• Atkins, N. T., A. McGee, R. Ducharme, R. M. Wakimoto, and J. Wurman, 2012: The LaGrange tornado during VORTEX2. Part II: Photogrammetric analysis of the tornado combined with dual-Doppler radar data. Mon. Wea. Rev., 140, 29392958, doi:10.1175/MWR-D-11-00285.1.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., M. M. French, R. L. Tanamachi, S. Frasier, K. Hardwick, F. Junyent, and A. L. Pazmany, 2007: Close-range observations of tornadoes in supercells made with a dual-polarization, X-band, mobile Doppler radar. Mon. Wea. Rev., 135, 15221543, doi:10.1175/MWR3349.1.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., M. M. French, I. PopStefanija, R. T. Bluth, and J. B. Knorr, 2010: A mobile, phased-array Doppler radar for the study of severe convective storms: The MWR-05XP. Bull. Amer. Meteor. Soc., 91, 579600, doi:10.1175/2009BAMS2914.1.

    • Search Google Scholar
    • Export Citation
  • Bringi, V. N., and V. Chandrasekar, 2001: Polarimetric Doppler Weather Radar: Principles and Applications. Cambridge University Press, 636 pp.

  • Bringi, V. N., V. Chandrasekar, N. Balakrishnan, and D. S. Zrnić, 1990: An examination of propagation effects in rainfall on radar measurements at microwave frequencies. J. Atmos. Oceanic Technol., 7, 829840, doi:10.1175/1520-0426(1990)007<0829:AEOPEI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Bringi, V. N., T. D. Keenan, and V. Chandrasekar, 2001: Correcting C-band radar reflectivity and differential reflectivity data for rain attenuation: A self-consistent method with constraints. IEEE Trans. Geosci. Remote Sens., 39, 19061915, doi:10.1109/36.951081.

    • Search Google Scholar
    • Export Citation
  • Burgess, D. W., C. M. Schwarz, J. Snyder, M. M. French, H. B. Bluestein, and C. L. Ziegler, 2012: A cyclic, tornadic supercell on 10 May 2010: Analysis of a VORTEX2 case. Proc. 35th Conf. on Radar Meteorology, Pittsburgh, PA, Amer. Meteor. Soc., 8B.4. [Available online at https://ams.confex.com/ams/35Radar/webprogram/Paper191297.html.]

  • Cao, Q., G. Zhang, E. Brandes, T. Schuur, A. Ryzhkov, and K. Ikeda, 2008: Analysis of video disdrometer and polarimetric radar data to characterize rain microphysics in Oklahoma. J. Appl. Meteor. Climatol., 47, 22382255, doi:10.1175/2008JAMC1732.1.

    • Search Google Scholar
    • Export Citation
  • Davidson, E. A., and D. W. Burgess, 2013: Radar studies of a VORTEX2 tornadic supercell. 36th Conf. on Radar Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 177. [Available online at https://ams.confex.com/ams/36Radar/webprogram/Paper228668.html.]

  • Davies-Jones, R. P., R. J. Trapp, and H. B. Bluestein, 2001: Tornadoes and tornadic storms. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 167–221.

  • Dawson, D. T., II, E. R. Mansell, Y. Jung, L. J. Wicker, M. R. Kumjian, and M. Xue, 2013: Comparisons of numerically simulated and observed low-level polarimetric signatures in supercells. 36th Conf. on Radar Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 12B.6. [Available online at https://ams.confex.com/ams/36Radar/webprogram/Paper229115.html.]

  • Delrieu, G., S. Caoudal, and J. D. Creutin, 1997: Feasibility of using mountain return for the correction of ground-based X-band weather radar data. J. Atmos. Oceanic Technol., 14, 368385, doi:10.1175/1520-0426(1997)014<0368:FOUMRF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dolan, B., and S. A. Rutledge, 2009: A theory-based hydrometeor identification algorithm for X-band polarimetric radars. J. Atmos. Oceanic Technol., 26, 20712088, doi:10.1175/2009JTECHA1208.1.

    • Search Google Scholar
    • Export Citation
  • Doswell, C. A., III, and D. W. Burgess, 1993: Tornadoes and tornadic storms: A review of conceptual models. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, Geophys. Monogr., Vol. 79, Amer. Geophys. Union, 161–172.

  • French, M. M., H. B. Bluestein, I. PopStefanija, C. A. Baldi, and R. T. Bluth, 2013: Reexamining the vertical development of tornadic vortex signature in supercells. Mon. Wea. Rev., 141, 45764601, doi:10.1175/MWR-D-12-00315.1.

    • Search Google Scholar
    • Export Citation
  • French, M. M., H. B. Bluestein, I. PopStefanija, C. A. Baldi, and R. T. Bluth, 2014: Mobile, phased-array, Doppler radar observations of tornadoes at X band. Mon. Wea. Rev., 142, 10101036, doi:10.1175/MWR-D-13-00101.1.

    • Search Google Scholar
    • Export Citation
  • Friedrich, K., E. A. Kalina, F. J. Masters, and C. R. Lopez, 2013: Drop-size distributions in thunderstorms measured by optical disdrometers during VORTEX2. Mon. Wea. Rev., 141, 11821203, doi:10.1175/MWR-D-12-00116.1.

    • Search Google Scholar
    • Export Citation
  • Gorgucci, E., G. Scarchilli, and V. Chandrasekar, 1999: A procedure to calibrate multiparameter weather radar using properties of the rain medium. IEEE Trans. Geosci. Remote Sens., 37, 269276, doi:10.1109/36.739161.

    • Search Google Scholar
    • Export Citation
  • Grzych, M. L., B. D. Lee, and C. A. Finley, 2007: Thermodynamic analysis of supercell rear-flank downdrafts from Project ANSWERS. Mon. Wea. Rev., 135, 240246, doi:10.1175/MWR3288.1.

    • Search Google Scholar
    • Export Citation
  • Homeyer, C. R., and M. R. Kumjian, 2015: Microphysical characteristics of overshooting convection from polarimetric radar observations. J. Atmos. Sci., 72, 870891, doi:10.1175/JAS-D-13-0388.1.

    • Search Google Scholar
    • Export Citation
  • Hubbert, J., and V. N. Bringi, 1995: An iterative filtering technique for the analysis of copolar differential phase and dual-frequency radar measurements. J. Atmos. Oceanic Technol., 12, 643648, doi:10.1175/1520-0426(1995)012<0643:AIFTFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kosiba, K. A., J. Wurman, Y. Richardson, P. Markowski, and P. Robinson, 2013: The genesis of the Goshen County, Wyoming, tornado (5 June 2009). Mon. Wea. Rev., 141, 11571181, doi:10.1175/MWR-D-12-00056.1.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2011: Precipitation properties of supercell hook echoes. Electron. J. Severe Storms Meteor., 6 (5). [Available online at http://ejssm.org/ojs/index.php/ejssm/article/viewArticle/93.]

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2013a: Principles and applications of dual-polarization weather radar. Part I: Description of the polarimetric radar variables. J. Oper. Meteor., 1, 226242, doi:10.15191/nwajom.2013.0119.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2013b: Principles and applications of dual-polarization weather radar. Part II: Warm- and cold-season applications. J. Oper. Meteor., 1, 243264, doi:10.15191/nwajom.2013.0120.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2013c: Principles and applications of dual-polarization weather radar. Part III: Artifacts. J. Oper. Meteor., 1, 265274, doi:10.15191/nwajom.2013.0121.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., and A. V. Ryzhkov, 2008a: Microphysical differences between tornadic and non-tornadic supercell rear-flank downdrafts revealed by dual-polarization radar measurements. Preprints, 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., 3B.4. [Available online at https://ams.confex.com/ams/pdfpapers/141912.pdf.]

  • Kumjian, M. R., and A. V. Ryzhkov, 2008b: Polarimetric signatures in supercell thunderstorms. J. Appl. Meteor. Climatol., 47, 19401961, doi:10.1175/2007JAMC1874.1.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., A. V. Ryzhkov, V. M. Melnikov, and T. J. Schuur, 2010: Rapid-scan super-resolution observations of a cyclic supercell with a dual-polarization WSR-88D. Mon. Wea. Rev., 138, 37623786, doi:10.1175/2010MWR3322.1.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Y. P. Richardson, 2009: Tornadogenesis: Our current understanding, forecasting considerations, and questions to guide future research. Atmos. Res., 93, 310, doi:10.1016/j.atmosres.2008.09.015.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Y. P. Richardson, 2014: The influence of environmental low-level shear and cold pools on tornadogenesis: Insights from idealized simulations. J. Atmos. Sci., 71, 243275, doi:10.1175/JAS-D-13-0159.1.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., J. M. Straka, and E. N. Rasmussen, 2002: Direct surface thermodynamic observations within the rear-flank downdrafts of nontornadic and tornadic supercells. Mon. Wea. Rev., 130, 16921721, doi:10.1175/1520-0493(2002)130<1692:DSTOWT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Coauthors, 2012a: The pretornadic phase of the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Part I: Evolution of kinematic and surface thermodynamic fields. Mon. Wea. Rev., 140, 28872915, doi:10.1175/MWR-D-11-00336.1.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Coauthors, 2012b: The pretornadic phase of the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Part II: Intensification of low-level rotation. Mon. Wea. Rev., 140, 29162938, doi:10.1175/MWR-D-11-00337.1.

    • Search Google Scholar
    • Export Citation
  • Marquis, J., Y. Richardson, P. Markowski, D. Dowell, and J. Wurman, 2012: Tornado maintenance investigated with high-resolution dual-Doppler and EnKF analysis. Mon. Wea. Rev., 140, 327, doi:10.1175/MWR-D-11-00025.1.

    • Search Google Scholar
    • Export Citation
  • Matrosov, S. Y., R. Cifelli, P. C. Kennedy, S. W. Nesbitt, S. A. Rutledge, V. N. Bringi, and B. E. Martner, 2006: A comparative study of rainfall retrievals based on specific differential phase shifts at X- and S-band radar frequencies. J. Atmos. Oceanic Technol., 23, 952963, doi:10.1175/JTECH1887.1.

    • Search Google Scholar
    • Export Citation
  • Melnikov, V., D. Zrnić, A. Ryzhkov, A. Zaharai, and J. Carter, 2009: Validation of attenuation correction at X band performed with collocated S-band polarimetric radar. Preprints, 34th Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., 11A.5. [Available online at https://ams.confex.com/ams/pdfpapers/155322.pdf.]

  • Palmer, R. D., and Coauthors, 2009: Weather radar education at the University of Oklahoma–An integrated interdisciplinary approach. Bull. Amer. Meteor. Soc., 90, 12771282, doi:10.1175/2009BAMS2738.1.

    • Search Google Scholar
    • Export Citation
  • Palmer, R. D., and Coauthors, 2011: Observations of the 10 May 2010 tornado outbreak using OU-PRIME: Potential for new science with high-resolution polarimetric radar. Bull. Amer. Meteor. Soc., 92, 871891, doi:10.1175/2011BAMS3125.1.

    • Search Google Scholar
    • Export Citation
  • Park, S.-G., V. N. Bringi, V. Chandrasekar, M. Maki, and K. Iwanami, 2005a: Correction of radar reflectivity and differential reflectivity for rain attenuation at X band. Part I: Theoretical and empirical basis. J. Atmos. Oceanic Technol., 22, 16211632, doi:10.1175/JTECH1803.1.

    • Search Google Scholar
    • Export Citation
  • Park, S.-G., M. Maki, K. Iwanami, V. N. Bringi, and V. Chandrasekar, 2005b: Correction of radar reflectivity and differential reflectivity for rain attenuation at X band. Part II: Evaluation and application. J. Atmos. Oceanic Technol., 22, 16331655, doi:10.1175/JTECH1804.1.

    • Search Google Scholar
    • Export Citation
  • Parker, M. D., 2014: Composite VORTEX2 supercell environments from near-storm soundings. Mon. Wea. Rev., 142, 508529, doi:10.1175/MWR-D-13-00167.1.

    • Search Google Scholar
    • Export Citation
  • Picca, J., and A. Ryzhkov, 2012: A dual-wavelength polarimetric analysis of the 16 May 2010 Oklahoma City extreme hailstorm. Mon. Wea. Rev., 140, 13851403, doi:10.1175/MWR-D-11-00112.1.

    • Search Google Scholar
    • Export Citation
  • Pruppacher, H. R., and K. V. Beard, 1970: A wind tunnel investigation of the internal circulation and shape of water drops falling at terminal velocity in air. Quart. J. Roy. Meteor. Soc., 96, 247256, doi:10.1002/qj.49709640807.

    • Search Google Scholar
    • Export Citation
  • Richardson, Y. P., P. Markowski, J. N. Marquis, J. Wurman, K. A. Kosiba, P. Robinson, D. W. Burgess, and C. C. Weiss, 2012: Tornado maintenance and demise in the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Proc. 26th Conf. on Severe Local Storms, Nashville, TN, Amer. Meteor. Soc., 13.3. [Available online at https://ams.confex.com/ams/pdfpapers/88063.pdf.]

  • Romine, G. S., D. W. Burgess, and R. B. Wilhelmson, 2008: A dual-polarization-radar-based assessment of the 8 May 2003 Oklahoma City area tornadic supercell. Mon. Wea. Rev., 136, 28492870, doi:10.1175/2008MWR2330.1.

    • Search Google Scholar
    • Export Citation
  • Ryzhkov, A. V., S. E. Giangrande, V. M. Melnikov, and T. J. Schuur, 2005a: Calibration issues of dual-polarization radar measurements. J. Atmos. Oceanic Technol., 22, 11381155, doi:10.1175/JTECH1772.1.

    • Search Google Scholar
    • Export Citation
  • Ryzhkov, A. V., T. J. Schuur, D. W. Burgess, and D. S. Zrnić, 2005b: Polarimetric tornado detection. J. Appl. Meteor., 44, 557570, doi:10.1175/JAM2235.1.

    • Search Google Scholar
    • Export Citation
  • Schuur, T. J., A. V. Ryzhkov, D. S. Zrnić, and M. Schönhuber, 2001: Drop size distributions measured by a 2D video disdrometer: Comparison with dual-polarization radar data. J. Appl. Meteor., 40, 10191034, doi:10.1175/1520-0450(2001)040<1019:DSDMBA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schwarz, C. M., and D. W. Burgess, 2011: Supercell polarimetric signatures at X-band: Data from VORTEX2. Preprints, 35th Conf. on Radar Meteorology, Pittsburgh, PA, Amer. Meteor. Soc., P60. [Available online at https://ams.confex.com/ams/35Radar/webprogram/Paper191298.html.]

  • Seliga, T. A., and V. N. Bringi, 1976: Potential use of radar differential reflectivity measurements at orthogonal polarizations for measuring precipitation. J. Appl. Meteor., 15, 6976, doi:10.1175/1520-0450(1976)015<0069:PUORDR>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Snyder, J. C., 2013: Observations and simulations of polarimetric X-band radar signatures in supercells. Ph.D. dissertation, University of Oklahoma, Norman, OK, 214 pp. [Available online at https://shareok.org/handle/11244/6398.]

  • Snyder, J. C., H. B. Bluestein, G. Zhang, and S. J. Frasier, 2010: Attenuation correction and hydrometeor classification of high-resolution, X-band, dual-polarized mobile radar measurements in severe convective storms. J. Atmos. Oceanic Technol., 27, 19792001, doi:10.1175/2010JTECHA1356.1.

    • Search Google Scholar
    • Export Citation
  • Snyder, J. C., H. B. Bluestein, V. Venkatesh, and S. J. Frasier, 2013: Observations of polarimetric signatures in supercells by an X-band mobile Doppler radar. Mon. Wea. Rev., 141, 329, doi:10.1175/MWR-D-12-00068.1.

    • Search Google Scholar
    • Export Citation
  • Tanamachi, R. L., H. B. Bluestein, J. B. Houser, S. J. Frasier, and K. M. Hardwick, 2012: Mobile, X-band, polarimetric Doppler radar observations of the 4 May 2007 Greensburg, Kansas, tornadic supercell. Mon. Wea. Rev., 140, 21032125, doi:10.1175/MWR-D-11-00142.1.

    • Search Google Scholar
    • Export Citation
  • Testud, J., E. Le Bouar, E. Obligis, and M. Ali-Mehenni, 2000: The rain profiling algorithm applied to polarimetric weather radar. J. Atmos. Oceanic Technol., 17, 332356, doi:10.1175/1520-0426(2000)017<0332:TRPAAT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Thompson, E. J., S. A. Rutledge, and B. Dolan, 2014: A dual-polarization radar hydrometeor classification algorithm for winter precipitation. J. Atmos. Oceanic Technol., 31, 14571481, doi:10.1175/JTECH-D-13-00119.1.

    • Search Google Scholar
    • Export Citation
  • Thompson, R. L., R. Edwards, J. A. Hart, K. L. Elmore, and P. Markowski, 2003: Close proximity soundings within supercell environments obtained from the Rapid Update Cycle. Wea. Forecasting, 18, 12431261, doi:10.1175/1520-0434(2003)018<1243:CPSWSE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Trapp, R. J., 1999: Observations of nontornadic low-level mesocyclones and attendant tornadogenesis failure during VORTEX. Mon. Wea. Rev., 127, 16931705, doi:10.1175/1520-0493(1999)127<1693:OONLLM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., N. T. Atkins, and J. Wurman, 2011: The LaGrange tornado during VORTEX2. Part I: Photogrammetry analysis of the tornado combined with single-Doppler radar data. Mon. Wea. Rev., 139, 22332258, doi:10.1175/2010MWR3568.1.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., P. Stauffer, W.-C. Lee, N. T. Atkins, and J. Wurman, 2012: Finescale structure of the LaGrange, Wyoming, tornado during VORTEX2: GBVTD and photogrammetric analyses. Mon. Wea. Rev., 140, 33973418, doi:10.1175/MWR-D-12-00036.1.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., D. Dowell, Y. Richardson, P. Markowski, E. Rasmussen, D. Burgess, L. Wicker, and H. B. Bluestein, 2012: The Second Verification of the Origins of Rotation in Tornadoes Experiment: VORTEX2. Bull. Amer. Meteor. Soc., 93, 11471170, doi:10.1175/BAMS-D-11-00010.1.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., K. Kosiba, and P. Robinson, 2013: In situ, Doppler radar, and video observations of the interior structure of a tornado and the wind–damage relationship. Bull. Amer. Meteor. Soc., 94, 835846, doi:10.1175/BAMS-D-12-00114.1.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., 2013: A diabatic Lagrangian technique for the analysis of convective storms. Part II: Application to a radar-observed storm. J. Atmos. Oceanic Technol., 30, 22662280, doi:10.1175/JTECH-D-13-00036.1.

    • Search Google Scholar
    • Export Citation
  • Zrnić, D. S., V. M. Melnikov, and J. K. Carter, 2006: Calibrating differential reflectivity on the WSR-88D. J. Atmos. Oceanic Technol., 23, 944951, doi:10.1175/JTECH1893.1.

    • Search Google Scholar
    • Export Citation
  • View in gallery
    Fig. 1.

    Conceptual model of the precipitation characteristics of supercell hook echoes overlaid with the locations of the updraft (red), RFD (blue), and primary rear-flank gust front (RFGF; black line). The region shaded in pink indicates large-drop zones with high ZDR (including the ZDR arc). The green shading represents regions of lower ZDR and smaller drop sizes. The dark-green region represents an enhancement in tiny drops transported by the occlusion downdraft. The location of the tornado is indicated by the black circle with a red T. [From Kumjian (2011).]

  • View in gallery
    Fig. 2.

    NOXP PPIs of (a),(e) (dBZ), (b),(f) ZH (dBZ), (c),(g) (dB), and (d),(h) ZDR (dB) of (top) a high-precipitation, nontornadic supercell near Sayre, Oklahoma, on 13 May 2010 and (bottom) a nontornadic supercell northeast of Sidney, Nebraska, on 6 Jun 2009. The ZPHI scheme was used in (b) and (f) to correct for attenuation, and the gamma method was used in (d) and (h) to correct for differential attenuation. To isolate the effects of attenuation correction and differential attenuation correction, no bias corrections have been performed on the data shown. The white circles enclose examples of areas where and were noticeably corrected. Range rings are every 5 km. All images are centered at the same location. The Sayre supercell dataset and others with similarly large amounts of attenuation and differential attenuation are not included in this study.

  • View in gallery
    Fig. 3.

    Comparative PPIs of (dBZ) in a supercell from 5 Jun 2009 as observed by (a) NOXP and (b) UMass X-Pol. The height of the center of the beam at the front edge of the hook echo is ~930 and ~910 m ARL for (a) and (b), respectively. The numbers in parentheses at the bottom of each image, from left to right, are the mean and median in the hook echo for ≥ 20 dBZ.

  • View in gallery
    Fig. 4.

    The ZHZDR scatterplot of assumed snow aggregates in NOXP (red circles) and UMass X-Pol (blue circles) data from 2220 to 2225 UTC 5 Jun 2009. Data are taken from levels above the bright band with heights ranging from 2 to 4 km ARL. NOXP ZH and ZDR data are binned every 0.5 dB and 0.0625 dB, respectively, during data collection.

  • View in gallery
    Fig. 5.

    NOXP PPIs of ZH (dBZ) in supercells from four cases in which hook echo data are examined for this study: tornadic supercells on (a) 5 Jun 2009 in Goshen County and (b) and 19 May 2010 near Kingfisher and nontornadic supercells on (c) 23 May 2011 near Watonga, Oklahoma and (d) 28 May 2012 near Rush Springs, Oklahoma. The supercell shown in (d) has a hook echo with anticyclonic curvature. The black lines mark the subjective isolation of the hook echo from the rest of the supercell echo. In the tornadic supercells, the location of the tornadic vortex signature (TVS) associated with the tornado is marked by a white circle. The demarcation line in (b) is not located farther north because of strong attenuation and differential attenuation at those locations. Range rings are every 2.5 km. The approximate heights of the observations are shown in Table 2.

  • View in gallery
    Fig. 6.

    Scatterplots of ZH vs ZDR for NOXP data from six tornadic hook echoes: (a) 5 Jun 2009 in Goshen County; (b) 19 May 2010 near Kingfisher; (c) 25 May 2010 in Tribune, Kansas; (d) 8 Jun 2010 near Scottsbluff, Nebraska; (e) 12 Jun 2010 near Limon, Colorado; and (f) 30 May 2012 near Crowell, Texas. The black line marks a relation between ZH and ZDR that was developed by Cao et al. (2008) and is based on a large amount of DSD data collected in Oklahoma rain. Only data with ρHV ≥ 0.95 are shown and analyzed. In (a), data from UMass X-Pol from the same approximate time and height as those from NOXP are shown for comparison. The number of data points with ZH ≥ 20 dBZ is shown at the top or bottom left (for convenience, gates with ZH ≥ 60 dBZ are not shown). The approximate center-beam height level ARL (m) of the hook echo is shown at the bottom or top right.

  • View in gallery
    Fig. 7.

    As in Fig. 6, but using data from three nontornadic hook echoes: (a) 23 May 2011 near Carnegie, Oklahoma; (b) 28 May 2012 near Rush Springs; and (c) 30 May 2012 near Kingfisher.

  • View in gallery
    Fig. 8.

    As in Fig. 6, but using data from six nontornadic hook echoes: (a) 6 Jun 2009 near Sidney; (b) 6 Jun 2009 near Mullen, Nebraska; (c) 7 Jun 2009 near Savannah, Missouri; (d) 9 Jun 2009 in Greensburg; (e) 14 Jun 2009 near Amarillo, Texas; and (f) 26 May 2010 near Hoyt, Colorado.

  • View in gallery
    Fig. 9.

    NOXP PPIs of (top) ZH (dBZ) and (bottom) Vr (m s−1) in three nontornadic supercells that were examined in this study: (a) 23 May 2011 and (b) 28 and (c) 30 May 2012. The black circles outline low-level mesocyclone-scale rotation in the hook echoes. The white circles in (b) and (c) outline smaller-scale rotation. Note in (b) that the rotation is anticyclonic in the left-moving supercell. The color-coded numbers are the maximum ΔV in the corresponding mesocyclones or vortex signatures. Range rings are every 1.0 km. The approximate heights of the observations at the range of the hook echo are shown in Table 2.

  • View in gallery
    Fig. 10.

    Scatterplots of environmental variables from near-storm soundings for the 11 VORTEX2 hook echo cases used in this study: (a) CAPE calculated from the surface to 9 km vs the BWD magnitude between the surface and 6 km and (b) SRH from the surface to 3 km vs the wind-difference magnitude between the surface and 1 km. Triangles (circles) indicate tornadic (nontornadic) hook echo cases. The calculations of CAPE include the virtual temperature correction and only progress up to 9 km AGL because the soundings used in the calculations terminated at different heights. The BWD calculations use the wind speed–magnitude data points that are closest to 10 m in height for the bottom of the layer. Storm motion was estimated by following the storm’s inflow notch or, if that feature was not able to be tracked reliably, the forward-flank reflectivity maximum over a 10–20-min period centered on the analysis time. All heights are AGL from the sounding launch location. Note that, for the VORTEX2 cases, all of the small-drop cases are tornadic cases.

  • View in gallery
    Fig. 11.

    Line plots of (a) near-surface RH (%) and (b) height of the LCL (m AGL) for the 15 hook echo cases used in this study. Triangles (circles) indicate tornadic (nontornadic) hook echo cases. LCL height calculations for the 2011–12 cases use Oklahoma Mesonet temperature and RH measurements at 1.5 m.

  • View in gallery
    Fig. 12.

    NOXP PPIs of (top) ZH (dBZ) and (bottom) ZDR (dB) in (a)–(e) five tornadic cases and (f) one nontornadic supercell example case used in this study. The black circles in (a)–(e) mark the locations of the TVSs associated with tornadoes. The arrows in (a)–(e) mark locations of local minima in ZDR, and the white lines in (d) and (f) mark the local maxima in ZDR on the inner half of the hook echo. The black lines in (d) outline the quadrants shown in Fig. 13 and discussed in the text, and the dotted black line is the approximate subjective cutoff for the hook echo. The only ZDR gates shown are those with ZH > 20 dBZ. Range rings are every 1.0 km. The approximate heights of the observations at the range of the hook echo are shown in Table 2.

  • View in gallery
    Fig. 13.

    As in Fig. 6, but with the gates color coded by where they were located relative to the TVS heading angle. The locations are centered on quadrants located to the forward right, rear right, rear left, and forward left relative to the location of the TVS associated with each tornado (black circle) and the TVS’s heading angle (black arrow; up points due north). An example of the quadrants is shown in the top left in (a). The centroid for each quadrant, calculated only for ZH ≥ 30 dBZ, is marked by a large colored circle. To avoid quadrants with fewer data points being blocked, the plots for each quadrant are layered such that the quadrant with the fewest (most) data points appears on the top (bottom). The percentage of total gates with ZH > 30 dBZ lying below the C08L is shown for the quadrants in each case.

  • View in gallery
    Fig. 14.

    As in Fig. 12, but for one tornadic case from 12 Jun 2010, shown every ~2 min from 0104 to 0114 UTC. The white arrow in (c) marks the area of low ZDR that developed prior to or as the tornado formed. The black arrows point to a cell that merged with the supercell.

  • View in gallery
    Fig. 15.

    As in Fig. 6, but for one case on 12 Jun 2010 at (a) 0102:20, (b) 0104:21, (c) 0106:19, (d) 0108:20, (e) 0110:18, (f) 0112:19, (g) 0114:21, (h) 0116:18, and (i) 0118:32 UTC.

  • View in gallery
    Fig. 16.

    Time series of hook echo (a) mean ZDR (dB) for three ranges of binned ZH values and (b) the percentage of radar gates characterized by large and small drops before, during, and after the formation of a tornado on 12 Jun 2010. Also shown is the maximum ΔV in the TVS associated with the tornado (ordinate scale on the right). The percentages of large- and small-drop gates were determined in the manner described in the text.

  • View in gallery
    Fig. 17.

    The number of radar gates characterized by (left) large and (right) small raindrops at several heights (a),(b) before and (c),(d) after the formation of a tornado on 12 Jun 2010. The numbers of small and large drops were binned every 200 m from 200 to 3100 m ARL. A bin was only included if at least one-half of the height level was sampled (e.g., the 400–600-m bin was included only if the lowest observation was from below 500 m ARL). Note the differing abscissa scales for the large and small-drop plots.

  • View in gallery
    Fig. 18.

    As in Fig. 16, but for the dissipation of tornadoes on (a),(b) 5 Jun 2009 and (c),(d) 19 May 2010. The TVS ΔV data are from the lowest-observed MWR-05XP level and are smoothed using a 1–2–1 filter.

  • View in gallery
    Fig. 19.

    (a) NOXP PPIs of (top) ZH (dBZ) and (bottom) VR (m s−1) every ~4 min from 2304 to 2318 UTC 26 May 2010 for a nontornadic supercell hook echo. (b),(c) As in Fig. 16, but during the evolution of a hook echo in a nontornadic supercell. In (a) the range rings are every 1.0 km, and the approximate heights of the observations at the range of the hook echo are shown in Table 2.

  • View in gallery
    Fig. 20.

    As in Fig. 19, but from 0045 to 0103 UTC 30 May 2012 for a nontornadic supercell hook echo. Distinct TVSs are enclosed by color-coded circles. Note that the ordinate scales differ from Fig. 19 in (b) and (c).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 327 141 17
PDF Downloads 412 285 5

Bulk Hook Echo Raindrop Sizes Retrieved Using Mobile, Polarimetric Doppler Radar Observations

Michael M. FrenchNOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Michael M. French in
Current site
Google Scholar
PubMed
Close
,
Donald W. BurgessCooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma, and NOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Donald W. Burgess in
Current site
Google Scholar
PubMed
Close
,
Edward R. MansellNOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Edward R. Mansell in
Current site
Google Scholar
PubMed
Close
, and
Louis J. WickerNOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Louis J. Wicker in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Polarimetric radar observations obtained by the NOAA/National Severe Storms Laboratory mobile, X-band, dual-polarization radar (NOXP) are used to investigate “hook echo” precipitation properties in several tornadic and nontornadic supercells. Hook echo drop size distributions (DSDs) were estimated using NOXP data obtained from 2009 to 2012, including during the second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2). Differences between tornadic and nontornadic hook echo DSDs are explored, and comparisons are made with previous observations of estimated hook echo DSDs made from stationary S- and C-band Doppler radars. Tornadic hook echoes consistently contain radar gates that are characterized by small raindrops; nontornadic hook echoes are mixed between those that have some small-drop gates and those that have almost no small-drop gates. In addition, the spatial distribution of DSDs was estimated using the high-spatial-resolution data afforded by NOXP. A unique polarimetric signature, an area of relatively low values of differential radar reflectivity factor ZDR south and east of the tornado, is observed in many of the tornadic cases. Also, because most data were obtained using 2-min volumetric updates, the evolution of approximated hook echo precipitation properties was studied during parts of the life cycles of three tornadoes. In one case, there is a large decrease in the percentage of large-raindrop gates and an increase in the percentage of small-raindrop gates in the minutes leading up to tornado formation. The percentage of large-drop gates generally increases prior to and during tornado dissipation. Near-storm environmental data are used to put forth possible relationships between bulk hook echo DSDs and tornado production and life cycle.

Corresponding author address: Michael M. French, NOAA/National Severe Storms Laboratory, National Weather Center, 120 David L. Boren Blvd., Norman, OK 73072. E-mail: michael.french@noaa.gov

Abstract

Polarimetric radar observations obtained by the NOAA/National Severe Storms Laboratory mobile, X-band, dual-polarization radar (NOXP) are used to investigate “hook echo” precipitation properties in several tornadic and nontornadic supercells. Hook echo drop size distributions (DSDs) were estimated using NOXP data obtained from 2009 to 2012, including during the second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2). Differences between tornadic and nontornadic hook echo DSDs are explored, and comparisons are made with previous observations of estimated hook echo DSDs made from stationary S- and C-band Doppler radars. Tornadic hook echoes consistently contain radar gates that are characterized by small raindrops; nontornadic hook echoes are mixed between those that have some small-drop gates and those that have almost no small-drop gates. In addition, the spatial distribution of DSDs was estimated using the high-spatial-resolution data afforded by NOXP. A unique polarimetric signature, an area of relatively low values of differential radar reflectivity factor ZDR south and east of the tornado, is observed in many of the tornadic cases. Also, because most data were obtained using 2-min volumetric updates, the evolution of approximated hook echo precipitation properties was studied during parts of the life cycles of three tornadoes. In one case, there is a large decrease in the percentage of large-raindrop gates and an increase in the percentage of small-raindrop gates in the minutes leading up to tornado formation. The percentage of large-drop gates generally increases prior to and during tornado dissipation. Near-storm environmental data are used to put forth possible relationships between bulk hook echo DSDs and tornado production and life cycle.

Corresponding author address: Michael M. French, NOAA/National Severe Storms Laboratory, National Weather Center, 120 David L. Boren Blvd., Norman, OK 73072. E-mail: michael.french@noaa.gov

1. Introduction

A recent focus of observational research involving supercells has been in determining the particle size distribution (PSD) within “hook echoes.” This focus is motivated by the relationship between storm microphysical processes and the near-surface thermodynamic state of supercell rear-flank downdraft (RFD) regions. Dynamic, thermodynamic, and microphysical processes all are likely to play an important role in determining the hook echo PSD, with feedbacks to the thermodynamic characteristics of the environment within the hook echo. In turn, many past studies have discussed the important influence that the thermodynamic attributes of supercell RFDs may have on supercell evolution, including, but not limited to, tornadogenesis (e.g., Markowski et al. 2002; Grzych et al. 2007) and tornado maintenance (e.g., Marquis et al. 2012). Information about PSDs has been obtained from convective storms mainly through two different methods: direct PSD data from disdrometers and estimated drop size distributions (DSDs) from polarimetric radar data.

Only a few studies have investigated supercell PSDs using disdrometer data. Schuur et al. (2001) used a 2D video disdrometer to obtain data on DSDs in a tornadic supercell, but the disdrometer was located ~1.5 km north of the hook echo of the eastward-moving supercell and therefore the storm’s rear flank was only partially sampled (see their Fig. 12). Friedrich et al. (2013) used data from optical disdrometers deployed during the second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2; Wurman et al. 2012) to analyze DSDs in different regions of three supercells. In one tornadic case, small drops were found in the “ball” of the hook echo at and near the location of the tornado and then larger particles were sampled upshear. In preliminary work using VORTEX2 disdrometer data, as discussed by Dawson et al. (2013), variable DSDs were found in hook echoes, with larger drops at the leading edge and smaller drops at the rear, in the “spine” of the hook echoes.

Despite the specific information about PSDs provided by disdrometers, such data generally are difficult to obtain and are lacking in their spatiotemporal coverage within a storm. An alternative is to use polarimetric radar data to estimate DSDs in a greater number of storms over a larger area. In preliminary work by Kumjian and Ryzhkov (2008a), nine supercell hook echoes (four tornadic and five nontornadic) were investigated for differences in approximated DSDs by using data from a polarimetric prototype Weather Surveillance Radar-1988 Doppler (WSR-88D) that is located in Norman, Oklahoma (KOUN). Nontornadic hook echoes contained radar gates that were characterized by larger median raindrop sizes relative to the set of tornadic hook echoes. The authors suggested that greater evaporation rates in nontornadic hook echoes could explain the presence of larger median drop sizes.

In Kumjian (2011, hereinafter K11), data in six different supercell hook echoes, five of which were tornadic, from KOUN and from the University of Oklahoma Polarimetric Radar for Innovations in Meteorology and Engineering (OU-PRIME; Palmer et al. 2011) were again used to assess approximate DSDs in hook echoes. It was found that median raindrop sizes decreased with distance from the inflow edge of the reflectivity gradient of hook echoes (Fig. 1). It was proposed that 1) size sorting of large drops near the storm updraft was responsible for large drops on the inflow sides of hook echoes (pink shading in Fig. 1) and 2) dynamically driven downdrafts from storm low levels brought large numbers of small drops to the surface in the rear portion of hook echoes and south and east (in the Northern Hemisphere) of the main circulation (green shading in Fig. 1). The small drops were hypothesized to originate from warm-rain processes initiated by forced shallow lifting of moist boundary layer air. Local enhancements of small drops would be found in the vicinity of an occlusion downdraft owing to stronger downward-directed vertical perturbation pressure gradient forces (VPPGFs). The small number of hook echo cases used in the study and the need for data from high-resolution dual-polarization radars was emphasized. In addition, Tanamachi et al. (2012) used data from an X-band, polarimetric mobile Doppler radar developed at the University of Massachusetts (UMass X-Pol; Bluestein et al. 2007) to infer small drops around several weak tornadoes and near dual-Doppler-analyzed downdrafts in the Greensburg, Kansas, supercell, providing possible evidence of mechanism 2. Several additional recent studies have investigated supercells by using polarimetric radar data from fixed or mobile radars, but the focus of such studies was not on hook echo DSDs (e.g., Ryzhkov et al. 2005b; Bluestein et al. 2007; Romine et al. 2008; Kumjian and Ryzhkov 2008b; Kumjian et al. 2010; Snyder et al. 2013).

Fig. 1.
Fig. 1.

Conceptual model of the precipitation characteristics of supercell hook echoes overlaid with the locations of the updraft (red), RFD (blue), and primary rear-flank gust front (RFGF; black line). The region shaded in pink indicates large-drop zones with high ZDR (including the ZDR arc). The green shading represents regions of lower ZDR and smaller drop sizes. The dark-green region represents an enhancement in tiny drops transported by the occlusion downdraft. The location of the tornado is indicated by the black circle with a red T. [From Kumjian (2011).]

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

K11 also analyzed one case by using “rapid scan” data from KOUN. An association between the cyclic mesocyclogenesis process and the sizes of drops was shown whereby the number of small-drop gates increased during an RFD outflow surge and the subsequent mesocyclone occlusion process. The number of large-drop gates increased upon the development of a new mesocyclone. Again, the need for more high-resolution radar observations to establish such relationships between storm processes and inferred drop sizes was discussed.

Despite the recent interest in quantifying the PSDs in supercell hook echoes, fewer than 20 cases have been investigated. In many of the cases, spatial and temporal mapping of PSDs was limited by the small amount of available data. In this study, data from 15 supercells, both tornadic and nontornadic, obtained by the NOAA/National Severe Storms Laboratory (NSSL) X-band, dual-polarization radar (NOXP; Melnikov et al. 2009; Palmer et al. 20091) are used to examine quantitatively estimated bulk DSDs in hook echoes. In addition, near-storm soundings from VORTEX2 are used to investigate environmental differences between the sets of cases exhibiting disparate DSDs. This is the first study to use mobile radar data, X-band radar data, or near-storm soundings to study hook echo DSDs specifically. The use of data with higher spatial resolution than that of the WSR-88D network allows for increased detail in mapping spatial changes in approximated DSDs. Also, the 2-min NOXP volumetric update time allows for changes in approximated DSDs to be analyzed during select tornadogenesis and tornado dissipation, among other, cases. Section 2 details information about the instruments used to collect data for this study and the datasets under examination. Section 3 presents observations of hook echo DSDs, their general spatial distribution, and their temporal evolution for several cases. The observations are summarized, interpreted, and discussed in the context of past work in section 4.

2. Data

a. NOXP

The main data sources for this study are radar data obtained in 15 supercells by NOXP from 2009 to 2012; many of these datasets were obtained during VORTEX2. NOXP is a ground-based, mobile, polarimetric, 3-cm-wavelength (X band) radar. Dual-polarization capabilities are accomplished using horizontally and vertically polarized waves transmitted and received simultaneously. Several NOXP system parameters are provided in Table 1. Moment data, including measured radar reflectivity factor at horizontal polarization and vertical polarization ,2 and mean radial velocity Vr were processed.

Table 1.

Several characteristics of NOXP. The values for adjustable parameters represent what was commonly used during data collection in 2009–12.

Table 1.

NOXP was operated as an updraft-scale or mesocyclone-scale radar in VORTEX2 (e.g., Wurman et al. 2012) with the focus being on obtaining data from the rear flanks of supercells. Distances to the target storms varied from very close (1–2 km) to very far (>30 km) depending on the configuration of the mesocyclone-scale radars at the time of deployment. The radar truck has a hydraulic leveling system so that the radar antenna was precisely leveled prior to data collection. Datasets were chosen such that data from the lowest observable level with usable data (i.e., minimal beam blockage) had a center-beam height in the lowest 1 km, as in K11. In most cases, the center-beam height was 300–700 m above radar level (ARL). Data from NOXP have been used in several past and ongoing studies of supercells (e.g., Melnikov et al. 2009; Burgess et al. 2012; Richardson et al. 2012; Davidson and Burgess 2013; Friedrich et al. 2013; Ziegler 2013).

1) Polarimetric variables

The polarimetric variables used for assessing precipitation properties in hook echoes are briefly explained here with a focus on their relevance to this study. For a more complete summary of polarimetric variables and their applications, see Bringi and Chandrasekar (2001) and Kumjian (2013ac, and references therein). For an overview of common polarimetric signatures observed in supercells at S band and X band, see Kumjian and Ryzhkov (2008b) and Snyder et al. (2013), respectively.

Differential radar reflectivity factor ZDR (Seliga and Bringi 1976) is the logarithm of the ratio of ZH to ZV; positive (negative) values of ZDR indicate more (less) backscattered power returned from the horizontally polarized wave than from the vertically polarized wave. As a raindrop’s diameter increases, it becomes more oblate (Pruppacher and Beard 1970), leading to more power returned in the horizontal plane than in the vertical plane. As a result, ZDR increases with drop diameter and is used as an estimate of the median raindrop size in a radar sample volume (e.g., K11).

The copolar cross-correlation coefficient at lag zero ρHV is a measure of the heterogeneity of scatterers within a radar resolution volume. One way it can be reduced from a value of 1.0 within a sample volume is by the presence of different types of hydrometeors. In this study, a ρHV threshold of 0.95 was used [based on the scattering simulations of Snyder et al. (2010)] to remove radar gates that likely contained hail or a mix of rain and hail. In addition, the total differential propagation phase shift ΦDP, which is the sum of the propagation differential phase ϕDP and the backscatter differential phase δ, was used in correcting for attenuation and differential attenuation. It measures the accrued relative change in phase that occurs for the horizontally polarized wave when compared with the vertically polarized wave as it encounters scatterers.

Resonance (Mie) scattering increases with increasing radar frequency and may have a significant effect on X-band polarimetric data. For small raindrops (equivolume diameters less than ~2 mm), neither ZH nor ZDR is sensitive to the radar frequencies used in this study. Drop diameters above ~2 mm begin to show resonance effects at X band, however. For these larger drops, ZDR at X band may exceed that at S band by ~0.5 dB for ~3-mm drops, with otherwise smaller differences at larger drop sizes (Matrosov et al. 2006; Snyder et al. 2010). Resonance scattering also affects δ by introducing additional phase shifts beyond those caused by the aforementioned properties of the hydrometeor; these phase shifts alter the assumed monotonic increases of ΦDP with range. Also, the presence of resonance scatterers at X band can reduce ρHV, even in pure rain, although such values should only be slightly below those at S band in pure rain (Snyder et al. 2010). Values of ZH and ZDR in hail at X and C band are extremely sensitive to hailstone diameter and fractional water content and are difficult to interpret. As a result, in this study, every effort was made to restrict observations to rain.

2) Attenuation correction

A significant challenge associated with using X-band radar data to study severe convection rather than using data from an S-band or C-band system is that power loss from rain/graupel/hail attenuation increases with increasing radar frequency. Attenuation typically preferentially reduces relative to because of the aforementioned tendency for raindrops and small melting hailstones to be oblate spheroids. In turn, this differential attenuation leads to anomalously low values. Datasets of obvious high-precipitation (HP; Doswell and Burgess 1993) supercells were generally not included in this study because they often had near-complete signal extinction that could only be partially corrected using the schemes to be described presently (e.g., Figs. 2a–d). Most hook echoes that are examined do not have identifiable signs of attenuation or differential attenuation (e.g., Figs. 2e,g). Because quantitative analysis of hook echo data is used to approximate DSDs, however, corrections for attenuation were applied to all cases used in this study.

Fig. 2.
Fig. 2.

NOXP PPIs of (a),(e) (dBZ), (b),(f) ZH (dBZ), (c),(g) (dB), and (d),(h) ZDR (dB) of (top) a high-precipitation, nontornadic supercell near Sayre, Oklahoma, on 13 May 2010 and (bottom) a nontornadic supercell northeast of Sidney, Nebraska, on 6 Jun 2009. The ZPHI scheme was used in (b) and (f) to correct for attenuation, and the gamma method was used in (d) and (h) to correct for differential attenuation. To isolate the effects of attenuation correction and differential attenuation correction, no bias corrections have been performed on the data shown. The white circles enclose examples of areas where and were noticeably corrected. Range rings are every 5 km. All images are centered at the same location. The Sayre supercell dataset and others with similarly large amounts of attenuation and differential attenuation are not included in this study.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

The and fields were corrected for specific attenuation AH (dB km−1) and specific differential attenuation ADP (dB km−1), respectively, on the basis of past work that applied attenuation-correction techniques on mobile, X-band polarimetric radar data as presented in Snyder et al. (2010). The “ZPHI” method (Testud et al. 2000) was used to estimate AH, and a “gamma” method (Bringi et al. 1990) was used to estimate ADP. The ZPHI method estimates attenuation through a path using the total radial change in ϕDP. NOXP measures ΦDP, and therefore ϕDP was estimated by removing δ from resonance scattering using an iterative filtering method as in Hubbert and Bringi (1995). Attenuation correction with the ZPHI method uses two constants—a coefficient αH in an AHKDP (KDP is specific differential phase) relation and an exponent b in an AHZH relation (e.g., Bringi et al. 1990). Values for both parameters (αH = 0.313 and b = 0.76) were estimated on the basis of the results of observations and scattering simulations (e.g., Delrieu et al. 1997; Testud et al. 2000; Park et al. 2005b). Past studies using NOXP data from VORTEX2 also used the ZPHI method for attenuation correction (Schwarz and Burgess 2011; Ziegler 2013).

The gamma method relates AH to ADP by a value γ that is the ratio of αDP to αH, where αDP is a coefficient in an ADPKDP relation (e.g., Bringi et al. 1990). The value of αDP is variable, similar to αH, and was estimated as 0.0483 (e.g., Bringi et al. 2001; Park et al. 2005b). Because both αH and αDP vary with drop sizes, among other dependencies, γ also varies with drop size. On the basis of results from -matrix scattering simulations at X band, this method likely underestimates ADP for large drops and overestimates ADP for small drops because of the increase of γ with drop size (Snyder 2013). The elimination of HP supercells resulted in only modest attenuation-correction and differential-attenuation-correction magnitudes in most cases used in this study (e.g., Figs. 2f,h). In some cases, there is minor “streakiness” in ZH after attenuation correction because of noise in the ΦDP data (e.g., Snyder et al. 2010). The use of attenuation-correction techniques mitigates, but does not eliminate, the possibility of quantitative errors in the ZH and ZDR data. More information about attenuation-correction techniques at X band can be found in Park et al. (2005a) and Snyder et al. (2010).

3) Calibration of ZH and ZDR

Radar calibration is an important consideration in the use of quantitative polarimetric radar data, particularly in using ZDR data (e.g., Zrnić et al. 2006). Despite efforts to keep NOXP well calibrated, and values in datasets from 2009 to 2012 consistently have a positive bias when compared with data from UMass X-Pol. During VORTEX2, UMass X-Pol was calibrated often, using vertically pointing scans in light precipitation (J. Snyder 2013, personal communication). In comparisons of between the two systems (e.g., Fig. 3), values consistently are higher in NOXP data by ~5 dB. Similarly, in light-precipitation comparisons between NOXP and WSR-88D systems in 2009–10 (not shown), a positive ZH bias of 3–4 dB was calculated. As a result, NOXP ZH values were reduced by 5 dB after attenuation correction was performed so as to mitigate the positive bias.

Fig. 3.
Fig. 3.

Comparative PPIs of (dBZ) in a supercell from 5 Jun 2009 as observed by (a) NOXP and (b) UMass X-Pol. The height of the center of the beam at the front edge of the hook echo is ~930 and ~910 m ARL for (a) and (b), respectively. The numbers in parentheses at the bottom of each image, from left to right, are the mean and median in the hook echo for ≥ 20 dBZ.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

In the absence of vertically pointing “birdbath scans” (e.g., Gorgucci et al. 1999), a “scatterer” method was chosen to calibrate NOXP ZDR in this study. A method was used that identifies dry snow aggregates above the bright band, as discussed in Ryzhkov et al. (2005a) for S-band data. In results of scattering simulations at X and S band in Dolan and Rutledge (2009) and Thompson et al. (2014), differences in ZH and ZDR owing to radar frequency for aggregates were found to be negligible. As a result, ZDR calibration followed the general implementation suggested in Ryzhkov et al. (2005a) and that used in Picca and Ryzhkov (2012) for a supercell storm. Dry snow aggregates were identified in the forward flank of the supercell in the ~1.5-km layer above the bright band by identifying gates with 20 ≤ ZH ≤ 35 dBZ (after attenuation and bias correction) and ρHV ≥ 0.97. Dry snow aggregates were isolated for three consecutive volume scans, ideally centered on the analysis time. The offset was then determined by subtracting the assumed true value of 0.2 dB from the mean calculated ZDR.

To evaluate the bias-correction method, data from UMass X-Pol were used to compare ZDR measurements between the two radar systems from nearly identical times (e.g., Fig. 4). Before comparing data from the two radar systems, UMass X-Pol data also were corrected for attenuation and the same criteria were used to isolate dry snow aggregates.3 The ZDR values of presumed aggregates typically ranged from −0.1 to 0.6 dB in UMass X-Pol data, and the mean ZDR was 0.2–0.3 dB. These values are consistent with those found in convection above the bright band in WSR-88D data (V. Melnikov 2013, personal communication) and with recent values found similarly in Oklahoma convection (Homeyer and Kumjian 2015). NOXP ZDR observations above the bright band were more variable and were consistently larger than those in UMass X-Pol data. The UMass X-Pol ZDR observations provide increased confidence that, in correcting for the NOXP ZDR bias, snow aggregates with a true mean ZDR of ~0.2 dB likely were being sampled. Calculated NOXP ZDR biases ranged from 0 to 1.4 dB, were similar within a season, and became progressively smaller from 2009 to 2012. Herein, any reference to ZH or ZDR refers to data that have been corrected for attenuation or differential attenuation and biases, unless otherwise noted.

Fig. 4.
Fig. 4.

The ZHZDR scatterplot of assumed snow aggregates in NOXP (red circles) and UMass X-Pol (blue circles) data from 2220 to 2225 UTC 5 Jun 2009. Data are taken from levels above the bright band with heights ranging from 2 to 4 km ARL. NOXP ZH and ZDR data are binned every 0.5 dB and 0.0625 dB, respectively, during data collection.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Despite the efforts to correct and calibrate ZH and ZDR in NOXP data, it is likely that there are still errors that complicate comparisons with results from previous studies. Scatterplots of ZH and ZDR are the main tools used to estimate hook echo DSDs (see section 3), and therefore quantitative errors or uncorrected biases in ZH and ZDR could lead to inaccurate assessments of the numbers and percentages of small or large drops. Therefore, we emphasize observed differences within the datasets studied here and acknowledge that differences with results from other studies may partially result from nonmeteorological factors (i.e., radar frequency effects and/or ZH and ZDR errors).

b. Mobile soundings

To investigate the near-storm environments of the 11 cases from VORTEX2, data from the National Center for Atmospheric Research (NCAR) and NSSL mobile sounding units were examined. For most cases, soundings were launched at four locations within or near the storm simultaneously every ~1 h: two locations in the inflow and one each in the forward and rear flanks of the storm (e.g., Wurman et al. 2012). All sounding data used were from launches within 75 min of the analysis time from the “far inflow” or “near inflow” soundings. The data were quality controlled by NCAR’s Earth Observing Laboratory; more information about the radiosondes and details about the quality controlling of sounding data can be found in Parker (2014). For the four datasets that were not obtained during VORTEX2 and lacked near-storm sounding data, surface data from the Oklahoma Mesonet were analyzed; these data were for within 5 min of the analysis time and were obtained from the closest station located in the warm sector downstream of the supercell (distances from the hook echoes to the Mesonet stations varied from 30 to 85 km).

3. Observations of hook echoes

The observations of supercell hook echoes presented herein are divided into three sections, each discussing estimated DSDs in hook echoes using ZDR as a proxy for the median raindrop size in a volume. To better ensure that radar gates being analyzed are characterized by rain, a threshold of ρHV ≥ 0.95 was applied to all hook echo data. In sensitivity tests using ρHV cutoffs ranging from 0.93 to 0.99, there were no significant changes in the ZH or ZDR distributions. The hook echo was subjectively isolated from the rest of the supercell by estimating the location where the change in the width of the spine of the hook echo with range or azimuth (depending on where it was located relative to the radar) was locally maximized (i.e., where the appendage meets the body of the storm and reflectivity flares out horizontally; Fig. 5).

Fig. 5.
Fig. 5.

NOXP PPIs of ZH (dBZ) in supercells from four cases in which hook echo data are examined for this study: tornadic supercells on (a) 5 Jun 2009 in Goshen County and (b) and 19 May 2010 near Kingfisher and nontornadic supercells on (c) 23 May 2011 near Watonga, Oklahoma and (d) 28 May 2012 near Rush Springs, Oklahoma. The supercell shown in (d) has a hook echo with anticyclonic curvature. The black lines mark the subjective isolation of the hook echo from the rest of the supercell echo. In the tornadic supercells, the location of the tornadic vortex signature (TVS) associated with the tornado is marked by a white circle. The demarcation line in (b) is not located farther north because of strong attenuation and differential attenuation at those locations. Range rings are every 2.5 km. The approximate heights of the observations are shown in Table 2.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

a. Generalized hook echo DSDs

To estimate DSDs in supercell hook echoes, a single scan each from 15 cases was chosen (Table 2). Time-averaged hook echo data from multiple volumes were not used because short deployment times and low data-quality consistency would lead to either uneven averaging among the 15 datasets or consistent averaging in a smaller sample of cases. As a result, the estimated hook echo DSDs in any single case may not be representative of the lifetime of the hook echo. Within datasets that had several scans of usable hook echo data, generally the scan that was closest to the radar was chosen so that spatial resolution was maximized and the height of the center of the beam above the ground was minimized. Of the hook echoes examined, five were tornadic at the time of data collection (e.g., Figs. 5a,b), one was tornadic ~6 min after the analyzed time,4 and the remaining nine were nontornadic (e.g., Figs. 5c,d). In some of the nontornadic cases, the supercells were tornadic at some point but not within ±35 min of the time analyzed.

Table 2.

A list of the hook echo datasets examined in this study along with several descriptors. The classification of tornadic or nontornadic pertains only to the time being examined. The EF rating of the tornado for the tornadic hook echo datasets is also provided.

Table 2.

To assess the estimated DSDs in hook echoes, scatterplots of ZH versus ZDR are shown for six tornadic cases (Fig. 6) and nine nontornadic cases (Figs. 7 and 8). In each of the scatterplots, there appears a line representing a relation between ZH and ZDR that was developed for Oklahoma precipitation (both stratiform and convective) by Cao et al. (2008). Points that lie above (below) the line represent radar gates containing drops with a median size that is larger (smaller) than what would be expected in Oklahoma precipitation given the ZH value.5 The relation was specifically derived at S band rather than X band, and so there may be inherent slight differences in interpretation that are being ignored in this study. Also, although many of the datasets that were analyzed here were not obtained in Oklahoma, most were obtained in similar geographic areas. The use of the Cao et al. (2008) line simplifies comparisons with the cases that were investigated in K11.

Fig. 6.
Fig. 6.

Scatterplots of ZH vs ZDR for NOXP data from six tornadic hook echoes: (a) 5 Jun 2009 in Goshen County; (b) 19 May 2010 near Kingfisher; (c) 25 May 2010 in Tribune, Kansas; (d) 8 Jun 2010 near Scottsbluff, Nebraska; (e) 12 Jun 2010 near Limon, Colorado; and (f) 30 May 2012 near Crowell, Texas. The black line marks a relation between ZH and ZDR that was developed by Cao et al. (2008) and is based on a large amount of DSD data collected in Oklahoma rain. Only data with ρHV ≥ 0.95 are shown and analyzed. In (a), data from UMass X-Pol from the same approximate time and height as those from NOXP are shown for comparison. The number of data points with ZH ≥ 20 dBZ is shown at the top or bottom left (for convenience, gates with ZH ≥ 60 dBZ are not shown). The approximate center-beam height level ARL (m) of the hook echo is shown at the bottom or top right.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Fig. 7.
Fig. 7.

As in Fig. 6, but using data from three nontornadic hook echoes: (a) 23 May 2011 near Carnegie, Oklahoma; (b) 28 May 2012 near Rush Springs; and (c) 30 May 2012 near Kingfisher.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Fig. 8.
Fig. 8.

As in Fig. 6, but using data from six nontornadic hook echoes: (a) 6 Jun 2009 near Sidney; (b) 6 Jun 2009 near Mullen, Nebraska; (c) 7 Jun 2009 near Savannah, Missouri; (d) 9 Jun 2009 in Greensburg; (e) 14 Jun 2009 near Amarillo, Texas; and (f) 26 May 2010 near Hoyt, Colorado.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

In all of the cases but one, the hook echoes of the examined datasets have a majority of points lying above the Cao et al. (2008) line (hereinafter C08L; Figs. 68). This is true for both tornadic and nontornadic cases and can be interpreted as an indication that most gates contained drops with a median size that was about that expected in Oklahoma precipitation or larger for a given ZH. As a check for the bias-corrected NOXP ZDR data, data from UMass X-Pol are shown for a time and height that is comparable to those from NOXP (Fig. 6a) for the first case. The pair of scatterplots shows very similar ZH and ZDR values and provides confidence that the attenuation-correction and bias-correction methods used on NOXP data that are detailed in section 2 are sound.

For the tornadic cases, a supercell hook echo from 5 June 2009 (Fig. 6a) does not have many points that lie below the C08L. However, the remaining five tornadic cases (Figs. 6b–f) all have several hundred data points that lie below the C08L. The nontornadic cases were split between those that had a large number of points below the C08L (Fig. 7) and those that had very few points below the C08L (Fig. 8). The datasets have widely varying numbers of hook echo gates (e.g., Fig. 7a vs Fig. 8b) owing to the variable amounts of precipitation in supercell hook echoes and the different ranges from NOXP to the hook echoes. As a result, the prevalence of small-drop gates was estimated by calculating mean ZDR values and the percentages of gates with relatively large ZH and small ZDR values (Table 3). All of the tornadic datasets except the 5 June 2009 case have at least 19% of their gates lying below the C08L, whereas only 3 of the 10 nontornadic datasets do. The results are qualitatively similar using percentages calculated requiring the gates to have ZDR values that are at least 0.5 and 1.0 dB below the C08L. In addition, the five cases with the largest mean ZDR for 40 ≤ ZH < 50 dbZ are all nontornadic. Hereinafter, the cases are split between “small drop” and “large drop” hook echo cases using a natural break between the cases with at least 19.1% and not more than 10.4% of points lying below the C08L (Table 3), respectively. The large gap in percentages of small-drop gates between the two groups mitigates the effects that any errors in ZH and ZDR may have on the subsequent interpretation.

Table 3.

A list of the percentages of eligible radar gates lying below various increments of the C08L, and mean ZDR values for the 15 hook echo datasets examined in this study. The tornadic datasets are shown first and appear in boldface.

Table 3.

The nontornadic cases are examined in greater detail to determine whether there are any obvious differences between the small-drop and large-drop subsets. For example, all nine of the nontornadic supercells have low-level rotation signatures in radial velocity, which is not unexpected given that they all have a low-level hook echo. In the three nontornadic, small-drop cases, there is easily identifiable low-level rotation (Fig. 9). In the latter two cases, from 28 May 2012 (a “left moving” supercell; Fig. 9b) and 30 May 2012 (Fig. 9c), there are also smaller-scale vortex signatures (white circles). In the 30 May 2012 case, the strength of the vortex signature at 40-km range (gate-to-gate ΔV of ~35 m s−1) is some evidence that the supercell may have been tornadic at this time (the storm was tornadic ~35 min later). The nontornadic large-drop cases have varying levels of low-level rotation (not shown) that range from strong (e.g., ΔV ≈ 30 m s−1 on 9 June 2009) to only broad and weak (e.g., ΔV ≈ 15 m s−1 in the second case from 6 June 2009).

Fig. 9.
Fig. 9.

NOXP PPIs of (top) ZH (dBZ) and (bottom) Vr (m s−1) in three nontornadic supercells that were examined in this study: (a) 23 May 2011 and (b) 28 and (c) 30 May 2012. The black circles outline low-level mesocyclone-scale rotation in the hook echoes. The white circles in (b) and (c) outline smaller-scale rotation. Note in (b) that the rotation is anticyclonic in the left-moving supercell. The color-coded numbers are the maximum ΔV in the corresponding mesocyclones or vortex signatures. Range rings are every 1.0 km. The approximate heights of the observations at the range of the hook echo are shown in Table 2.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

The near-storm environments of the 11 supercell cases from VORTEX2 also are examined. Several thermodynamic and kinematic quantities were calculated using data from the sounding that is nearest to the analysis time used in this study. Using all of the available data, scatterplots and line graphs of selected environmental variables are shown (Figs. 10 and 11). There is no obvious difference in surface-based CAPE (calculated from 0 to 9 km AGL) or 0–6-km bulk wind difference (BWD) between the large- and small-drop cases (Fig. 10a). The small-drop cases (all tornadic) do constitute three of the five highest 0–1-km BWD values, four of the five highest 0–3-km storm-relative helicity (SRH) values (Fig. 10b), and four of the six highest 0–1-km SRH values (not shown).

Fig. 10.
Fig. 10.

Scatterplots of environmental variables from near-storm soundings for the 11 VORTEX2 hook echo cases used in this study: (a) CAPE calculated from the surface to 9 km vs the BWD magnitude between the surface and 6 km and (b) SRH from the surface to 3 km vs the wind-difference magnitude between the surface and 1 km. Triangles (circles) indicate tornadic (nontornadic) hook echo cases. The calculations of CAPE include the virtual temperature correction and only progress up to 9 km AGL because the soundings used in the calculations terminated at different heights. The BWD calculations use the wind speed–magnitude data points that are closest to 10 m in height for the bottom of the layer. Storm motion was estimated by following the storm’s inflow notch or, if that feature was not able to be tracked reliably, the forward-flank reflectivity maximum over a 10–20-min period centered on the analysis time. All heights are AGL from the sounding launch location. Note that, for the VORTEX2 cases, all of the small-drop cases are tornadic cases.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Fig. 11.
Fig. 11.

Line plots of (a) near-surface RH (%) and (b) height of the LCL (m AGL) for the 15 hook echo cases used in this study. Triangles (circles) indicate tornadic (nontornadic) hook echo cases. LCL height calculations for the 2011–12 cases use Oklahoma Mesonet temperature and RH measurements at 1.5 m.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Environmental parameters could not be calculated for the four cases from 2011 to 2012 because there were no near-storm soundings. Three of the cases constitute the entire nontornadic, small-drop hook echo set, and therefore comparisons could not be made with the nontornadic, large-drop set. All four cases from 2011 to 2012 occurred in or near Oklahoma, however, allowing for near-storm surface data from the Oklahoma Mesonet to be analyzed. The surface relative humidity (RH; Fig. 11a) and lifting condensation level (LCL) heights (Fig. 11b) are shown for all 15 cases. The LCL heights for cases from 2011 to 2012 were estimated using iterative calculations. The small-drop cases, both tornadic and nontornadic, consistently have higher RHs and lower LCL heights than do the large-drop cases.

In summary, while acknowledging both the small sample of cases (11–15) and the sensitivity of environmental parameters in VORTEX2 cases to the distance from the sounding to the storm (Parker 2014), the near-storm environments of the small-drop cases generally have higher 0–1-km BWD, 0–3- and 0–1-km SRH, and near-surface RH and lower LCL heights than those in the large-drop cases.

b. Spatial distribution of DSDs

Within the hook echo, the spatial distribution of approximated DSDs is examined.6 The ZDR plots of the hook echo are shown for five of the tornadic supercells (Figs. 12a–e). In several cases, the largest ZDR values are found on the inner (forward) edge of the hook echo, usually collocated with maxima in ZH (e.g., white lines in Fig. 12d). The remaining cases do not have a clear gradient in ZDR within the hook echo. The same is true for the nontornadic cases (e.g., Fig. 12f), whereby some hook echoes have a clearly defined maximum in ZDR at the inner edge of the hook collocated with a gradient in ZH and several do not. Some of the nontornadic hook echoes have a “nontraditional” ZH appearance in which, instead of a cyclonic ZH appendage, there is an elongated appendage (e.g., Fig. 9c) or a ZH field in the rear of the storm that takes on a more complicated shape. These cases in particular do not have clearly defined areas of ZDR maxima.

Fig. 12.
Fig. 12.

NOXP PPIs of (top) ZH (dBZ) and (bottom) ZDR (dB) in (a)–(e) five tornadic cases and (f) one nontornadic supercell example case used in this study. The black circles in (a)–(e) mark the locations of the TVSs associated with tornadoes. The arrows in (a)–(e) mark locations of local minima in ZDR, and the white lines in (d) and (f) mark the local maxima in ZDR on the inner half of the hook echo. The black lines in (d) outline the quadrants shown in Fig. 13 and discussed in the text, and the dotted black line is the approximate subjective cutoff for the hook echo. The only ZDR gates shown are those with ZH > 20 dBZ. Range rings are every 1.0 km. The approximate heights of the observations at the range of the hook echo are shown in Table 2.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

In each of the five tornadic cases shown, there is a local minimum in ZDR south and/or southeast of the tornado (white arrows in Figs. 12a–e). In some of these cases, the local minimum is striking in its relatively low ZDR value and the feature’s geographic isolation within the hook echo (e.g., Fig. 12d). In most of the nontornadic cases, there is either no obvious local minimum in ZDR within the hook echo or there is one that is not to the south or southeast of the main circulation (e.g., Fig. 12f). In three of the tornadic cases in which local ZDR minima are identified, there are data before and/or after the scan in question, and therefore the temporal continuity of the local ZDR minima can be investigated (see section 3c).

A disadvantage of looking only at “texture” features in identifying the local minimum south and southeast of the tornado in the tornadic cases is that ZH and ρHV information is left out. For the tornadic cases, ZHZDR scatterplots were reanalyzed (using the same ρHV cutoff) with the points color coded on the basis of their location within the hook echo (Fig. 13). Four quadrants were identified relative to the location of the TVS and its heading angle in the hook echo for each case: centered northeast, southeast, southwest, and northwest of the TVS (e.g., Fig. 12d). TVS motion was calculated by tracking the TVS or the low-level rotation signature that preceded it over a 4-min period centered on the analysis time. In the 5 June 2009 case (Fig. 13a), the area of depressed ZDR is associated with an area of low ZH rather than a region with DSDs that are skewed toward small raindrops. In the other four cases (Figs. 13b–e), most of the points that lie below the C08L lines occur in the forward-right and rear-right quadrants, consistent with the texture features identified previously. Note that in some cases a large percentage of points within the rear-left quadrant lie below the C08L (e.g., Figs. 13d,e), likely owing to the inclusion of points in the rear of the spine of the hook echo, where drop sizes diminish (Fig. 1; Fig. 12d). The forward-left quadrant consistently has a small percentage of gates with assumed small median drop sizes.

Fig. 13.
Fig. 13.

As in Fig. 6, but with the gates color coded by where they were located relative to the TVS heading angle. The locations are centered on quadrants located to the forward right, rear right, rear left, and forward left relative to the location of the TVS associated with each tornado (black circle) and the TVS’s heading angle (black arrow; up points due north). An example of the quadrants is shown in the top left in (a). The centroid for each quadrant, calculated only for ZH ≥ 30 dBZ, is marked by a large colored circle. To avoid quadrants with fewer data points being blocked, the plots for each quadrant are layered such that the quadrant with the fewest (most) data points appears on the top (bottom). The percentage of total gates with ZH > 30 dBZ lying below the C08L is shown for the quadrants in each case.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

c. DSD relationship with tornado life cycle

Of the sample of six tornadic supercells used in this study, tornadogenesis (tornado dissipation) data were obtained in one7 case (three cases). In one case with both tornadogenesis and tornado dissipation data (12 June 2010), the tornado was short lived and was rated as category 0 on the enhanced Fujita scale (EF0). In the remaining two cases (5 June 2009 and 19 May 2010), the tornadoes were longer lived and were rated EF2 and EF1, respectively. In these three cases, we examined whether there is any obvious relationship between the tornado life cycle and the estimated median hook echo drop size in qualifying radar volumes.

In the tornadogenesis case, NOXP data collection of the storm’s hook echo began ~8 min prior to tornadogenesis. From NOXP data and the report in the NOAA Storm Data publication, the tornado likely lasted less than 10 min, and so the progression of the approximated hook echo DSD during the tornado’s life cycle was captured in the lowest 1 km every 2 min. In this case, a prominent area of locally reduced ZDR, likely relatively small drops, was identified to the southwest of the tornado (Figs. 12d and 13d). The progression of the hook echo ZDR before, during, and after tornado formation8 is shown in Fig. 14. More than ~5 min prior to tornado formation, there was not an area of reduced ZDR in the hook echo (Fig. 14a). An area of low ZDR was subsequently identified in the southwest portion of the hook echo (Fig. 14b), which then expanded rapidly prior to or during tornado formation (Figs. 14c,d). The area of low ZDR remained south of the TVS in association with the tornado (Fig. 14e) before the texture feature becomes nebulous by 0114 UTC (Fig. 14f). The TVS weakens substantially by 0116 UTC before it no longer can be identified in the NOXP data by 0118 UTC (not shown). Also note an additional area of precipitation, perhaps a new cell, to the west and southwest of the hook echo beginning at 0110 UTC (black arrows in Figs. 14d,e) that eventually merges with the hook echo in the next 4 min (Figs. 14e,f).

Fig. 14.
Fig. 14.

As in Fig. 12, but for one tornadic case from 12 Jun 2010, shown every ~2 min from 0104 to 0114 UTC. The white arrow in (c) marks the area of low ZDR that developed prior to or as the tornado formed. The black arrows point to a cell that merged with the supercell.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

The change in the estimated DSD in the hook echo prior to and during tornado formation was also examined quantitatively9 (Figs. 15 and 16). A series of nine ZHZDR scatterplots are shown that begin ~8 min prior to tornadogenesis and go through tornado dissipation (0102–0118 UTC; Fig. 15). Both the total number of hook echo gates and the number and percentage of points lying below the C08L increase in the minutes leading up to tornado formation (Figs. 15a–f). As the tornado dissipates (Figs. 15g–i) and the cell merger is completed, the number of small- and large-drop gates continues to increase. The temporal progression of the estimated DSDs for this case is more clearly seen in time series of mean ZDR and the percentage of small and large drops in the hook echo (Fig. 16). Because of the C08L’s increasingly positive slope with increasing ZH, mean ZDR was calculated at three binned ZH ranges: 30–40, 40–45, and 45–50 dBZ. Consistent with K11, small- (large-) drop gates are defined as being at least 1.0 dB below (above) the C08L given the gate’s ZH value. The mean ZDR at all three ZH ranges decreases by ~1.0 dB, and the percentage of large- (small-) drop gates decreases (increases) substantially in the ~8 min leading up to the time of tornadogenesis. The decrease in the percentage of large-drop gates from ~55% to ~20% is exactly coincident with an increase in small-drop gates from ~1% to ~12%. These trends are consistent with the appearance of the region of depressed ZDR to the southwest of the main low-level rotation in the hook echo shown in Fig. 14. In addition, mean ZDR at 30 ≤ ZH < 40 dBZ and the percentage of large-drop gates both increase before and as the tornado is dissipating and during the cell merger.

Fig. 15.
Fig. 15.

As in Fig. 6, but for one case on 12 Jun 2010 at (a) 0102:20, (b) 0104:21, (c) 0106:19, (d) 0108:20, (e) 0110:18, (f) 0112:19, (g) 0114:21, (h) 0116:18, and (i) 0118:32 UTC.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Fig. 16.
Fig. 16.

Time series of hook echo (a) mean ZDR (dB) for three ranges of binned ZH values and (b) the percentage of radar gates characterized by large and small drops before, during, and after the formation of a tornado on 12 Jun 2010. Also shown is the maximum ΔV in the TVS associated with the tornado (ordinate scale on the right). The percentages of large- and small-drop gates were determined in the manner described in the text.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

The data obtained by NOXP prior to tornadogenesis extend up to 7.0°-elevation angle or to heights of near 3 (2) km at 0102 (0114) UTC (not shown). Volumetric data were examined to see how the progression in the number of large- and small-drop gates changes with time and height. The data were separated into 200-m height bins, and the numbers of large- and small-drop gates were calculated (Fig. 17). Prior to tornado formation, the number of large-drop gates (Fig. 17a) decreases at progressively lower levels with time, decreasing first above 1.5 km ARL at 0102–0104 UTC (times refer to the time at which a volume scan began), then at 1–2 km at 0104–0106 UTC, and then last in the lowest 1 km from 0106 to 0108 UTC. At the same time, the number of small-drop gates (Fig. 17b) increases at progressively lower levels as tornadogenesis nears. The observed local maximum in small-drop gates is at 2–3 km at 0102–0104 UTC, then at 1.5–2 km at 0106 UTC, and then in the lowest 1 km at 0108 UTC. For the small-drop set in particular, the “rate of descent” of the small-drop maximum is larger than the fall speeds of raindrops, providing some evidence of advection of the drops by a downdraft. During tornado dissipation, the increase in large-drop gates (Fig. 17c) is first observed at ~1.5 km at 0112 UTC; this is followed by a large increase in large-drop gates in the lowest ~1.25 km at 0114 UTC. The number of small-drop gates (Fig. 17d) remains similar in the lowest 1 km during tornado dissipation and the cell merger.

Fig. 17.
Fig. 17.

The number of radar gates characterized by (left) large and (right) small raindrops at several heights (a),(b) before and (c),(d) after the formation of a tornado on 12 Jun 2010. The numbers of small and large drops were binned every 200 m from 200 to 3100 m ARL. A bin was only included if at least one-half of the height level was sampled (e.g., the 400–600-m bin was included only if the lowest observation was from below 500 m ARL). Note the differing abscissa scales for the large and small-drop plots.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

In the other two cases, an EF2 tornado on 5 June 2009 in Goshen County, Wyoming (e.g., Wakimoto et al. 2011, 2012; Markowski et al. 2012a,b; Atkins et al. 2012; Wurman et al. 2013; Kosiba et al. 2013; French et al. 2013, 2014), and an EF1 tornado on 19 May 2010 near Kingfisher, Oklahoma (e.g., French et al. 2014), NOXP data were only obtained during the dissipation phases of the tornadoes. Both of these cases also were examined to determine changes in mean ZDR using binned ZH values and changes in the percentage of small- and large-drop gates (Fig. 18).10 The ZH-binned ZDR and the percentage of large-drop gates both increase prior to tornado dissipation in the Goshen County case (Figs. 18a,b) while the percentage of small-drop gates shows little change. In the Kingfisher case, mean ZDR for ZH ≥ 40 dBZ increases prior to and during TVS dissipation (Fig. 18c), but the percentage of small- and large-drop gates stays nearly constant (Fig. 18d).

Fig. 18.
Fig. 18.

As in Fig. 16, but for the dissipation of tornadoes on (a),(b) 5 Jun 2009 and (c),(d) 19 May 2010. The TVS ΔV data are from the lowest-observed MWR-05XP level and are smoothed using a 1–2–1 filter.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

As a means of comparison for the tornado-life-cycle cases, the time evolution of bulk DSDs also was investigated for two nontornadic cases: one large-drop case (26 May 2010) and one small-drop case (30 May 2012). In the former case (Fig. 19), the hook echo becomes more cyclonically curved in the 10 min leading up to the analysis time and low-level rotation becomes stronger and more organized (Fig. 19a). From 2310 to 2318 UTC (Figs. 19b,c), mean ZDR decreases and the number of large drops decreases in each consecutive volume; the percentage of small drops increases during this time period, but the values are small (<5%). In the 30 May 2012 case (Fig. 20), the amount of hook echo curvature changes often and there are several distinct TVSs identified during the analysis time (Fig. 20a). During this time period, mean ZDR fluctuates by ~0.5 dB; the percentages of small and large drops also vary accordingly (Figs. 20c,d). The changing values and the multiple areas of rotation may be an indication that the supercell was undergoing cyclic mesocyclogenesis at this time (e.g., K11), but the long range from NOXP to the storm and the 3-min volumetric update time hinder data interpretation. The time evolution of other nontornadic cases could not be investigated because of the small number of scans with a hook echo, variable amounts of precipitation at the radar site, or poor data quality.

Fig. 19.
Fig. 19.

(a) NOXP PPIs of (top) ZH (dBZ) and (bottom) VR (m s−1) every ~4 min from 2304 to 2318 UTC 26 May 2010 for a nontornadic supercell hook echo. (b),(c) As in Fig. 16, but during the evolution of a hook echo in a nontornadic supercell. In (a) the range rings are every 1.0 km, and the approximate heights of the observations at the range of the hook echo are shown in Table 2.

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

Fig. 20.
Fig. 20.

As in Fig. 19, but from 0045 to 0103 UTC 30 May 2012 for a nontornadic supercell hook echo. Distinct TVSs are enclosed by color-coded circles. Note that the ordinate scales differ from Fig. 19 in (b) and (c).

Citation: Journal of Applied Meteorology and Climatology 54, 2; 10.1175/JAMC-D-14-0171.1

4. Summary and discussion

The observations shown in sections 3a and 3b are largely consistent with those discussed in previous studies. Similar to results from Kumjian and Ryzhkov (2008a), tornadic cases generally have lower mean/median ZDR for a given ZH than do the nontornadic cases. In addition, as in K11, there is an area of presumed small raindrops located to the south of the tornado in several of the tornadic hook echo cases. There are some minor differences in the location of the large drops in this study in comparison with K11: they are collocated with the highest ZH rather than along the ZH gradient. The greater ZH for large raindrops at X band than at S band is one possible explanation. Also, cases with cyclonic curvature in the hook echo (and therefore strong low-level rotation) are more likely to have the large-drop signature, which is consistent with the idea posited in K11 that the large-drop signature is caused by size sorting in the low-level updraft.

A more prominent difference from the results shown in K11 is that the relative frequency of points lying below the C08L in the cases examined here is comparatively low. Possibilities for these differences include the geographic impacts on the C08L for cases from different regions and the small sample of cases used in both studies. As mentioned previously, the effects of radar frequency on ZH and ZDR at X band and uncorrected biases in NOXP ZH and ZDR values also cannot be ruled out as the cause for the observed differences. Investigating these possibilities requires examination of additional hook echo cases beyond the tornadic cases already studied here and in previous works.

Two topics that were not examined in detail in previous studies are the near-storm environments of small-drop and large-drop hook echoes and the relationship between estimated hook echo raindrop sizes and the tornado life cycle. The small-drop cases all have easily identifiable low-level rotation and, in the cases in which sounding data were available, relatively high 0–1-km BWD and 0–3- and 0–1-km SRH. In addition, they generally have larger surface RHs and lower LCL heights than the large-drop cases. In one case, prior to tornado formation the number of small-drop gates increases at progressively lower levels at a rate greater than the drop fall speed, consistent with vertical advection of small raindrops by a downdraft. In three tornado dissipation cases, the raindrop size generally increases before and during tornado decay, although one of these cases was also affected by a cell merger. In one large-drop, nontornadic case, median drop sizes decrease as low-level rotation increases, and, in one small-drop, nontornadic case, bulk DSD changes may be related to storm cycling.

In Kumjian and Ryzhkov (2008a), it was suggested that nontornadic hook echoes may have greater median ZDR and more large-drop gates than do tornadic hook echoes because of greater rates of evaporation in the attendant storm RFD. Further, in K11 a hypothesis was presented for the presence of small drops in hook echoes and their location to the south and east of the tornado: dynamically driven downdrafts that bring small drops to the surface in a relatively moist boundary layer. An occlusion downdraft would be associated with an enhanced region of smaller drops. The observations discussed in this study are consistent with both ideas, although without direct evidence linking dynamically driven downdrafts or the evaporation rate to the small-drop regions.

The formation of supercell tornadoes is thought to require the formation of a downdraft that advects existing or developing vertical vorticity to near the surface (e.g., Davies-Jones et al. 2001; Markowski and Richardson 2009). A dynamically driven downdraft can occur when strong low-level rotation develops, which aids in generating downward-directed VPPGFs. Also, tornado formation is more likely when RFD air is more positively buoyant (Markowski et al. 2002) and low-level dynamic lifting is large (e.g., Markowski and Richardson 2014). Consistent with these ideas, past studies have shown that two of the best discriminators between tornadic and nontornadic supercells are lower (higher) LCL height and larger (smaller) 0–1-km SRH (e.g., Thompson et al. 2003). Lower LCL heights are associated with less evaporation of rain in an RFD and more positively buoyant RFD outflow air. Higher 0–1-km SRH is likely related to larger upward-directed VPPGFs and more stretching of existing vertical vorticity. The relationship between tornado dissipation and the thermodynamic state of RFD outflow air is not as well understood, but changes in RFD strength have been shown to precede tornado demise in some cases (Marquis et al. 2012).

On the basis of the observations presented here and in Kumjian and Ryzhkov (2008a) and K11, we speculate generally on evaporation rates in a dynamically driven and/or occlusion downdraft as a link between the near-storm environment, bulk hook echo DSD, and tornado production and life cycle. In tornadic supercells, there is a necessary amount of low-level wind shear (e.g., Fig. 10b) for a strong low-level updraft to be established; in turn, the rotation in the low-level mesocyclone aids in forcing a dynamic downdraft. The storm environment also exhibits a moist boundary layer with low LCL heights (e.g., Fig. 11, black triangles), and so the downdraft can bring small drops to progressively lower levels quickly (e.g., Fig. 17b) with relatively little evaporation or without completely evaporating the drops (e.g., Fig. 6) until they reach the surface (e.g., Figs. 1416). The relatively low evaporation rates responsible for the small drops also allow for positively buoyant (or smaller magnitudes of negatively buoyant) air to reach the surface. In addition, the environment also has relatively large 0–1-km SRH, which allows for upward-directed VPPGFs that are strong enough to stretch existing vertical vorticity. The combination of necessary amounts of positively (or small negatively) buoyant air and low-level dynamic lifting may produce a tornado if there is rotation near the ground,11 as discussed in Markowski and Richardson (2014). If the evaporation rate increases (e.g., because of a variable storm environment or a change in the strength of the downdraft), there is an increase in the median drop size (e.g., Figs. 16 and 18). This air is likely to be more negatively buoyant and to be associated with a change in the strength of the RFD, which can contribute to tornado demise.

In other supercells, particularly those with strong low-level rotation and low LCL heights (e.g., Fig. 11, black circles), a relatively high number of small drops also can be brought to the surface (e.g., Fig. 7) by a downdraft; these conditions are not, by themselves, sufficient for tornado formation (e.g., Trapp 1999), however. Rotation may not be established near the surface or low-level VPPGFs may be too weak for sufficient amounts of vertical vorticity stretching if low-level wind shear diminishes. In still other cases, weaker low-level rotation inhibits downdraft formation or higher LCL heights (e.g., Fig. 11, blue circles) lead to more evaporation in the downdraft. Fewer small drops are brought to the surface (e.g., Fig. 8), and colder RFD temperatures lessen the probability of tornado formation.

We caution again that acceptance of these ideas requires more than the circumstantial evidence presented in K11 and this overview study. Future work should focus on two separate questions that are similar in spirit but that are ideally addressed using two different sets of polarimetric radar data. The first question to address is What is the statistical relationship, if any, between the estimated sizes of raindrops in supercell hook echoes and the storm’s propensity to produce tornadoes? This includes the possibilities 1) of a diagnostic relationship between small hook echo drop sizes and tornadoes and 2) that an increase in the number of small drops can portend tornado development in some cases. Statistical relationships are best assessed using a large enough sample of supercell hook echo data from the WSR-88D network that a climatological description can be constructed. Also, quantitative problems like the ones encountered in this study are likely to be of far less concern when using data from an S-band radar network.

The second question to address is: What is the dynamic relationship between the estimated sizes of raindrops in supercell hook echoes and the storm’s propensity to produce tornadoes? The idea that the production of large numbers of small raindrops in some hook echoes may be caused by low evaporation rates in dynamically driven downdrafts currently lacks any direct observational evidence. In the ideal case, the large amount of integrated data obtained in VORTEX2 and other field projects, including high-resolution polarimetric radar, dual-Doppler radar, disdrometer, and thermodynamic data, can be used to investigate the relationship between estimated and directly observed DSDs and relevant storm dynamic and thermodynamic processes. Despite the limited number of cases with fully integrated data, we hope that selected case studies will provide more conclusive answers as to the cause of hook echo DSD heterogeneities.

Acknowledgments

The authors appreciate helpful discussions with Joey Picca, Valery Melnikov, Christopher Schwarz, and Jeffrey Snyder. The latter also provided an attenuation-correction code. Howie Bluestein provided UMass X-Pol data for comparison purposes. Sounding data were provided by NCAR/EOL under sponsorship of the National Science Foundation. Thanks are given also to Patrick Skinner for assistance with sounding data. We are indebted to the VORTEX2 crews for their work in obtaining the radar and sounding data used in this study. Constructive feedback on this manuscript from Matthew Kumjian and two anonymous reviewers improved several important points. Funding was provided by NOAA/OAR under NOAA–University of Oklahoma Cooperative Agreement NA11OAR4320072, U.S. Department of Commerce. This work was completed while the first author was a National Research Council postdoctoral research associate at the National Severe Storms Laboratory.

REFERENCES

  • Atkins, N. T., A. McGee, R. Ducharme, R. M. Wakimoto, and J. Wurman, 2012: The LaGrange tornado during VORTEX2. Part II: Photogrammetric analysis of the tornado combined with dual-Doppler radar data. Mon. Wea. Rev., 140, 29392958, doi:10.1175/MWR-D-11-00285.1.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., M. M. French, R. L. Tanamachi, S. Frasier, K. Hardwick, F. Junyent, and A. L. Pazmany, 2007: Close-range observations of tornadoes in supercells made with a dual-polarization, X-band, mobile Doppler radar. Mon. Wea. Rev., 135, 15221543, doi:10.1175/MWR3349.1.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., M. M. French, I. PopStefanija, R. T. Bluth, and J. B. Knorr, 2010: A mobile, phased-array Doppler radar for the study of severe convective storms: The MWR-05XP. Bull. Amer. Meteor. Soc., 91, 579600, doi:10.1175/2009BAMS2914.1.

    • Search Google Scholar
    • Export Citation
  • Bringi, V. N., and V. Chandrasekar, 2001: Polarimetric Doppler Weather Radar: Principles and Applications. Cambridge University Press, 636 pp.

  • Bringi, V. N., V. Chandrasekar, N. Balakrishnan, and D. S. Zrnić, 1990: An examination of propagation effects in rainfall on radar measurements at microwave frequencies. J. Atmos. Oceanic Technol., 7, 829840, doi:10.1175/1520-0426(1990)007<0829:AEOPEI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Bringi, V. N., T. D. Keenan, and V. Chandrasekar, 2001: Correcting C-band radar reflectivity and differential reflectivity data for rain attenuation: A self-consistent method with constraints. IEEE Trans. Geosci. Remote Sens., 39, 19061915, doi:10.1109/36.951081.

    • Search Google Scholar
    • Export Citation
  • Burgess, D. W., C. M. Schwarz, J. Snyder, M. M. French, H. B. Bluestein, and C. L. Ziegler, 2012: A cyclic, tornadic supercell on 10 May 2010: Analysis of a VORTEX2 case. Proc. 35th Conf. on Radar Meteorology, Pittsburgh, PA, Amer. Meteor. Soc., 8B.4. [Available online at https://ams.confex.com/ams/35Radar/webprogram/Paper191297.html.]

  • Cao, Q., G. Zhang, E. Brandes, T. Schuur, A. Ryzhkov, and K. Ikeda, 2008: Analysis of video disdrometer and polarimetric radar data to characterize rain microphysics in Oklahoma. J. Appl. Meteor. Climatol., 47, 22382255, doi:10.1175/2008JAMC1732.1.

    • Search Google Scholar
    • Export Citation
  • Davidson, E. A., and D. W. Burgess, 2013: Radar studies of a VORTEX2 tornadic supercell. 36th Conf. on Radar Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 177. [Available online at https://ams.confex.com/ams/36Radar/webprogram/Paper228668.html.]

  • Davies-Jones, R. P., R. J. Trapp, and H. B. Bluestein, 2001: Tornadoes and tornadic storms. Severe Convective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 167–221.

  • Dawson, D. T., II, E. R. Mansell, Y. Jung, L. J. Wicker, M. R. Kumjian, and M. Xue, 2013: Comparisons of numerically simulated and observed low-level polarimetric signatures in supercells. 36th Conf. on Radar Meteorology, Breckenridge, CO, Amer. Meteor. Soc., 12B.6. [Available online at https://ams.confex.com/ams/36Radar/webprogram/Paper229115.html.]

  • Delrieu, G., S. Caoudal, and J. D. Creutin, 1997: Feasibility of using mountain return for the correction of ground-based X-band weather radar data. J. Atmos. Oceanic Technol., 14, 368385, doi:10.1175/1520-0426(1997)014<0368:FOUMRF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Dolan, B., and S. A. Rutledge, 2009: A theory-based hydrometeor identification algorithm for X-band polarimetric radars. J. Atmos. Oceanic Technol., 26, 20712088, doi:10.1175/2009JTECHA1208.1.

    • Search Google Scholar
    • Export Citation
  • Doswell, C. A., III, and D. W. Burgess, 1993: Tornadoes and tornadic storms: A review of conceptual models. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, Geophys. Monogr., Vol. 79, Amer. Geophys. Union, 161–172.

  • French, M. M., H. B. Bluestein, I. PopStefanija, C. A. Baldi, and R. T. Bluth, 2013: Reexamining the vertical development of tornadic vortex signature in supercells. Mon. Wea. Rev., 141, 45764601, doi:10.1175/MWR-D-12-00315.1.

    • Search Google Scholar
    • Export Citation
  • French, M. M., H. B. Bluestein, I. PopStefanija, C. A. Baldi, and R. T. Bluth, 2014: Mobile, phased-array, Doppler radar observations of tornadoes at X band. Mon. Wea. Rev., 142, 10101036, doi:10.1175/MWR-D-13-00101.1.

    • Search Google Scholar
    • Export Citation
  • Friedrich, K., E. A. Kalina, F. J. Masters, and C. R. Lopez, 2013: Drop-size distributions in thunderstorms measured by optical disdrometers during VORTEX2. Mon. Wea. Rev., 141, 11821203, doi:10.1175/MWR-D-12-00116.1.

    • Search Google Scholar
    • Export Citation
  • Gorgucci, E., G. Scarchilli, and V. Chandrasekar, 1999: A procedure to calibrate multiparameter weather radar using properties of the rain medium. IEEE Trans. Geosci. Remote Sens., 37, 269276, doi:10.1109/36.739161.

    • Search Google Scholar
    • Export Citation
  • Grzych, M. L., B. D. Lee, and C. A. Finley, 2007: Thermodynamic analysis of supercell rear-flank downdrafts from Project ANSWERS. Mon. Wea. Rev., 135, 240246, doi:10.1175/MWR3288.1.

    • Search Google Scholar
    • Export Citation
  • Homeyer, C. R., and M. R. Kumjian, 2015: Microphysical characteristics of overshooting convection from polarimetric radar observations. J. Atmos. Sci., 72, 870891, doi:10.1175/JAS-D-13-0388.1.

    • Search Google Scholar
    • Export Citation
  • Hubbert, J., and V. N. Bringi, 1995: An iterative filtering technique for the analysis of copolar differential phase and dual-frequency radar measurements. J. Atmos. Oceanic Technol., 12, 643648, doi:10.1175/1520-0426(1995)012<0643:AIFTFT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Kosiba, K. A., J. Wurman, Y. Richardson, P. Markowski, and P. Robinson, 2013: The genesis of the Goshen County, Wyoming, tornado (5 June 2009). Mon. Wea. Rev., 141, 11571181, doi:10.1175/MWR-D-12-00056.1.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2011: Precipitation properties of supercell hook echoes. Electron. J. Severe Storms Meteor., 6 (5). [Available online at http://ejssm.org/ojs/index.php/ejssm/article/viewArticle/93.]

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2013a: Principles and applications of dual-polarization weather radar. Part I: Description of the polarimetric radar variables. J. Oper. Meteor., 1, 226242, doi:10.15191/nwajom.2013.0119.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2013b: Principles and applications of dual-polarization weather radar. Part II: Warm- and cold-season applications. J. Oper. Meteor., 1, 243264, doi:10.15191/nwajom.2013.0120.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., 2013c: Principles and applications of dual-polarization weather radar. Part III: Artifacts. J. Oper. Meteor., 1, 265274, doi:10.15191/nwajom.2013.0121.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., and A. V. Ryzhkov, 2008a: Microphysical differences between tornadic and non-tornadic supercell rear-flank downdrafts revealed by dual-polarization radar measurements. Preprints, 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., 3B.4. [Available online at https://ams.confex.com/ams/pdfpapers/141912.pdf.]

  • Kumjian, M. R., and A. V. Ryzhkov, 2008b: Polarimetric signatures in supercell thunderstorms. J. Appl. Meteor. Climatol., 47, 19401961, doi:10.1175/2007JAMC1874.1.

    • Search Google Scholar
    • Export Citation
  • Kumjian, M. R., A. V. Ryzhkov, V. M. Melnikov, and T. J. Schuur, 2010: Rapid-scan super-resolution observations of a cyclic supercell with a dual-polarization WSR-88D. Mon. Wea. Rev., 138, 37623786, doi:10.1175/2010MWR3322.1.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Y. P. Richardson, 2009: Tornadogenesis: Our current understanding, forecasting considerations, and questions to guide future research. Atmos. Res., 93, 310, doi:10.1016/j.atmosres.2008.09.015.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Y. P. Richardson, 2014: The influence of environmental low-level shear and cold pools on tornadogenesis: Insights from idealized simulations. J. Atmos. Sci., 71, 243275, doi:10.1175/JAS-D-13-0159.1.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., J. M. Straka, and E. N. Rasmussen, 2002: Direct surface thermodynamic observations within the rear-flank downdrafts of nontornadic and tornadic supercells. Mon. Wea. Rev., 130, 16921721, doi:10.1175/1520-0493(2002)130<1692:DSTOWT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Coauthors, 2012a: The pretornadic phase of the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Part I: Evolution of kinematic and surface thermodynamic fields. Mon. Wea. Rev., 140, 28872915, doi:10.1175/MWR-D-11-00336.1.

    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., and Coauthors, 2012b: The pretornadic phase of the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Part II: Intensification of low-level rotation. Mon. Wea. Rev., 140, 29162938, doi:10.1175/MWR-D-11-00337.1.

    • Search Google Scholar
    • Export Citation
  • Marquis, J., Y. Richardson, P. Markowski, D. Dowell, and J. Wurman, 2012: Tornado maintenance investigated with high-resolution dual-Doppler and EnKF analysis. Mon. Wea. Rev., 140, 327, doi:10.1175/MWR-D-11-00025.1.

    • Search Google Scholar
    • Export Citation
  • Matrosov, S. Y., R. Cifelli, P. C. Kennedy, S. W. Nesbitt, S. A. Rutledge, V. N. Bringi, and B. E. Martner, 2006: A comparative study of rainfall retrievals based on specific differential phase shifts at X- and S-band radar frequencies. J. Atmos. Oceanic Technol., 23, 952963, doi:10.1175/JTECH1887.1.

    • Search Google Scholar
    • Export Citation
  • Melnikov, V., D. Zrnić, A. Ryzhkov, A. Zaharai, and J. Carter, 2009: Validation of attenuation correction at X band performed with collocated S-band polarimetric radar. Preprints, 34th Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., 11A.5. [Available online at https://ams.confex.com/ams/pdfpapers/155322.pdf.]

  • Palmer, R. D., and Coauthors, 2009: Weather radar education at the University of Oklahoma–An integrated interdisciplinary approach. Bull. Amer. Meteor. Soc., 90, 12771282, doi:10.1175/2009BAMS2738.1.

    • Search Google Scholar
    • Export Citation
  • Palmer, R. D., and Coauthors, 2011: Observations of the 10 May 2010 tornado outbreak using OU-PRIME: Potential for new science with high-resolution polarimetric radar. Bull. Amer. Meteor. Soc., 92, 871891, doi:10.1175/2011BAMS3125.1.

    • Search Google Scholar
    • Export Citation
  • Park, S.-G., V. N. Bringi, V. Chandrasekar, M. Maki, and K. Iwanami, 2005a: Correction of radar reflectivity and differential reflectivity for rain attenuation at X band. Part I: Theoretical and empirical basis. J. Atmos. Oceanic Technol., 22, 16211632, doi:10.1175/JTECH1803.1.

    • Search Google Scholar
    • Export Citation
  • Park, S.-G., M. Maki, K. Iwanami, V. N. Bringi, and V. Chandrasekar, 2005b: Correction of radar reflectivity and differential reflectivity for rain attenuation at X band. Part II: Evaluation and application. J. Atmos. Oceanic Technol., 22, 16331655, doi:10.1175/JTECH1804.1.

    • Search Google Scholar
    • Export Citation
  • Parker, M. D., 2014: Composite VORTEX2 supercell environments from near-storm soundings. Mon. Wea. Rev., 142, 508529, doi:10.1175/MWR-D-13-00167.1.

    • Search Google Scholar
    • Export Citation
  • Picca, J., and A. Ryzhkov, 2012: A dual-wavelength polarimetric analysis of the 16 May 2010 Oklahoma City extreme hailstorm. Mon. Wea. Rev., 140, 13851403, doi:10.1175/MWR-D-11-00112.1.

    • Search Google Scholar
    • Export Citation
  • Pruppacher, H. R., and K. V. Beard, 1970: A wind tunnel investigation of the internal circulation and shape of water drops falling at terminal velocity in air. Quart. J. Roy. Meteor. Soc., 96, 247256, doi:10.1002/qj.49709640807.

    • Search Google Scholar
    • Export Citation
  • Richardson, Y. P., P. Markowski, J. N. Marquis, J. Wurman, K. A. Kosiba, P. Robinson, D. W. Burgess, and C. C. Weiss, 2012: Tornado maintenance and demise in the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Proc. 26th Conf. on Severe Local Storms, Nashville, TN, Amer. Meteor. Soc., 13.3. [Available online at https://ams.confex.com/ams/pdfpapers/88063.pdf.]

  • Romine, G. S., D. W. Burgess, and R. B. Wilhelmson, 2008: A dual-polarization-radar-based assessment of the 8 May 2003 Oklahoma City area tornadic supercell. Mon. Wea. Rev., 136, 28492870, doi:10.1175/2008MWR2330.1.

    • Search Google Scholar
    • Export Citation
  • Ryzhkov, A. V., S. E. Giangrande, V. M. Melnikov, and T. J. Schuur, 2005a: Calibration issues of dual-polarization radar measurements. J. Atmos. Oceanic Technol., 22, 11381155, doi:10.1175/JTECH1772.1.

    • Search Google Scholar
    • Export Citation
  • Ryzhkov, A. V., T. J. Schuur, D. W. Burgess, and D. S. Zrnić, 2005b: Polarimetric tornado detection. J. Appl. Meteor., 44, 557570, doi:10.1175/JAM2235.1.

    • Search Google Scholar
    • Export Citation
  • Schuur, T. J., A. V. Ryzhkov, D. S. Zrnić, and M. Schönhuber, 2001: Drop size distributions measured by a 2D video disdrometer: Comparison with dual-polarization radar data. J. Appl. Meteor., 40, 10191034, doi:10.1175/1520-0450(2001)040<1019:DSDMBA>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schwarz, C. M., and D. W. Burgess, 2011: Supercell polarimetric signatures at X-band: Data from VORTEX2. Preprints, 35th Conf. on Radar Meteorology, Pittsburgh, PA, Amer. Meteor. Soc., P60. [Available online at https://ams.confex.com/ams/35Radar/webprogram/Paper191298.html.]

  • Seliga, T. A., and V. N. Bringi, 1976: Potential use of radar differential reflectivity measurements at orthogonal polarizations for measuring precipitation. J. Appl. Meteor., 15, 6976, doi:10.1175/1520-0450(1976)015<0069:PUORDR>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Snyder, J. C., 2013: Observations and simulations of polarimetric X-band radar signatures in supercells. Ph.D. dissertation, University of Oklahoma, Norman, OK, 214 pp. [Available online at https://shareok.org/handle/11244/6398.]

  • Snyder, J. C., H. B. Bluestein, G. Zhang, and S. J. Frasier, 2010: Attenuation correction and hydrometeor classification of high-resolution, X-band, dual-polarized mobile radar measurements in severe convective storms. J. Atmos. Oceanic Technol., 27, 19792001, doi:10.1175/2010JTECHA1356.1.

    • Search Google Scholar
    • Export Citation
  • Snyder, J. C., H. B. Bluestein, V. Venkatesh, and S. J. Frasier, 2013: Observations of polarimetric signatures in supercells by an X-band mobile Doppler radar. Mon. Wea. Rev., 141, 329, doi:10.1175/MWR-D-12-00068.1.

    • Search Google Scholar
    • Export Citation
  • Tanamachi, R. L., H. B. Bluestein, J. B. Houser, S. J. Frasier, and K. M. Hardwick, 2012: Mobile, X-band, polarimetric Doppler radar observations of the 4 May 2007 Greensburg, Kansas, tornadic supercell. Mon. Wea. Rev., 140, 21032125, doi:10.1175/MWR-D-11-00142.1.

    • Search Google Scholar
    • Export Citation
  • Testud, J., E. Le Bouar, E. Obligis, and M. Ali-Mehenni, 2000: The rain profiling algorithm applied to polarimetric weather radar. J. Atmos. Oceanic Technol., 17, 332356, doi:10.1175/1520-0426(2000)017<0332:TRPAAT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Thompson, E. J., S. A. Rutledge, and B. Dolan, 2014: A dual-polarization radar hydrometeor classification algorithm for winter precipitation. J. Atmos. Oceanic Technol., 31, 14571481, doi:10.1175/JTECH-D-13-00119.1.

    • Search Google Scholar
    • Export Citation
  • Thompson, R. L., R. Edwards, J. A. Hart, K. L. Elmore, and P. Markowski, 2003: Close proximity soundings within supercell environments obtained from the Rapid Update Cycle. Wea. Forecasting, 18, 12431261, doi:10.1175/1520-0434(2003)018<1243:CPSWSE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Trapp, R. J., 1999: Observations of nontornadic low-level mesocyclones and attendant tornadogenesis failure during VORTEX. Mon. Wea. Rev., 127, 16931705, doi:10.1175/1520-0493(1999)127<1693:OONLLM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., N. T. Atkins, and J. Wurman, 2011: The LaGrange tornado during VORTEX2. Part I: Photogrammetry analysis of the tornado combined with single-Doppler radar data. Mon. Wea. Rev., 139, 22332258, doi:10.1175/2010MWR3568.1.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., P. Stauffer, W.-C. Lee, N. T. Atkins, and J. Wurman, 2012: Finescale structure of the LaGrange, Wyoming, tornado during VORTEX2: GBVTD and photogrammetric analyses. Mon. Wea. Rev., 140, 33973418, doi:10.1175/MWR-D-12-00036.1.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., D. Dowell, Y. Richardson, P. Markowski, E. Rasmussen, D. Burgess, L. Wicker, and H. B. Bluestein, 2012: The Second Verification of the Origins of Rotation in Tornadoes Experiment: VORTEX2. Bull. Amer. Meteor. Soc., 93, 11471170, doi:10.1175/BAMS-D-11-00010.1.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., K. Kosiba, and P. Robinson, 2013: In situ, Doppler radar, and video observations of the interior structure of a tornado and the wind–damage relationship. Bull. Amer. Meteor. Soc., 94, 835846, doi:10.1175/BAMS-D-12-00114.1.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., 2013: A diabatic Lagrangian technique for the analysis of convective storms. Part II: Application to a radar-observed storm. J. Atmos. Oceanic Technol., 30, 22662280, doi:10.1175/JTECH-D-13-00036.1.

    • Search Google Scholar
    • Export Citation
  • Zrnić, D. S., V. M. Melnikov, and J. K. Carter, 2006: Calibrating differential reflectivity on the WSR-88D. J. Atmos. Oceanic Technol., 23, 944951, doi:10.1175/JTECH1893.1.

    • Search Google Scholar
    • Export Citation
1

In the two referenced studies, the same radar system was referred to as the X-Band Experimental Radar for the Examination of Storms (XERES) and “NO-XP,” respectively.

2

In this study, and refer to measured reflectivity factor and differential reflectivity factor, respectively. Conversely, ZH and ZDR in the text or figures refer to the same variables except that they have been corrected for attenuation/differential attenuation using the schemes discussed in the text.

3

There were more NOXP data than UMass X-Pol data that met the criteria for aggregates because of slightly higher NOXP ρHV values. Calculations using a different ρHV cutoff criterion for snow aggregates (e.g., 0.95 or 0.99) had little effect on mean ZDR calculations for either radar.

4

The hook echo from 25 May 2010 is considered to be tornadic because of the short amount of time between the analysis time and the time of tornadogenesis. The scan used for this case is from the last NOXP volume collected in the deployment. Conversely, a report of a brief, weak tornado in the 9 June 2009 hook echo case could not be confirmed by VORTEX2 teams, and the case is considered to be nontornadic (e.g., Wurman et al. 2012).

5

Several assumptions have to be made to directly relate a ZDR value to a median drop-diameter value. The focus in this study is on changes in ZDR using the Cao et al. (2008) ZHZDR relation rather than estimated changes in equivalent drop diameters. An estimate for relationships between ZDR and drop diameter at X band can be found in Fig. 1 of Snyder et al. (2010).

6

The emphasis in this section is on tornadic cases so that comparisons can be made with the tornadic cases studied in K11.

7

NOXP tornadogenesis data were obtained in two additional cases analyzed in sections 3a and 3b. In one case, from 8 June 2010, the hook echo formed coincident with the tornado, however, and in another case, from 30 May 2012, attenuation and differential attenuation degraded data quality in several of the scans.

8

A small-scale vortex signature is identified first at ~0108 UTC, but with only weak ΔV. At ~0110 UTC, the signature has ΔV of >30 m s−1. Storm Data lists both the tornado formation and dissipation times as 0110 UTC.

9

The additional cell to the rear of the hook echo was not included in the analyses until 0116 UTC when the cell merged with the hook echo and the two were indistinguishable.

10

For both of these dissipation cases, radial velocity data from the Meteorological Weather Radar 2005 X-band Phased Array (MWR-05XP; Bluestein et al. 2010) were used to track the TVSs associated with the tornadoes. The radar’s faster volumetric update time relative to that of NOXP (by an order of magnitude) allows for a more accurate assessment of the tornado life cycle.

11

The generation of rotation near the ground is a pivotal step in supercell tornado development, but, given that only single-Doppler radar data are analyzed, the topic is beyond the scope of this study.

Save