential minimal optimization.” In: Advances in KernelMethods. Support Vector Learning. Ed. by B. Schölkopf, C. J. C. Burges, and A. J. Smola. MIT Press, 1999, pp. 185–208. isbn: 9780262194167.  
  38. [38] R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria, 2014. url: http://www.R-project.org/.  
  39. [39] S. Rajasekaran, S. Gayathri, and T.-L. Lee. “Support vector regression methodology for storm surge predictions.” In: Ocean Engineering 35.16 (2008), pp. 1578–1587. doi: 10.1016/j.oceaneng.2008.08.004. [Crossref] 
  40. [40] H. A. Ramsay, M. B. Richman, and L. M. Leslie. “Seasonal tropical cyclone predictions using optimized combinations of ENSO regions: Application to the Coral Sea basin.” In: Journal of Climate 27.22 (2014), pp. 8527–8542. doi: 10.1175/JCLI-D- 14-00017.1. [Crossref] 
  41. [41] M. B. Richman and L. M. Leslie. “Adaptive machine learning approaches to seasonal prediction of tropical cyclones.” In: Procedia Computer Science 12 (2012), pp. 276–281. doi: 10.1016/j.procs.2012.09.069. [Crossref] 
  42. [42] K. K. Saha and S. A.Wasimi. “An index to assess the propensity of landfall in Australia of a tropical cyclone.” In: Natural Hazards 72.2 (2014), pp. 1111–1121. doi: 10.1007/s11069-014-1058-y. [Crossref] 
  43. [43] J. Shao. Mathematical statistics. 2nd ed. Springer, 2003. isbn: 978-0-387-95382-3.  
  44. [44] K. E. Trenberth. “Signal versus noise in the Southern Oscillation.” In: Monthly Weather Review 112.2 (1984), pp. 326–332. doi: 10.1175/1520-0493(1984)112<0326:SVNITS>2.0.CO;2. [Crossref] 
  45. [45] K. E. Trenberth. “The definition of El Niño.” In: Bulletin of the AmericanMeteorological Society 78.12 (1997), pp. 2771–2777. doi: 10.1175/1520-0477(1997)078<2771:TDOENO>2.0.CO;2. [Crossref] 
  46. [46] V. Vapnik. The nature of statistical learning theory. Springer-Verlag New York, 2000. isbn: 978-0-387-98780-4. doi: 10.1007/ 978-1-4757-3264-1.  Zbl0934.62009
  47. [47] P. J. Webster, G. J. Holland, J. A. Curry, and H.-R. Chang. “Changes in tropical cyclone number, duration, and intensity in a warming environment.” In: Science 309.5742 (2005), pp. 1844–1846. doi: 10.1126/science.1116448. [Crossref] 
  48. [48] J. S.Wijnands, K. Shelton, and Y. Kuleshov. “Improving the operational methodology of tropical cyclone seasonal prediction in the Australian and the South Pacific Ocean regions.” In: Advances in Meteorology (2014), pp. 1–8. doi: 10.1155/ 2014/838746. [Crossref] 
  49. [49] D. S.Wilks. Statistical methods in the atmospheric sciences. 2nd ed. International Geophysics Series. Elsevier, 2006. isbn: 978-0-12-751966-1.  
  50. [50] K.Wolter and M. S. Timlin. “Monitoring ENSO in COADS with a seasonally adjusted principal component index.” In: Proceedings of the 17th Annual Climate DiagnosticsWorkshop. Norman, OK: NOAA/NMC/CAC, NSSL, Oklahoma Clim. Survey, CIMMS and the School of Meteor., Univ. of Oklahoma, 1993, pp. 52–57.  
  51. [51] Z. Zeng,W.W. Hsieh, A. Shabbar, andW. R. Burrows. “Seasonal prediction of winter extreme precipitation over Canada by support vector regression.” In: Hydrology and Earth System Sciences 15.1 (2011), pp. 65–74. doi: 10.5194/hess-15-65-2011. [Crossref] 

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

                
Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.