FY-4A LMI Observed Lightning Activity in Super Typhoon Mangkhut (2018) in Comparison with WWLLN Data

+ Author Affiliations + Find other works by these authors
  • Corresponding author: Wenjuan ZHANG, zwj@cma.gov.cn
  • Funds:

    Supported by the National Key Research and Development Program of China (2017YFC1501502), Open Fund of Key Laboratory of Radiometric Calibration and Validation for Environmental Satellites of National Satellite Meteorological Center, and National Natural Science Foundation of China (41405004 and 41875001)

  • doi: 10.1007/s13351-020-9500-4

PDF

  • Using lightning observations from the Fengyun-4A Lightning Mapping Imager (FY-4A LMI), best-track data from the Shanghai Typhoon Institute, bright temperature (TBB) data from Himawari-8 satellite, and composite reflectivity (CR) data from the South China radar network, we investigate the spatiotemporal distribution of lightning activity and convective evolution during the landfall of Super Typhoon Mangkhut, the strongest landing typhoon in China in 2018. Three stages of active total lightning are observed, and differences of lightning characteristics between the inner core and the outer rainbands are present. The onset of inner-core lightning outbreak is about 4 h ahead of the maximum intensity of the storm, providing indicative information on the change of typhoon intensity. Lightning rates in the outer rainbands increase rapidly 12 h before the landfall, and lightning activity is mainly confined in the outer rainbands after the landfall. A good correlation in hourly variation is shown between lightning rates from the LMI and TBBs from the satellite. The averaged TBB within the inner core reaches its minimum (–80°C) when the inner-core lightning outbreak occurs, indicating the occurrence and enhancement of deep convection there. Lightning locations observed by the LMI has a good spatial correspondence with regions of low TBBs and high CRs, revealing the monitoring capability of the LMI to lightning activity and deep convection in landing typhoons. Comparisons between the World Wide Lightning Location Network (WWLLN) and the LMI reveal that the spatial distribution, temporal evolution, and radial pattern of lightning activity in Mangkhut observed by the two systems are consistent. Furthermore, due to the detection capability of total lightning, the LMI has advantages in revealing the higher ratio of intra-cloud lightning within the inner core in typhoon. The continuous and real-time observation of FY-4A LMI provides an unprecedented platform for monitoring total lightning and deep convection in landing typhoons in China, which will promote the generation of new research and applications in the future.
  • 加载中
  • Fig. 1.  Schematic diagram of the observation field of FY-4A LMI (Huang and Hui, 2014).

    Fig. 2.  Best tracks and intensities of Super Typhoon Mangkhut during the period of analysis (0000 UTC 14–1200 UTC 17 September 2018) and the observation field of LMI. Storm intensities are highlighted by color lines. The bold black line indicates the observational range of LMI. Shadings represent the topography (DEM; m).

    Fig. 3.  Distributions of lightning densities (shadings) from the LMI for (a) events, (b) groups, and (c) flashes, and (d) median TBBs (shading; °C) of the storm from Himawari-8 with the best tracks and intensities (denoted by color lines) during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut.

    Fig. 4.  Hourly evolutions of lightning rates (left-hand axis) from the LMI for (a) events, (b) groups, and (c) flashes within the inner core (solid bars) and outer rainbands (hollow bars) during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut overlaid with storm intensity and a curve for pressure (right-hand axis; hPa) on each panel. Note that lightning rates of the inner core are multiplied by 10. The arrow points to the time of landfall

    Fig. 5.  Lightning density (shading) for the LMI flashes as a function of storm intensity and distance from the storm center (left-hand axis; km) with a bold black curve for pressure (right-hand axis; hPa) during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. The arrow points to the time of landfall.

    Fig. 6.  Time variation of LMI flash rates (left-hand axis) within the inner core (denoted by red bars) and outer rainbands (denoted by blue bars), superimposed with median TBBs (right-hand axis; red line for inner core and blue line for outer rainbands) and the maximum sustained wind speed (right-hand axis;black line) from Himawari-8 during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. The gray shadings indicate three stages of active lightning. The arrow points to the time of landfall.

    Fig. 7.  TBB imageries (shading; °C) from Himawari-8 overlaid with the LMI (red dots) and WWLLN flashes (blue dots) within ± 30 min of the satellite observation time: (a) 1500 UTC 14, (b) 2300 UTC 15, (c) 1600 UTC 16, and (d) 2200 UTC 16 September during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. ◇ indicates the typhoon center.

    Fig. 8.  Composite reflectivity (CR; dBZ; shading) from South China radar network overlaid with the LMI (dark crosses) and WWLLN flashes (blue crosses) within ± 5 min of the radar observation time: (a) 0400 UTC, (b) 0900 UTC, (c) 1600 UTC, and (d) 2200 UTC 16 September during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. ◇ indicates the typhoon center, and • indicates radar station.

    Fig. 9.  Radial distribution of lightning densities detected by the LMI and WWLLN during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut.

    Fig. 10.  As in Fig. 4, but for WWLLN lightning rates of (a) strokes and (b) flashes.

    Fig. 11.  Normalized scatterplots of the hourly lightning rates recorded by the LMI and WWLLN. (a) LMI group and WWLLN stroke; (b) LMI flash and WWLLN flash. The black line shows a 1 : 1 relationship, and the Pearson correlation coefficient (r) is indicated in the lower right cornner of each panel.

    Table 1.  Lightning numbers detected by the LMI and WWLLN during the landfall period of Super Typhoon Mangkhut (2018)

    AreaAveraged range (km)WWLLN LMI
    StrokeFlash EventGroupFlash
    Inner core0–10014611087930173
    Outer rainbands260–49030,33917,36879,62623,2075405
    Typhoon0–49030,91817,77781,34523,8185571
    Note: The landfall period is from 0000 UTC 14 to 1200 UTC 17 September 2018. The two datasets are selected within the same definition of storm ranges.
    Download: Download as CSV
  • [1]

    Abarca, S. F., K. L. Corbosiero, and T. J. Galarneau, 2010: An evaluation of the Worldwide Lightning Location Network (WWLLN) using the National Lightning Detection Network (NLDN) as ground truth. J. Geophys. Res. Atmos., 115, D18206. doi: 10.1029/2009JD013411.
    [2]

    Abarca, S. F., K. L. Corbosiero, and D. Vollaro, 2011: The World Wide Lightning Location Network and convective activity in tropical cyclones. Mon. Wea. Rev., 139, 175–191. doi: 10.1175/2010MWR3383.1.
    [3]

    Aich, V., R. Holzworth, S. J. Goodman, et al., 2018: Lightning: A new essential climate variable. EOS, 99, doi: 10.1029/2018EO104583. Published on 7 September 2018; accessed on 5 November 2019.
    [4]

    Albrecht, R. I., S. J. Goodman, D. E. Buechler, et al., 2016: Where are the lightning hotspots on Earth? Bull. Amer. Meteor. Soc., 97, 2051–2068. doi: 10.1175/BAMS-D-14-00193.1.
    [5]

    Bovalo, C., C. Barthe, N. Yu, et al., 2014: Lightning activity within tropical cyclones in the South West Indian Ocean. J. Geophys. Res. Atmos., 119, 8231–8244. doi: 10.1002/2014JD021651.
    [6]

    Cao, D. J., 2016: The development of product algorithm of the Fengyun-4 geostationary lightning mapping imager. Adv. Meteor. Sci. Technol., 6, 94–98. doi: 10.3969/j.issn.2095-1973.2016.01.014. (in Chinese)
    [7]

    Carey, L. D., and S. A. Rutledge, 1996: A multiparameter radar case study of the microphysical and kinematic evolution of a lightning producing storm. Meteor. Atmos. Phys., 59, 33–64. doi: 10.1007/BF01032000.
    [8]

    Cecil, D. J., and E. J. Zipser, 1999: Relationships between tropical cyclone intensity and satellite-based indicators of inner core convection: 85-GHz ice-scattering signature and lightning. Mon. Wea. Rev., 127, 103–123. doi: 10.1175/1520-0493(1999)127<0103:RBTCIA>2.0.CO;2.
    [9]

    Cecil, D. J., E. J. Zipser, and S. W. Nesbitt, 2002: Reflectivity, ice scattering, and lightning characteristics of hurricane eyewalls and rainbands. Part I: Quantitative description. Mon. Wea. Rev., 130, 769–784. doi: 10.1175/1520-0493(2002)130<0769:RISALC>2.0.CO;2.
    [10]

    Chen, H. B., and D. R. Lyu, 2001: Observations of lightning from spcae: A review. Acta Meteor. Sinica, 59, 377–383. doi: 10.3321/j.issn:0577-6619.2001.03.014. (in Chinese)
    [11]

    Chen, L. S., Z. Y. Meng, and C. H. Cong, 2017: An overview on the research of typhoon rainfall distribution. J. Marine Meteor., 37, 1–7. doi: 10.19513/j.cnki.issn2096-3599.2017.04.001. (in Chinese)
    [12]

    Corbosiero, K. L., and J. Molinari, 2002: The effects of vertical wind shear on the distribution of convection in tropical cyclones. Mon. Wea. Rev., 130, 2110–2123. doi: 10.1175/1520-0493(2002)130<2110:TEOVWS>2.0.CO;2.
    [13]

    Corbosiero, K. L., and J. Molinari, 2003: The relationship between storm motion, vertical wind shear, and convective asymmetries in tropical cyclones. J. Atmos. Sci., 60, 366–376. doi: 10.1175/1520-0469(2003)060<0366:TRBSMV>2.0.CO;2.
    [14]

    DeMaria, M., R. T. DeMaria, J. A. Knaff, et al., 2012: Tropical cyclone lightning and rapid intensity change. Mon. Wea. Rev., 140, 1828–1842. doi: 10.1175/MWR-D-11-00236.1.
    [15]

    DeMaria, M., C. R. Sampson, J. A. Knaff, et al., 2014: Is tropical cyclone intensity guidance improving? Bull. Amer. Meteor. Soc., 95, 387–398. doi: 10.1175/BAMS-D-12-00240.1.
    [16]

    Dong, M. Y., L. S. Chen, P. J. Zheng, et al., 2009: Research progress on abrupt intensification of heavy rainfall and super heavy rainfall associated with landfalling tropical cyclones. J. Trop. Meteor., 25, 495–502. doi: 10.3969/j.issn.1004-4965.2009.04.016. (in Chinese)
    [17]

    Fan, P. L., D. Zheng, Y. J. Zhang, et al., 2018: A performance evaluation of the world wide lightning location network (WWLLN) over the Tibetan Plateau. J. Atmos. Oceanic Technol., 35, 927–939. doi: 10.1175/JTECH-D-17-0144.1.
    [18]

    Fierro, A. O., L. Leslie, E. Mansell, et al., 2007: A high-resolution simulation of microphysics and electrification in an idealized hurricane-like vortex. Meteor. Atmos. Phys., 98, 13–33. doi: 10.1007/s00703-006-0237-0.
    [19]

    Fierro, A. O., X. M. Shao, T. Hamlin, et al., 2011: Evolution of eyewall convective events as indicated by intracloud and cloud-to-ground lightning activity during the rapid intensification of hurricanes Rita and Katrina. Mon. Wea. Rev., 139, 1492–1504. doi: 10.1175/2010MWR3532.1.
    [20]

    Fierro, A. O., S. N. Stevenson, and R. M. Rabin, 2018: Evolution of GLM-observed total lightning in Hurricane Maria (2017) during the period of maximum intensity. Mon. Wea. Rev., 146, 1641–1666. doi: 10.1175/MWR-D-18-0066.1.
    [21]

    Franklin, C. N., G. J. Holland, and P. T. May, 2005: Sensitivity of tropical cyclone rainbands to ice-phase microphysics. Mon. Wea. Rev., 133, 2473–2493. doi: 10.1175/MWR2989.1.
    [22]

    Goodman, S. J., R. J. Blakeslee, W. J. Koshak, et al., 2013: The GOES-R Geostationary Lightning Mapper (GLM). Atmos. Res., 125–126, 34–49. doi: 10.1016/j.atmosres.2013.01.006.
    [23]

    Griffin, E. M., T. J. Schuur, D. R. MacGorman, et al., 2014: An electrical and polarimetric analysis of the overland reintensification of Tropical storm Erin (2007). Mon. Wea. Rev., 142, 2321–2344. doi: 10.1175/MWR-D-13-00360.1.
    [24]

    Guimond, S. R., G. M. Heymsfield, and F. J. Turk, 2010: Multiscale observations of Hurricane Dennis (2005): The effects of hot towers on rapid intensification. J. Atmos. Sci., 67, 633–654. doi: 10.1175/2009JAS3119.1.
    [25]

    Holle, R. L., K. L. Cummins, and W. A. Brooks, 2016: Seasonal, monthly, and weekly distributions of NLDN and GLD360 cloud-to-ground lightning. Mon. Wea. Rev., 144, 2855–2870. doi: 10.1175/MWR-D-16-0051.1.
    [26]

    Huang, F. X., 2007: Lightning imaging sensor on FY-4 meteorological satellite: Mission and challenge. Meteor. Sci. Technol., 35, 35–42. doi: 10.3969/j.issn.1671-6345.2007.z1.008. (in Chinese)
    [27]

    Huang, F., and W. Hui, 2014: Analysis of Application Requirements for Fengyun-4 Lightning Mapping Imager. Technical Report of the National Satellite Meteorological Center, 26 pp. (in Chinese)
    [28]

    Hui, W., F. X. Huang, and Q. Guo, 2015: Filtering of false signals in lightning detection by geostationary satellite. Meteor. Sci. Technol., 43, 805–813, 828. doi: 10.3969/j.issn.1671-6345.2015.05.006. (in Chinese)
    [29]

    Hui, W., F. X. Huang, and J. Zhu, 2016: Technology of optical detection of lightning from space. Opt. Precis. Eng., 24, 361–369. doi: 10.3788/OPE.20162413.0361. (in Chinese)
    [30]

    Hutchins, M. L., R. H. Holzworth, K. S. Virts, et al., 2013: Radiated VLF energy differences of land and oceanic lightning. Geophys. Res. Lett., 40, 2390–2394. doi: 10.1002/grl.50406.
    [31]

    Jacobson, A. R., R. Holzworth, J. Harlin, et al., 2006: Performance assessment of the World wide Lightning Location Network (WWLLN), using the Los Alamos Sferic Array (LASA) as ground truth. J. Atmos. Oceanic Technol., 23, 1082–1092. doi: 10.1175/JTECH1902.1.
    [32]

    Lang, T. J., and S. A. Rutledge, 2002: Relationships between convective storm kinematics, precipitation, and lightning. Mon. Wea. Rev., 130, 2492–2506. doi: 10.1175/1520-0493(2002)130<2492:RBCSKP>2.0.CO;2.
    [33]

    Lei, X. T., Y. J. Zhang, and M. Ma, 2009: The preliminary analysis of the lightning characteristics and the relationship between lightning and intensity of tropical cyclone in northwest Pacific. Acta Oceanol. Sinica, 31, 29–38. doi: 10.3321/j.issn:0253-4193.2009.04.004. (in Chinese)
    [34]

    Liang, H., S. L. Bao, Q. Chen, et al., 2017: Design and implementation of FY-4 geostationary lightning imager. Aerospace Shanghai, 34, 43–51. doi: 10.19328/j.cnki.1006-1630.2017.04.006. (in Chinese)
    [35]

    Marks, F. D., L. K. Shay, G. Barnes, et al, 1998: Landfalling tropical cyclones: Forecast problems and associated research opportunities. Bull. Amer. Meteor. Soc., 79, 305–323. doi: 10.1175/1520-0477(1998)079<0305:LTCFPA>2.0.CO;2.
    [36]

    Molinari, J., P. K. Moore, V. P. Idone, et al., 1994: Cloud-to-ground lightning in Hurricane Andrew. J. Geophys. Res. Atmos., 99, 16665–16676. doi: 10.1029/94JD00722.
    [37]

    Molinari, J., P. Moore, and V. Idone, 1999: Convective structure of hurricanes as revealed by lightning locations. Mon. Wea. Rev., 127, 520–534. doi: 10.1175/1520-0493(1999)127<0520:csohar>2.0.co;2.
    [38]

    Pan, L. X., and X. S. Qie, 2010: Lightning activity in super typhoon Sepat (0709). Chinese J. Atmos. Sci., 34, 1088–1098. doi: 10.3321/j.issn:0577-6619.2009.05.012. (in Chinese)
    [39]

    Pan, L. X., X. S. Qie, D. X. Liu, et al., 2010: The lightning activities in super typhoons over the Northwest Pacific. Sci. China Earth Sci., 53, 1241–1248. doi: 10.1007/s11430-010-3034-z.
    [40]

    Pan, L. X., X. S. Qie, and D. F. Wang, 2014: Lightning activity and its relation to the intensity of typhoons over the Northwest Pacific Ocean. Adv. Atmos. Sci., 31, 581–592. doi: 10.1007/s00376-013-3115-y.
    [41]

    Petersen, W. A., R. C. Cifelli, S. A. Rutledge, et al., 1999: Shipborne dual-Doppler operations during TOGA COARE: Integrated observations of storm kinematics and electrification. Bull. Amer. Meteor. Soc., 80, 81–98. doi: 10.1175/1520-0477(1999)080<0081:SDDODT>2.0.CO;2.
    [42]

    Price, C., M. Asfur, and Y. Yair, 2009: Maximum hurricane intensity preceded by increase in lightning frequency. Nat. Geosci., 2, 329–332. doi: 10.1038/ngeo477.
    [43]

    Ranalkar, M. R., S. D. Pawar, and P. P. Kumar, 2017: Characteristics of lightning activity in tropical cyclones developed over North Indian Ocean basin during 2010−2015. Atmos. Res., 187, 16–32. doi: 10.1016/j.atmosres.2016.12.003.
    [44]

    Rodger, C. J., J. B. Brundell, R. L. Dowden, et al., 2004: Location accuracy of long distance VLF lightning location network. Ann. Geophys., 22, 747–758. doi: 10.5194/angeo-22-747-2004.
    [45]

    Rogers, R. F., P. D. Reasor, and J. A. Zhang, 2015: Multiscale structure and evolution of Hurricane Earl (2010) during rapid intensification. Mon. Wea. Rev., 143, 536–562. doi: 10.1175/MWR-D-14-00175.1.
    [46]

    Rudlosky, S. D., S. J. Goodman, K. S. Virts, et al., 2019: Initial geostationary lightning mapper observations. Geophys. Res. Lett., 46, 1097–1104. doi: 10.1029/2018GL081052.
    [47]

    Solorzano, N. N., J. N. Thomas, M. L. Hutchins, et al., 2016: WWLLN lightning and satellite microwave radiometrics at 37 to 183 GHz: Thunderstorms in the broad tropics. J. Geophys. Res. Atmos., 121, 12298–12318. doi: 10.1002/2016JD025374.
    [48]

    Squires, K., and S. Businger, 2008: The morphology of eyewall lightning outbreaks in two category 5 hurricanes. Mon. Wea. Rev., 136, 1706–1726. doi: 10.1175/2007MWR2150.1.
    [49]

    Stevenson, S. N., K. L. Corbosiero, and J. Molinari, 2014: The convective evolution and rapid intensification of Hurricane Earl (2010). Mon. Wea. Rev., 142, 4364–4380. doi: 10.1175/MWR-D-14-00078.1.
    [50]

    Stevenson, S. N., K. L. Corbosiero, and S. F. Abarca, 2016: Lightning in eastern North Pacific tropical cyclones: A comparison to the North Atlantic. Mon. Wea. Rev., 144, 225–239. doi: 10.1175/MWR-D-15-0276.1.
    [51]

    Stevenson, S. N., K. L. Corbosiero, M. DeMaria, et al., 2018: A 10-year survey of tropical cyclone inner-core lightning bursts and their relationship to intensity change. Wea. Forecasting, 33, 23–36. doi: 10.1175/WAF-D-17-0096.1.
    [52]

    Susca-Lopata, G., J. Zawislak, and E. J. Zipser, 2015: The role of observed environmental conditions and precipitation evolution in the rapid intensification of Hurricane Earl (2010). Mon. Wea. Rev., 143, 2207–2223. doi: 10.1175/MWR-D-14-00283.1.
    [53]

    Takahashi, T., 1978: Riming electrification as a charge generation mechanism in thunderstorms. J. Atmos. Sci., 35, 1536–1548. doi: 10.1175/1520-0469(1978)035<1536:reaacg>2.0.co;2.
    [54]

    Thomas, J. N., N. N. Solorzano, S. A. Cummer, et al., 2010: Polarity and energetics of inner core lightning in three intense North Atlantic hurricanes. J. Geophys. Res. Space Phys., 115, A00E15. doi: 10.1029/2009JA014777.
    [55]

    Virts, K. S., J. M. Wallace, M. L. Hutchins, et al., 2013: Highlights of a new ground-based, hourly global lightning climatology. Bull. Amer. Meteor. Soc., 94, 1381–1391. doi: 10.1175/BAMS-D-12-00082.1.
    [56]

    Wadler, J. B., R. F. Rogers, and P. D. Reasor, 2018: The relationship between spatial variations in the structure of convective bursts and tropical cyclone intensification as determined by airborne Doppler radar. Mon. Wea. Rev., 146, 761–780. doi: 10.1175/MWR-D-17-0213.1.
    [57]

    Wang, F., X. S. Qie, and X. D. Cui, 2017: Climatological characteristics of lightning activity within tropical cyclones and its relationship to cyclone intensity change over the Northwest Pacific. Chinese J. Atmos. Sci., 41, 1167–1176. doi: 10.3878/j.issn.1006-9895.1704.17102. (in Chinese)
    [58]

    Wang, Y., D. Zheng, and Y. J. Zhang, 2011: Characteristics of lightning activities and precipitation in typhoon processes making landfall in China from 2000 to 2007. J. Appl. Meteor. Sci., 22, 321–328. doi: 10.3969/j.issn.1001-7313.2011.03.008. (in Chinese)
    [59]

    Wen, Y., Y. Zhao, G. Zhao, et al., 2015: Characteristics of lightning activity and radar reflectivity during Typhoon “Morakot” landfall. Plateau Meteor., 34, 832–841. doi: 10.7522/j.issn.1000-0534.2015.0036. (in Chinese)
    [60]

    Williams, E. R., 1988: The electrification of thunderstorms. Sci. Amer., 259, 88–99. doi: 10.1038/scientificamerican1188-88.
    [61]

    Xu, L. T., Y. J. Zhang, W. J. Zhang, et al., 2016: Simulation of the evolution of the charge structure during the landfall of Typhoon Molave (2009). Acta Meteor. Sinica, 74, 1002–1016. doi: 10.11676/qxxb2016.071. (in Chinese)
    [62]

    Xu, W. X., S. A. Rutledge, and W. J. Zhang, 2017: Relationships between total lightning, deep convection, and tropical cyclone intensity change. J. Geophys. Res. Atmos., 122, 7047–7063. doi: 10.1002/2017JD027072.
    [63]

    Yang, J., Z. Q. Zhang, C. Y. Wei, et al., 2017: Introducing the new generation of Chinese geostationary weather satellites, Fengyun-4. Bull. Amer. Meteor. Soc., 98, 1637–1658. doi: 10.1175/BAMS-D-16-0065.1.
    [64]

    Zhang, W. J., Y. J. Zhang, D. Zheng, et al., 2012: Lightning distribution and eyewall outbreaks in tropical cyclones during landfall. Mon. Wea. Rev., 140, 3573–3586. doi: 10.1175/MWR-D-11-00347.1.
    [65]

    Zhang, W. J., Y. J. Zhang, and X. J. Zhou, 2013: Lightning activity and precipitation characteristics of Typhoon Molave (2009) around its landfall. Acta Meteor. Sinica, 27, 742–757. doi: 10.1007/s13351-013-0510-3.
    [66]

    Zhang, W. J., Y. J. Zhang, D. Zheng, et al., 2015: Relationship between lightning activity and tropical cyclone intensity over the northwest Pacific. J. Geophys. Res. Atmos., 120, 4072–4089. doi: 10.1002/2014JD022334.
    [67]

    Zhang, W. J., Y. J. Zhang, D. Zheng, et al., 2018: Lightning climatology over the Northwest Pacific region: An 11-year study using data from the World Wide Lightning Location Network. Atmos. Res., 210, 41–57. doi: 10.1016/j.atmosres.2018.04.013.
    [68]

    Zhang, W. J., S. A. Rutledge, W. X. Xu, et al., 2019: Inner-core lightning outbreaks and convective evolution in Super Typhoon Haiyan (2013). Atmos. Res., 219, 123–139. doi: 10.1016/j.atmosres.2018.12.028.
    [69]

    Zhao, F., C. X. Ji, S. T. Gao, et al., 2012: Study of the structure and characteristics associated with landfalling typhoons on the southeastern coast of Zhejiang province in China using Doppler radar data. Acta Meteor. Sinica, 70, 15–29. doi: 10.11676/qxxb2012.002. (in Chinese)
    [70]

    Zhao, Z., 2019: Multi-satellite observations on the structure characteristics of typhoon Meranti in 2016. Plateau Meteor., 38, 156–164. doi: 10.7522/j.issn.1000-0534.2018.00065. (in Chinese)
  • 11-中文稿-张文娟.pdf

  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

FY-4A LMI Observed Lightning Activity in Super Typhoon Mangkhut (2018) in Comparison with WWLLN Data

    Corresponding author: Wenjuan ZHANG, zwj@cma.gov.cn
  • 1. State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081
  • 2. Key Laboratory of Radiometric Calibration and Validation for Environmental Satellites, National Satellite Meteorological Center, China Meteorological Administration, Beijing 100081
  • 3. Ecological and Meteorological Center of Heilongjiang Province, Harbin 150030
  • 4. Institute of Atmospheric Science, Fudan University, Shanghai 200438
Funds: Supported by the National Key Research and Development Program of China (2017YFC1501502), Open Fund of Key Laboratory of Radiometric Calibration and Validation for Environmental Satellites of National Satellite Meteorological Center, and National Natural Science Foundation of China (41405004 and 41875001)

Abstract: Using lightning observations from the Fengyun-4A Lightning Mapping Imager (FY-4A LMI), best-track data from the Shanghai Typhoon Institute, bright temperature (TBB) data from Himawari-8 satellite, and composite reflectivity (CR) data from the South China radar network, we investigate the spatiotemporal distribution of lightning activity and convective evolution during the landfall of Super Typhoon Mangkhut, the strongest landing typhoon in China in 2018. Three stages of active total lightning are observed, and differences of lightning characteristics between the inner core and the outer rainbands are present. The onset of inner-core lightning outbreak is about 4 h ahead of the maximum intensity of the storm, providing indicative information on the change of typhoon intensity. Lightning rates in the outer rainbands increase rapidly 12 h before the landfall, and lightning activity is mainly confined in the outer rainbands after the landfall. A good correlation in hourly variation is shown between lightning rates from the LMI and TBBs from the satellite. The averaged TBB within the inner core reaches its minimum (–80°C) when the inner-core lightning outbreak occurs, indicating the occurrence and enhancement of deep convection there. Lightning locations observed by the LMI has a good spatial correspondence with regions of low TBBs and high CRs, revealing the monitoring capability of the LMI to lightning activity and deep convection in landing typhoons. Comparisons between the World Wide Lightning Location Network (WWLLN) and the LMI reveal that the spatial distribution, temporal evolution, and radial pattern of lightning activity in Mangkhut observed by the two systems are consistent. Furthermore, due to the detection capability of total lightning, the LMI has advantages in revealing the higher ratio of intra-cloud lightning within the inner core in typhoon. The continuous and real-time observation of FY-4A LMI provides an unprecedented platform for monitoring total lightning and deep convection in landing typhoons in China, which will promote the generation of new research and applications in the future.

1.   Introduction
  • Great progress of track forecasts on tropical cyclone (TC) has been achieved in the last decade due to the development of detection technology and understanding of the mechanisms. However, forecasts of TC intensity still need further improvement (Marks et al., 1998; DeMaria et al., 2014). TC intensity change is not only related to large-scale environment and dynamic processes, but also to internal meso- and small-scale convections (Guimond et al., 2010; Wadler et al., 2018). Most of the present observational platforms (e.g., the ground-based observation, aircraft observation, and polar-orbiting satellites) cannot continuously observe the outbreak and enhancement of small-scale convection in TCs (especially over the sea), limiting the prediction of rapid intensity change caused by internal physical processes (Fierro et al., 2011).

    Lightning activity is closely related to the dynamic and microphysical processes as well as precipitation structure in thunderstorms (Petersen et al., 1999). The collision between graupel and ice particle in the mixed-phase region is the major process of noninductive charging mechanism (Takahashi, 1978) that leads to lightning in thunderstorms (Williams, 1988). Lightning can be used to indicate the location of meso- and small-scale convection as well as strong updrafts in TCs (Molinari et al., 1994). The enhancement of lightning activity can indicate the volume increase of graupel and hail particles in TCs, characterizing the development of deep convection (Fierro et al., 2007). Therefore, the combination of lightning detection network with radar, satellite, and other observational platforms would aid to further understand the convective structures and intensity changes of TCs.

    Previous studies on lightning activity in TCs are mainly based on ground-based lightning location networks, such as the World Wide Lightning Location Network (WWLLN; Rodger et al., 2004) and the Global Lightning Detection network 360 (GLD360; Holle et al., 2016). These studies have promoted the understanding of temporal and spatial distribution of TC lightning and its relationship with TC intensity change (Price et al., 2009; Pan et al., 2010; Abarca et al., 2011; DeMaria et al., 2012; Zhang et al., 2015; Wang et al., 2017). The studies also found a positive correlation between the inner-core lightning and TC intensificaiton (Squires and Businger, 2008; Stevenson et al., 2014; Susca-Lopata et al., 2015; Zhang et al., 2015). However, the ground-based lightning location network works in low and very low frequency (VLF), and mainly detects cloud-to-ground (CG) lightning. In fact, intra-cloud (IC) lightning accounts for a higher proportion of total lightning (both CG and IC), and thus these ground-based networks limit the comprehensive and in-depth understanding of lightning activity in TCs.

    Satellite lightning imaging opens a new era of global (land and ocean) observation for lightning activity in TCs. As total lightning is more closely related to convective properties (content of graupel and hail, volume of updrafts, the maximum vertical velocity, etc.) of thunderstorm (Carey and Rutledge, 1996; Lang and Rutledge, 2002), satellite observation is considered to be the best platform for monitoring lightning activity in global TCs (Chen and Lyu, 2001; Albrecht et al., 2016). Using lightning data from the low-orbit lightning imaging sensor (LIS) on the tropical rainfall measuring mission (TRMM), in contrast, DeMaria et al. (2012), Stevenson et al. (2016), and Xu et al. (2017) revealed a negative correlation between the inner-core lightning and TC intensity change and suggested that lightning increase in the outer rainbands was an indicator of intensification. The inconsistency of these conclusions may be related to the differences in lightning types and detection efficiencies of the systems. Recent studies have shown that there are different lightning types in the inner core and outer rainbands. The outer rainbands is dominated by CG lightning (Griffin et al., 2014), while the inner core by IC lightning (Fierro et al., 2018). Until now, a clear relationship between lightning activity, especially inner-core lightning activity and TC intensity changes has not been established yet. Therefore, the combination of ground-based and space-based lightning measurements can provide a more complete depiction of lightning activity within TCs and establish representative relationships between lightning activity and TC intensity changes.

    The Fengyun-4A (FY-4A) Lightning Mapping Imager (LMI; Yang et al., 2017) provides unprecedented real-time and continuous observations for total lightning and convections of landing typhoons in China. Based on lightning observations from the FY-4A LMI, this study investigates the spatiotemporal characteristics of lightning activity and convective properties during the landfall period of Super Typhoon Mangkhut, the strongest landing typhoon in China in 2018. The study intends to answer the following questions: (1) What are the temporal and spatial distributions of total lightning observed by the LMI during the typhoon’s landfall? (2) What are the convective properties during different stages of lightning activity? (3) Is there any difference in lightning characteristics observed by the LMI and the ground-based lightning location network? The detailed analysis on this study would provide insight for the application of FY-4A LMI lightning data in the monitoring and tracking of convection in landing typhoons in China.

2.   FY-4A LMI
  • The geostationary satellite lightning imager is considered to be the most effective means of lightning detection, enabling continuous and real-time monitoring of strong convection in thunderstorms. So far, there are two geostationary satellite lightning imagers launched globally: the Geostationary Lightning Mapper (GLM) on the Geostationary Operational Environmental Satellites R-series (GOES-R) in the United States and the LMI on the new generation of geostationary satellite FY-4A in China. Launched in November 2016, the GLM aims to map total lightning activity continuously with storm scale spatial resolution in the Western Hemisphere. This will aid in monitoring, tracking, and forecasting severe storms and convective weather, and also improving the warning for tornadoes (Goodman et al., 2013). Since lightning now has been designated as an essential climate variable of the Global Climate Observing System (Aich et al., 2018), the GLM will extend measurements of lightning variability and establish a long-term and continuous global dataset of optical lightning (Rudlosky et al., 2019).

    China’s new generation geostationary meteorological satellite FY-4A was launched in December 2016. After nearly 1-yr on-orbit testing and evaluation, it was put into operation in September 2017. As the first satellite lightning imager in China (Huang, 2007), FY-4A LMI achieves the continuous observation of total lightning in stationary orbit in Asia and Oceania (Yang et al., 2017). With 777.4 nm as the detection wavelength, the LMI adopts a 1-nm bandwidth ultra-narrow band filter and dual lens splicing to implement large field of view. Using a 400 × 600 pixel CCD array detector, the imaging rate is set to 500 fps to minimize the pulse splitting in lightning. For the area of cloud top illumination corresponding to a typical thunderstorm cell, the spatial resolution of the sub-satellite point is set to 7.8 km. The on-board real-time event processor can complete multi-frame background removal and lightning event extraction of focal plane data by pixels within 2 min, and the upper limit of output of lightning events per frame is 120 (Liang et al., 2017). The designed detection efficiency of lightning events is greater than 90%, false alarm rate less than 10%, and location accuracy better than 1 pixel (Huang, 2007). Through real-time background evaluation (Hui et al., 2016), false signal filtering (Hui et al., 2015), and cluster analysis algorithms (Cao, 2016), the LMI generates lightning products of events, groups, and flashes, and gives lightning information on the occurrence time, location, and intensity (energy density). The observation field of the LMI covers China and its adjacent sea areas (Fig. 1). The continuous and real-time monitoring of the occurrence and development of lightning activity and deep convection by the LMI will provide important information for early warning of severe weather in China and its coastal areas.

    Figure 1.  Schematic diagram of the observation field of FY-4A LMI (Huang and Hui, 2014).

3.   Data and method
  • The lightning data used in this study are total lightning from the FY-4A LMI (http://satellite.nsmc.org.cn/) and lightning strokes from the WWLLN (http://wwlln.net/). The level 2 one-minute products from the LMI are used, which gives information on the occurrence time, duration, radiation center location, radiation intensity, and area of the event, group, and flash. An event is the basic detection unit obtained by the LMI. By comparing the background radiation to the radiation threshold, the pixels that exceed the threshold are extracted by the real-time event processor and are determined as an event. Through the clustering−filtering algorithms, events are clustered into products of the group and flash.

    The WWLLN was established by the University of Washington in 2004. At present, there are over 70 sensors around the world to detect global lightning activity (Hutchins et al., 2013). The network detects the VLF (3–30 kHz) electromagnetic radiation signal generated by lightning and uses the method of time of group arrival (TOGA) to determine the location. Each sensor obtains the precise time of the VLF signal from GPS, analyzes the measured waveform, and sends the arrival time to the central station in real-time. When at least five sensors detect the same VLF signal, the central station locates the stroke, and gives information on time, longitude, and latitude of the stroke (Rodger et al., 2004). WWLLN can detect both CG and IC lightning strokes as long as the peak current exceeds 30 kA (Jacobson et al., 2006). Since the peak current of CG lightning is usually larger than that of IC lightning, the detection efficiency of WWLLN for CG is higher than that for IC strokes (Abarca et al., 2010). In recent years, with the increase of sensor number and the upgrade of locating algorithms, the detection performance of WWLLN has been improved constantly. It is reported that the average location accuracy is about 5 km, and the detection efficiency is about 11% (Hutchins et al., 2013; Virts et al., 2013). In this study, lightning strokes detected by WWLLN are grouped into flashes with temporal and spatial thresholds of 0.5 s and 30 km, respectively (Fan et al., 2018).

    The regions of typhoon are determined according to the wind radius data from the Joint Typhoon Warning Center (JTWC). Combined with images from infrared satellite and ground-based Doppler radar, it is found that the radius of the storm is approximately equivalent to the radius of 34-kt winds and the radius of inner core to the radius of 64-kt winds. The region of outer rainbands is located between the radius of 50-kt winds and the radius of 34-kt winds. According to this definition, lightning activity within the storm, inner core and outer rainbands are determined. The radius of Mangkhut ranges from 426.2 to 556.0 km during its landfall period, which is consistent with TC size in previous studies (Corbosiero and Molinari, 2002, 2003; Abarca et al., 2011; DeMaria et al., 2012; Zhang et al., 2015; Xu et al., 2017; Fierro et al., 2018).

  • Super Typhoon Mangkhut (2018) formed in North-west Pacific at 1200 UTC 7 September and developed into a super typhoon at 0000 UTC 11 September. It entered the South China Sea after crossing Luzon Island of the Philippines in the early morning of 15 September and then moved northwestward. At 0900 UTC 16 September, it made landfall at Jiangmen in Guangdong Province, with the minimum central pressure of 960 hPa and the maximum wind speed of 42 m s−1. The storm gradually weakened after the landfall, and dissipated in Guangxi Region at 1200 UTC 17 September. Mangkhut was the strongest typhoon that landed in China in 2018, causing heavy rainfall in Guangdong, Guangxi, and Hainan.

    TC track and intensity data are from the best-track dataset of the Shanghai Typhoon Institute of China Meteorological Administration (http://tcdata.typhoon.org.cn/). This dataset provides with interval of 6 h (sometimes 3 h), the time (year, month, day, and hour), location, minimum central pressure (hPa), and maximum wind speed (m s−1) of the storm. The hourly intensity and location are obtained by using the cubic spline interpolation to estimate lightning relative to typhoon center. The analyzed period (totally 85 h) in this study is from 0000 UTC 14 September when Mangkhut entered the LMI observation field to 1200 UTC 17 September when the storm ceased. The best tracks and intensities of Mangkhut during the study period overlaid the LMI observation field are shown in Fig. 2.

    Figure 2.  Best tracks and intensities of Super Typhoon Mangkhut during the period of analysis (0000 UTC 14–1200 UTC 17 September 2018) and the observation field of LMI. Storm intensities are highlighted by color lines. The bold black line indicates the observational range of LMI. Shadings represent the topography (DEM; m).

  • The black body temperature (TBB) data are obtained from the Himawari-8, a new-generation geostationary meteorological satellite of Japan. Himawari-8 contains 16 channels, including 3 visible, 3 near infrared, and 10 infrared channels. The high temporal and spatial resolutions enable its capability for detecting convective weather system in small scale. The satellite operates in the region 20ºS–70ºN, 70º–160ºE, covering East Asia and the western Pacific. TBB data retrieved by the near-infrared channel at 0.05° spatial and 1 h temporal resolutions are used in this study.

  • The composite reflectivity (CR) data are from the new-generation Doppler weather radar network in South China. The network consists of 20 CINRAD-SA radars in Xiamen, Ganzhou, Yongzhou, Guangzhou, Shaoguan, Yangjiang, Meizhou, Zhanjiang, Shenzhen, Nanning, Liuzhou, Baise, Guilin, Shanwei, Heyuan, Wuzhou, Haikou, Longyan, Shantou, and Fuzhou. The volume scanning time for radar reflectivity is 6 min at an effective detection range of 460 km. The spatial resolution of the CR data is 0.01° × 0.01°.

4.   Characteristics of lightning activity observed by the LMI
  • The lightning activity of Mangkhut during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) is effectively observed by the LMI. Figure 3 presents the distribution of lightning density from the LMI event, group, and flash, and the spatial distribution of median TBBs observed by Himawari-8. The spatial distribution of lightning densities for the event (Fig. 3a), group (Fig. 3b), and flash (Fig. 3c) are consistent. Lightning activity is weak and mainly located in the outer rainbands before the typhoon landed in the Philippines. After entering the South China Sea, lightning activity in Mangkhut begins to increase. The maximum lightning density is located over the sea east of Hainan Island and the offshore area of southern Hainan and southeastern Guangdong provinces. The regions of high lightning density is consistent with the low-TBB regions observed by the satellite, and lightning mainly occurs in the areas where the median TBB is below −60°C (Fig. 3d). The TBBs increase sharply after the storm made landfall, causing the rapid decreases of lightning densities.

    Figure 3.  Distributions of lightning densities (shadings) from the LMI for (a) events, (b) groups, and (c) flashes, and (d) median TBBs (shading; °C) of the storm from Himawari-8 with the best tracks and intensities (denoted by color lines) during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut.

  • Figure 4 shows the hourly evolution of lightning rates within the inner core and outer rainbands observed by the LMI, overlaid with the storm intensity. Three stages of active lightning are identified during the landfall period. The first stage is from 1000 to 1800 UTC 14 September, during which Mangkhut was a super typhoon and strengthened gradually. Lightning activities (events, groups, and flashes) increase significantly with storm intensity. Previous studies (e.g., Zhao, 2019) have observed that the maximum echo reflectivity can reach 57 dBZ, and the cloud top can reach 17 km in a landing typhoon. The strong convection in the hot tower can lead to a rapid increase in lightning activity during the landfall. Mangkhut experiences an outbreak of inner-core lightning 4 h ahead of its maximum intensity (908 hPa, 1400 UTC 14 September). The indication of inner-core lightning outbreak on the TC enhancement is consistent with previous studies. Observations in the Atlantic Ocean showed that there is a positive correlation between the inner-core (or eyewall) lightning outbreak and hurricane intensification (Squires and Businger, 2008; Fierro et al., 2011; Stevenson et al., 2018). For a hurricane in stable or deepening state, an outbreak of lightning within the inner core may indicate its peak intensity or imminent end of intensification (Molinari et al., 1999). This phenomena was also observed in typhoons in Northwest Pacific. Lightning density within the inner core increases significantly in developing typhoons, and the outbreak of inner-core lightning is ahead of typhoon’s peak intensity for several hours (Pan et al., 2010; Pan and Qie, 2010; Zhang et al., 2012, 2015).

    The second stage is from 1800 UTC 15 to 0000 UTC 16 September when Mangkhut crossed the Philippines and entered the South China Sea. During this period, the storm weakened to a severe typhoon and remained in a stable state. Due to the interaction between the outer rainbands and the land, total lightning activity observed by the LMI turns active again, but is mainly concentrated in the outer rainbands. The third stage of active lightning is from 0600 UTC 16 September to 0000 UTC 17 September when lightning activity increased rapidly and reached its maximum rate. As in the second stage, lightning activity is still concentrated in the outer rainbands, and little lightning occurs within the inner core. The increase of lightning activity at this stage is due to the landfall of the super typhoon. The friction between the spiral rainbands and the land and the influence of environmental airflow strengthen convection in the rainbands, and thus enhance lightning activity there. The structure of outer rainbands of landing typhoons has convective precipitation property. The strong updrafts, high concentration of precipitation and ice particles in the mixed-phase region in the outer rainbands, cause stronger lightning activity than that in the inner core and inner rainbands (Zhang et al., 2013; Xu et al., 2016). After the third stage, Mangkhut weakened into a tropical depression, and lightning activity within the storm decreased rapidly. During the landfall period, the evolutions of lightning activity are consistent for events, groups, and flashes observed by the LMI, except for the values of lightning rates (Fig. 4).

    Figure 4.  Hourly evolutions of lightning rates (left-hand axis) from the LMI for (a) events, (b) groups, and (c) flashes within the inner core (solid bars) and outer rainbands (hollow bars) during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut overlaid with storm intensity and a curve for pressure (right-hand axis; hPa) on each panel. Note that lightning rates of the inner core are multiplied by 10. The arrow points to the time of landfall

    Lightning density for LMI flashes as a function of typhoon intensity and distance from the storm center is shown in Fig. 5. Taking 50 km as distance bin, the averaged lightning density of each radial ring is calculated as lightning density for that radial distance. When the storm is at super typhoon intensity, lightning within the inner core is frequent, while lightning density in the outer rainbands is much lower than that in the inner core. When the storm reaches its peak intensity, the first stage of active lightning activity occurs (1200–1800 UTC 14 September, shown in Fig. 4a), and the maximum lightning density locates within the eyewall region (radius about 50 km from the storm center). This lightning pattern during the period of maximum intensity is consistent with the radial distribution of lightning activity in Super Typhoon Haiyan (2013) (Zhang et al., 2019). The second stage of active lightning (from 1800 UTC 15 to 0000 UTC 16 September) is correspond to a stable state of the storm during the pre-landfall period. In this stage, lightning activity within the inner core ceases while lightning density in the outer rainbands gradually increased with the maximum density located at the radius of about 350 km (Fig. 5). The lightning density reaches its maximum at 6 h after the landfall, at the radius of 400–500 km, corresponding to the third stage of active lightning (from 0600 UTC 16 to 0000 UTC 17 September). Wang et al. (2017) studied the climatological characteristics of typhoon lightning over Northwest Pacific and found an asymmetrical distribution of lightning density in typhoons. Their study indicated that the maximum lightning density presents in southern quadrant of the outer rainbands, at the radius of 500–600 km from storm center.

    Figure 5.  Lightning density (shading) for the LMI flashes as a function of storm intensity and distance from the storm center (left-hand axis; km) with a bold black curve for pressure (right-hand axis; hPa) during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. The arrow points to the time of landfall.

    Three distinct regions of lightning activity are shown during the landfall period of Mangkhut, a significant lightning in the eyewall, little lightning in the inner rainbands, and a strong maximum in the outer rainbands (Fig. 5). This pattern is consistent with previous studies using ground-based lightning data (Molinari et al., 1994, 1999; Pan et al., 2010; Zhang et al., 2012; Wang et al., 2017) and satellite lightning data (Cecil and Zipser, 1999; Cecil et al., 2002; Lei et al., 2009; Wang et al., 2011). The different characteristics of lightning activity within the inner core, inner rainbands, and outer rainbands reflect the differences of convective structure, and dynamic and micro-physical processes in these regions. The inner-core structure is similar to weakly electrified deep convection in monsoon period. This type of convection contains a relatively low lightning rate, and the outbreak of lightning only occurs when strong convection presents (Molinari et al., 1994). The inner rainbands is associated with the fallout of frozen hydrometeors that ejected from the eyewall and often contains downdrafts and suppressed convection. Therefore, convection in the inner rainbands is weak, and lightning rarely occurs there (Molinari et al., 1994). Affected by the environmental airflow, the structure of outer rainbands is similar to continental deep convection during monsoon break periods. Due to strong updrafts and downdrafts and the simultaneous presence of liquid water, graupel, and ice particles above the melting level, there is a high lightning rate in the outer rainbands in TCs (Molinari et al., 1999).

  • The spatiotemporal distribution of lightning activity described above shows that there are three stages of active lightning during the landfall of Mangkhut. In this subsection, using TBB data from satellite and CR data from radar, the evolution of deep convection is investigated and the relationship between lightning activity and convective evolution is analyzed. Figure 6 shows the time variation of LMI flash rates within the inner core and outer rainbands, superimposed on median TBBs from Himawari-8 and the maximum sustained wind speed. The gray shadings indicate three stages of active lightning. The first two stages are associated with obvious decreases in TBB, indicating the enhanced convection in the storm. It is worth noting when the inner-core lightning outbreaks, the median TBB in the inner core reaches its minimum value (−80°C), suggesting the occurrence of deep convection there. The rapid increase of lightning rate within the inner core indicates the increase of ice particle content in this region. The presence of a large number of ice particles enhances the latent heat in the core and contributes to the further intensification of the storm (Guimond et al., 2010; Rogers et al., 2015). As the storm weakens and makes landfall, the median TBBs in the inner core and outer rainbands increase rapidly (from −55 to −30°C). Nevertheless, strong lightning activity still occurs in the outer rainbands, and the third stages of active lightning occurs. The friction between the storm and the land weakens the overall convection in the typhoon (Fig. 3d), but the occurrence and development of convective cells in the outer rainbands still produce strong lightning activity (Fig. 7d).

    Figure 6.  Time variation of LMI flash rates (left-hand axis) within the inner core (denoted by red bars) and outer rainbands (denoted by blue bars), superimposed with median TBBs (right-hand axis; red line for inner core and blue line for outer rainbands) and the maximum sustained wind speed (right-hand axis;black line) from Himawari-8 during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. The gray shadings indicate three stages of active lightning. The arrow points to the time of landfall.

    Figure 7.  TBB imageries (shading; °C) from Himawari-8 overlaid with the LMI (red dots) and WWLLN flashes (blue dots) within ± 30 min of the satellite observation time: (a) 1500 UTC 14, (b) 2300 UTC 15, (c) 1600 UTC 16, and (d) 2200 UTC 16 September during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. ◇ indicates the typhoon center.

    Figure 7 shows several infrared satellite imageries from Himawari-8 during the three stages of active lightning, overlaid with lightning flashes observed by the LMI and WWLLN. At 1500 UTC 14 September (Fig. 7a), during the first lightning stage, the storm was at super typhoon intensity and reached its maximum intensity. The structure of the storm is obvious: the typhoon eye is clearly visible, and the TBB values are relatively low; the eyewall is surrounded by a dense cloud area, and the outer area is surrounded by compact and thick spiral cloud. At this time, lightning activity is observed both within the inner core and the outer rainbands. The inner-core lightning exhibits in deep convective cloud with the TBBs below −80°C on the south side of the typhoon eye. Lightning in the outer rainbands occurs on the right front quadrant of the moving direction and concentrates in the convective region with low TBBs. Figure 7b shows lightning flashes at 2300 UTC 15 September during the second lightning stage. The storm is in a weakening process at this time, and the typhoon eye is observed to be filled. Lightning within the inner core ceases while clustered lightning in the outer rainbands still occurs. Lightning mainly presents in deep convection (the minimum TBB of −85°C) of the spiral rainbands on the southwest of the storm. Figures 7c and 7d show images of the third lightning stage at 1600 and 2200 UTC 16 September. During this stage, the storm has made landfall and began to gradually weaken. The storm eye is collapsed and the cloud around the storm center presents a distinct multiple-rainband structure. The land friction and dry–cold air greatly weaken the intensity of convection within the inner core (Pan et al., 2010), thus no lightning occurs within this region. However, lightning activity still occurs in the areas with low TBB in the outer rainbands, where convection is still active due to the water vapor transport from the sea. In addition, Fig. 7 shows that lightning locations detected by the LMI and WWLLN have a good correspondence.

    Based on radar observations from the South China Doppler radar network, Fig. 8 shows several CR imageries with lightning flashes observed by the LMI and WWLLN. At 5 h before the landfall (Fig. 8a), although strong CRs are observed in the spiral rainbands on the north side of the typhoon, no lightning activity occurs in the rainbands. Wen et al. (2015) found that during the landfall of Typhoon Morakot (2015), strong echo of the rainbands is observed in northwest of the typhoon center, but no lightning is observed. This may be attributed to the fact that the altitude of deep convection is low and the height of strong updraft fails to reach the mixed-phase region. The radar echoes in the eyewall strengthen significantly when the storm makes landfall (Fig. 8b), which is also observed in landing Typhoons Saomai (2006), Wipha (2007), and Krosa (2007) (Zhao et al., 2012). No lightning is observed within the eyewall in spite of the enhanced radar echoes at this time. Lightning activity only sporadically occurs in the southeastern spiral rainbands about 500 km from the storm center. After the landfall (Figs. 8c, d), the CRs in the eyewall and the outer rainbands weakens rapidly. Strong echoes only appear in the outer rainbands on the south and southeast sides of the center. Lightning is observed in strong convective areas with CRs > 40 dBZ. In addition, Figs. 8c and 8d show that the location and spatial pattern of lightning detected by the LMI and WWLLN are consistent. Both systems can effectively detect lightning activity in Mangkhut and track the evolution of deep convection during its landfall.

    Figure 8.  Composite reflectivity (CR; dBZ; shading) from South China radar network overlaid with the LMI (dark crosses) and WWLLN flashes (blue crosses) within ± 5 min of the radar observation time: (a) 0400 UTC, (b) 0900 UTC, (c) 1600 UTC, and (d) 2200 UTC 16 September during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut. ◇ indicates the typhoon center, and • indicates radar station.

    The evolution of radar echoes during the landfall described above indicates that the deep convection within the eyewall and the outer rainbands obviously weaken as the structure of the storm breaks and the intensity of the storm decreases. However, strong echoes and deep convections still appear on the south side of the storm, and lightning activity mainly occurs in the outer rainbands with high reflectivity. Previous studies have shown that formation and enhancement of meso- and small-scale convective systems in the outer rainbands are the main causes of heavy rainfall during typhoon’s landfall (Chen et al., 2017). Although in a weakening typhoon circulation, strong convective clouds develop and strengthen in the outer rainbands, and thus lead to an obvious increase of precipitation during the landfall (Dong et al., 2009). The content of ice particle in the spiral rainbands is an important factor affecting precipitation amount in landing typhoons (Franklin et al., 2005). As lightning is closely related to the increase in volume of ice particles, the enhancement of updrafts and the development of convections (Williams, 1988), combination of lightning data with radar observational data can be used to monitoring strong convection and heavy precipitation in landing typhoons.

5.   Comparisons between the LMI and WWLLN
  • The WWLLN is a ground-based global lightning detection network with the longest continuous observation period, and has provided lightning dataset for more than 10 yr. Although WWLLN mainly detects CG lightning and the detection efficiency still needs to be improved, this dataset has been widely used to study the spatiotemporal characteristics of lightning activity in TCs (Pan et al., 2010; Abarca et al., 2011; Bovalo et al., 2014; Ranalkar et al., 2017; Zhang et al., 2018). WWLLN has the advantage of long-distance detection and is capable of monitoring TC lightning in the whole life cycle. The dataset helps to determine TC intensity changes (Price et al., 2009; Thomas et al., 2010; DeMaria et al., 2012; Pan et al., 2014) and quantitatively indicate the distribution of strong convections in TCs (Stevenson et al., 2014; Susca-lopata et al., 2015; Solorzano et al., 2016). In this study, by using lightning data of Mangkhut from the WWLLN, events, groups and flashes observed by the LMI are compared with strokes and flashes observed by the WWLLN. The differences of spatiotemporal distribution of lightning between the two datasets are analyzed, and the ability of LMI for monitoring lightning activity in a landing typhoon is test.

    Figure 9 shows the radial distribution of lightning densities detected by the LMI and WWLLN during the landfall of Mangkhut. The radial pattern of lightning activity is observed by both systems. It is worth noting that the ratio of lightning density in the inner core to that in the outer rainbands observed by the LMI is larger than that observed by WWLLN. In other words, when same lightning density is observed in the outer rainbands by the two systems, more inner-core lightning is observed by the LMI, which may indicate higher proportion of IC flashes within the inner core in typhoon. Recent studies have revealed the differences in the ratio of IC to CG flashes between the inner core and outer rainbands in TCs. Griffin et al. (2014) suggested that in Tropical Storm Erin (2007), the ratio of IC to CG flashes within the inner core is higher than that in the outer rainbands. Using lightning observations from the GLM for Hurricane Maria (2017), Fierro et al. (2018) further pointed out that different lightning types exist between the inner core and the outer rainbands, and the inner core may have a larger ratio of IC to CG flashes.

    Figure 9.  Radial distribution of lightning densities detected by the LMI and WWLLN during the landfall period (0000 UTC 14–1200 UTC 17 September 2018) of Super Typhoon Mangkhut.

    The time variations of stroke and flash rates detected by WWLLN are shown in Fig. 10, overlaid with the storm intensity. By comparing the variation observed by the LMI in Fig. 4, the evolution characteristics of lightning rates observed by the two systems are consistent. The three stages of active lightning and the maximum lightning rate during the post-landfall observed by the LMI are also observed by WWLLN. In particular, the inner-core lightning outbreaks at the storm’s maximum intensity are both detected by the two systems. However, two differences of lightning activity have been noted. First, WWLLN observes increased lightning frequency from 0000 to 1200 UTC 14 September, but the frequency of lightning activity from the LMI is low. This may be related to the fact that the storm had just entered the LMI observation field and most of its area is outside the observation field. Second, from 0000 to 0600 UTC 15 September, the secondary peak of lightning rate in the inner core is detected by WWLLN but not observed by the LMI. This may be associated with the different lightning types detected by the two systems. The increased CG lightning within the inner core observed by WWLLN (Fig. 10; note that lightning rates of the inner-core enlarged by 10 times) is not shown in the time variation of lightning rates by the LMI (Fig. 4). Although the values of lightning rate observed by the two systems are different, there is a significant correlation of lightning rates between the two systems. The Pearson correlation coefficient between LMI group and WWLLN stroke is 0.75, and that between LMI flash and WWLLN flash is 0.78 (Fig. 11).

    Figure 10.  As in Fig. 4, but for WWLLN lightning rates of (a) strokes and (b) flashes.

    Figure 11.  Normalized scatterplots of the hourly lightning rates recorded by the LMI and WWLLN. (a) LMI group and WWLLN stroke; (b) LMI flash and WWLLN flash. The black line shows a 1 : 1 relationship, and the Pearson correlation coefficient (r) is indicated in the lower right cornner of each panel.

    During the landfall period of Mangkhut (85 h), 81,345 events, 23,818 groups, and 5571 flashes are observed by the LMI, while 30,918 strokes and 17,777 flashes are observed by WWLLN (Table 1). The number of flashes observed by the LMI is less than that observed by WWLLN. The event : group : flash ratio observed by the LMI for Mangkhut is 14 : 4 : 1. For the GLM, this ratio is 34 : 11 : 1 for Hurricane Maria (2017) (Fierro et al., 2018) and 42 : 16 : 1 (Rudlosky et al., 2019) for the first nine-month observation (December 2017–August 2018). In addition, it is found that for the whole typhoon and the outer rainbands, the number of events detected by the LMI is about three times the number of strokes detected by WWLLN, and the number of groups is close to that of strokes. However, for the inner core, the number of events is six times that of strokes, and the number of groups is twice that of strokes. The ratio of lightning number between the LMI and WWLLN within the inner core is higher than that in the outer rainbands and the whole typhoon region. This may indicate that Mangkhut’s inner core may has a higher ratio of IC to CG lightning number than the outer rainbands, which is consistent with the results in Fig. 9.

    AreaAveraged range (km)WWLLN LMI
    StrokeFlash EventGroupFlash
    Inner core0–10014611087930173
    Outer rainbands260–49030,33917,36879,62623,2075405
    Typhoon0–49030,91817,77781,34523,8185571
    Note: The landfall period is from 0000 UTC 14 to 1200 UTC 17 September 2018. The two datasets are selected within the same definition of storm ranges.

    Table 1.  Lightning numbers detected by the LMI and WWLLN during the landfall period of Super Typhoon Mangkhut (2018)

6.   Conclusions
  • Using lightning observations from the FY-4A LMI, best-track data from the Shanghai Typhoon Institute of the China Meteorological Administration, high-resolution TBB data from Himawari-8 satellite, and CR data from the South China radar network, this study investigates lightning characteristics and convective evolution of Super Typhoon Mangkhut, the strongest typhoon making landfall in China in 2018. The spatiotemporal distributions of lightning activity and the relationship between lightning and convective evolution are analyzed. In addition, using lightning data from the WWLLN, the differences of lightning patterns detected by the two systems are compared, and the ability of the LMI for monitoring lightning and convection in landing typhoons is tested.

    The lightning activity during the landfall period (85 h) of Mangkhut is effectively observed by the LMI. Lightning rate begins to increase when the storm enters the South China Sea. The maximum lightning density locates on the sea east of Hainan Island and in the offshores of southern Hainan and southeastern Guangdong provinces. The lightning density decreases rapidly after the typhoon makes landfall. The spatial distributions of lightning density of LMI events, groups and flashes are consistent, and the region with high lightning densities matches the region with low TBBs observed by the satellite.

    There are three stages of active lightning during the landfall period, and differences in the characteristic of lightning activity are observed between the inner core and the outer rainbands. Lightning activity within the inner core has different characteristics at different periods of typhoon development. At the developing period, the inner-core lightning increases rapidly and is mainly confined in the eyewall region. When the storm reaches its maximum intensity, the inner-core lightning outbreaks within 4 h ahead of the maximum intensity, showing the prediction of inner-core lightning to typhoon intensification. The inner-core lightning decreases rapidly after the landfall. For the outer rainbands, the lightning rate and lightning density increase rapidly 12 h before the landfall and reach their maximum after the landfall. Lightning density in the outer rainbands at the weakening stage after the landfall far exceeds that at the maximum intensity stage before the landfall. The radial distribution of lightning in Mangkhut is consistent with the results in previous studies, suggesting the different convective structure as well as dynamic and microphysical processes in different typhoon regions.

    A good correlation is shown between lightning rates from the LMI and the TBBs from the Himawari-8 satellite. The first two stages of active lightning correspond to the significant decreases of TBBs during the pre-landfall. The inner-core lightning outbreaks when the median TBB reaches the lowest value (−80°C), indicating the enhancement of deep convection in the inner core. The TBBs increase rapidly after the landfall, showing the weakening of convention and the storm. At this time, lightning mainly occurs in strong convective cells with low TBBs in the outer rainbands. The radar observations show that echoes in the eyewall and the outer rainbands gradually weaken after the landfall, and lightning is mainly confined in the area of high reflectivity. Lightning locations observed by the LMI have a good spatial correspondence with low-TBB areas observed by the Himawari-8 satellite and high-CR areas observed by the radar, which verifies the good ability of the LMI for monitoring and tracking lightning and strong convections in typhoons.

    The comparisons with WWLLN data show that the spatial distribution and temporal evolution of lightning activity during the landfall of Mangkhut observed by the two systems are consistent. A good correlation in lightning rates is shown for the two systems with correlation coefficient being 0.7. Both the LMI and WWLLN can observe variations of lightning activity and evolution of strong convection in landing typhoons in China. Due to the detection capability of total lightning, the LMI has advantages in revealing the characteristics of IC lightning within the inner core.

    The FY-4A LMI can effectively detect total lightning in thunderstorms in its observation field, and carry out continuous and real-time monitoring for the occurrence and development of convective systems. It will provide important information for predicting and early warning of severe weather in China and its coastal areas. The combination of lightning data from the LMI and other observation data (radar, visible, infrared, and microwave data) from meteorological satellites will improve the forecasting capability of existing tools and promote the generation of new research and applications in the future.

    Acknowledgments. The authors wish to thank the World Wide Lightning Location Network (http://wwlln.net) personnel for providing the lightning location data used in this study. The FY-4A LMI data are obtained from the National Satellite Meteorological Center, and the best-track data are from the Shanghai Typhoon Institute of the China Meteorological Administration. We also thank the three reviewers for their helpful comments on this manuscript.

Reference (70)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return