Lightning Activity Predictions For Single Buoy Moorings

Ee78d71d8c39eaf0880a1e9ed7187a95?s=47 datawookie
September 12, 2013

Lightning Activity Predictions For Single Buoy Moorings

A short talk that I gave at the LIGHTS 2013 Conference (Johannesburg, 12 September 2013). The slides are a little short on text because I like the audience to hear the content rather than read it. The objective with this project was to develop a model which would predict the occurrence of lightning in the vicinity of a Single Buoy Mooring (SBM). I used data from the World Wide Lightning Location Network (WWLLN). I considered four possible models: Neural Network, Conditional Inference Tree, Support Vector Machine and Random Forest. Of the four, Random Forests produced the best performance. The preliminary results of the model are very promising: there is good agreement between predicted and observed lightning occurrence in the vicinity of the SBM.

Ee78d71d8c39eaf0880a1e9ed7187a95?s=128

datawookie

September 12, 2013
Tweet

Transcript

  1. 1.

    Lightning Activity Predictions For Single Buoy Moorings Andrew B. Collier

    Exegetic Analytics andrew@exegetic.biz http://1.bp.blogspot.com/
  2. 2.

    2 Oceanic lightning... … is difficult to investigate. Brooks, C.

    E. P. (1925). The Distribution of Thunderstorms over the Globe. Geophysical Memoirs, 3(24), 147–164. “... not much happening over the oceans...”
  3. 3.

    3 Trent, E. M., & Gathman, S. G. (1972). Oceanic

    thunderstorms. Pure and Applied Geophysics, 100(1), 60–69. doi:10.1007/BF00880227. ocean:land area ratio between 3:1 and 1:1 72% of the Earth covered by ocean sparse lightning over ocean
  4. 5.

    6 Christian, H. J., Blakeslee, R. J., Boccippio, D. J.,

    Boeck, W. L., Buechler, D. E., Driscoll, K. T., Goodman, S. J., et al. (2003). Global frequency and distribution of lightning as observed from space by the Optical Transient Detector. Journal of Geophysical Research, 108(D1). doi:10.1029/2002JD002347.
  5. 6.

    7 Füllekrug, M., Price, C., Yair, Y., & Williams, E.

    R. (2002). Intense oceanic lightning. Annales Geophysicae, 20, 133–137. doi:10.5194/angeo- 20-133-2002
  6. 9.

    10

  7. 10.

    11

  8. 12.

    13

  9. 13.

    14 WWLLN data partitioned into • 15 min time bins

    • 10, 20, 50, 100 and 200 km • 4 quadrants Model • Input: lightning counts • Output: lightning presence
  10. 14.

    16 A B1 B2 B3 B4 C1 C2 C3 C4

    D1 D2 D3 D4 E1 E2 E3 E4 2506 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2507 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2509 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2726 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2729 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2733 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2796 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2880 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2882 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2884 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2886 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 2887 FALSE 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 1 2890 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 2891 FALSE 0 0 0 0 0 0 2 0 0 2 3 0 0 1 0 0 2892 FALSE 0 0 0 0 0 0 3 0 0 0 2 0 0 1 0 0 2893 TRUE 0 0 0 0 0 0 1 1 0 5 2 4 0 1 0 1 2894 TRUE 0 0 3 0 0 0 4 5 0 3 1 2 0 3 0 0 2895 TRUE 0 0 3 4 0 2 1 6 1 3 0 1 0 17 0 0 2896 TRUE 2 0 0 2 0 11 1 0 0 1 0 0 0 23 0 0 2897 FALSE 0 0 0 0 2 12 0 0 1 18 0 1 0 6 0 0 2898 FALSE 0 0 0 0 3 4 0 0 17 15 0 3 0 7 0 1 2899 FALSE 1 0 0 0 0 1 0 0 16 29 0 0 0 8 0 0 2900 FALSE 0 0 0 0 1 0 0 0 14 30 0 0 5 14 0 2 2901 FALSE 0 0 0 0 1 0 0 0 16 28 0 0 5 11 0 0 2902 FALSE 0 0 0 0 0 1 0 0 14 8 0 0 12 41 1 0 2903 FALSE 0 0 0 0 0 0 0 0 5 5 0 0 15 64 0 1 2904 FALSE 0 0 0 0 0 0 0 0 3 0 0 0 24 59 0 1 2905 FALSE 0 0 0 0 0 0 0 0 1 0 0 0 45 32 1 0 2906 FALSE 0 0 0 0 0 0 0 0 0 0 0 0 26 17 0 0 2907 FALSE 0 0 0 0 0 0 0 0 0 1 0 0 66 10 0 0 2908 FALSE 0 0 0 0 0 0 0 0 0 1 0 0 31 2 0 0
  11. 15.

    18 Confusion Matrices Neural Network observed predicted FALSE TRUE FALSE

    12681 65 TRUE 15 48 Conditional Inference Tree observed predicted FALSE TRUE FALSE 12690 92 TRUE 6 21 Random Forest observed predicted FALSE TRUE FALSE 12696 32 TRUE 0 81 Support Vector Machine observed predicted FALSE TRUE FALSE 12696 22 TRUE 0 91 false positive false negative
  12. 17.

    20

  13. 18.

    21

  14. 22.