• Allen, J. T., and D. Karoly, 2014: A climatology of Australian severe thunderstorm environments 1979–2011: Inter-annual variability and the ENSO influence. Int. J. Climatol., 34, 8197, https://doi.org/10.1002/joc.3667.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, J. T., and E. R. Allen, 2016: A review of severe thunderstorms in Australia. Atmos. Res., 178–179, 347366, https://doi.org/10.1016/j.atmosres.2016.03.011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, J. T., A. B. Pezza, and M. T. Black, 2010: Explosive cyclogenesis: A global climatology comparing multiple reanalyses. J. Climate, 23, 64686484, https://doi.org/10.1175/2010JCLI3437.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, J. T., D. Karoly, and G. A. Mills, 2011: A severe thunderstorm climatology for Australia and associated thunderstorm environments. Aust. Meteor. Oceanogr. J., 61, 143158, https://doi.org/10.22499/2.6103.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Aumann, H. H., and A. Ruzmaikin, 2013: Frequency of deep convective clouds in the tropical zone from 10 years of AIRS data. Atmos. Chem. Phys., 13, 10 79510 806, https://doi.org/10.5194/acp-13-10795-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barnes, G., 2001: Severe local storms in the tropics. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 359–432.

    • Crossref
    • Export Citation
  • Bedka, K. M., 2011: Overshooting cloud top detections using MSG SEVIRI infrared brightness temperatures and their relationship to severe weather over Europe. Atmos. Res., 99, 175189, https://doi.org/10.1016/j.atmosres.2010.10.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., 2018: Multi-Function Transport Satellite (MTSAT) Japanese Advanced Meteorological Imager (JAMI) Overshooting Cloud Top Pixel Detection Database. NASA Langley Research Center, https://clouds.larc.nasa.gov/cgi-bin/site/showdoc?docid=262&c=dataset-description.

    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., and K. Khlopenkov, 2016: A probabilistic multispectral pattern recognition method for detection of overshooting cloud tops using passive satellite imager observations. J. Appl. Meteor. Climatol., 55, 19832005, https://doi.org/10.1175/JAMC-D-15-0249.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., J. Brunner, R. Dworak, W. Feltz, J. Otkin, and T. Greenwald, 2010: Objective satellite-based overshooting top detection using infrared window channel brightness temperature gradients. J. Appl. Meteor. Climatol., 49, 181202, https://doi.org/10.1175/2009JAMC2286.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., R. Dworak, J. Brunner, and W. Feltz, 2012: Validation of satellite-based objective overshooting cloud top detection methods using CloudSat Cloud Profiling Radar observations. J. Appl. Meteor. Climatol., 51, 18111822, https://doi.org/10.1175/JAMC-D-11-0131.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., C. Wang, R. Rogers, L. D. Carey, W. Feltz, and J. Kanak, 2015: Examining deep convective cloud evolution using total lightning, WSR-88D, and GOES-14 super rapid scan datasets. Wea. Forecasting, 30, 571590, https://doi.org/10.1175/WAF-D-14-00062.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bessho, K., and Coauthors, 2016: An introduction to Himawari-8/9— Japan’s new generation geostationary meteorological satellites. J. Meteor. Soc. Japan, 94, 151183, https://doi.org/10.2151/jmsj.2016-009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Blair, S. F., and Coauthors, 2017: High-resolution hail observations: Implications for NWS warning operations. Wea. Forecasting, 32, 11011119, https://doi.org/10.1175/WAF-D-16-0203.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cecil, D. J., and C. B. Blankenship, 2012: Toward a global climatology of severe hailstorms as estimated by satellite passive microwave imagers. J. Climate, 25, 687703, https://doi.org/10.1175/JCLI-D-11-00130.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cecil, D. J., D. E. Buechler, and R. J. Blakeslee, 2014: Gridded lightning climatology from TRMM-LIS and OTD: Dataset description. Atmos. Res., 135–136, 404414, https://doi.org/10.1016/j.atmosres.2012.06.028.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cecil, D. J., D. E. Buechler, and R. J. Blakeslee, 2015: TRMM LIS climatology of thunderstorm occurrence and conditional lightning flash rates. J. Climate, 28, 65366547, https://doi.org/10.1175/JCLI-D-15-0124.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chambers, C. R. S., G. B. Brassington, K. Walsh, and I. Simmonds, 2015: Sensitivity of the distribution of thunderstorms to sea surface temperatures in four Australian east coast lows. Meteor. Atmos. Phys., 127, 499517, https://doi.org/10.1007/s00703-015-0382-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cintineo, J. L., T. M. Smith, V. Lakshmanan, H. E. Brooks, and K. L. Ortega, 2012: An objective high-resolution hail climatology of the contiguous United States. Wea. Forecasting, 27, 12351248, https://doi.org/10.1175/WAF-D-11-00151.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Collis, S., A. Protat, P. T. May, and C. Williams, 2013: Statistics of storm updraft velocities from TWP-ICE including verification with profiling measurements. J. Appl. Meteor. Climatol., 52, 19091922, https://doi.org/10.1175/JAMC-D-12-0230.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dance, S., and R. Potts, 2002: Microburst detection using agent networks. J. Atmos. Oceanic Technol., 19, 646653, https://doi.org/10.1175/1520-0426(2002)019<0646:MDUAN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • de Laat, A., and Coauthors, 2017: Analysis of geostationary satellite derived cloud parameters associated with high ice water content environments. Atmos. Meas. Tech., 10, 13591371, https://doi.org/10.5194/amt-10-1359-2017.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dotzek, N., P. Groenemeijer, B. Feuerstein, and A. M. Holzer, 2009: Overview of ESSL's severe convective storms research using the European Severe Weather Data-base ESWD. Atmos. Res., 93, 575586, https://doi.org/10.1016/j.atmosres.2008.10.020.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dowdy, A. J., and Y. Kuleshov, 2014: Climatology of lightning activity in Australia: spatial and seasonal variability. Aust. Meteor. Oceanogr. J., 6, 914.

    • Search Google Scholar
    • Export Citation
  • Dworak, R., K. M. Bedka, J. Brunner, and W. Feltz, 2012: Comparison between GOES-12 overshooting top detections, WSR-88D radar reflectivity, and severe storm reports. Wea. Forecasting, 27, 684699, https://doi.org/10.1175/WAF-D-11-00070.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ferraro, R., J. Beauchamp, D. Cecil, and G. Heymsfield, 2015: A prototype hail detection algorithm and hail climatology developed with the Advanced Microwave Sounding Unit (AMSU). Atmos. Res., 163, 2435, https://doi.org/10.1016/j.atmosres.2014.08.010.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gensini, V. A., and W. S. Ashley, 2011: Climatology of potentially severe convective environments from North American Regional Reanalysis. Electron. J. Severe Storms Meteor, 6 (8), http://www.ejssm.org/ojs/index.php/ejssm/article/view/85/68.

    • Search Google Scholar
    • Export Citation
  • Goodman, S. J., and Coauthors, 2013: The GOES-R Geostationary Lightning Mapper (GLM). Atmos. Res., 125–126, 3449, https://doi.org/10.1016/j.atmosres.2013.01.006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gravelle, C. M., J. R. Mecikalski, W. E. Line, K. M. Bedka, R. A. Petersen, J. M. Sieglaff, G. T. Stano, and S. J. Goodman, 2016: Demonstration of a GOES-R satellite convective toolkit to “bridge the gap” between severe weather watches and warnings: An example from the 20 May 2013 Moore, Oklahoma, tornado outbreak. Bull. Amer. Meteor. Soc., 97, 6984, https://doi.org/10.1175/BAMS-D-14-00054.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Griffin, S., K. M. Bedka, and C. Velden, 2016: A method for calculating the height of overshooting convective cloud tops using satellite-based IR imager and CloudSat Cloud Profiling Radar observations. J. Appl. Meteor. Climatol., 55, 479491, https://doi.org/10.1175/JAMC-D-15-0170.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Holland, G. J., J. L. McBride, R. K. Smith, D. Jasper, and T. D. Keenan, 1986: The BMRC Australian Monsoon Experiment: AMEX. Bull. Amer. Meteor. Soc., 67, 14661472, https://doi.org/10.1175/1520-0477(1986)067<1466:TBAMEA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and K. I. Hodges, 2005: A new perspective on Southern Hemisphere storm tracks. J. Climate, 18, 41084129, https://doi.org/10.1175/JCLI3570.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Keenan, T. D., and R. E. Carbone, 2008: Propagation and diurnal evolution of warm season cloudiness in the Australian and Maritime Continent region. Mon. Wea. Rev., 136, 973994, https://doi.org/10.1175/2007MWR2152.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Keenan, T. D., and Coauthors, 2000: The Maritime Continent Thunderstorm Experiment (MCTEX): Overview and some results. Bull. Amer. Meteor. Soc., 81, 24332455, https://doi.org/10.1175/1520-0477(2000)081<2433:TMCTEM>2.3.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Knight, C., and N. Knight, 2001: Hailstorms. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 223–248.

    • Crossref
    • Export Citation
  • Kuleshov, Y., 2012: Thunderstorm and lightning climatology of Australia. Modern Climatology, S.-Y. Wang, Ed., InTech, https://doi.org/10.5772/35075.

    • Crossref
    • Export Citation
  • Kunkel, K. E., and Coauthors, 2013: Monitoring and understanding trends in extreme storms: State of knowledge. Bull. Amer. Meteor. Soc., 94, 499514, https://doi.org/10.1175/BAMS-D-11-00262.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Laflin, J. M., S. F. Blair, C. Gravelle, K. M. Bedka, B. Scarino, J. Apke, and C. Homeyer, 2016: Super rapid scan imagery analysis of two hailstorms sampled by HailSTONE. 28th Conf. on Severe and Local Storms, Portland, OR, Amer. Meteor. Soc., https://ams.confex.com/ams/28SLS/webprogram/Manuscript/Paper300948/extendedSLSabstract_JLaflin_HailSTONE_SRSOR.pdf.

  • Lay, E. H., A. R. Jacobson, R. H. Holzworth, C. J. Rodger, and R. L. Dowden, 2007: Local time variation in land/ocean lightning flash density as measured by the World Wide Lightning Location Network. J. Geophys. Res., 112, D13111, https://doi.org/10.1029/2006JD007944.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • May, P. T., and T. D. Keenan, 2005: Evaluation of microphysical retrievals from polarimetric radar with wind profiler data. J. Appl. Meteor., 44, 827838, https://doi.org/10.1175/JAM2230.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • May, P. T., C. N. Long, and A. Protat, 2012: The diurnal cycle of the boundary layer, convection, clouds, and surface radiation in a coastal monsoon environment (Darwin, Australia). J. Climate, 25, 53095326, https://doi.org/10.1175/JCLI-D-11-00538.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Climate, 16, 14561475, https://doi.org/10.1175/1520-0442-16.10.1456.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Peter, J. R., M. J. Manton, R. J. Potts, P. T. May, S. M. Collis, and L. Wilson, 2015: Radar-derived statistics of convective storms in southeast Queensland. J. Appl. Meteor. Climatol., 54, 19852008, https://doi.org/10.1175/JAMC-D-13-0347.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pope, M., C. Jakob, and M. J. Reeder, 2008: Convective systems of the north Australian monsoon. J. Climate, 21, 50915112, https://doi.org/10.1175/2008JCLI2304.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Potts, R. J., T. D. Keenan, and P. T. May, 2000: Radar characteristics of storms in the Sydney area. Mon. Wea. Rev., 128, 33083319, https://doi.org/10.1175/1520-0493(2000)128<3308:RCOSIT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Proud, S. R., 2014: Analysis of overshooting top detections by Meteosat Second Generation: A 5-year dataset. Quart. J. Roy. Meteor. Soc., 141, 909915, https://doi.org/10.1002/qj.2410.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Punge, H. J., and M. Kunz, 2016: Hail observations and hailstorm characteristics in Europe: A review. Atmos. Res., 176–177, 159184, https://doi.org/10.1016/j.atmosres.2016.02.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Punge, H. J., K. M. Bedka, M. Kunz, and A. Werner, 2014: A new physically based stochastic event catalog for hail in Europe. Nat. Hazards, 73, 16251645, https://doi.org/10.1007/s11069-014-1161-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Punge, H. J., K. M. Bedka, M. Kunz, and A. Reinbold, 2017: Hail frequency estimation across Europe based on a combination of overshooting top detections and the ERA-Interim reanalysis. Atmos. Res., 198, 3443, https://doi.org/10.1016/j.atmosres.2017.07.025.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Richter, H., J. Peter, and S. Collis, 2014: Analysis of a destructive wind storm on 16 November 2008 in Brisbane, Australia. Mon. Wea. Rev., 142, 30383060, https://doi.org/10.1175/MWR-D-13-00405.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rienecker, M., and Coauthors, 2011: MERRA: NASA’s Modern-Era Retrospective Analysis for Research and Applications. J. Climate, 24, 36243648, https://doi.org/10.1175/JCLI-D-11-00015.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Romps, D. M., and Z. Kuang, 2009: Overshooting convection in tropical cyclones. Geophys. Res. Lett., 36, L09804, https://doi.org/10.1029/2009GL037396.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schmit, T. J., M. M. Gunshor, W. P. Menzel, J. Li, S. Bachmeier, and J. J. Gurka, 2005: Introducing the next-generation Advanced Baseline Imager (ABI) on GOES-R. Bull. Amer. Meteor. Soc., 86, 10791096, https://doi.org/10.1175/BAMS-86-8-1079.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Slingo, A., K. I. Hodges, and G. J. Robinson, 2004: Simulation of the diurnal cycle in a climate model and its evaluation using data from Meteosat 7. Quart. J. Roy. Meteor. Soc., 130, 14491467, https://doi.org/10.1256/qj.03.165.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Soderholm, J. S., H. McGowan, H. Richter, K. Walsh, T. M. Weckwerth, and M. Coleman, 2017: An 18-year climatology of hailstorm trends and related drivers across southeast Queensland, Australia. Quart. J. Roy. Meteor. Soc., 143, 11231135, https://doi.org/10.1002/qj.2995.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thiery, W., E. L. Davin, S. I. Seneviratne, K. M. Bedka, S. Lhermitte, and N. van Lipzig, 2016: Hazardous thunderstorm intensification over Lake Victoria. Nat. Commun., 7, 12786, http://dx.doi.org/10.1038/ncomms12786.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thiery, W., L. Gudmundsson, K. M. Bedka, F. H. M. Semazzi, S. Lhermitte, P. Willems, N. P. M. van Lipzig, and S. I. Seneviratne, 2017: Early warnings of hazardous storms over Lake Victoria. Environ. Res. Lett., 12, 074012, https://doi.org/10.1088/1748-9326/aa7521.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Toracinta, E. R., D. J. Cecil, E. J. Zipser, and S. W. Nesbitt, 2002: Radar, passive microwave, and lightning characteristics of precipitating systems in the tropics. Mon. Wea. Rev., 130, 802824, https://doi.org/10.1175/1520-0493(2002)130<0802:RPMALC>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vaughan, G., K. Bower, C. Schiller, A. R. MacKenzie, T. Peter, H. Schlager, N. R. Harris, and P. T. May, 2008: SCOUT-O3/ACTIVE: High-altitude aircraft measurements around deep tropical convection. Bull. Amer. Meteor. Soc., 89, 647662, https://doi.org/10.1175/BAMS-89-5-647.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H. Holzworth, 2013: Highlights of a new ground-based, hourly global lightning climatology. Bull. Amer. Meteor. Soc., 94, 13811391, https://doi.org/10.1175/BAMS-D-12-00082.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Walsh, K., and Coauthors, 2016: Natural hazards in Australia: Storms, wind and hail. Climatic Change, 139, 5567, https://doi.org/10.1007/s10584-016-1737-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Webster, P. J., and R. A. Houze, 1991: The Equatorial Mesoscale Experiment (EMEX): An overview. Bull. Amer. Meteor. Soc., 72, 14811505, https://doi.org/10.1175/1520-0477(1991)072<1481:TEMEAO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Williams, E. R., S. A. Rutledge, S. G. Geotis, N. Renno, E. Rasmussen, and T. Rickenbach, 1992: A radar and electrical study of tropical “hot towers”. J. Atmos. Sci., 49, 13861395, https://doi.org/10.1175/1520-0469(1992)049<1386:ARAESO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yost, C. R., and Coauthors, 2018: A prototype method for diagnosing high ice water content probability using satellite imager data. Atmos. Meas. Tech., 11, 16151637, https://doi.org/10.5194/amt-11-1615-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Young, A. H., J. J. Bates, and J. A. Curry, 2012: Complementary use of passive and active remote sensing for detection of penetrating convection from CloudSat, CALIPSO, and Aqua MODIS. J. Geophys. Res., 117, D13205, https://doi.org/10.1029/2011JD016749.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zipser, E. J., and K. R. Lutz, 1994: The vertical profile of radar reflectivity of convective cells: A strong indicator of storm intensity and lightning probability? Mon. Wea. Rev., 122, 17511759, https://doi.org/10.1175/1520-0493(1994)122<1751:TVPORR>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zipser, E. J., J. C. Daniel, C. Liu, S. W. Nesbitt, and D. P. Yorty, 2006: Where are the most intense thunderstorms on Earth? Bull. Amer. Meteor. Soc., 87, 10571071, https://doi.org/10.1175/BAMS-87-8-1057.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • View in gallery

    Land surface elevation over Australia (m) from the ETOPO1 dataset. The locations of cities and regions discussed in the text and analyzed in Fig. 4b are noted.

  • View in gallery

    (a) A map of the mean number of OT pixel detections per year within each 0.25° grid box using all available MTSAT JAMI scans. City and region labels defined in Fig. 1 are also included to facilitate discussion in the main text. (b) The fraction of daily OT detections occurring from 1100 to 2259 LST. A value of 1 indicates that OTs always occur during daylight or early evening hours. (c) The yearly mean number of OT detections from 1100 to 2259 LST. (d) As in (c), but for 2300–1059 LST.

  • View in gallery

    Maps of OT pixel detection counts per year within each 0.50° grid box during each month of year: (top) July–October, (middle) November–February, and (bottom) March–June. Counts are divided by 4 to make the values and color table equivalent to the 0.25° maps shown in Fig. 2.

  • View in gallery

    (a) The fraction of daily OT pixel detections occurring during each hour over Australia (black line) and ocean (gray line) using the hourly OT detection dataset. (b) The mean number of OT pixel detections per year occurring within 2-h bins at 10 individual sites identified in Fig. 1. Sites over land (ocean) are colored in black (gray). The Darwin site data have been reduced by a factor of 4 and the Indian Ocean site data have been enhanced by a factor of 10 to fit within the range of the other sites. The annual mean number of detections per site is shown in the legend.

  • View in gallery

    Empirical cumulative distribution function of selected variables of ERA-Interim for locations of hail reports from the Severe Storm Archive (red solid line) and OTs (blue solid line). Blue areas mark filtered OTs. For reference, the distribution for hail reports > 5 cm is included (red dashed line). Shown are (a) CAPE, (b) 0–6-km bulk wind shear, (c) freezing-level height.

  • View in gallery

    Percentage of OTs eliminated by the filter for each ERA-Interim variable: (a) CAPE < 0 J kg−1, (b) 0–6-km DLS < 1.5 ms−1, and (c) freezing-level height < 4845 m.

  • View in gallery

    (a) OT pixels retained per year per 50 km2 after application of the hail environment filter and smoothed three times by a 75-km box filter. Values less than 0.1 are colored white. (b) OT pixels retained per year per 50 km2 after application of the median CAPE (736 J kg−1) and DLS value (15.3 ms−1) as a “likely” hail event filter and smoothed in the same way as in (a). Note that the color table has been scaled by a factor of 10 to account for the fact that a large fraction of the OT database was filtered using the median CAPE and DLS values. (c) Distribution of BoM Severe Storm Archive hail reports in Australia (1900–2014) and (d) population index both, on a 1° grid.

  • View in gallery

    Raw (blue) and hail-filtered (red) OT count east of 142°E as a function of latitude. Hail report count (orange) relates to the population index q (violet).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 66 66 15
PDF Downloads 49 49 14

A Long-Term Overshooting Convective Cloud-Top Detection Database over Australia Derived from MTSAT Japanese Advanced Meteorological Imager Observations

View More View Less
  • 1 Science Directorate, NASA Langley Research Center, Hampton, Virginia
  • 2 Department of Earth and Atmospheric Sciences, Central Michigan University, Mount Pleasant, Michigan
  • 3 Institute of Meteorology and Climate Research, Karlsruhe Institute of Technology, Karlsruhe, Germany
  • 4 Willis Re Australia, Sydney, New South Wales, Australia
© Get Permissions
Full access

ABSTRACT

A 10-yr geostationary (GEO) overshooting cloud-top (OT) detection database using Multifunction Transport Satellite (MTSAT) Japanese Advanced Meteorological Imager (JAMI) observations has been developed over the Australian region. GEO satellite imagers collect spatially and temporally detailed observations of deep convection, providing insight into the development and evolution of hazardous storms, particularly where surface observations of hazardous storms and deep convection are sparse and ground-based radar or lightning sensor networks are limited. Hazardous storms often produce one or more OTs that indicate the location of strong updrafts where weather hazards are typically concentrated, which can cause substantial impacts on the ground such as hail, damaging winds, tornadoes, and lightning and to aviation such as turbulence and in-flight icing. The 10-yr OT database produced using an automated OT detection algorithm is demonstrated for analysis of storm frequency, diurnally, spatially, and seasonally relative to known features such as the Australian monsoon, expected regions of hazardous storms along the southeastern coastal regions of southern Queensland and New South Wales, and the preferential extratropical cyclone track along the Indian Ocean and southern Australian coast. A filter based on atmospheric instability, deep-layer wind shear, and freezing level was used to identify OTs that could have produced hail. The filtered OT database is used to generate a hail frequency estimate that identifies a region extending from north of Brisbane to Sydney and the Goldfields–Esperance region of eastern Western Australia as the most hail-prone regions.

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Kristopher Bedka, kristopher.m.bedka@nasa.gov

ABSTRACT

A 10-yr geostationary (GEO) overshooting cloud-top (OT) detection database using Multifunction Transport Satellite (MTSAT) Japanese Advanced Meteorological Imager (JAMI) observations has been developed over the Australian region. GEO satellite imagers collect spatially and temporally detailed observations of deep convection, providing insight into the development and evolution of hazardous storms, particularly where surface observations of hazardous storms and deep convection are sparse and ground-based radar or lightning sensor networks are limited. Hazardous storms often produce one or more OTs that indicate the location of strong updrafts where weather hazards are typically concentrated, which can cause substantial impacts on the ground such as hail, damaging winds, tornadoes, and lightning and to aviation such as turbulence and in-flight icing. The 10-yr OT database produced using an automated OT detection algorithm is demonstrated for analysis of storm frequency, diurnally, spatially, and seasonally relative to known features such as the Australian monsoon, expected regions of hazardous storms along the southeastern coastal regions of southern Queensland and New South Wales, and the preferential extratropical cyclone track along the Indian Ocean and southern Australian coast. A filter based on atmospheric instability, deep-layer wind shear, and freezing level was used to identify OTs that could have produced hail. The filtered OT database is used to generate a hail frequency estimate that identifies a region extending from north of Brisbane to Sydney and the Goldfields–Esperance region of eastern Western Australia as the most hail-prone regions.

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Kristopher Bedka, kristopher.m.bedka@nasa.gov

1. Introduction and background

An understanding of the climatological distribution of hazardous thunderstorms over the Australian continent has been derived primarily from a variety of surface observational records [see Allen and Allen (2016), Walsh et al. (2016), and references therein]. However, a larger body of research using remote sensing has focused on nonsevere storms (those rarely producing hail, damaging winds, and tornadoes) within the northern tropical monsoonal region. Storms over this region have been studied using a combination of ground-, airborne-, and satellite-based instruments in a number field campaigns and at a fixed observation site in Darwin operated by the U.S. Department of Energy Atmospheric Radiation Measurement (DOE ARM) program, beginning in the 1980s (e.g., Holland et al. 1986; Webster and Houze 1991; Keenan et al. 2000; Vaughan et al. 2008; May et al. 2012, and references therein). Weather radar observations have also been periodically used to study convective storm characteristics, particularly those producing hail and damaging winds, near major cities such as Sydney and Brisbane (Potts et al. 2000; Dance and Potts 2002; Richter et al. 2014; Peter et al. 2015; Soderholm et al. 2017).

The Australian region lacks a climate-quality radar data network, requiring the use of other ground-based, reanalysis proxy, and spaceborne datasets to develop hazardous convective storm climatologies. Observations of thunderstorms from automated observing stations were used to construct a 10-yr thunder-day map, which agrees well with lightning flash detections from ground-based networks (Virts et al. 2013; Dowdy and Kuleshov 2014). Building on this understanding of thunderstorm occurrence, recent efforts have analyzed the frequency of hail- and tornado-producing storms using spotter observations (Allen and Allen 2016) and environments favorable to the development of such storms as a proxy for how frequently these storms could occur (Allen et al. 2011; Allen and Karoly 2014).

Geostationary (GEO) satellites provide frequent and spatially detailed observations critical for forecasting hazardous thunderstorms especially over regions without contiguous weather radar or lightning detection network coverage, as is the case over much of Australia and its offshore waters. Hazards such as damaging wind, hail, tornadoes, and lightning are concentrated near intense updrafts that typically produce an overshooting cloud-top (OT) signature in GEO imagery (Dworak et al. 2012; Bedka et al. 2015). OT updrafts also pose threats to aviation such as turbulence or in-flight icing (Bedka et al. 2010; de Laat et al. 2017; Yost et al. 2018). Rapid hazardous storm evolution and day-to-day differences in storm coverage make it difficult for weather forecasters and climate researchers to understand where and when these storms occur most frequently. Studies have shown that OT detections derived from long-term GEO imager infrared (IR) brightness temperature (BT) data records can be used to produce high-quality, spatially detailed deep convective storm frequency maps useful for climate analysis (Bedka et al. 2010, hereinafter B2010; Bedka 2011; Proud 2014; Punge et al. 2014; Thiery et al. 2016, 2017).

Space-based remote sensing observations have been used to better understand the Australian convection climatology. The primary benefits of these observations are the longevity of the satellite data record and detailed spatial sampling that can resolve individual updraft regions responsible for generating hazardous weather conditions. Spaceborne lightning sensor data were used to develop a 16-yr climatology based on Tropical Rainfall Measuring Mission (TRMM) Lightning Imaging Sensor data (LIS; Cecil et al. 2014, 2015), while TRMM and SSM/I passive microwave imager data were used to identify strong convection as a proxy for severe hail events (Cecil and Blankenship 2012; Ferraro et al. 2015). While these efforts have provided useful insight, hail proxies have been shown to overestimate the frequency of hail in tropical regions, suggesting a need to explore other satellite-derived metrics (Allen and Allen 2016). Through analysis of TRMM LIS, Microwave Imager (TMI), and Precipitation Radar (PR) data, Zipser et al. (2006) found that the most intense storms across the continent occur in the Northern Territory, the northern and east-central regions of Western Australia, and the eastern coast of Queensland and New South Wales. Convection over Australia has also been examined using IR BT thresholding, IR BT comparisons with tropopause temperature, and multispectral IR BT differences (Pope et al. 2008; Romps and Kuang 2009; Young et al. 2012; Aumann and Ruzmaikin 2013). Although these IR-based analyses are relatively consistent with those derived from lightning and microwave data, their approaches often identify a large region of nonovershooting anvil cloud useful for research goals such as analysis of mesoscale convective system areal coverage, but not necessarily useful for discriminating updraft regions where hazardous weather conditions typically occur (B2010; Bedka et al. 2012).

Although these diverse ground- and space-based datasets are extremely valuable, many of the observing systems used to construct them provide only a few observations per day, have limited spatial coverage, are not publically available (e.g., ground-based lightning detection), and/or do not have the range to sample storms over both land and ocean. The TRMM satellite orbited between 35°N and 35°S collecting ~4000 LIS observations at any point near the equator and ~13 000 observations in the subtropics throughout the TRMM lifetime (Cecil et al. 2015). The narrow LIS swath (600 km) did not enable repeated observations of storms throughout their lifetime. In contrast, over its 10+-yr operational lifetime, the Multifunction Transport Satellite (MTSAT) Japanese Advanced Meteorological Imager (JAMI) collected ~120 000 observations (32 images per day on average) of the Australian continent and offshore waters with a pixel spacing of ~4–7 km, comparable to that of the TRMM LIS. The improved image frequency and relatively high image detail enables new opportunities for understanding the regional and temporal distribution of storms over the continent.

This article describes a 10-yr OT detection database over Australia derived from MTSAT JAMI observations and reanalysis data. This short-term climatology is used to show the regional, seasonal, and diurnal distribution of tropopause-penetrating updrafts that generate distinct cold areas within cirrus anvils in JAMI IR imagery. To illustrate the utility of this climatology, we combine OT detections with atmospheric reanalyses to derive a hail frequency estimate. Hail can occur near strong updrafts that generate OTs if the ambient environment is supportive of hail formation. We analyze the environmental conditions present near historical hailstorms over Australia to filter OTs unlikely to have produced hail. The results of this study provide a unique perspective on the diurnal and seasonal evolution of hazardous storms with high spatial detail that complements previous analyses.

2. Data and method

a. MTSAT JAMI imagery database

The JAMI was flown aboard the MTSAT-1R and MTSAT-2 satellites. MTSAT-1R (140°E nadir position) was the operational meteorological imaging satellite from 28 June 2005 to 30 June 2010. MTSAT-2 (145°E nadir) was operational from 1 July 2010 to 4 December 2015. The study time period, from 1 July 2005 to 30 June 2015, encompasses almost the full MTSAT operational lifetime. The study domain (Fig. 1) is observed hourly by JAMI, but during the 0000, 0600, 1200, and 1800 UTC hours, the region is scanned 3 times at ~15-min intervals, providing a total of 32 images per day. Given that these four higher-frequency imaging periods are distributed evenly throughout the day and are aligned with maxima in storm activity over land during day and ocean during night, we do not expect that including these extra images biases the results. The JAMI 10.8-μm IR channel pixel size is 4 km at nadir and ~7 km along the southern edge of the domain. JAMI data were acquired from the University of Wisconsin–Madison Space Science and Engineering Data Center (UW–SSEC).

Fig. 1.
Fig. 1.

Land surface elevation over Australia (m) from the ETOPO1 dataset. The locations of cities and regions discussed in the text and analyzed in Fig. 4b are noted.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

b. Automated overshooting cloud-top detection

The B2010 OT detection algorithm assumes that OTs are small clusters of JAMI 10.8-μm IR channel pixels colder than the surrounding anvil cloud and tropopause. BT minima colder than 217.5 K and at least within 5 K of the NASA Modern-Era Retrospective Analysis for Research and Applications (MERRA; Rienecker et al. 2011) tropopause temperature are used to identify OT candidates. Pixels warmer than the tropopause were included to account for the fact that 4–7-km‐resolution JAMI imagery cannot resolve the coldest BT truly present in OTs, especially for those that do not fill an entire JAMI pixel. Requiring BTs colder than the tropopause would cause these small and/or weak OTs to be missed. An OT candidate ≥ 6.0 K colder than the anvil is classified as an OT, and surrounding pixels also colder than the local anvil mean are accumulated. A reduction of the threshold from 6.5 K in B2010 to 6.0 K in this study is justified based on a combination of feedback from the operational forecasting community (Gravelle et al. 2016) and practical experience with processing relatively coarse GEO imagery where OT signatures are not as prominent as they would appear in higher-resolution imagery captured by polar-orbiting satellite instruments such as MODIS, AVHRR, or VIIRS. The B2010 probability of OT detection ranges from 35% to 57% based on “truth” defined by human analyses of MODIS imagery and CloudSat Cloud Profiling Radar data (Bedka et al. 2012; Bedka and Khlopenkov 2016). The false detection rate ranges from 16% to 25%. Bedka and Khlopenkov (2016) discuss some of the challenges associated with IR-based OT detection and validation, providing context for these accuracy statistics.

The OT pixel database was assigned to a 0.25° grid, and the OT frequency per year was computed per grid box. OT pixels were corrected for parallax based on the OT cloud height (Griffin et al. 2016). OT detection times were converted to a local solar time (LST) by dividing the longitude by 15° and adding this offset to the JAMI image time. OTs from all JAMI scans were used to construct maps shown in Figs. 2 and 3, but only hourly OTs were used to analyze select locations throughout the domain (see Figs. 1 and 4). JAMI images with noise or bad scanlines were not processed, which amounted to <0.2% of all images. Overshooting cloud-top detection maps were compared with land surface elevation from the 1-minute gridded elevations/bathymetry for the world (ETOPO1) dataset. The Australia OT pixel database can be acquired online (Bedka 2018).

Fig. 2.
Fig. 2.

(a) A map of the mean number of OT pixel detections per year within each 0.25° grid box using all available MTSAT JAMI scans. City and region labels defined in Fig. 1 are also included to facilitate discussion in the main text. (b) The fraction of daily OT detections occurring from 1100 to 2259 LST. A value of 1 indicates that OTs always occur during daylight or early evening hours. (c) The yearly mean number of OT detections from 1100 to 2259 LST. (d) As in (c), but for 2300–1059 LST.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

Fig. 3.
Fig. 3.

Maps of OT pixel detection counts per year within each 0.50° grid box during each month of year: (top) July–October, (middle) November–February, and (bottom) March–June. Counts are divided by 4 to make the values and color table equivalent to the 0.25° maps shown in Fig. 2.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

Fig. 4.
Fig. 4.

(a) The fraction of daily OT pixel detections occurring during each hour over Australia (black line) and ocean (gray line) using the hourly OT detection dataset. (b) The mean number of OT pixel detections per year occurring within 2-h bins at 10 individual sites identified in Fig. 1. Sites over land (ocean) are colored in black (gray). The Darwin site data have been reduced by a factor of 4 and the Indian Ocean site data have been enhanced by a factor of 10 to fit within the range of the other sites. The annual mean number of detections per site is shown in the legend.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

c. Filtering OT detections based on hailstorm environments

Using hail report data from the European Severe Weather Database (ESWD; Dotzek et al. 2009), Bedka (2011) showed that the majority of hail-producing storms in Europe were collocated with OTs detected within Meteosat Second Generation (MSG) SEVIRI imagery. Despite the prevalence of OTs in hailstorms, the frequency of hail near any OT is difficult to derive because of the lack of complete and homogeneous observations of hail, even over the United States where a well-established hail report database exists (Kunkel et al. 2013). OTs cannot directly be used as a hail proxy on a global scale given that they are most frequent over the tropics where hail is very rare. This is also the case over Australia, where large hail is reported only very rarely in the Northern Territory, for example, despite the fact that the region is frequently impacted by OT-producing storms, as will be discussed later in this article (see Figs. 2a and 7c). The lack of hail events in tropical regions can be attributed to a generally unfavorable environment with weak wind shear and high freezing levels that inhibits storm organization or causes hail to melt into large raindrops before it reaches the ground (e.g., Barnes 2001; May and Keenan 2005). In addition, not all OT detections are correct. Cloud-top patterns can look like OTs, but in reality they are not; a recently decayed OT can remain cold and still resemble an OT as it advects laterally through the cirrus anvil, sometimes triggering detection (Punge et al. 2017).

An 11-yr database of MSG SEVIRI OT detections filtered by IR BT was used by Punge et al. (2014) to generate an initial hail frequency estimate over Europe. Use of this satellite-derived proxy for strong convective updrafts yields a spatially uniform map of regions where hail could have occurred over Europe and offshore waters, spanning national boundaries. A satellite-derived map offers distinct advantages for estimating hail frequency relative to hail reports from the general public that are strongly biased by population density and variations in technological capability to submit hail reports. Nevertheless, despite the fact that we do not know exactly where hail is occurring at any time, analysis of well-observed hailstorms indicates that OT minimum IR BT alone is insufficient for discriminating hail-producing storms (Laflin et al. 2016; Blair et al. 2017).

A recent improvement to the Punge et al. (2014) European hail frequency estimate was achieved by using a hail-specific filter of the OT database based on atmospheric variables defined by reanalysis in close proximity to historical European hail events (Punge et al. 2017). Thresholds for convective available potential energy (CAPE), 0–6-km deep-layer shear (DLS), and freezing-level height Z0°C indicating where hail is possible were defined and then applied to filter OT data, removing some errant OT detections and eliminating OTs where hail was extremely unlikely. CAPE was defined as a mixed-layer CAPE for parcels ascending from the lowest 30 hPa. The filter thresholds indicate the 2nd percentile (in the case of CAPE and DLS) or 98th percentile (Z0°C) values present in hailstorm environments. The filter required nonzero CAPE and DLS with a Z0°C ≤ 4340 m. The net result is a database of satellite-observed cold pixels with characteristics similar to OTs, present in unstable environments with at least some shear, which combine to increase the likelihood for hail. This simple filtering procedure eliminated 21% of the OTs and yielded a hail frequency estimate (expressed as number of OTs per year per 50 km2 area) across Europe and North Africa consistent with previous country-specific studies (Punge and Kunz 2016 and references therein). Another filter requiring CAPE and DLS to exceed the median value (613 J kg−1 and 11.9 ms−1, respectively) across all historical European hail events highlighted known hail maxima along and downwind of mountain ranges such as the Pyrenees, Massif Central, and Alps.

The Australian Bureau of Meteorology’s Severe Storm Archive contains reports of hail and other severe weather types. The geographical distribution of reports is strongly biased toward populated areas (Allen and Allen 2016; see their Figs. 7b and 7c), similar to the situation over Europe. Of the 880 hail reports in the 10-yr time period covered by the JAMI OT database, 246 (28.0%) were accompanied by an OT in a range of 50 km within 1 h of the report time and 387 (44.0%) were on the same day, allowing us to contrast the environments of OT- and hail-producing storms. These ranges were chosen to compensate for the relatively high timing uncertainty in the report data and the hourly temporal resolution of the JAMI data for 20 of the 24 h in a day over Australia. This 28% severe hail–OT agreement is quite a bit lower than the ~50% agreement over Europe (Bedka 2011) and the United States (Dworak et al. 2012). We attribute this discrepancy to a combination of 1) hail report timing uncertainty over Australia; 2) some differences in reporting standards in that especially small hail from weaker storms without a prominent OT signature may be included in the Australian database; and 3) coarse JAMI temporal sampling, namely, with 7.5–15-min GOES and SEVIRI imagery there are more opportunities for OT detection and thus a greater likelihood for match with a hail report. Many OTs truly present in nature are missed with hourly MTSAT data, which is a major limitation for using this data for hazardous storm analysis and nowcasting over the Southern Hemisphere, yet the sampling is far more frequent than that provided by TRMM. As an example of timing uncertainty, 13.2% of these hail reports had exactly 0000 UTC as the time of report, but only 0.1% and 0.7% of the reports occurred in the preceding and following hours. Allen et al. (2011) suggests that these 0000 UTC reports were either errantly assigned or assigned to 0000 UTC by default. The environmental conditions at 0600 UTC were determined by Allen et al. (2011) to be more representative than 0000 UTC for hailstorm events, so all reports with exactly a 0000 UTC time stamp were assigned to 0600 UTC.

We then applied the Punge et al. (2017) hail-filter method to the Australian OT dataset, using reports of hail over Australia to define filter thresholds. We expect the Australian filter thresholds to differ from those derived over Europe because a substantial fraction of the continent is in the tropics. Nevertheless, CAPE and DLS are common ingredients for hazardous storms (Allen and Karoly 2014), but a favorable environment alone does not guarantee these storms. This is the case when there is a strong capping inversion producing convective inhibition (CIN). This high CAPE, high CIN environment has been documented across southern Texas (Gensini and Ashley 2011) but is also likely to be present in other tropical and subtropical regions. Satellite-based OT detection indicates that CIN has been overcome and strong convective updrafts are present. These updrafts could generate hail in a favorable CAPE, DLS, and Z0°C environment. The 0600 UTC ERA-Interim data were used to construct the CAPE filter, which is the closest ERA-Interim time step to the peak land-based daytime temperature and convection activity over the region (Allen et al. 2011). The DLS and Z0°C filters were based on the closest ERA-Interim profile in time to the hail report. The filter threshold values for CAPE, DLS, and Z0°C are listed in Table 1.

Table 1.

Variables and thresholds used for the OT hail filter.

Table 1.

3. Results

a. Regional, diurnal, and seasonal OT distributions

The 0.25° gridded OT detection frequency map shows that hazardous storms were most frequent along the northern edge of the Northern Territory (NT), the Kimberley Coast, and the western coast of the Cape York Peninsula, with an average of over 200 OT pixels per year per 0.25° grid box (size of grid box omitted hereafter) over each of these regions (Fig. 2a). OT frequencies of greater than 100 pixels per year are quite common over land and ocean north of 20°S latitude. A regional OT maximum of up to 40 pixels per year is also present along the southeastern coast of the continent and offshore waters and along the northwestern coast of Western Australia (WA). Another local OT maximum (~20 pixels per year) is located over the Indian Ocean south of 40°S latitude.

OTs occur most frequently over land during the 1600–1700 LST time frame and over ocean with a broader peak at 0400–0500 LST (Fig. 4a). The timing and overall shape of these peaks generally agrees with previous studies (see Nesbitt and Zipser 2003). OT frequency increases over land at 1100 LST and dissipates after 2200 LST, a time frame we define as “day” because storms present during this time frame were a by-product of solar heating. The remaining 12 h are considered “night.” OT frequency at individual locations (Fig. 4b) shows that all major cities except Cairns have a 1600–1700 LST peak. Cairns and Darwin show slightly enhanced night activity, perhaps because they are located in the tropics where a conditionally unstable environment is more often present that could enable sporadic nighttime storm activity. The Timor Sea, Gulf of Carpentaria, and Indian Ocean (IO; see symbol in the lower left of Fig. 1 corresponding to the location of data shown in Fig. 4b) regions clearly show peaks during the night. The Indian Ocean OT peak and minimum precede the other two oceanic sites by ~4 h, reflecting the cooler troposphere and reduced convective inhibition associated with nocturnal intensification of extratropical cyclones. Regional variations of up to 5 h in the timing of peak lightning flash density have also been noted by Lay et al. (2007), so some variability in results is expected.

Maps of nighttime OT frequency (Fig. 2d) and daytime OT fraction (Fig. 2b) show very clearly the enhanced storm frequency over ocean and reduced frequency over or downstream of elevated topography at night. Approximately 70% of OTs were present over land during day, and a comparable fraction of OTs were present over ocean at night. A sharp gradient in nighttime OT frequency is present along coastlines, approaching 100 pixels per year over a 150-km distance in the most extreme case over the northeast corner of the NT. Two other distinct OT minima were present at night over the eastern half of the Cape York Peninsula and inland from the Kimberley Coast. These three regions are collocated with local land elevation maxima (up to ~500 m; Fig. 1) that have cooler and drier nocturnal boundary layers, making the environment unfavorable for nocturnal storm formation. In contrast, at lower elevations along the coast and inland, OT-producing storms were frequent (50 pixels per year) at night in several locations. Night activity is also clearly enhanced over the ocean east of New South Wales and southeastern Queensland associated with storms initiated over land that move out to sea or initiation via cyclone development or trough passage associated with the cooling midtroposphere during the nocturnal hours. In the winter months, the presence of east coast lows also contributes to this signal, typically producing intense convection during their development (see Fig. 3 and Chambers et al. 2015). Enhanced night OT activity is present along the southern edge of the domain, likely associated with convection along the typical track of intensifying extratropical cyclones in the region (Hoskins and Hodges 2005; Allen et al. 2010).

OT-producing storms were most common during day along elevation gradients and coastal regions (Fig. 2c). A sea-breeze circulation causes the maximum over the northwestern coast of WA. Sea breezes are well known for initiating convection over Australian tropical regions. (Keenan and Carbone 2008; Soderholm et al. 2017). The key terrain feature over the southeastern coast is the Great Dividing Range, which stretches from west of Brisbane, Sydney, and Melbourne, providing a localized source of initiation and forcing for OT-producing storms. OTs also often (25 pixels per year) occur along the southeastern coast at elevations typically below 100 m. One notable exception is the daytime maximum in the eastern region of WA where elevations of ~500 m are common. This region was also identified as a local thunder-day and lightning maximum by Kuleshov (2012) and is known to be associated with a locally high number of hail- and tornado-producing storms (Allen and Karoly 2014; Allen and Allen 2016).

The distribution of land OT detections agrees quite well with the thunder-day and lightning climatologies, but the agreement between OT and lightning frequency over the tropical ocean is poor. Williams et al. (1992) and Zipser and Lutz (1994) indicate that vertical velocities in oceanic cumulonimbus tend to be weaker than those over land. As a result of these weaker updrafts, supercooled liquid water, large ice particles, and ice–ice collisions may not be present in the mixed-phase region in sufficient concentrations to produce storm electrification (Zipser and Lutz 1994).

The B2010 OT–anvil mean BT difference (BTD) serves as a proxy for updraft strength to investigate differences between land and oceanic storms. Griffin et al. (2016) found that OTs cool at an average rate of 7.3 K km−1 as they ascend above the anvil based on comparisons of MODIS BTD and CloudSat radar profiles. GEO imager data are 4 times as coarse as MODIS and observe BTDs that are 2.4–3.9 K less than MODIS (Griffin et al. 2016). We found a mean OT BTD over land (ocean) of −11.05 K (−10.43 K) within the 10°–20°S and 122°–147°E domain where the number of overall OT detections are comparable for land and ocean regions. This indicates that the average land OT penetrates 0.13–0.18 km higher above the anvil than an oceanic OT using the 3.4–4.9 K km−1 GEO imager lapse rate range. The interquartile range over land (ocean) is from −7 to −12.8 K (from −6.9 to 12.2 K) showing that fewer storms with extreme updrafts are present over ocean.

These results provide some indication that oceanic storms have weaker updrafts near cloud top that could be correlated with weaker updrafts in the mixed-phase region where charge separation occurs. But these small mean cloud-top height differences would not explain the large differences in lightning activity and implied and measured vertical motions far beneath cloud top between tropical land and ocean (Keenan and Carbone 2008; Collis et al. 2013). Analysis of TRMM echo-top height distributions for tropical oceanic and land storms has indicated relatively small differences in 20-dBZ echo top (Toracinta et al. 2002). Tropical land and ocean storms extend to similar heights (and thus feature similar IR BT), with land features reaching extreme heights with slightly greater frequency than tropical oceanic features, consistent with our findings from MTSAT JAMI. For a given maximum 20-dBZ echo-top height, features over land typically have 30- and 40-dBZ echo-top heights that extend several kilometers higher than those in tropical oceanic systems, indicating differences in in-cloud microphysics that control storm electrification that would not be depicted in IR imagery (Toracinta et al. 2002). Thus, while especially cold and intense (inferred from BTD) OTs may produce enhanced lightning flash rates regardless of whether the storm is over land or ocean, there is insufficient information in IR imagery alone to reliably infer lightning flash rates within any particular updraft.

Monthly OT analyses show differences in the distribution of hazardous convection throughout the year (Fig. 3). OTs are almost never found over the northern third of continental land from May to September during the dry season. Storms then begin to develop in October over the NT and are significantly more frequent over land relative to the offshore ocean from October to November, prior to the summer monsoonal period. OT activity peaks in this region over both land and ocean in December and January associated with the monsoon period and frequent tropical cyclones. Storms with OTs can occur in almost any month over the southeastern continent and nearby offshore regions with activity peaking over land in the November–December time frame, reflecting the peak season of both ordinary convection (Dowdy and Kuleshov 2014) and severe convection (Allen and Karoly 2014). Storms can occur throughout the year over ocean, aided by the warm water transported southward by the East Australian Current and periodic extratropical disturbances such as transitioning cyclones or east coast lows (Chambers et al. 2015).

OT activity peaks along the southern edge of the domain during the winter months of June–August. This coincides with the presence of frequent midlatitude cyclones and frontal systems that tend to produce convection more often during the night based on Fig. 4b (Allen et al. 2010). Proud (2014) shows that B2010 OT errors can occur within complex cirrus patterns that “look like” OTs from a computer algorithm’s perspective. While this may be occurring to some extent, the facts that 1) OT detections are most frequent during night similar to other oceanic regions, 2) Virts et al. (2013) show an area of enhanced lightning frequency in this region, and 3) this region is associated with a local maxima in the tracks of storms undergoing explosive cyclogenesis that often produces deep convection (Hoskins and Hodges 2005; Allen et al. 2010) combine to suggest that the detections found over the Indian Ocean region are reasonable.

b. Hail frequency estimation

We apply the filters listed in Table 1 to the JAMI OT database to derive a hail frequency estimate based on the satellite-derived climatology. The cumulative distribution functions for the respective variables given in Fig. 5 and maps in Fig. 6 show the fractions of OTs filtered by these variables per grid box (0.3° × 0.5°). The CAPE distribution for OT events is quite similar to hail events that include small hail (Fig. 5a). The median CAPE present near storms that produced large hail (>5 cm; 1026 J kg−1) is about 40% greater than the entire hailstorm population. OTs occur in much lower DLS environments than hailstorms, which is dominated by the signal from storms over the tropics where shear is much weaker on average than the heavily populated regions in the subtropical southeast region where most of the hail was reported. OTs occurred in much higher freezing-level environments than hailstorms, again dominated by storms over the tropics where the atmospheric column is warmer than in the southeast. In addition to the high freezing level over the tropics, hailstorms are also infrequent because the CAPE is often distributed over a much greater depth (i.e., skinny CAPE) than what is typical in midlatitude environments. This results in weaker updrafts and a reduced likelihood for large hail.

Fig. 5.
Fig. 5.

Empirical cumulative distribution function of selected variables of ERA-Interim for locations of hail reports from the Severe Storm Archive (red solid line) and OTs (blue solid line). Blue areas mark filtered OTs. For reference, the distribution for hail reports > 5 cm is included (red dashed line). Shown are (a) CAPE, (b) 0–6-km bulk wind shear, (c) freezing-level height.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

Fig. 6.
Fig. 6.

Percentage of OTs eliminated by the filter for each ERA-Interim variable: (a) CAPE < 0 J kg−1, (b) 0–6-km DLS < 1.5 ms−1, and (c) freezing-level height < 4845 m.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

The overall contributions by each filter to the total vary between 1.4% (CAPE) and 84.8% (Z0°C). The filter shows that hail events occur in an environment with some mixed-layer CAPE, but only 1.4% of the OTs were in a stable environment. The fraction locally exceeds 50% in the southern half of the continent (Fig. 6a). The DLS threshold value is also very low at 1.5 ms−1, meaning that hail occurs frequently in low-shear environments. DLS is more often a limiting factor for hailstorm formation in the tropics than other regions (Knight and Knight 2001). The filtered fraction is quite homogeneous across the northern half of the continent and reaches up to 60% in the tropical seas (Fig. 6b). As would be expected, the Z0°C filter also predominantly impacts tropical regions where >90% of the OTs are filtered (Fig. 6c). The Z0°C threshold is about 500 m higher than that over Europe (Punge et al. 2017) likely because of the warmer surface temperatures in hailstorm environments over Australia. It is important to note that only 1.7% of the hail reports in Australia were north of 20°S, so either hail reaching the ground is uncommon in this region or it is typically unreported because of extreme surface temperatures and a sparse population, consistent with other tropical regions (Barnes 2001). Hail reports in this region typically occur during the winter months when synoptic-scale forcing and approaching upper-level troughs provide adequate deep-layer shear relative to local instability to generate hail-producing supercells (Allen and Allen 2016).

The distribution of the OTs after filtering is shown in Fig. 7a. The maximum along the southeast coast is plausible given that meteorological conditions are often favorable for severe thunderstorms (Allen and Karoly 2014), and hail reports in the BoM Severe Storm Archive are most common in this region (Fig. 7b). Nearly two hail events per year are estimated to occur south of Brisbane. The Darwin region, Kimberley Coast, the southern edge of the Gulf of Carpentaria, and the Goldfields–Esperance region of eastern WA are also considered hail prone, as is the Indian Ocean region along the southern edge of the domain. The AMSR-E hail climatology from Cecil and Blankenship (2012) highlights many of the same areas with the exception of the Indian Ocean region, despite the fact that AMSR-E only observes at 0130 and 1330 LST, which misses peaks in convection activity that are captured by JAMI observations.

Fig. 7.
Fig. 7.

(a) OT pixels retained per year per 50 km2 after application of the hail environment filter and smoothed three times by a 75-km box filter. Values less than 0.1 are colored white. (b) OT pixels retained per year per 50 km2 after application of the median CAPE (736 J kg−1) and DLS value (15.3 ms−1) as a “likely” hail event filter and smoothed in the same way as in (a). Note that the color table has been scaled by a factor of 10 to account for the fact that a large fraction of the OT database was filtered using the median CAPE and DLS values. (c) Distribution of BoM Severe Storm Archive hail reports in Australia (1900–2014) and (d) population index both, on a 1° grid.

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

When the OTs are filtered using median CAPE and DLS (736 J kg−1 and 15.3 ms−1) for hail reports instead of the minimum thresholds from Table 1, indicating “likely” hail events, the maxima along the east coast and WA are preserved while the tropical and Indian Ocean regions highlighted in Fig. 7a are mostly excluded. A region near Canberra and Melbourne also is deemphasized based on this median environment filter. This may point to a limitation of the OT detection method; in regions with lower CAPE—because of altitude and orography in Canberra and lesser low-level moisture in Melbourne—hailstorms would produce lower, warmer cloud tops, which would make detection less likely and cause an underestimation of hail frequency. The Goldfields–Esperance, WA, maximum may seem questionable given the lack of hail reports in the Severe Storm Archive. The fact that a significant number of OTs were present in environments exceeding the median CAPE and DLS indicates that this region definitely has storms that could produce large hail. It also seems to be reasonable given that the lightning frequency there is similar to New South Wales and seems to be an aggregate of convection in both the cool and warm seasons. However, some uncertainty exists because the region is extremely remote and warm summertime lower-tropospheric temperatures may lead to excessive melting, thus only a few of the hailstones that do form reach the surface. More observation and study of this region is encouraged to validate the Fig. 7 results.

The hail report counts are strongly biased toward populated regions (Fig. 7c). Taking the number of towns with more than 1000 inhabitants from the GeoNames database (p; http://www.geonames.org/), we attributed a weight w = max[1; log (p/1000)] to reflect the fact that hailstorms that impact cities are more likely to be reported than those that impact less populated areas and combined these into a population index defined by q = pw. This illustrates the spatial distribution of population likely to influence hail reporting (Fig. 7d). To compare the longitudinal performance of the hail-filtered OT count with reports, the population p of 405 towns and cities east of 142°E with more than 1000 inhabitants was taken and condensed to a longitudinal cross section (Fig. 8). As would be expected, the peaks of hail reporting coincide with the higher population regions, particularly Melbourne, Sydney, and Brisbane. However, this signal is noticeably reduced for Brisbane, which suggests a clear peak remains in the Brisbane area (27°–28°S). Comparing reports with the OT filtered hail frequency data, there is good agreement regarding the distribution of the two datasets. Based on the OT data, the zone of maximum hail frequency extends from about 34° to 23°S, which is considerably farther equatorward than in North America (32°–43°N; Cintineo et al. 2012) and Europe (36°–53°N; Punge et al. 2017).

Fig. 8.
Fig. 8.

Raw (blue) and hail-filtered (red) OT count east of 142°E as a function of latitude. Hail report count (orange) relates to the population index q (violet).

Citation: Journal of Applied Meteorology and Climatology 57, 4; 10.1175/JAMC-D-17-0056.1

Because hail reporting is deemed more complete in larger cities than elsewhere, we compare the hail-filtered OT count as given in Fig. 7 with the number of hail reports and days with hail reports for major cities of Australia (Table 2). For the latter, a radius of 10 km around the city center was chosen as the reference area. While the OT hail index is of the same magnitude as the report counts, some deviations are apparent. Most notably, reports are more frequent in Brisbane and Perth than estimated by OT proxy, and reports are lacking in Canberra, Hobart, and Darwin. These deviations can be explained by local anomalies in hail frequency not depicted in the OT database, perhaps due to relatively coarse JAMI temporal resolution (typically 1 h), and deficiencies in the OT detection approach such as the failure to detect in warmer (>217.5 K), low-topped storms that are relatively common in these regions (Allen and Allen 2016).

Table 2.

For major cities of Australia, annual mean OT-based hail events at the nearest grid point to the city, annual mean hail report count (1980–2009), and annual mean hail report days (1980–2009) within 10 km from the city center.

Table 2.

4. Summary

This article describes a unique 10-yr MTSAT JAMI OT detection database over the Australia region, providing insight into the distribution of hazardous convection beyond the existing climatologies derived through storm environment proxy or ground-based observations. The results show OT distributions over land that generally agree with previous analyses of hazardous storm activity. OT frequency is greatest over land north of 20°S latitude, which continues to raise the question as to whether storms produce severe hail or damaging winds along the northern margins of the continent, an area with scant observations (Allen and Allen 2016). A distinct diurnal OT variation between land and ocean was present, with ~70% of storms occurring over land during day and a comparable percentage occurring over ocean at night. The nocturnal maximum over ocean found here is consistent with previous studies (Slingo et al. 2004; May et al. 2012). The high spatial resolution of the database revealed interesting details such as the impact of land surface elevation and elevation gradients on OT-producing storm frequency. OTs were detected more frequently over the tropical ocean (north of 20°S) at night than would be inferred from lightning-based storm analyses. Updrafts near cloud top were inferred to be slightly stronger in tropical land-based storms than nearby oceanic storms, which we assume to be correlated with stronger updrafts within the mixed-phase region where electrification typically occurs. But there is insufficient information in IR imagery alone to infer lightning flash rates because electrification is dictated by a combination of dynamical and in-cloud microphysical processes not observable in the IR.

A filter applied to the OT database requiring at least some CAPE and DLS, and a Z0°C threshold derived based on historical hail events over Australia from the BoM Severe Storm Archive, provides a reasonable estimate of hail frequency. Despite having much lower overall OT detection counts than tropical regions, the southeast coastal region of the continent has the highest estimated hail frequency, which agrees quite well with the BoM archive. Use of relatively modest median hailstorm CAPE and DLS values to indicate “likely” hail events identified the southeast coast and portions of WA as the most hail-prone regions. In contrast with other hazard indicators, this approach is based on spatially homogeneous data sources, with the addition of ground truth data in the form of hail reports, and hence is less prone to biases due to sensor calibration errors (e.g., radar) or exposure and vulnerability (e.g., insured loss data). Therefore, this provides an avenue for application as a reference for hail frequency over Australia in the insurance market.

Advances in OT detection capability (Bedka and Khlopenkov 2016) coupled with more frequent and detailed data provided by new GEO imagers such as the GOES-16 Advanced Baseline Imager (Schmit et al. 2005) and Himawari-8/-9 Advanced Himawari Imager (AHI; Bessho et al. 2016) will increase the quality and utility of OTs in future weather and climate applications. Initial comparisons between coincident MTSAT JAMI and AHI OT observations show AHI IR BTs that are 3–6 K colder than JAMI, with the greatest differences occurring in small OTs that do not fill a JAMI pixel. The two sensors observed anvil IR BTs within 1–2 K of each other. OTs are therefore more prominent relative to the anvil in AHI IR imagery. Hail-producing storms evolve quite rapidly, and the typically hourly JAMI imaging frequency is insufficient for resolving sharp OT temperature fluctuations that often occur within minutes of hail fall (Bedka et al. 2015). The AHI collects observations of Australia at 10-min intervals, so this improved sampling coupled with enhanced OT prominence will improve automated OT detection capability. AHI OT datasets could provide critical real-time hazardous thunderstorm analyses over the coming decades along with complementary climatologies to supplement limited surface observational records and relatively sparsely distributed radar sites (Allen and Allen 2016; Walsh et al. 2016). An assessment of wind or tornado risk over Australia could be derived through filtering using a similar process to the hail-based approach described in this study. GEO OT detections will also provide a valuable complement to the more frequent but spatially coarser GEO lightning observations from instruments such as the GOES-16 Geostationary Lightning Mapper (Goodman et al. 2013) for recognizing potentially hazardous updraft regions, especially for aviation interests over ocean.

Acknowledgments

Generation of the MTSAT JAMI Australia region OT database was funded by Willis Limited via the NASA Space Act Agreement (UK-0533-0). Analysis of the OT database by K. Bedka was supported by the GOES-R Risk Reduction Research program. Author Punge is supported by the Willis Research Network.

REFERENCES

  • Allen, J. T., and D. Karoly, 2014: A climatology of Australian severe thunderstorm environments 1979–2011: Inter-annual variability and the ENSO influence. Int. J. Climatol., 34, 8197, https://doi.org/10.1002/joc.3667.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, J. T., and E. R. Allen, 2016: A review of severe thunderstorms in Australia. Atmos. Res., 178–179, 347366, https://doi.org/10.1016/j.atmosres.2016.03.011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, J. T., A. B. Pezza, and M. T. Black, 2010: Explosive cyclogenesis: A global climatology comparing multiple reanalyses. J. Climate, 23, 64686484, https://doi.org/10.1175/2010JCLI3437.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, J. T., D. Karoly, and G. A. Mills, 2011: A severe thunderstorm climatology for Australia and associated thunderstorm environments. Aust. Meteor. Oceanogr. J., 61, 143158, https://doi.org/10.22499/2.6103.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Aumann, H. H., and A. Ruzmaikin, 2013: Frequency of deep convective clouds in the tropical zone from 10 years of AIRS data. Atmos. Chem. Phys., 13, 10 79510 806, https://doi.org/10.5194/acp-13-10795-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Barnes, G., 2001: Severe local storms in the tropics. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 359–432.

    • Crossref
    • Export Citation
  • Bedka, K. M., 2011: Overshooting cloud top detections using MSG SEVIRI infrared brightness temperatures and their relationship to severe weather over Europe. Atmos. Res., 99, 175189, https://doi.org/10.1016/j.atmosres.2010.10.001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., 2018: Multi-Function Transport Satellite (MTSAT) Japanese Advanced Meteorological Imager (JAMI) Overshooting Cloud Top Pixel Detection Database. NASA Langley Research Center, https://clouds.larc.nasa.gov/cgi-bin/site/showdoc?docid=262&c=dataset-description.

    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., and K. Khlopenkov, 2016: A probabilistic multispectral pattern recognition method for detection of overshooting cloud tops using passive satellite imager observations. J. Appl. Meteor. Climatol., 55, 19832005, https://doi.org/10.1175/JAMC-D-15-0249.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., J. Brunner, R. Dworak, W. Feltz, J. Otkin, and T. Greenwald, 2010: Objective satellite-based overshooting top detection using infrared window channel brightness temperature gradients. J. Appl. Meteor. Climatol., 49, 181202, https://doi.org/10.1175/2009JAMC2286.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., R. Dworak, J. Brunner, and W. Feltz, 2012: Validation of satellite-based objective overshooting cloud top detection methods using CloudSat Cloud Profiling Radar observations. J. Appl. Meteor. Climatol., 51, 18111822, https://doi.org/10.1175/JAMC-D-11-0131.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bedka, K. M., C. Wang, R. Rogers, L. D. Carey, W. Feltz, and J. Kanak, 2015: Examining deep convective cloud evolution using total lightning, WSR-88D, and GOES-14 super rapid scan datasets. Wea. Forecasting, 30, 571590, https://doi.org/10.1175/WAF-D-14-00062.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bessho, K., and Coauthors, 2016: An introduction to Himawari-8/9— Japan’s new generation geostationary meteorological satellites. J. Meteor. Soc. Japan, 94, 151183, https://doi.org/10.2151/jmsj.2016-009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Blair, S. F., and Coauthors, 2017: High-resolution hail observations: Implications for NWS warning operations. Wea. Forecasting, 32, 11011119, https://doi.org/10.1175/WAF-D-16-0203.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cecil, D. J., and C. B. Blankenship, 2012: Toward a global climatology of severe hailstorms as estimated by satellite passive microwave imagers. J. Climate, 25, 687703, https://doi.org/10.1175/JCLI-D-11-00130.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cecil, D. J., D. E. Buechler, and R. J. Blakeslee, 2014: Gridded lightning climatology from TRMM-LIS and OTD: Dataset description. Atmos. Res., 135–136, 404414, https://doi.org/10.1016/j.atmosres.2012.06.028.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cecil, D. J., D. E. Buechler, and R. J. Blakeslee, 2015: TRMM LIS climatology of thunderstorm occurrence and conditional lightning flash rates. J. Climate, 28, 65366547, https://doi.org/10.1175/JCLI-D-15-0124.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chambers, C. R. S., G. B. Brassington, K. Walsh, and I. Simmonds, 2015: Sensitivity of the distribution of thunderstorms to sea surface temperatures in four Australian east coast lows. Meteor. Atmos. Phys., 127, 499517, https://doi.org/10.1007/s00703-015-0382-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cintineo, J. L., T. M. Smith, V. Lakshmanan, H. E. Brooks, and K. L. Ortega, 2012: An objective high-resolution hail climatology of the contiguous United States. Wea. Forecasting, 27, 12351248, https://doi.org/10.1175/WAF-D-11-00151.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Collis, S., A. Protat, P. T. May, and C. Williams, 2013: Statistics of storm updraft velocities from TWP-ICE including verification with profiling measurements. J. Appl. Meteor. Climatol., 52, 19091922, https://doi.org/10.1175/JAMC-D-12-0230.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dance, S., and R. Potts, 2002: Microburst detection using agent networks. J. Atmos. Oceanic Technol., 19, 646653, https://doi.org/10.1175/1520-0426(2002)019<0646:MDUAN>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • de Laat, A., and Coauthors, 2017: Analysis of geostationary satellite derived cloud parameters associated with high ice water content environments. Atmos. Meas. Tech., 10, 13591371, https://doi.org/10.5194/amt-10-1359-2017.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dotzek, N., P. Groenemeijer, B. Feuerstein, and A. M. Holzer, 2009: Overview of ESSL's severe convective storms research using the European Severe Weather Data-base ESWD. Atmos. Res., 93, 575586, https://doi.org/10.1016/j.atmosres.2008.10.020.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dowdy, A. J., and Y. Kuleshov, 2014: Climatology of lightning activity in Australia: spatial and seasonal variability. Aust. Meteor. Oceanogr. J., 6, 914.

    • Search Google Scholar
    • Export Citation
  • Dworak, R., K. M. Bedka, J. Brunner, and W. Feltz, 2012: Comparison between GOES-12 overshooting top detections, WSR-88D radar reflectivity, and severe storm reports. Wea. Forecasting, 27, 684699, https://doi.org/10.1175/WAF-D-11-00070.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ferraro, R., J. Beauchamp, D. Cecil, and G. Heymsfield, 2015: A prototype hail detection algorithm and hail climatology developed with the Advanced Microwave Sounding Unit (AMSU). Atmos. Res., 163, 2435, https://doi.org/10.1016/j.atmosres.2014.08.010.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gensini, V. A., and W. S. Ashley, 2011: Climatology of potentially severe convective environments from North American Regional Reanalysis. Electron. J. Severe Storms Meteor, 6 (8), http://www.ejssm.org/ojs/index.php/ejssm/article/view/85/68.

    • Search Google Scholar
    • Export Citation
  • Goodman, S. J., and Coauthors, 2013: The GOES-R Geostationary Lightning Mapper (GLM). Atmos. Res., 125–126, 3449, https://doi.org/10.1016/j.atmosres.2013.01.006.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gravelle, C. M., J. R. Mecikalski, W. E. Line, K. M. Bedka, R. A. Petersen, J. M. Sieglaff, G. T. Stano, and S. J. Goodman, 2016: Demonstration of a GOES-R satellite convective toolkit to “bridge the gap” between severe weather watches and warnings: An example from the 20 May 2013 Moore, Oklahoma, tornado outbreak. Bull. Amer. Meteor. Soc., 97, 6984, https://doi.org/10.1175/BAMS-D-14-00054.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Griffin, S., K. M. Bedka, and C. Velden, 2016: A method for calculating the height of overshooting convective cloud tops using satellite-based IR imager and CloudSat Cloud Profiling Radar observations. J. Appl. Meteor. Climatol., 55, 479491, https://doi.org/10.1175/JAMC-D-15-0170.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Holland, G. J., J. L. McBride, R. K. Smith, D. Jasper, and T. D. Keenan, 1986: The BMRC Australian Monsoon Experiment: AMEX. Bull. Amer. Meteor. Soc., 67, 14661472, https://doi.org/10.1175/1520-0477(1986)067<1466:TBAMEA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hoskins, B. J., and K. I. Hodges, 2005: A new perspective on Southern Hemisphere storm tracks. J. Climate, 18, 41084129, https://doi.org/10.1175/JCLI3570.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Keenan, T. D., and R. E. Carbone, 2008: Propagation and diurnal evolution of warm season cloudiness in the Australian and Maritime Continent region. Mon. Wea. Rev., 136, 973994, https://doi.org/10.1175/2007MWR2152.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Keenan, T. D., and Coauthors, 2000: The Maritime Continent Thunderstorm Experiment (MCTEX): Overview and some results. Bull. Amer. Meteor. Soc., 81, 24332455, https://doi.org/10.1175/1520-0477(2000)081<2433:TMCTEM>2.3.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Knight, C., and N. Knight, 2001: Hailstorms. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 223–248.

    • Crossref
    • Export Citation
  • Kuleshov, Y., 2012: Thunderstorm and lightning climatology of Australia. Modern Climatology, S.-Y. Wang, Ed., InTech, https://doi.org/10.5772/35075.

    • Crossref
    • Export Citation
  • Kunkel, K. E., and Coauthors, 2013: Monitoring and understanding trends in extreme storms: State of knowledge. Bull. Amer. Meteor. Soc., 94, 499514, https://doi.org/10.1175/BAMS-D-11-00262.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Laflin, J. M., S. F. Blair, C. Gravelle, K. M. Bedka, B. Scarino, J. Apke, and C. Homeyer, 2016: Super rapid scan imagery analysis of two hailstorms sampled by HailSTONE. 28th Conf. on Severe and Local Storms, Portland, OR, Amer. Meteor. Soc., https://ams.confex.com/ams/28SLS/webprogram/Manuscript/Paper300948/extendedSLSabstract_JLaflin_HailSTONE_SRSOR.pdf.

  • Lay, E. H., A. R. Jacobson, R. H. Holzworth, C. J. Rodger, and R. L. Dowden, 2007: Local time variation in land/ocean lightning flash density as measured by the World Wide Lightning Location Network. J. Geophys. Res., 112, D13111, https://doi.org/10.1029/2006JD007944.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • May, P. T., and T. D. Keenan, 2005: Evaluation of microphysical retrievals from polarimetric radar with wind profiler data. J. Appl. Meteor., 44, 827838, https://doi.org/10.1175/JAM2230.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • May, P. T., C. N. Long, and A. Protat, 2012: The diurnal cycle of the boundary layer, convection, clouds, and surface radiation in a coastal monsoon environment (Darwin, Australia). J. Climate, 25, 53095326, https://doi.org/10.1175/JCLI-D-11-00538.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Climate, 16, 14561475, https://doi.org/10.1175/1520-0442-16.10.1456.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Peter, J. R., M. J. Manton, R. J. Potts, P. T. May, S. M. Collis, and L. Wilson, 2015: Radar-derived statistics of convective storms in southeast Queensland. J. Appl. Meteor. Climatol., 54, 19852008, https://doi.org/10.1175/JAMC-D-13-0347.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pope, M., C. Jakob, and M. J. Reeder, 2008: Convective systems of the north Australian monsoon. J. Climate, 21, 50915112, https://doi.org/10.1175/2008JCLI2304.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Potts, R. J., T. D. Keenan, and P. T. May, 2000: Radar characteristics of storms in the Sydney area. Mon. Wea. Rev., 128, 33083319, https://doi.org/10.1175/1520-0493(2000)128<3308:RCOSIT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Proud, S. R., 2014: Analysis of overshooting top detections by Meteosat Second Generation: A 5-year dataset. Quart. J. Roy. Meteor. Soc., 141, 909915, https://doi.org/10.1002/qj.2410.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Punge, H. J., and M. Kunz, 2016: Hail observations and hailstorm characteristics in Europe: A review. Atmos. Res., 176–177, 159184, https://doi.org/10.1016/j.atmosres.2016.02.012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Punge, H. J., K. M. Bedka, M. Kunz, and A. Werner, 2014: A new physically based stochastic event catalog for hail in Europe. Nat. Hazards, 73, 16251645, https://doi.org/10.1007/s11069-014-1161-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Punge, H. J., K. M. Bedka, M. Kunz, and A. Reinbold, 2017: Hail frequency estimation across Europe based on a combination of overshooting top detections and the ERA-Interim reanalysis. Atmos. Res., 198, 3443, https://doi.org/10.1016/j.atmosres.2017.07.025.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Richter, H., J. Peter, and S. Collis, 2014: Analysis of a destructive wind storm on 16 November 2008 in Brisbane, Australia. Mon. Wea. Rev., 142, 30383060, https://doi.org/10.1175/MWR-D-13-00405.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rienecker, M., and Coauthors, 2011: MERRA: NASA’s Modern-Era Retrospective Analysis for Research and Applications. J. Climate, 24, 36243648, https://doi.org/10.1175/JCLI-D-11-00015.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Romps, D. M., and Z. Kuang, 2009: Overshooting convection in tropical cyclones. Geophys. Res. Lett., 36, L09804, https://doi.org/10.1029/2009GL037396.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schmit, T. J., M. M. Gunshor, W. P. Menzel, J. Li, S. Bachmeier, and J. J. Gurka, 2005: Introducing the next-generation Advanced Baseline Imager (ABI) on GOES-R. Bull. Amer. Meteor. Soc., 86, 10791096, https://doi.org/10.1175/BAMS-86-8-1079.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Slingo, A., K. I. Hodges, and G. J. Robinson, 2004: Simulation of the diurnal cycle in a climate model and its evaluation using data from Meteosat 7. Quart. J. Roy. Meteor. Soc., 130, 14491467, https://doi.org/10.1256/qj.03.165.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Soderholm, J. S., H. McGowan, H. Richter, K. Walsh, T. M. Weckwerth, and M. Coleman, 2017: An 18-year climatology of hailstorm trends and related drivers across southeast Queensland, Australia. Quart. J. Roy. Meteor. Soc., 143, 11231135, https://doi.org/10.1002/qj.2995.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thiery, W., E. L. Davin, S. I. Seneviratne, K. M. Bedka, S. Lhermitte, and N. van Lipzig, 2016: Hazardous thunderstorm intensification over Lake Victoria. Nat. Commun., 7, 12786, http://dx.doi.org/10.1038/ncomms12786.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thiery, W., L. Gudmundsson, K. M. Bedka, F. H. M. Semazzi, S. Lhermitte, P. Willems, N. P. M. van Lipzig, and S. I. Seneviratne, 2017: Early warnings of hazardous storms over Lake Victoria. Environ. Res. Lett., 12, 074012, https://doi.org/10.1088/1748-9326/aa7521.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Toracinta, E. R., D. J. Cecil, E. J. Zipser, and S. W. Nesbitt, 2002: Radar, passive microwave, and lightning characteristics of precipitating systems in the tropics. Mon. Wea. Rev., 130, 802824, https://doi.org/10.1175/1520-0493(2002)130<0802:RPMALC>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vaughan, G., K. Bower, C. Schiller, A. R. MacKenzie, T. Peter, H. Schlager, N. R. Harris, and P. T. May, 2008: SCOUT-O3/ACTIVE: High-altitude aircraft measurements around deep tropical convection. Bull. Amer. Meteor. Soc., 89, 647662, https://doi.org/10.1175/BAMS-89-5-647.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H. Holzworth, 2013: Highlights of a new ground-based, hourly global lightning climatology. Bull. Amer. Meteor. Soc., 94, 13811391, https://doi.org/10.1175/BAMS-D-12-00082.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Walsh, K., and Coauthors, 2016: Natural hazards in Australia: Storms, wind and hail. Climatic Change, 139, 5567, https://doi.org/10.1007/s10584-016-1737-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Webster, P. J., and R. A. Houze, 1991: The Equatorial Mesoscale Experiment (EMEX): An overview. Bull. Amer. Meteor. Soc., 72, 14811505, https://doi.org/10.1175/1520-0477(1991)072<1481:TEMEAO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Williams, E. R., S. A. Rutledge, S. G. Geotis, N. Renno, E. Rasmussen, and T. Rickenbach, 1992: A radar and electrical study of tropical “hot towers”. J. Atmos. Sci., 49, 13861395, https://doi.org/10.1175/1520-0469(1992)049<1386:ARAESO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yost, C. R., and Coauthors, 2018: A prototype method for diagnosing high ice water content probability using satellite imager data. Atmos. Meas. Tech., 11, 16151637, https://doi.org/10.5194/amt-11-1615-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Young, A. H., J. J. Bates, and J. A. Curry, 2012: Complementary use of passive and active remote sensing for detection of penetrating convection from CloudSat, CALIPSO, and Aqua MODIS. J. Geophys. Res., 117, D13205, https://doi.org/10.1029/2011JD016749.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zipser, E. J., and K. R. Lutz, 1994: The vertical profile of radar reflectivity of convective cells: A strong indicator of storm intensity and lightning probability? Mon. Wea. Rev., 122, 17511759, https://doi.org/10.1175/1520-0493(1994)122<1751:TVPORR>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zipser, E. J., J. C. Daniel, C. Liu, S. W. Nesbitt, and D. P. Yorty, 2006: Where are the most intense thunderstorms on Earth? Bull. Amer. Meteor. Soc., 87, 10571071, https://doi.org/10.1175/BAMS-87-8-1057.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save