• Asai, T., 1988: Meso-scale features of heavy snowfalls in Japan Sea coastal regions of Japan (in Japanese). Tenki, 35 , 156161.

  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2003a: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part I: Tornadogenesis. Mon. Wea. Rev., 131 , 29542967.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., W. C. Lee, M. Bell, C. C. Weiss, and A. L. Pazmany, 2003b: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part II: Tornado-vortex structure. Mon. Wea. Rev., 131 , 29682984.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2004: Doppler radar observations of dust devils in Texas. Mon. Wea. Rev., 132 , 209224.

    • Search Google Scholar
    • Export Citation
  • Brown, R. A., V. T. Wood, and D. Sirmans, 2002: Improved tornado detection using simulated and actual WSR-88D data with enhanced resolution. J. Atmos. Oceanic Technol., 19 , 17591771.

    • Search Google Scholar
    • Export Citation
  • Church, C. R., and J. T. Snow, 1993: Laboratory models of tornadoes. The Tornado: Its Structure, Dynamics, Prediction and Hazards, Geophys. Monogr., Vol. 79, Amer. Geophys. Union, 277–295.

    • Search Google Scholar
    • Export Citation
  • Dessens, J., 1972: Influence of ground roughness on tornadoes: A laboratory simulation. J. Appl. Meteor., 11 , 7275.

  • Dowell, D. C., C. R. Alexander, J. M. Wurman, and L. J. Wicher, 2005: Centrifuging of hydrometeors and debris in tornadoes: Radar-reflectivity patterns and wind-measurement errors. Mon. Wea. Rev., 133 , 15011524.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1971: Proposed characterization of tornadoes and hurricanes by area and intensity. SMRP Research Paper 91, University of Chicago, 15 pp.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38 , 15111534.

  • Groenemeijer, P., 2003: Three events of strong deep moist convection in the Netherlands. Internal Rep., KNMI, 76 pp.

  • Hanstrum, B. N., G. A. Mills, A. Watson, J. P. Monteverdi, and C. A. Doswell, 2002: The cool-season tornadoes of California and southern Australia. Wea. Forecasting, 17 , 705722.

    • Search Google Scholar
    • Export Citation
  • Kato, W., H. Suzuki, M. Shimamura, K. Kusunoki, and T. Hayashi, 2007: The design and initial testing of an X-band Doppler radar for monitoring hazardous winds for railroad system. Preprints, 33rd Conf. on Radar Meteorology, Cairns, Australia, Amer. Meteor. Soc., P13A.15. [Available online at http://ams.confex.com/ams/33Radar/techprogram/paper_123530.htm].

    • Search Google Scholar
    • Export Citation
  • Kawashima, M., and Y. Fujiyoshi, 2005: Shear instability wave along a snowband: Instability structure, evolution, and energetics derived from dual-Doppler radar data. J. Atmos. Sci., 62 , 351370.

    • Search Google Scholar
    • Export Citation
  • Kitagawa, N., and K. Michimoto, 1994: Meteorological and electrical aspects of winter thunderclouds. J. Geophys. Res., 99 , (D5). 1071310721.

    • Search Google Scholar
    • Export Citation
  • Klemp, J. B., and R. Rotunno, 1983: A study of the tornadic region within a supercell thunderstorm. J. Atmos. Sci., 40 , 359377.

  • Kobayashi, F., K. Kikuchi, and H. Uyeda, 1996: Life cycle of the Chitose tornado of September 22, 1988. J. Meteor. Soc. Japan, 74 , 125140.

    • Search Google Scholar
    • Export Citation
  • Kobayashi, F., Y. Sugimoto, T. Suzuki, T. Maesaka, and Q. Moteki, 2007: Doppler radar observation of a tornado generated over the Japan Sea coast during a cold air outbreak. J. Meteor. Soc. Japan, 85 , 321334.

    • Search Google Scholar
    • Export Citation
  • Kusunoki, K., and Coauthors, 2008: An overview of the Shonai area railroad weather project and early outcomes. Preprints, Fifth European Conf. on Radar in Meteorology and Hydrology, Helsinki, Finland, European Meteorological Society, P12.1.

    • Search Google Scholar
    • Export Citation
  • Laird, N. F., L. J. Miller, and D. A. R. Kristovich, 2001: Synthetic dual-doppler analysis of a winter mesoscale vortex. Mon. Wea. Rev., 129 , 312331.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., and R. B. Whilhelmson, 1997b: The numerical simulation of non-supercell tornadogenesis. Part II: Evolution of a family of tornadoes along a weak outflow boundary. J. Atmos. Sci., 54 , 23872415.

    • Search Google Scholar
    • Export Citation
  • Lee, W.-C., and J. Wurman, 2005: Diagnosed three-dimensional axisymmetric structure of the Mulhall tornado on 3 May 1999. J. Atmos. Sci., 62 , 23732393.

    • Search Google Scholar
    • Export Citation
  • Lemon, L. R., and C. A. Doswell, 1979: Severe thunderstorm evolution and mesocyclone structure as related to tornadogenesis. Mon. Wea. Rev., 107 , 11841197.

    • Search Google Scholar
    • Export Citation
  • Leslie, F. W., 1977: Surface roughness effects on suction vortex formation: A laboratory simulation. J. Atmos. Sci., 34 , 10221027.

  • Marquis, J. N., Y. P. Richardson, and J. M. Wurman, 2007: Kinematic observations of misocyclones along boundaries during IHOP. Mon. Wea. Rev., 135 , 17491768.

    • Search Google Scholar
    • Export Citation
  • Miyazawa, S., 1967: On vertical mesoscale disturbances observed during the period of heavy snow or rain in the Hokuriku district. J. Meteor. Soc. Japan, 45 , 166176.

    • Search Google Scholar
    • Export Citation
  • Nagata, M., 1993: Meso-β-scale vortices developing along the Japan-Sea polar-airmass convergence zone (JPCZ) cloud band: Numerical simulation. J. Meteor. Soc. Japan, 71 , 4357.

    • Search Google Scholar
    • Export Citation
  • Niino, H., O. Suzuki, H. Nirasawa, T. Fujitani, H. Ohno, I. Takayabu, N. Kinoshita, and Y. Ogura, 1993: Tornadoes in Chiba prefecture on 11 December 1990. Mon. Wea. Rev., 121 , 30013018.

    • Search Google Scholar
    • Export Citation
  • Niino, H., T. Fujitani, and N. Watanabe, 1997: A statistical study of tornadoes and waterspouts in Japan from 1961 to 1993. J. Climate, 10 , 17301752.

    • Search Google Scholar
    • Export Citation
  • Ninomiya, K., and K. Hoshino, 1990: Evolution process and multi-scale structure of a polar low developed over the Japan Sea on 11–12 December 1985. Part II: Meso-β-scale low in meso-α-scale polar low. J. Meteor. Soc. Japan, 68 , 307318.

    • Search Google Scholar
    • Export Citation
  • Ohno, H., and T. Takemi, 2010: Mechanisms for intensification and maintenance of numerically simulated dust devils. Atmos. Sci. Lett., 11 , 2732. doi:10.1002/asl.249.

    • Search Google Scholar
    • Export Citation
  • Roberts, R. D., and J. W. Wilson, 1995: The genesis of three nonsupercell tornadoes observed with dual-Doppler radar. Mon. Wea. Rev., 123 , 34083436.

    • Search Google Scholar
    • Export Citation
  • Suzuki, O., H. Niino, H. Ohno, and H. Nirasawa, 2000: Tornado-producing mini supercells associated with Typhoon 9019. Mon. Wea. Rev., 128 , 18681882.

    • Search Google Scholar
    • Export Citation
  • Tanamachi, R. L., H. B. Bluestein, W. C. Lee, M. Bell, and A. Pazmany, 2007: Ground-based velocity track display (GBVTD) analysis of W-band Doppler radar data in a tornado near Stockton, Kansas, on 15 May 1999. Mon. Wea. Rev., 135 , 783800.

    • Search Google Scholar
    • Export Citation
  • Tsuboki, K., and T. Asai, 2004: The multi-scale structure and development mechanism of mesoscale cyclones over the Sea of Japan in winter. J. Meteor. Soc. Japan, 82 , 597621.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., and J. W. Wilson, 1989: Non-supercell tornadoes. Mon. Wea. Rev., 117 , 11131140.

  • Wilczak, J., D. Wolfe, R. Zamora, B. Stankov, and T. Christian, 1992: Observations of a Colorado tornado. Part I: Mesoscale environment and tornadogenesis. Mon. Wea. Rev., 120 , 497521.

    • Search Google Scholar
    • Export Citation
  • Wilkins, E. M., Y. Sasaki, and H. L. Johnson, 1975: Surface friction effects on thermal convection in a rotating fluid: A laboratory simulation. Mon. Wea. Rev., 103 , 305317.

    • Search Google Scholar
    • Export Citation
  • Wilson, J. W., 1986: Tornadogenesis by nonprecipitation induced wind shear lines. Mon. Wea. Rev., 114 , 270284.

  • Winn, W. P., S. J. Hunyady, and G. D. Aulich, 1999: Pressure at the ground in a large tornado. J. Geophys. Res., 104 , (D18). 2206722082.

    • Search Google Scholar
    • Export Citation
  • Wood, V. T., and R. A. Brown, 1997: Effects of radar sampling on single-Doppler velocity signatures of mesocyclones and tornadoes. Wea. Forecasting, 12 , 928938.

    • Search Google Scholar
    • Export Citation
  • Wood, V. T., and R. A. Brown, 2000: Oscillations in mesocyclone signatures with range owing to azimuthal radar sampling. J. Atmos. Oceanic Technol., 17 , 9095.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., 2002: The multiple-vortex structure of a tornado. Wea. Forecasting, 17 , 473505.

  • Wurman, J., and S. Gill, 2000: Finescale radar observations of the Dimmitt, Texas (2 June 1995), tornado. Mon. Wea. Rev., 128 , 21352164.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., and C. R. Alexander, 2005: The 30 May 1998 Spencer, South Dakota, storm. Part II: Comparison of observed damage and radar-derived winds in the tornadoes. Mon. Wea. Rev., 133 , 97119.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., J. M. Straka, and E. N. Rasmussen, 1996: Fine-scale Doppler radar observations of tornadoes. Science, 272 , 17741777.

  • Yamauchi, H., O. Suzuki, and K. Akaeda, 2006: A hybrid multi-PRI method to dealias Doppler velocities. SOLA, 2 , 9295.

  • View in gallery

    Map of the Shonai area, Yamagata Prefecture, Japan. (a) Cross and range circles represent the location of the JR-EAST radar and its observational range at 10-km intervals in radius. Contours represent topography (100-, 200-, and 500-m intervals). Dots denote the locations of surface automated weather stations. (b) Crosses denote the locations of damaged houses (A–D), the solid inverted triangle indicates the location of the Sakata Fire Department, and the diamond indicates the location of the JMA’s Sakata Weather Station.

  • View in gallery

    Schematic diagram of the Doppler spectrum (solid curve) and the estimated mode and mean Doppler velocities (bold arrows).

  • View in gallery

    (a) Sea level pressure and surface wind vectors from MANAL at 1500 UTC 1 Dec 2007. The solid lines are the isobars at 2-hPa intervals. (b) MTSAT satellite IR1 image for the same time. The solid square and the dot represent the locations of the JR-EAST radar and the JMA operational Niigata radar, respectively. The dashed rectangular region indicates the region depicted in Fig. 4.

  • View in gallery

    Time series of the radar reflectivity of the JMA Niigata radar at an elevation angle of 0.0° and surface winds (barbed arrows) by Automated Meteorological Data Acquisition System (AMeDAS). The parent storm of the tornado is indicated by the arrow. The bold open square represents the location of the JR-EAST radar.

  • View in gallery

    Hodograph of the mean wind profiler observation at Sakata Weather Station before (1610–1630 UTC, solid line with circles) and after (1700–1730 UTC, dotted line with squares) the passage of the parent storm on 1 Dec. The grayscale represents the observation height in MSL.

  • View in gallery

    (left) Doppler velocity fields and (right) reflectivity fields of the storm at (a) 1623:51, (b) 1632:07, and (c) 1640:24 UTC. Four misocyclones (A–D) are indicated by couplets of (left) Doppler velocity maximum and minimum as viewed from the JR-EAST radar (cross). Wind barbs measured at automated weather stations are also depicted (one barb denotes 5 m s−1). Bold dashed lines and arrows in (b) and (c) denote the estimated convergence line. The dashed rectangular region in (c) indicates the location of the region depicted in Fig. 10. The dots are the same as those in Fig. 1. The gray line represents the coastline.

  • View in gallery

    Contours of (a) surface-temperature deviation fields and (b) surface-pressure deviation fields of the storm at 1634:04 and 1646:43 UTC. Reflectivity of the storm is represented by shading. Contour intervals are at 0.25 K in temperature deviation and 0.1 hPa in pressure deviation. Wind barbs measured at automated weather stations are also depicted (a barb denotes 5 m s−1). Dots and the gray line are the same as those in Fig. 6.

  • View in gallery

    Tracks of misocyclones A–D between 1612:10 and 1650:07 UTC derived from the couplets of Doppler velocity maximum and minimum for each PPI scan at 30-s intervals. The size of the shaded circle represents the diameter of each misocyclone, and the grayscale denotes the observation time. The locations of the damaged houses are depicted by bold crosses. Range marks are 5- and 10-km intervals from the JR-EAST radar (cross).

  • View in gallery

    Temporal variations of (a) core diameter, peak tangential velocity, and (b) vertical vorticity of misocyclone A. The bold lines represent the estimated core diameter, peak tangential velocity, and vertical vorticity and the thin gray lines and dots represent Φ and ΔV/2 for each PPI scan. Time of misocyclone A’s landfall and period of misocyclone A passage over the damage path are also indicated. Beam height of misocyclone A for each scan is also plotted in (b).

  • View in gallery

    Close-up views of misocyclones A and D at (a) 1627:15, (b) 1630:11, (c) 1633:06, (d) 1636:01, (e) 1638:56, and (f) 1641:51 UTC during the passage over the damage path denoted by the dashed rectangle in Fig. 6c. (left) Doppler velocity fields and (right) reflectivity fields are shown. Dots, crosses, the inverted triangle, and the diamond are the same as those in Fig. 1.

  • View in gallery

    As in Fig. 9, but for misocyclone D.

  • View in gallery

    Azimuthal profile of Doppler velocity (dots) and radar reflectivity (gray open squares) through the core of misocyclone A at 1635:02 UTC. Profiles from the conceptual vortex models are denoted by dotted (Rankine) and dashed (modified Rankine: VtR−0.7 in the outer region) lines for comparison.

  • View in gallery

    Time histories of (a) wind speed, (b) wind direction, maximum/minimum wind direction, (c) surface pressure, and (d) temperature (bold line) and relative humidity (thin line) at Sakata Weather Station between 1625 and 1650 UTC.

  • View in gallery

    Close-up views of misocyclone A at 1635:02 UTC, and surface pressure (solid line) and wind (barb) estimated through time-to-space conversion of surface observation data of Sakata Weather Station. (left) Doppler velocity field and (right) reflectivity field are shown.

  • View in gallery

    Positions of misocyclones A and D (open circles) during their interaction (between 1631:09 and 1639:54 UTC). The size of each circle represents the size of each misocyclone. The dots represent the estimated orbit centers that divide each line segment between misocyclones A and D by the ratio of 1.57. The dotted line connecting the orbit centers at 1631:09 and 1639:54 UTC is shown for reference.

  • View in gallery

    True angular distance Ψ0 vs the average angular distance Φ for the simulated mode (dots) and mean (squares) Doppler velocity signatures. The black solid line is the approximated distribution of best fit to the calculated data points for the mode Doppler velocity signature. The black dotted lines correspond to the examples given in the text.

  • View in gallery

    True angular distance Ψ0 vs the average velocity ratio γυ for the simulated mode (dots) and mean (squares) Doppler velocity signatures. The solid line is the approximated distribution of best fit to the calculated data points for the mode Doppler velocity signature. The black dotted lines correspond to the examples given in the text.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 233 118 8
PDF Downloads 195 114 6

Finescale Doppler Radar Observation of a Tornado and Low-Level Misocyclones within a Winter Storm in the Japan Sea Coastal Region

View More View Less
  • 1 * Meteorological Research Institute, Tsukuba, Japan
  • | 2 East Japan Railway Company, Saitama, Japan
  • | 3 Railway Technical Research Institute, Tokyo, Japan
  • | 4 Disaster Prevention Research Institute, Kyoto University, Kyoto, Japan
Full access

Abstract

Life histories of low-level misocyclones, one of which corresponded to a tornado vortex within a winter storm in the Japan Sea coastal region on 1 December 2007, were observed from close range by X-band Doppler radar of the East Japan Railway Company. Continuous plan position indicator (PPI) observations at 30-s intervals at the low-elevation angle revealed at least four cyclonic misocyclones within the head of the comma-shaped echo of the vortical disturbance under winter monsoon conditions. The meso-β-scale vortical disturbance developed within the weak frontal zone at the leading edge of cold-air outbreaks.

High-resolution observation of misocyclones revealed the detailed structures of these misocyclones and their temporal evolution. As the parent storm evolved, a low-level convergence line was observed at the edge of the easternmost misocyclone. This convergence line was considered to be important for the initiation and development of the misocyclones and the tornado through vortex stretching. The strongest misocyclone gradually intensified as its diameter contracted until landfall, and then began to dissipate soon after landfall. The temporal evolution of the misocyclones through landfall is discussed.

Surface wind and pressure variations suggested a cyclonic vortex passage, which was consistent with the passage of the radar-derived misocyclone. The observed pressure drop was also consistent with that computed from the cyclostrophic equation for the modified Rankine vortex. The observed behavior of two adjacent misocyclones was primarily consistent with the rotational flow associated with the other misocyclone. The generation and development processes of the tornado and misocyclones are discussed.

Corresponding author address: Hanako Y. Inoue, Meteorological Research Institute, 1-1 Nagamine, Tsukuba, 305-0052, Japan. Email: hainoue@mri-jma.go.jp

Abstract

Life histories of low-level misocyclones, one of which corresponded to a tornado vortex within a winter storm in the Japan Sea coastal region on 1 December 2007, were observed from close range by X-band Doppler radar of the East Japan Railway Company. Continuous plan position indicator (PPI) observations at 30-s intervals at the low-elevation angle revealed at least four cyclonic misocyclones within the head of the comma-shaped echo of the vortical disturbance under winter monsoon conditions. The meso-β-scale vortical disturbance developed within the weak frontal zone at the leading edge of cold-air outbreaks.

High-resolution observation of misocyclones revealed the detailed structures of these misocyclones and their temporal evolution. As the parent storm evolved, a low-level convergence line was observed at the edge of the easternmost misocyclone. This convergence line was considered to be important for the initiation and development of the misocyclones and the tornado through vortex stretching. The strongest misocyclone gradually intensified as its diameter contracted until landfall, and then began to dissipate soon after landfall. The temporal evolution of the misocyclones through landfall is discussed.

Surface wind and pressure variations suggested a cyclonic vortex passage, which was consistent with the passage of the radar-derived misocyclone. The observed pressure drop was also consistent with that computed from the cyclostrophic equation for the modified Rankine vortex. The observed behavior of two adjacent misocyclones was primarily consistent with the rotational flow associated with the other misocyclone. The generation and development processes of the tornado and misocyclones are discussed.

Corresponding author address: Hanako Y. Inoue, Meteorological Research Institute, 1-1 Nagamine, Tsukuba, 305-0052, Japan. Email: hainoue@mri-jma.go.jp

1. Introduction

The detailed investigation of tornado structures and evolution processes is intrinsically difficult since tornadoes are characterized by their short lifetime and small horizontal scale. Mobile Doppler radars have been used to obtain high-resolution observations of tornadoes, and several detailed observational studies of tornadoes and tornadic storms have been reported. However, they have focused mainly on typical supercell storms and tornadoes in the United States (e.g., Wurman and Gill 2000; Bluestein et al. 2003a,b). Few detailed observational studies of tornadoes and tornadic storms have been conducted in Japan because of their lower frequencies. Moreover, there have been few opportunities to observe tornadoes and tornadic storms in detail with only ground-based Doppler radar.

A statistical study of tornadoes and waterspouts in Japan from 1961 to 1993 (Niino et al. 1997) indicated that tornado occurrence in Japan was concentrated in the coastal regions. It also demonstrated that the tornado occurrence on the Japan Sea side had two maxima in frequency: one in September and the other in January. The latter maximum was characterized by the winter monsoon tornadoes associated with enhanced cumulus cloud activities during cold-air outbreaks over the warm sea surface. Such winter monsoon tornadoes made up 12% of all tornadoes.

In the Japan Sea coastal region, various types of mesoscale disturbances associated with cold-air outbreaks from the Eurasian continent and airmass transformation over the warm sea surface frequently develop in the winter. Such disturbances are characterized by cloud tops lower than those of common cumulonimbus clouds. Nevertheless, these disturbances are also characterized by hazardous features, such as peculiar lightning activity (Kitagawa and Michimoto 1994), and associated tornadoes or other gusts. A recent observational study showed that small-scale snow clouds could produce tornadoes over the Japan Sea coast under uniform winter monsoon conditions (Kobayashi et al. 2007). Kobayashi et al. (2007) discussed that the tilting of the horizontal vorticity due to the large vertical shear was important as a possible formation mechanism for the tornado, based on the environmental conditions. However, the formation mechanism, detailed structure, and evolution process of winter tornadoes on the Japan Sea coastal region are not understood well.

Some other observational studies of tornadic storms have also been performed in Japan, such as Doppler radar studies of nearly classic supercell storms (Niino et al. 1993), supercell tornadoes with a mesocyclone in the parent storm (Kobayashi et al. 1996), and mini supercells associated with a typhoon (Suzuki et al. 2000). However, no Doppler radar observation of winter tornadoes has been conducted in the Japan Sea coastal region in sufficient high temporal and spatial resolution to clarify their detailed structure and temporal evolution.

Since winter monsoon tornadoes can cause considerable damage in the Japan Sea coastal region, it is extremely important to clarify their detailed structure and their generation and evolution mechanisms in order to prevent and mitigate wind disasters in this region. Thus, as part of a research project for developing an automatic strong gust detection system for railroads, the Shonai Area Railroad Weather Project, field observation (Kusunoki et al. 2008) was conducted in the Shonai area, Yamagata Prefecture, Japan (Fig. 1) between October 2007 and March 2010 in order to study the finescale structure and time evolution of strong gusts and associated storm dynamics around the Japan Sea coastal region.

Between 1600 and 1700 UTC 1 December 2007 [UTC is local standard time (LST) minus 9 h] during this field observation, an intense wavelike precipitation system developed over the Japan Sea and propagated to the Shonai area, causing F0 scale (Fujita 1971) tornado damage in Sakata city, Yamagata Prefecture, Japan. Associated with the passage of this system, several low-level misocyclones were detected by the continuous plan position indicator (PPI) observation of the X-band Doppler radar of the East Japan Railway Company (hereafter referred to as the JR-EAST radar) at 30-s intervals. One of these misocyclones corresponded to a tornado vortex and approached as close as 6 km to the radar. The dataset collected on 1 December 2007 was the first X-band Doppler radar dataset to document the detailed structure and evolution of misocyclones and a tornado within a winter vortical disturbance in the Japan Sea coastal region in high spatial and temporal resolutions throughout their lifetimes. This dataset also contains information on the time-dependent behavior of misocyclones and a tornado through landfall, which had not yet been observed. The purpose of this paper is to describe in detail the temporal evolution of misocyclones and to discuss the generation and development process of a tornado within a winter vortical disturbance.

The rest of this paper is organized as follows. Section 2 introduces the observational instruments, data analysis, and tornado damage characteristics. Section 3 provides the synoptic situation. Section 4 provides a detailed description of the observed structure and characteristics of the parent storm and low-level misocyclones. Section 5 discusses generation and development processes of the tornado and the low-level misocyclones. Finally, section 6 provides the conclusions of our findings.

2. Instrumentation, data analysis, and tornado damage characteristics

a. Instrumentation and data analysis

The field observation was conducted between October 2007 and March 2010 in the Shonai area, Yamagata Prefecture, Japan, which is located on the Japan Sea side and provides a suitable setting for studying tornadoes and other gusts in the winter (Fig. 1). The major facilities for the observation include the X-band Doppler radar (JR-EAST radar) and a network of automated surface weather transmitters.

The JR-EAST radar was installed at Amarume Station, Yamagata Prefecture, Japan (Fig. 1) in March 2007 and has operated continuously to assess the utility of Doppler radar in operational railroad warning systems (Kato et al. 2007). It is a 9.77-GHz X-band Doppler radar with a maximum observation range of 30 km. The antenna is 1.2 m in diameter, resulting in a beamwidth of 2.0°. The 0.5-μs pulse width provides independent data points at 75-m intervals in the radial direction. Table 1 summarizes the JR-EAST radar characteristics.

The JR-EAST radar is operated in a PPI mode at 2 rpm at the low-elevation angle (3.0°) so as to provide reflectivity and Doppler velocity fields as close to the ground as possible since it is necessary to observe wind gusts successively. The instrument, as utilized, did not observe the three-dimensional structures of storms, but a series of PPI scans taken at 30-s intervals provides a unique dataset to document temporal and spatial variations of low-level circulation in high resolution.

Surface weather stations were distributed in the study area to characterize and validate the finescale structure of gusts and storms near the surface. Twenty-six weather transmitters were installed to cover the Shonai plain around the JR-EAST radar at 4-km intervals (Fig. 1). The sensors are mounted on the top of steel poles about 5 m high. The instrument measures wind direction and wind speed at 1-s intervals, and precipitation, pressure, temperature and humidity at 10-s intervals. These sensors are hereafter referred to as automated weather stations.

We also used surface observation data at 10-s intervals measured at Sakata Weather Station of the Japan Meteorological Agency (JMA) and surface meteorological data acquired at the Sakata Fire Department (Fig. 1). Upper-air wind data at 10-minute intervals was available from the JMA operational 1.3-GHz wind profiler observation at Sakata Weather Station.

1) JR-EAST radar data processing

The JR-EAST radar is the noncoherent magnetron type and provides radar reflectivity and Doppler velocity data. The definition for estimating the commonly used mean Doppler velocity υd is υd = −(λ/2)∫fSn( f )df, where Sn( f ) is the Doppler spectral density at the Doppler frequency f and λ is the wavelength. However, in the JR-EAST radar, the Doppler velocity is estimated from the Doppler frequency fk where the Doppler spectral density Sn( fk) is largest. We define “mode” Doppler velocity υmd as the Doppler velocity having the largest spectral density: υmd = −(λ/2) fk (Fig. 2). It is the mode of the reflectivity-weighted radial motion of all the radar scatterers within the resolution volume. If the Doppler spectrum is symmetric and unimodal, the mode Doppler velocity is equal to the mean Doppler velocity. This mode Doppler velocity will be referred to as the Doppler velocity in this paper.

The JR-EAST radar observation is performed using dual pulse repetition frequency (PRF) sampling; pulses are alternately transmitted at a single PRF of 900 pps within a block and transmitted at 1200 pps within another block. Doppler velocity and reflectivity data were estimated with FFT using 64 pulses for each block. With an antenna scanning rate of 2 rpm and a pulse repetition frequency of 900 (1200) pps, a series of 64 pulses yields 0.85° (0.64°) in the azimuthal direction. This results in a sampling resolution in the azimuthal direction of 0.75°, oversampling the beamwidth by approximately a factor of 3. Dual PRF sampling of 900/1200 pps yields a maximum unfolding velocity of ±27.5 m s−1. The hybrid multi-PRI method (Yamauchi et al. 2006) was used to dealias Doppler velocities, and further correction was performed manually. The removal of ground clutter contamination around the misocyclones was also performed manually.

2) Vortex detection

The small-scale vortex was detected manually by identifying the couplet of Doppler velocity maximum (Vmax) and minimum (Vmin) and tracking the size and location of each vortex for each PPI scan. For the estimation of core diameter and peak tangential velocity of the misocyclone, we used the observed azimuthal angular distance Φ between Vmax and Vmin and one-half the difference ΔV/2 [=(VmaxVmin)/2] between Vmax and Vmin.

Brown et al. (2002) demonstrated that the measurement of the vortex core diameter and peak tangential velocity by a single Doppler radar depends heavily on the ratio of the vortex diameter to the effective radar beamwidth θe. Their result is based on the simulated mean Doppler velocity field. However, the JR-EAST radar uses the mode Doppler velocity instead of the mean Doppler velocity. We performed simulations of the mode Doppler velocity of vortices similar to that described in Wood and Brown (1997) and investigated the relationship between Φ, ΔV/2, and vortex parameters. Then, we estimated the true core diameters and peak tangential velocities of vortices from the observed Φ and ΔV/2 based on the simulation results. The detailed simulation results and estimation procedure is shown in the appendix. The vorticity of the misocyclone was calculated from the estimated peak tangential velocity and diameter, assuming solid-body rotation. Thus, the estimation of vorticities depends on the estimation of both core diameters and peak tangential velocities. The estimated core diameters, peak tangential velocities, and vorticities were used for the analysis.

b. Tornado damage characteristics

Figure 1 depicts the locations of damaged houses, based on the damage survey conducted by the JMA. The damage included scattering of tin roofs, roof tiles, and signboards. The damage scale was ranked as F0 based on the damage survey. The intermittent damage pathlength was about 3 km, and the damage width was about 100 m. While the wind direction inferred from the damage was mostly westerly (locations A, B, and C), some damage was estimated to be due to easterly winds (location D). Although there was no eye witness (it was around midnight, between 0100 and 0200 LST), JMA concluded that the damage was caused by a tornado based on the JR-EAST Doppler radar observation, the damage survey, and meteorological records such as pressure drop and wind variation at the Sakata Fire Department (maximum wind speed of 38 m s−1 and pressure drop of about 3 hPa at 1634 UTC) and Sakata Weather Station (1636 UTC, section 4c). The tornado occurred at around 1630 UTC based on the surface observation.

3. Synoptic situation

Figure 3 presents the surface pressure and wind field from the JMA mesoscale objective analysis data (MANAL) with a horizontal resolution of 10 km (available every 3 h) and the infrared (IR1) image taken by Multifunctional Transport Satellite 1 (MTSAT1) at 1500 UTC 1 December 2007, which was about half and an hour before the tornado occurred. The surface-pressure pattern indicated a moderate southwest–northeast pressure gradient, which is typical of the winter monsoon situation around Japan. The satellite image indicates a well-developed cloud system near the Japan Sea coastal region (dashed rectangular region in Fig. 3). The MANAL showed that this cloud system corresponded to the leading edge of the cold-air outbreak and accompanied a low-level horizontal wind shear and moderate temperature gradient (not shown), though it was not strong enough to appear on the synoptic weather map. The JMA operational Niigata radar (Fig. 4) showed that several rainbands were embedded within the cloud system associated with the large-scale frontal zone. These rainbands exhibited a chain pattern of comma-shaped clouds. The parent storm of the tornado, which is indicated by the arrow in Fig. 4, passed over the Shonai area around 1600 UTC.

According to the wind profiler observation at Sakata Weather Station, strong south-southwesterly wind below 1 km MSL prevailed before the passage of this well-developed cloud system (Fig. 5). Low-level bulk shear below 1 km MSL was between 0.7 and 1.2 × 10−2 s−1 during this period. The wind below 1.5 km turned from west-southwesterly to northwesterly at 1500 and 1630 UTC (Fig. 5), in association with the passage of the rainbands. The surface observation at automated weather stations along the coast and the Sakata Weather Station also recorded similar change of wind direction and the temperature drop around 1500 and 1630 UTC. These features suggest that the observed comma-shaped rainbands were associated with the mesoscale frontal zone.

No upper-air sounding associated with the passage of the frontal system was available because of the sparsity of upper-air soundings of thermodynamic quantities in time and space. Thus, we calculated the thermodynamic quantities from MANAL. The CAPE around the large-scale frontal zone was between 100 and 200 J kg−1 at 1500 UTC 1 December.

4. Characteristics of the parent storm and the low-level misocyclones

a. Characteristics of the parent storm

Reflectivity and Doppler velocity fields of the parent storm are depicted in Fig. 6. The parent storm had a comma-shaped echo and an enhanced reflectivity region within the head of the storm. The horizontal scale was about 30 km, and the maximum echo-top height was estimated to be 6 km, according to the JMA operational radar observations. The storm developed over the Japan Sea, rapidly moved eastward, and propagated inland around 1620 UTC. As the storm propagated, reflectivity within the head intensified, and the curvature of the storm increased. The surface wind exhibited a cyclonically rotating wind field associated with this parent storm.

Some thermodynamic features of the storm were also recognized from the surface observation at the automated weather stations. The temporal average of the surface temperature was higher in the coastal region than that at the inland stations during the passage of the rainbands. It is likely that warm southwesterly or northwesterly, which had been transformed over the warm sea surface, advected to the coastal region and then cooled over the land as it advected farther inland. To examine the temperature field related to the parent storm by excluding the average spatial difference due to the advection, we calculated the deviation of surface temperature from its temporal average between 1500 and 1800 UTC at each station. The deviation of surface pressure was calculated in the same manner.

Figure 7 illustrates the deviation of surface temperature and pressure field during the passage of the storm. The temperature deviation exhibited positive values before and during the passage of the parent storm but became negative after it passed over most of the automated weather stations, which is consistent with the temperature gradient shown by MANAL. The positive temperature deviation was distinct while the center of the parent storm passed over some of the stations. The pressure deviation was mostly negative before and during the passage of the storm, but became positive after it passed over the Shonai area. This suggests that the parent storm had a warm, low pressure vortex center.

b. Observed misocyclones and their temporal evolutions

High-resolution Doppler radar observations detected at least four vortex signatures within the head of the parent storm (Fig. 6). As the horizontal scales of the four vortex signatures were less than 4 km, they will hereafter be referred to as misocyclones (Fujita 1981). They were labeled A–D. All four misocyclones rotated in a cyclonic direction, consistent with that of the parent storm. These misocyclones aligned along a line near the edge of the parent vortical disturbance. Two misocyclones generated at the eastern edge of the head of the parent storm at 1620:27 and 1623:51 UTC. The Doppler velocity field began to exhibit a clear convergence line a few minutes after the generation of the easternmost misocyclone (see Figs. 6b,c). The low-level convergence increased to about 1.0 × 10−2 s−1 as the storm evolved. This convergence line was associated with cyclonic shear flow. The convergence line extended from the easternmost misocyclone and moved with this vortex, suggesting that the misocyclones developed along the low-level convergence line.

Tracks of the detected misocyclones are depicted in Fig. 8. Each misocyclone was first detected over the Japan Sea. Misocyclones A–C moved east-southeastward, but misocyclone D changed its translational direction and began to approach misocyclone A around 1631 UTC, which is possibly due to the interaction with misocyclone A and is investigated in section 4d. The misocyclone A passed over the damage path between 1634 and 1638 UTC, which was consistent with the tornado occurrence. These facts suggest that a tornado was associated with misocyclone A.

Figure 9 presents time histories of the estimated core diameter, peak tangential velocity, and vorticity of misocyclone A. Both Φ and ΔV/2 for each PPI are also shown for reference. A short time-scale variation was observed in both Φ and ΔV/2. Since the magnitude of the oscillation of Φ was almost the same or twice as large as the sampling resolution during most of the time, this short time-scale oscillation may be an artifact due to the discrete azimuthal sampling of the JR-EAST radar observations (Wood and Brown 1997, 2000).

The estimated peak tangential velocity of misocyclone A indicated that it gradually increased until misocyclone A crossed the coastline and then began to decrease soon after landfall. Coincident with the peak tangential velocity, the diameter contracted before landfall, but it did not show much variation after crossing the coastline. The vorticity variation resulted in the increase before landfall and the decrease after landfall. A notable point is that the peak tangential velocity started to decrease and the vortex began to decay just after it crossed the coastline, which is discussed in section 5a.

Close-up views of misocyclone A (Fig. 10) showed a low-reflectivity “eye” and enhanced reflectivity surrounding it (high-reflectivity tube), which were analogous to those reported in past radar observations of tornadoes (e.g., Wurman and Gill 2000; Bluestein et al. 2003b; Tanamachi et al. 2007). It is likely that this low-reflectivity eye was due to the centrifuging of hydrometeors and dynamically induced downward motion (e.g., Wurman et al. 1996; Dowell et al. 2005). However, we could not find any evidence of divergence in the Doppler velocity pattern. The low-reflectivity eye was clearly visible during the most vigorous stage of misocyclone A (between around 1630:11 and 1636:01 UTC). Dowell et al. (2005) demonstrated that the important processes for the development of a low-reflectivity eye and a high-reflectivity tube around that include centrifugal ejection of hydrometeors and recycling of small or medium raindrops. As the high-reflectivity tube of misocyclone A passed over the Sakata Weather Station, graupel and small hail (7 mm in diameter) were observed. This suggests the possibility of such recycling process.

The core diameter, peak tangential velocity, and vorticity of misocyclones B, C, and D were also estimated. The core diameters of misocyclones B and C were almost constant at first and then expanded with time until both misocyclones dissipated over the sea (not shown). The peak tangential velocities of these misocyclones were lower than that of misocyclone A and almost constant with time. It was suggested that misocyclones B and C were in their decaying stage throughout the observation.

The peak tangential velocity of misocyclone D rapidly decreased after it crossed the coastline (Fig. 11). Such variation was similar to that of misocyclone A, suggesting the influence of landfall. The diameter of the misocyclone D contracted after landfall, which was different from that of misocyclone A and is discussed in section 5a. Another noticeable feature of misocyclone D was that its diameter did not expand in the decaying stage, in strong contrast with the other three misocyclones and with the temporal variation found in previous studies (e.g., Bluestein et al. 2003b). It is possible that such variation was also related to the interaction between misocyclones A and D.

Characteristics of the four misocyclones are summarized in Table 2. The lifetime of each misocyclone was 15–30 min. The core diameters ranged from 400 to 2400 m, and their peak tangential velocities ranged from 9 to 18 m s−1, resulting in estimated vorticities on the order of 10−2–10−1 s−1. The strongest and longest-lived of the four misocyclones was misocyclone A, the only one that was associated with tornado damage. Therefore, we focus on misocyclone A in the following sections.

c. Detailed structure of the misocyclone

To examine the velocity structure of misocyclone A, we compared its Doppler velocity field with that of conceptual models. The tangential velocity Vt of the modified Rankine vortex is
i1520-0493-139-2-351-e1
where R is the distance from the center of the vortex and V0 is the peak tangential velocity at R = R0, and the α is a constant. Previous observational studies determined that α in Eq. (1) ranged from 0.5 to 0.7 in most cases (e.g., Wurman 2002; Wurman and Gill 2000; Wurman and Alexander 2005). These results implied frictional loss of angular momentum (e.g., Wurman and Alexander 2005).

The spatial variations of Doppler velocity in the azimuthal direction across the misocyclone A at 1635:02 UTC is indicated in Fig. 12. The wind profile from the conceptual models, to which the translational velocity of the misocyclone was added, is superimposed for comparison. The Doppler velocity profile presents maximum and minimum through the core. The observed flow of misocyclone A in the outer region fitted better with that of a modified Rankine vortex (α = 0.7) rather than with a Rankine vortex, which is consistent with the previous studies. This result indicated that the velocity structure of the observed misocyclone was similar to those of the previously studied tornadoes. It is considered that the observed distance between the Doppler velocity extremes and the difference between them underestimated the true core diameter and peak tangential velocity, so the better fit with a modified Rankine vortex may be partially due to the underestimation of these parameters.

Next, we examine the relationship between the Doppler velocity and surface observation associated with misocyclone A. We used data from Sakata Weather Station (Fig. 1), over which the misocyclone A passage was clearly identified from the JR-EAST radar observation. Figure 13 depicts the time histories of surface wind speed, wind direction, surface pressure, temperature, and relative humidity measured between 1625 and 1650 UTC. Since the wind measurement at 0.25-s intervals are processed to output the wind speed and direction at 10-s intervals, the minimum and the maximum wind directions during each 10-s interval are also plotted to examine the observed wind direction in detail. The observed increase of wind speed (∼6 m s−1) and sudden change of wind direction (SSE → W → NNW) from 1635:00 to 1636:40 UTC implied the passage of a cyclonic vortex. The large differences between the maximum and the minimum wind directions for the same period suggested that the wind direction varied significantly. The sharp pressure dip of 3–3.4 hPa from 1635:20 to 1635:50 UTC also indicated the passage of a vortex. As mentioned in section 3, the wind direction shifted from southwesterly to westerly or west-northwesterly and the surface temperature gradually decreased in association with the passage of this vortical storm.

Figure 14 presents close-up views of misocyclone A superimposed on the surface pressure and wind through the use of time-to-space conversion of surface observation data. The running mean of the translational velocity of misocyclone A was used as a uniform translation speed. The diameter of the core region was estimated to be 320 m from the strong turbulent wind variation and 480 m from the pressure dip at the surface, which were almost the same scale as that detected by the Doppler velocities at about 500 m MSL [520 (500) m at 1635:02 (1635:32) UTC]. The damage width was about 100 m based on the damage survey (section 2b). Since the translational velocity of misocyclone A was nearly the same as its peak tangential velocity, it is highly probable that the region of strong westerly winds, which is capable of causing damage, was narrower than the core diameter. It is suggested that the damage width was consistent with the core diameter estimated by surface observation. Misocyclone A was located slightly leeward of the surface vortex, possibly due to the vertical shear of the environmental wind field.

From Doppler velocity data, we estimated that the peak tangential velocities of misocyclone A were 15.7 (15.5) m s−1 at 1635:02 (1635:32) UTC. From these estimates, the amplitude of the pressure dip was calculated for the modified Rankine vortex expressed in Eq. (1). The pressure field of such a vortex is calculated by integrating the cyclostrophic equation (e.g., Winn et al. 1999; Lee and Wurman 2005) and is expressed as follows:
i1520-0493-139-2-351-e2
Here, P is the pressure at a great distance from the vortex, and ρ is the air density. By using the estimated constants ρ = 1.25 kg m−3 and α = 0.7, the pressure dip from P was estimated as 3.6–3.7 hPa at the vortex center. Although the observed pressure dip was measured from a certain distance from the vortex, the potential error is likely to be smaller than 0.3 hPa when the distance was more than 5 times greater than the radius of peak tangential velocity (Lee and Wurman 2005). The estimated pressure drop agreed well with the observed value of 3–3.4 hPa.

Comparisons between the Doppler velocity and the surface data in the previous paragraphs indicated the same size and magnitude of pressure gradient in misocyclone A and the surface circulation. The size of the surface circulation was also consistent with the surface damage width, suggesting that misocyclone A corresponded to the tornado vortex itself.

d. Interaction between adjacent misocyclones

As misocyclone D approached misocyclone A, they started to rotate cyclonically around each other (Fig. 10). Bluestein et al. (2004) documented such interaction between two adjacent dust devils and suggested the consistency with the Fujiwhara effect. Marquis et al. (2007) also demonstrated similar interaction during the merger of the adjacent boundary layer misocyclones. Ohno and Takemi (2010) numerically indicated that the merger of multiple vortices is an important process in intensifying dust devils in a convective boundary layer. Thus, we quantitatively investigated the interaction between misocyclones A and D.

When like-signed vortices A and D with separation LAD and circulations ΓA and ΓD interact with each other, vortices A and D rotate about a point that divides a line segment connecting the centers of vortices by the ratio of ΓDA with velocities of ΓD/(2πLAD) and ΓA/(2πLAD). Thus, we estimated the rotation velocities of each misocyclone (VestA, VestD) from the observed peak tangential velocity and compared them with the observed rotation velocities (VobsA, VobsD). Using the peak tangential velocity of a modified Rankine vortex (α = 0.7), we calculated the circulation (ΓA, ΓD) for a certain area outside the core region of each misocyclone. The estimated rotation velocities of the misocyclones were VestA = 2.66 m s−1 and VestD = 4.18 m s−1, and the ratio of circulations ΓAD was 1.57.

We assumed that the orbit center divided a line segment connecting the centers of misocyclones A and D by the ratio of ΓAD = 1.57 and estimated the observed rotation velocities of each misocyclone around the orbit center. Figure 15 depicts the positions of misocyclones A and D during the interaction. While misocyclones A and D rotated 121.7° cyclonically about the estimated orbit center, the orbit center moved east-southeastward at almost constant translational velocities of u = 12.0 m s−1 and υ = −3.0 m s−1. This translational velocities were similar to those of misocyclones A, B, and C. Using the average separation LAD = 1600 m, we estimated the observed rotation velocities of each misocyclone to be VobsA = 2.52 m s−1 and VobsD = 3.95 m s−1. The estimated rotation velocities of each misocyclone were 106% of the observed values. These results suggest that the behavior of misocyclones A and D was primarily consistent with the rotational flow associated with the other vortex. The misocyclones A, B, and C did not appear to interact. Since the separations of misocyclones A, B, and C are larger than that of misocyclones A and D, and thus the interaction among them becomes much weaker, it is suggested that it was difficult to recognize them.

The reflectivity field associated with misocyclones A and D during their interaction looked similar to the high-amplitude boundary inflections associated with merging misocyclones in Marquis et al. (2007). The low-level convergence line also seems to be transformed during the interaction (Fig. 6). The estimated diameter and the peak tangential velocity of misocyclone D started decreasing as it began to interact with misocyclone A (Fig. 11). It is possible that the dissipation of misocyclone D was partly due to its interaction with misocyclone A (Lee and Whilhelmson 1997b). Many previous studies indicated that misocyclone mergers were observed just prior to the nonsupercell tornado occurrences and contributed to the development of the active moist convection (e.g., Wilson 1986; Wilczak et al. 1992; Roberts and Wilson 1995). Unlike them, the interaction in the present case started after the active convection developed and the misocyclone A reached tornadic intensity.

5. Discussion

a. Generation and development of the tornado and the misocyclones

We found several important features related to the generation of the tornado and the misocyclones. A clear convergence line was observed at the edge of the easternmost misocyclone (Figs. 6b,c). The surface observation at Sakata Weather Station also implied that misocyclone A was located at the surface wind shear boundary (Fig. 13). It is possible that low-level horizontal wind shear was important for initiating the misocyclones.

The vorticity of misocyclone A continued to increase to the order of 1.0 × 10−1 s−1 before landfall (section 4b). Since the reflectivity around misocyclone A continued to increase before its landfall, it is speculated that such increase of the vertical vorticity was due to the vortex stretching by active convection. Thus, we roughly evaluated the development of vertical vorticity of misocyclone A due to the vortex stretching based on the simplified vorticity equation after Bluestein et al. (2003b). Constant convergence δ acting on an air parcel passing into the vortex is estimated from the increase of vorticity from ζ1 to ζ2 over a time period of Δt in the following equation: −δ = (1/Δt) ln(ζ2/ζ1). The estimated vertical vorticity of misocyclone A increased from 3.9 × 10−2 s−1 (1622:53 UTC) to 1.6 × 10−1 s−1 (1631:38 UTC). The estimated convergence was 2.7 × 10−3 s−1, comparable to the average convergence estimated from the Doppler velocity (3.0 to 4.5 × 10−3 s−1). It is suggested that the stretching of the vertical vorticity was the primary factor in misocyclone A’s development and the tornado occurrence. Such development seemed to be similar to the nonsupercell tornadoes (e.g., Wakimoto and Wilson 1989; Lee and Whilhelmson 1997b).

Since many of the misocyclones generated over the Japan Sea in winter travel toward the coastal region, it is important to know their behavior during landfall. The peak tangential velocities of both misocyclones A and D decreased after landfall (section 4b). It has been noted that an increase in surface roughness causes an increased frictional dissipation in the surface layer and causes transition to a lower swirl configuration (e.g., Leslie 1977). The effect of increasing roughness results in reduced peak tangential velocities and increased core diameters (e.g., Church and Snow 1993). The observed decrease of peak tangential velocities through landfall (increase of surface roughness) is consistent with previous studies (e.g., Dessens 1972; Wilkins et al. 1975). The diameter of the misocyclone D decreased after the landfall, which is different from the previous studies and may be due to some other factors.

b. Comparison with other tornadic storms

The parent storm formed within the mesoscale weak frontal zone and developed into a meso-β-scale vortical disturbance. Occurrence of such small-scale (meso-γ to -β scale) vortical disturbances is characteristic feature of winter disturbances over the Japan Sea (e.g., Miyazawa 1967; Asai 1988; Nagata 1993). The horizontal shear instability has been implicated in the initiation and development of such disturbances (e.g., Nagata 1993). Since the parent storm developed within the wind shear zone, it is possible that the horizontal shear instability was also important for the development of the parent storm. In this region, small-scale (meso-β or meso-γ scale) vortical storms often develop in association with larger scale vortical disturbances (Ninomiya and Hoshino 1990; Tsuboki and Asai 2004; Kawashima and Fujiyoshi 2005). The development of the misocyclones within the mesoscale vortical disturbance in this study seems similar to such multiscale structures.

The morphology of the observed parent storm resembled a typical supercell storm with cold air converging into a rear-flank gust front (e.g., Lemon and Doswell 1979; Klemp and Rotunno 1983). In contrast with such a typical gust front feature, the observed vortical disturbance had a warm, low pressure core in the center of the circulation (section 4a). The previous studies reported such winter vortical disturbances with warm cores (e.g., Laird et al. 2001; Kawashima and Fujiyoshi 2005), and some of them attributed it to the dynamically induced downward motion. However, we do not have enough data to substantiate that.

The storm environment was characterized by moderate to relatively large low-level vertical wind shear and small CAPE (section 3). These environmental conditions are similar to those for other winter tornadoes in the Japan Sea coastal region (e.g., Kobayashi et al. 2007) and for cool-season tornadoes of other areas (Hanstrum et al. 2002), but differed from the large wind shear and large CAPE environments that are typical of the supercell storms over the Great Plains in the United States.

6. Conclusions

Characteristics of a tornado, low-level misocyclones, and the parent vortical disturbance in the Japan Sea coastal region on 1 December 2007 were examined by high-resolution JR-EAST X-band Doppler radar observation and analysis.

The parent storm developed within the mesoscale frontal zone at the leading edge of a cold-air outbreak. The morphology of the parent storm somewhat resembled a supercell, but a warm, low pressure core at the rotation center contrasted with the gust front structure of typical supercells. Environmental conditions with small CAPE and moderate to relatively large low-level wind shear were similar to those for cool-season tornadoes.

The time-dependent behavior of the misocyclones indicated that the peak tangential velocity began to decrease after landfall. It was suggested that such vortex dissipation during landfall were consistent with the greater surface roughness over land. A low-level convergence line appeared along the easternmost misocyclone, suggesting that it had an important role in developing the tornado and the misocyclones through vortex stretching.

The misocyclones showed several structures presented in previous observation studies on tornadoes, such as a clear low-reflectivity “eye” and a velocity profile similar to that of a modified Rankine vortex. Surface wind and pressure variations were consistent with the passage of the radar-derived misocyclone. The observed pressure drop was also consistent with that computed from the cyclostrophic equation for the modified Rankine vortex. The behavior of the two adjacent vortices was primarily consistent with the circulation associated with the other vortex.

The vortical disturbance is a typical configuration of winter storms over the Japan Sea, and such winter storms have also been observed over the Great Lakes in the United States. Small vortices that are potentially hazardous have also been observed in other areas (e.g., Groenemeijer 2003). How the tornadoes occur and develop in association with winter mesoscale vortex should be investigated systematically in the future.

Acknowledgments

The authors would like to express their thanks to the Sakata Fire Department for providing surface weather data. The authors are grateful to Prof. Niino of the University of Tokyo and Prof. Fujiyoshi and Dr. Kawashima of Hokkaido University for their beneficial comments. Thanks are extended to Dr. Shimose for his assistance in the surface data analysis. The authors are grateful to Dr. R. J. Trapp and three anonymous reviewers for their helpful comments. Doppler radar data analyses were performed using the Doppler radar analysis tool “Draft” of MRI. This research is supported by the Program for Promoting Fundamental Transport Technology Research from the Japan Railway Construction, Transport and Technology Agency (JRTT).

REFERENCES

  • Asai, T., 1988: Meso-scale features of heavy snowfalls in Japan Sea coastal regions of Japan (in Japanese). Tenki, 35 , 156161.

  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2003a: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part I: Tornadogenesis. Mon. Wea. Rev., 131 , 29542967.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., W. C. Lee, M. Bell, C. C. Weiss, and A. L. Pazmany, 2003b: Mobile Doppler radar observations of a tornado in a supercell near Bassett, Nebraska, on 5 June 1999. Part II: Tornado-vortex structure. Mon. Wea. Rev., 131 , 29682984.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., C. C. Weiss, and A. L. Pazmany, 2004: Doppler radar observations of dust devils in Texas. Mon. Wea. Rev., 132 , 209224.

    • Search Google Scholar
    • Export Citation
  • Brown, R. A., V. T. Wood, and D. Sirmans, 2002: Improved tornado detection using simulated and actual WSR-88D data with enhanced resolution. J. Atmos. Oceanic Technol., 19 , 17591771.

    • Search Google Scholar
    • Export Citation
  • Church, C. R., and J. T. Snow, 1993: Laboratory models of tornadoes. The Tornado: Its Structure, Dynamics, Prediction and Hazards, Geophys. Monogr., Vol. 79, Amer. Geophys. Union, 277–295.

    • Search Google Scholar
    • Export Citation
  • Dessens, J., 1972: Influence of ground roughness on tornadoes: A laboratory simulation. J. Appl. Meteor., 11 , 7275.

  • Dowell, D. C., C. R. Alexander, J. M. Wurman, and L. J. Wicher, 2005: Centrifuging of hydrometeors and debris in tornadoes: Radar-reflectivity patterns and wind-measurement errors. Mon. Wea. Rev., 133 , 15011524.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1971: Proposed characterization of tornadoes and hurricanes by area and intensity. SMRP Research Paper 91, University of Chicago, 15 pp.

    • Search Google Scholar
    • Export Citation
  • Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38 , 15111534.

  • Groenemeijer, P., 2003: Three events of strong deep moist convection in the Netherlands. Internal Rep., KNMI, 76 pp.

  • Hanstrum, B. N., G. A. Mills, A. Watson, J. P. Monteverdi, and C. A. Doswell, 2002: The cool-season tornadoes of California and southern Australia. Wea. Forecasting, 17 , 705722.

    • Search Google Scholar
    • Export Citation
  • Kato, W., H. Suzuki, M. Shimamura, K. Kusunoki, and T. Hayashi, 2007: The design and initial testing of an X-band Doppler radar for monitoring hazardous winds for railroad system. Preprints, 33rd Conf. on Radar Meteorology, Cairns, Australia, Amer. Meteor. Soc., P13A.15. [Available online at http://ams.confex.com/ams/33Radar/techprogram/paper_123530.htm].

    • Search Google Scholar
    • Export Citation
  • Kawashima, M., and Y. Fujiyoshi, 2005: Shear instability wave along a snowband: Instability structure, evolution, and energetics derived from dual-Doppler radar data. J. Atmos. Sci., 62 , 351370.

    • Search Google Scholar
    • Export Citation
  • Kitagawa, N., and K. Michimoto, 1994: Meteorological and electrical aspects of winter thunderclouds. J. Geophys. Res., 99 , (D5). 1071310721.

    • Search Google Scholar
    • Export Citation
  • Klemp, J. B., and R. Rotunno, 1983: A study of the tornadic region within a supercell thunderstorm. J. Atmos. Sci., 40 , 359377.

  • Kobayashi, F., K. Kikuchi, and H. Uyeda, 1996: Life cycle of the Chitose tornado of September 22, 1988. J. Meteor. Soc. Japan, 74 , 125140.

    • Search Google Scholar
    • Export Citation
  • Kobayashi, F., Y. Sugimoto, T. Suzuki, T. Maesaka, and Q. Moteki, 2007: Doppler radar observation of a tornado generated over the Japan Sea coast during a cold air outbreak. J. Meteor. Soc. Japan, 85 , 321334.

    • Search Google Scholar
    • Export Citation
  • Kusunoki, K., and Coauthors, 2008: An overview of the Shonai area railroad weather project and early outcomes. Preprints, Fifth European Conf. on Radar in Meteorology and Hydrology, Helsinki, Finland, European Meteorological Society, P12.1.

    • Search Google Scholar
    • Export Citation
  • Laird, N. F., L. J. Miller, and D. A. R. Kristovich, 2001: Synthetic dual-doppler analysis of a winter mesoscale vortex. Mon. Wea. Rev., 129 , 312331.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., and R. B. Whilhelmson, 1997b: The numerical simulation of non-supercell tornadogenesis. Part II: Evolution of a family of tornadoes along a weak outflow boundary. J. Atmos. Sci., 54 , 23872415.

    • Search Google Scholar
    • Export Citation
  • Lee, W.-C., and J. Wurman, 2005: Diagnosed three-dimensional axisymmetric structure of the Mulhall tornado on 3 May 1999. J. Atmos. Sci., 62 , 23732393.

    • Search Google Scholar
    • Export Citation
  • Lemon, L. R., and C. A. Doswell, 1979: Severe thunderstorm evolution and mesocyclone structure as related to tornadogenesis. Mon. Wea. Rev., 107 , 11841197.

    • Search Google Scholar
    • Export Citation
  • Leslie, F. W., 1977: Surface roughness effects on suction vortex formation: A laboratory simulation. J. Atmos. Sci., 34 , 10221027.

  • Marquis, J. N., Y. P. Richardson, and J. M. Wurman, 2007: Kinematic observations of misocyclones along boundaries during IHOP. Mon. Wea. Rev., 135 , 17491768.

    • Search Google Scholar
    • Export Citation
  • Miyazawa, S., 1967: On vertical mesoscale disturbances observed during the period of heavy snow or rain in the Hokuriku district. J. Meteor. Soc. Japan, 45 , 166176.

    • Search Google Scholar
    • Export Citation
  • Nagata, M., 1993: Meso-β-scale vortices developing along the Japan-Sea polar-airmass convergence zone (JPCZ) cloud band: Numerical simulation. J. Meteor. Soc. Japan, 71 , 4357.

    • Search Google Scholar
    • Export Citation
  • Niino, H., O. Suzuki, H. Nirasawa, T. Fujitani, H. Ohno, I. Takayabu, N. Kinoshita, and Y. Ogura, 1993: Tornadoes in Chiba prefecture on 11 December 1990. Mon. Wea. Rev., 121 , 30013018.

    • Search Google Scholar
    • Export Citation
  • Niino, H., T. Fujitani, and N. Watanabe, 1997: A statistical study of tornadoes and waterspouts in Japan from 1961 to 1993. J. Climate, 10 , 17301752.

    • Search Google Scholar
    • Export Citation
  • Ninomiya, K., and K. Hoshino, 1990: Evolution process and multi-scale structure of a polar low developed over the Japan Sea on 11–12 December 1985. Part II: Meso-β-scale low in meso-α-scale polar low. J. Meteor. Soc. Japan, 68 , 307318.

    • Search Google Scholar
    • Export Citation
  • Ohno, H., and T. Takemi, 2010: Mechanisms for intensification and maintenance of numerically simulated dust devils. Atmos. Sci. Lett., 11 , 2732. doi:10.1002/asl.249.

    • Search Google Scholar
    • Export Citation
  • Roberts, R. D., and J. W. Wilson, 1995: The genesis of three nonsupercell tornadoes observed with dual-Doppler radar. Mon. Wea. Rev., 123 , 34083436.

    • Search Google Scholar
    • Export Citation
  • Suzuki, O., H. Niino, H. Ohno, and H. Nirasawa, 2000: Tornado-producing mini supercells associated with Typhoon 9019. Mon. Wea. Rev., 128 , 18681882.

    • Search Google Scholar
    • Export Citation
  • Tanamachi, R. L., H. B. Bluestein, W. C. Lee, M. Bell, and A. Pazmany, 2007: Ground-based velocity track display (GBVTD) analysis of W-band Doppler radar data in a tornado near Stockton, Kansas, on 15 May 1999. Mon. Wea. Rev., 135 , 783800.

    • Search Google Scholar
    • Export Citation
  • Tsuboki, K., and T. Asai, 2004: The multi-scale structure and development mechanism of mesoscale cyclones over the Sea of Japan in winter. J. Meteor. Soc. Japan, 82 , 597621.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., and J. W. Wilson, 1989: Non-supercell tornadoes. Mon. Wea. Rev., 117 , 11131140.

  • Wilczak, J., D. Wolfe, R. Zamora, B. Stankov, and T. Christian, 1992: Observations of a Colorado tornado. Part I: Mesoscale environment and tornadogenesis. Mon. Wea. Rev., 120 , 497521.

    • Search Google Scholar
    • Export Citation
  • Wilkins, E. M., Y. Sasaki, and H. L. Johnson, 1975: Surface friction effects on thermal convection in a rotating fluid: A laboratory simulation. Mon. Wea. Rev., 103 , 305317.

    • Search Google Scholar
    • Export Citation
  • Wilson, J. W., 1986: Tornadogenesis by nonprecipitation induced wind shear lines. Mon. Wea. Rev., 114 , 270284.

  • Winn, W. P., S. J. Hunyady, and G. D. Aulich, 1999: Pressure at the ground in a large tornado. J. Geophys. Res., 104 , (D18). 2206722082.

    • Search Google Scholar
    • Export Citation
  • Wood, V. T., and R. A. Brown, 1997: Effects of radar sampling on single-Doppler velocity signatures of mesocyclones and tornadoes. Wea. Forecasting, 12 , 928938.

    • Search Google Scholar
    • Export Citation
  • Wood, V. T., and R. A. Brown, 2000: Oscillations in mesocyclone signatures with range owing to azimuthal radar sampling. J. Atmos. Oceanic Technol., 17 , 9095.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., 2002: The multiple-vortex structure of a tornado. Wea. Forecasting, 17 , 473505.

  • Wurman, J., and S. Gill, 2000: Finescale radar observations of the Dimmitt, Texas (2 June 1995), tornado. Mon. Wea. Rev., 128 , 21352164.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., and C. R. Alexander, 2005: The 30 May 1998 Spencer, South Dakota, storm. Part II: Comparison of observed damage and radar-derived winds in the tornadoes. Mon. Wea. Rev., 133 , 97119.

    • Search Google Scholar
    • Export Citation
  • Wurman, J., J. M. Straka, and E. N. Rasmussen, 1996: Fine-scale Doppler radar observations of tornadoes. Science, 272 , 17741777.

  • Yamauchi, H., O. Suzuki, and K. Akaeda, 2006: A hybrid multi-PRI method to dealias Doppler velocities. SOLA, 2 , 9295.

APPENDIX

Estimation of Core Diameters and Peak Tangential Velocities of Misocyclones

With the JR-EAST radar, the estimation of the Doppler velocity is different from that with the commonly used mean Doppler velocity. The Doppler velocity is estimated from the Doppler frequency fk at which the Doppler spectral density Sn( fk) is at its largest. We define “mode” Doppler velocity υmd as υmd = −(λ/2) fk, where λ is the wavelength. By using the analytical model simulations, we demonstrated how the vortex signatures are exemplified in the mode Doppler velocity field and also how we estimated the core diameters and peak tangential velocities of the vortices from the observed vortex signatures.

We used an analytical model similar to that in Wood and Brown (1997, hereafter WB97) to generate the simulated Doppler velocity measurements of the vortices. For simplicity, we treated a simplified two-dimensional problem following the assumptions made in WB97. The assumptions were 1) the velocity field is uniform in height, 2) reflectivity is uniform across the vortex, and 3) the beam axis is horizontal. We used the modified Rankine vortex model given by
i1520-0493-139-2-351-ea1
where Vt is the tangential velocity, R is the distance from the center of the vortex, V0 is the peak tangential velocity at the core radius R0, and α is a constant. In Eq. (A1), V0, R0, and α were first set to be 20 m s−1, 150 m, and 0.7, respectively.
Following WB97, the mean Doppler velocity υd(θ0, r0) at the center azimuth angle θ0 and range r0 of the effective resolution volume of the radar beam can be written as follows:
i1520-0493-139-2-351-ea2
where I and J are odd-number subpoints in azimuth θi and range rj centered on the effective resolution volume and υd(θ, r) is the Doppler velocity component at each subpoint. I = 42 601 and J = 75 are used. Here |W(r)|2 is the magnitude of the two-way range-weighting function with a 6-dB width r6. Here f4(θ) is the two-way antenna pattern-weighting function with an effective beamwidth θe. They are expressed as follows:
i1520-0493-139-2-351-ea3
i1520-0493-139-2-351-ea4
where r6 = 75 m and θe = 2.13° (from a beamwidth of 2.0° and an azimuthal sampling resolution ΔAZ of 0.75°) are used for the JR-EAST radar.

For the calculation of mode Doppler velocity υmd(θ0, r0), we used 301 velocity bins with increments of 0.2 m s−1. The spectral power of each bin was calculated by summing the composite antenna-weighting function |W(r)|2f4(θ) at each subpoint within the resolution volume for the Doppler velocities within the specific velocity bin. The mode Doppler velocity is estimated by finding the Doppler velocity bin where the spectral power is largest.

We investigated the relationships between the simulated vortex signature parameters and vortex model parameters. The azimuthal angular distance Φ between the Doppler velocity maximum (Vmax) and minimum (Vmin), and one-half the difference ΔV/2 [=(VmaxVmin)/2] between Vmax and Vmin were calculated for the vortex signature parameters. The true azimuthal angular distance Ψ0, representing R0 (e.g., Ψ0 ∼ 2R0/r, where r is the distance from the radar), and V0 were used as the vortex model parameters. As mentioned in Brown et al. (2002), Φ and ΔV/2 depend on the angular separation between the center of the vortex and the center of the closest resolution volume. To take all the possible angular separations into account and estimate the most likely Φ and ΔV/2, we calculated their average (Φ and ΔV/2) by computing Φ and ΔV/2 for 150 different angular separations from 0.0° to 0.75° (=ΔAZ).

Figure A1 illustrates Ψ0 versus Φ for the simulated mean and mode Doppler velocity fields. When Ψ0 becomes smaller than θe, Φ value approaches θe in the mean Doppler velocity field (Brown et al. 2002). In the mode Doppler velocity field, Φ becomes approximately ΔAZ. These results indicate that it is difficult to estimate the true core diameter of smaller vortices from Φ in both the mean and mode Doppler velocity field. However, when Ψ0 is larger than θe, Φ of the mode Doppler velocity signature underestimates Ψ0, whereas Φ of the mean Doppler velocity signature overestimates Ψ0. The relationship between Φ and Ψ0 in the mode Doppler velocity signature in this case was estimated and is shown in Fig. A1.

Figure A2 illustrates Ψ0 versus the average velocity ratio γυ(=ΔV/2/V0). In both the mean and mode Doppler velocity fields, ΔV/2 underestimated V0 (e.g., γυ < 1). As Ψ0 approached θe, γυ decreased. The important point is that ΔV/2 in the mode Doppler velocity field is less degraded than in the mean Doppler velocity field [e.g., γυ (mode) > γυ (mean)]. The relationship between Ψ0 and γυ in the mode Doppler velocity signature was estimated and is shown in Fig. A2.

We estimated R0 and V0 from the observed Φ and ΔV/2 based on the relationships shown in Figs. A1 and A2. The observed Φ and ΔV/2 were estimated by calculating the running mean of 5 PPI scans for each variable to reduce the dependence on the angular separations between the center of the vortex and the center of the closest resolution volume. We show two examples (Figs. A1 and A2). When the radar observed a vortex signature in which Φ is 1.26° (i.e., 550 m) and ΔV/2 is 15.0 m s−1 at a range of 25 km, the estimated true core diameter was 990 m (i.e., Ψ0 = 2.26°) and the true peak tangential velocity was 16.8 m s−1. In the same manner, from a vortex signature of Φ = 2.29° (i.e., 400 m) and ΔV/2 = 15.0ms−1 at a range of 10 km, the true core diameter was estimated to be 570 m (i.e., Ψ0 = 3.28°), and the true peak tangential velocity 15.9 m s−1.

The peak tangential velocities and core radii in our observation were approximately V0 = 10–20 m s−1 and R0 = 200–1200 m, respectively. Therefore, we performed simulations using different values for V0 (10 and 30 m s−1) and R0 (75, 112.5, 300, 600, and 1200 m) to show the estimation error. By comparing the simulated results obtained using different values for V0 and R0 (not shown), the estimation error was calculated to be within 5% for both Ψ0 and V0 when R0 > r6 (which was satisfied in our case).

Fig. 1.
Fig. 1.

Map of the Shonai area, Yamagata Prefecture, Japan. (a) Cross and range circles represent the location of the JR-EAST radar and its observational range at 10-km intervals in radius. Contours represent topography (100-, 200-, and 500-m intervals). Dots denote the locations of surface automated weather stations. (b) Crosses denote the locations of damaged houses (A–D), the solid inverted triangle indicates the location of the Sakata Fire Department, and the diamond indicates the location of the JMA’s Sakata Weather Station.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 2.
Fig. 2.

Schematic diagram of the Doppler spectrum (solid curve) and the estimated mode and mean Doppler velocities (bold arrows).

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 3.
Fig. 3.

(a) Sea level pressure and surface wind vectors from MANAL at 1500 UTC 1 Dec 2007. The solid lines are the isobars at 2-hPa intervals. (b) MTSAT satellite IR1 image for the same time. The solid square and the dot represent the locations of the JR-EAST radar and the JMA operational Niigata radar, respectively. The dashed rectangular region indicates the region depicted in Fig. 4.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 4.
Fig. 4.

Time series of the radar reflectivity of the JMA Niigata radar at an elevation angle of 0.0° and surface winds (barbed arrows) by Automated Meteorological Data Acquisition System (AMeDAS). The parent storm of the tornado is indicated by the arrow. The bold open square represents the location of the JR-EAST radar.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 5.
Fig. 5.

Hodograph of the mean wind profiler observation at Sakata Weather Station before (1610–1630 UTC, solid line with circles) and after (1700–1730 UTC, dotted line with squares) the passage of the parent storm on 1 Dec. The grayscale represents the observation height in MSL.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 6.
Fig. 6.

(left) Doppler velocity fields and (right) reflectivity fields of the storm at (a) 1623:51, (b) 1632:07, and (c) 1640:24 UTC. Four misocyclones (A–D) are indicated by couplets of (left) Doppler velocity maximum and minimum as viewed from the JR-EAST radar (cross). Wind barbs measured at automated weather stations are also depicted (one barb denotes 5 m s−1). Bold dashed lines and arrows in (b) and (c) denote the estimated convergence line. The dashed rectangular region in (c) indicates the location of the region depicted in Fig. 10. The dots are the same as those in Fig. 1. The gray line represents the coastline.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 7.
Fig. 7.

Contours of (a) surface-temperature deviation fields and (b) surface-pressure deviation fields of the storm at 1634:04 and 1646:43 UTC. Reflectivity of the storm is represented by shading. Contour intervals are at 0.25 K in temperature deviation and 0.1 hPa in pressure deviation. Wind barbs measured at automated weather stations are also depicted (a barb denotes 5 m s−1). Dots and the gray line are the same as those in Fig. 6.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 8.
Fig. 8.

Tracks of misocyclones A–D between 1612:10 and 1650:07 UTC derived from the couplets of Doppler velocity maximum and minimum for each PPI scan at 30-s intervals. The size of the shaded circle represents the diameter of each misocyclone, and the grayscale denotes the observation time. The locations of the damaged houses are depicted by bold crosses. Range marks are 5- and 10-km intervals from the JR-EAST radar (cross).

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 9.
Fig. 9.

Temporal variations of (a) core diameter, peak tangential velocity, and (b) vertical vorticity of misocyclone A. The bold lines represent the estimated core diameter, peak tangential velocity, and vertical vorticity and the thin gray lines and dots represent Φ and ΔV/2 for each PPI scan. Time of misocyclone A’s landfall and period of misocyclone A passage over the damage path are also indicated. Beam height of misocyclone A for each scan is also plotted in (b).

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 10.
Fig. 10.

Close-up views of misocyclones A and D at (a) 1627:15, (b) 1630:11, (c) 1633:06, (d) 1636:01, (e) 1638:56, and (f) 1641:51 UTC during the passage over the damage path denoted by the dashed rectangle in Fig. 6c. (left) Doppler velocity fields and (right) reflectivity fields are shown. Dots, crosses, the inverted triangle, and the diamond are the same as those in Fig. 1.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 11.
Fig. 11.

As in Fig. 9, but for misocyclone D.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 12.
Fig. 12.

Azimuthal profile of Doppler velocity (dots) and radar reflectivity (gray open squares) through the core of misocyclone A at 1635:02 UTC. Profiles from the conceptual vortex models are denoted by dotted (Rankine) and dashed (modified Rankine: VtR−0.7 in the outer region) lines for comparison.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 13.
Fig. 13.

Time histories of (a) wind speed, (b) wind direction, maximum/minimum wind direction, (c) surface pressure, and (d) temperature (bold line) and relative humidity (thin line) at Sakata Weather Station between 1625 and 1650 UTC.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 14.
Fig. 14.

Close-up views of misocyclone A at 1635:02 UTC, and surface pressure (solid line) and wind (barb) estimated through time-to-space conversion of surface observation data of Sakata Weather Station. (left) Doppler velocity field and (right) reflectivity field are shown.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. 15.
Fig. 15.

Positions of misocyclones A and D (open circles) during their interaction (between 1631:09 and 1639:54 UTC). The size of each circle represents the size of each misocyclone. The dots represent the estimated orbit centers that divide each line segment between misocyclones A and D by the ratio of 1.57. The dotted line connecting the orbit centers at 1631:09 and 1639:54 UTC is shown for reference.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. A1.
Fig. A1.

True angular distance Ψ0 vs the average angular distance Φ for the simulated mode (dots) and mean (squares) Doppler velocity signatures. The black solid line is the approximated distribution of best fit to the calculated data points for the mode Doppler velocity signature. The black dotted lines correspond to the examples given in the text.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Fig. A2.
Fig. A2.

True angular distance Ψ0 vs the average velocity ratio γυ for the simulated mode (dots) and mean (squares) Doppler velocity signatures. The solid line is the approximated distribution of best fit to the calculated data points for the mode Doppler velocity signature. The black dotted lines correspond to the examples given in the text.

Citation: Monthly Weather Review 139, 2; 10.1175/2010MWR3247.1

Table 1.

Parameters and scanning mode of JR-EAST radar.

Table 1.
Table 2.

Characteristics of misocyclones A–D derived from the JR-EAST radar data. Lifetime (T), and average eastward and northward translational velocities (υx, υy) are listed together with min–max of estimated diameter (D), peak tangential velocity (Vt), vertical vorticity (Vor) and F-scale rating of surface damage for each misocyclone.

Table 2.
Save