• Bister, M., , and K. A. Emanuel, 1997: The genesis of Hurricane Guillermo: TEXMEX analyses and a modeling study. Mon. Wea. Rev., 125 , 26622682.

    • Search Google Scholar
    • Export Citation
  • Bracken, W. E., 1999: A multiscale examination of tropical cyclogenesis. Ph.D. dissertation, The University at Albany, State University of New York, 340 pp. [Available from Dept. of Atmospheric Science, The University at Albany, State University of New York, Albany, NY 12222.].

  • Bracken, W. E., , and L. F. Bosart, 2000: The role of synoptic-scale flow during tropical cyclogenesis over the North Atlantic Ocean. Mon. Wea. Rev., 128 , 353376.

    • Search Google Scholar
    • Export Citation
  • Brandes, E. A., 1990: Evolution and structure of the 6–7 May 1985 mesoscale convective system and associated vortex. Mon. Wea. Rev., 118 , 109127.

    • Search Google Scholar
    • Export Citation
  • Davis, C. A., , and L. F. Bosart, 2001: Numerical simulations of the genesis of Hurricane Diana (1984). Part I: Control simulation. Mon. Wea. Rev., 129 , 18591881.

    • Search Google Scholar
    • Export Citation
  • Frank, W. M., , and E. A. Ritchie, 1999: Effects of environmental flow upon tropical cyclone structure. Mon. Wea. Rev., 127 , 20442061.

  • Frank, W. M., , and E. A. Ritchie, 2001: Effects of vertical wind shear on the intensity and structure of numerically simulated hurricanes. Mon. Wea. Rev., 129 , 22492269.

    • Search Google Scholar
    • Export Citation
  • Gamache, J. F., 1998: Evaluation of a fully three-dimensional variational Doppler analysis technique. Preprints, 28th Conf. on Radar Meteorology, Austin, TX, Amer. Meteor. Soc., 422–423.

  • Gamache, J. F., , F. D. Marks, , and F. Roux, 1995: Comparison of three airborne Doppler sampling techniques with airborne in situ wind observations in Hurricane Gustav (1990). J. Atmos. Oceanic Technol., 12 , 171181.

    • Search Google Scholar
    • Export Citation
  • Gerrish, H., 1991: Preliminary report: Hurricane Guillermo 4–10 August 1991. National Hurricane Center Rep., 7 pp. [Available online at ftp://ftp.nhc.noaa.gov/pub/storm_archives/pacific/prelimep/ep1991/guillerm.].

  • Hendricks, E. A., , M. T. Montgomery, , and C. A. Davis, 2004: On the role of “vortical” hot towers in tropical cyclone formation. J. Atmos. Sci., 61 , 12091232.

    • Search Google Scholar
    • Export Citation
  • Hertenstein, R., , and W. H. Schubert, 1991: Potential vorticity anomalies associated with squall lines. Mon. Wea. Rev., 119 , 16631672.

    • Search Google Scholar
    • Export Citation
  • Johnson, R. H., , and D. L. Bartels, 1992: Circulations associated with a mature-to-decaying midlatitude mesoscale convective system. Part II: Upper-level features. Mon. Wea. Rev., 120 , 13011321.

    • Search Google Scholar
    • Export Citation
  • Jorgensen, D. P., , P. H. Hildebrand, , and C. L. Frush, 1983: Feasibility test of an airborne pulse-Doppler meteorological radar. J. Appl. Meteor., 22 , 744757.

    • Search Google Scholar
    • Export Citation
  • Mapes, B. E., , and R. A. Houze, 1995: Diabatic divergence profiles in Western Pacific mesoscale convective systems. J. Atmos. Sci., 52 , 18071828.

    • Search Google Scholar
    • Export Citation
  • May, P. T., , and G. J. Holland, 1999: The role of potential vorticity generation in tropical cyclone rainbands. J. Atmos. Sci., 56 , 12241228.

    • Search Google Scholar
    • Export Citation
  • Molinari, J., , D. Vollaro, , and K. L. Corbosiero, 2004: Tropical cyclone formation in a sheared environment: A case study. J. Atmos. Sci., 61 , 24932509.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , and R. Kallenbach, 1997: A theory for vortex Rossby-waves and its application to spiral bands and intensity changes in hurricanes. Quart. J. Roy. Meteor. Soc., 123 , 435465.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , and J. Enagonio, 1998: Tropical cyclogenesis via convectively forced vortex Rossby waves in a three-dimensional quasigeostrophic model. J. Atmos. Sci., 55 , 31763207.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , M. E. Nicholls, , T. A. Cram, , and A. B. Saunders, 2005: A vortical hot tower route to tropical cyclogenesis. J. Atmos. Sci., , in press.

    • Search Google Scholar
    • Export Citation
  • Ooyama, K. V., 1982: Conceptual evolution of the theory and modeling of the tropical cyclone. J. Meteor. Soc. Japan, 60 , 369379.

  • Rappaport, E. N., 1996: Preliminary report: Hurricane Dolly 19–25 August 1996. National Hurricane Center Rep., 7 pp. [Available online at http://www.nhc.noaa.gov/1996dolly.html.].

  • Raymond, D. J., , and H. Jiang, 1990: A theory for long-lived mesoscale convective systems. J. Atmos. Sci., 47 , 30673077.

  • Raymond, D. J., , C. Lopez-Carrillo, , and L. Lopez Cavazos, 1998: Case-studies of developing east Pacific easterly waves. Quart. J. Roy. Meteor. Soc., 124 , 20052034.

    • Search Google Scholar
    • Export Citation
  • Reasor, P. D., , and M. T. Montgomery, 2001: Three-dimensional alignment and corotation of weak, TC-like vortices via linear vortex Rossby waves. J. Atmos. Sci., 58 , 23062330.

    • Search Google Scholar
    • Export Citation
  • Reasor, P. D., , M. T. Montgomery, , F. D. Marks Jr., , and J. F. Gamache, 2000: Low-wavenumber structure and evolution of the hurricane inner core observed by airborne dual-Doppler radar. Mon. Wea. Rev., 128 , 16531680.

    • Search Google Scholar
    • Export Citation
  • Reasor, P. D., , M. T. Montgomery, , and L. D. Grasso, 2004: A new look at the problem of tropical cyclones in vertical shear flow: Vortex resiliency. J. Atmos. Sci., 61 , 322.

    • Search Google Scholar
    • Export Citation
  • Riehl, H., , and J. Malkus, 1958: On the heat balance in the equatorial trough zone. Geophysica, 6 , 503538.

  • Ritchie, E. A., , and G. J. Holland, 1997: Scale interactions during the formation of Typhoon Irving. Mon. Wea. Rev., 125 , 13771396.

  • Rotunno, R., , and K. A. Emanuel, 1987: An air–sea interaction theory for tropical cyclones. Part II: Evolutionary study using a nonhydrostatic axisymmetric numerical model. J. Atmos. Sci., 44 , 542561.

    • Search Google Scholar
    • Export Citation
  • Schubert, W. H., , and J. J. Hack, 1982: Inertial stability and tropical cyclone development. J. Atmos. Sci., 39 , 16881697.

  • Simpson, J., , E. Ritchie, , G. J. Holland, , J. Halverson, , and S. Stewart, 1997: Mesoscale interactions in tropical cyclone genesis. Mon. Wea. Rev., 125 , 26432661.

    • Search Google Scholar
    • Export Citation
  • Wang, Y., , and G. J. Holland, 1996: Tropical cyclone motion and evolution in vertical shear. J. Atmos. Sci., 53 , 33133332.

  • Zehr, R., 1992: Tropical cyclogenesis in the western North Pacific. NOAA Tech. Rep. NESDIS 61, 181 pp.

  • Zhang, D-L., , and J. M. Fritsch, 1987: Numerical simulation of the meso-β scale structure and evolution of the 1977 Johnstown flood. Part II: Inertially stable warm-core vortex and the mesoscale convective complex. J. Atmos. Sci., 44 , 25932612.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Best track for Tropical Cyclone Guillermo (1991) from NHC (Gerrish 1991). The TS and H labels denote the locations where Guillermo was upgraded to a Tropical Storm and hurricane, respectively. Also shown are the TEXMEX observational domains on 3 and 4 Aug.

  • View in gallery

    Location of all mesoscale cyclonic circulations observed with airborne Doppler radar inside the TEXMEX domains shown in Fig. 1. The times indicated are in UTC, and M denotes the 5–7-km midlevel layer and L denotes the 1–3-km low-level layer. The center of circulation is used to define the vortex locations.

  • View in gallery

    Best track for Tropical Cyclone Dolly (1996) from NHC (Rappaport 1996). The TS and H labels denote the locations where Dolly was upgraded to a tropical storm and hurricane, respectively. Also shown are the observational domains on 15, 17, and 19 Aug.

  • View in gallery

    Flight tracks and flight-level winds within the pre-Dolly disturbance from the NOAA P3 aircraft at 570 hPa on 15 Aug and at 640 hPa on 17 Aug (see Fig. 3 for the location of the observations relative to Dolly’s storm track). In all figures a barb equals 2.5 m s−1 and a flag equals 10 m s−1. Labels C and V1 denote cyclonic circulations of interest. See text for further details.

  • View in gallery

    Vertical structure of winds and relative vorticity associated with low-level vortex V1 at 1914 UTC 17 Aug (FS1). Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

  • View in gallery

    Flight tracks and flight-level winds within the pre-Dolly disturbance from (a) the Air Force C-130 at 975 hPa and (b) the NOAA P3 aircraft at 480 hPa on 19 Aug. Labels C and V2–4 denote cyclonic circulations of interest. Barbs equal 2.5 m s−1 and flags equal 10 m s−1. See text for further details.

  • View in gallery

    GOES infrared satellite images of the Dolly disturbance on 19 Aug at (top left) 2145, (top right) 2245, and (bottom left) 2345 UTC. Two regions of deep convection are observed north and south of 19°N. The convective regions propagate westward, merging to form Hurricane Dolly 18 h later around (bottom right) 1745 UTC 20 Aug. An emerging convective cell and vortex V4 are identified in the northern region to the northwest of midlevel vortex V3. In the southern region, a midlevel cyclonic circulation C is identified. See text for further details. The airborne Doppler observation domain is contained within the heavy box.

  • View in gallery

    Wind and relative vorticity at (left) 1- and (right) 6-km height at (top) 1956 (FS2) and (bottom) 2230 (FS3) UTC 19 Aug within the southern convective region identified in Fig. 7. No low-level cyclonic circulations are evident, but two midlevel circulations, V2 and C, are present. The heavy solid line denotes the cross section through V2 in Fig. 9, and X indicates the location of V2 and C on the 1-km wind fields. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

  • View in gallery

    East–west cross section (see Fig. 8) of reflectivity through V2 at 1956 UTC 19 Aug. The peak tangential circulation is located near 5-km height at V2 within a stratiform region. The xz flow within the cross section is also shown. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

  • View in gallery

    Wind and relative vorticity (1- and 5-km height) at 2128 (FS4), 2245 (FS5), and 2345 (FS6) UTC 19 Aug within the northern convective region identified in Fig. 7. (a) Midlevel vortex V3 moves to the south-southwest, and (b) the evolution of a low-level cyclonic vortex V4 is observed. The heavy solid line denotes the cross section through V4 displayed in Fig. 11, and X denotes the location of V3 on the 1-km wind fields. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

  • View in gallery

    (Continued)

  • View in gallery

    East–west cross section (see Fig. 10b) of vertical velocity at 2128 UTC 19 Aug associated with the convective cell identified in infrared imagery in Fig. 7. The 1-km location of vortex V4 is identified. Contour interval is 1 m s−1. Negative values are dashed and the zero contour is not shown. The xz flow within the cross section is also shown. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

  • View in gallery

    Three-dimensional iso-surface of the vertical component of relative vorticity (15 × 10−4 s−1) showing the development of the low-level vortex at 2128, 2245, and 2345 UTC (clockwise from top). The depth shown is from 1 to 7 km. The feeder band about vortex V4 intensifies and spirals in toward the vortex core.

  • View in gallery

    Radius–height structure of azimuthal mean (using the estimated center of circulation at 5-km height) tangential wind (V, contour interval 1 m s−1) for vortex V3 in the northern convective region on 19 Aug. The tangential wind decreases in magnitude with time indicating a decay of the vortex circulation as it moves away from low-level vortex V4.

  • View in gallery

    As in Fig. 13 but for low-level vortex V4 centered at 1-km height. In addition to tangential wind, radial wind (U, contour interval 0.2 m s−1 and inward flow dashed) and the vertical component of relative vorticity (ζ, contour interval 0.5 × 10−3 s−1) are shown. The peak tangential wind increases from 7 to 8 m s−1 and moves down from a height near 2 km to below 1 km. The peak vorticity takes on a more vertically coherent structure, extending from near surface to above 7 km by the end of the observation period. The radial flow becomes increasingly dominated by inflow as time progresses.

  • View in gallery

    As in Fig. 12 but for reflectivity (30 dBZ). Shown are the primary convective cell near the center of V4 and the development of a feeder band coincident with the band of vorticity in Fig. 12.

  • View in gallery

    Absolute vertical vorticity from the RAMS TC genesis control simulation of M05. A number of VHTs are observed at the time shown, but one in particular stands out having converged elevated vorticity from previous convective events. The horizontal diameter of the dominant VHT core at 1-km height is 10–20 km with a peak absolute vorticity value of 50 × 10−4 s−1.

  • View in gallery

    Time series of 1-s P3 flight-level θe during FS6 (480 hPa) shown as a function of longitude. The elevated regions of θe correspond to transects through the spiral band near V4 and the feeder band to its east (see the 2345 UTC panel of Fig. 10 near 19.2°N for reference).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 0 0 0
PDF Downloads 0 0 0

Mesoscale Observations of the Genesis of Hurricane Dolly (1996)

View More View Less
  • 1 Department of Atmospheric Science, Colorado State University, Fort Collins, Colorado
  • 2 Department of Earth and Atmospheric Sciences, The University at Albany, State University of New York, Albany, New York
© Get Permissions
Full access

Abstract

Recent numerical studies of tropical cyclone genesis suggest a new paradigm for how the surface vortex is established based on a highly nonaxisymmetric mechanism involving the interaction of low-level cyclonic circulations generated by deep cumulonimbus convection. A reexamination of mesoscale observations during the genesis of Hurricane Guillermo (1991) confirms the presence of multiple cyclonic circulations. More recently, airborne Doppler radar wind observations during the genesis of Atlantic Hurricane Dolly (1996) also reveal multiple lower-to-middle-tropospheric mesoscale cyclonic circulations during sequential 15–20-min compositing periods.

A particularly well-organized, but initially weak (mean tangential wind of 7 m s−1), low-level cyclonic vortex embedded within the pre-Dolly tropical disturbance is observed coincident with deep, vertically penetrating cumulonimbus convection. The earliest observations of this vortex show the peak circulation near 2-km height with a mean diameter of 30–40 km. The circulation undergoes a slight intensification over a 2-h period, with the maximum tangential winds ultimately peaking below 1-km height. Approximately 18 h after these observations Dolly is classified as a hurricane by the National Hurricane Center.

A synthesis of observations during the early development of Dolly supports a stochastic view of tropical cyclone genesis in which multiple lower-to-middle-tropospheric mesoscale cyclonic circulations are involved in building the surface cyclonic circulation. It is suggested that, in particular, the interaction of low-level circulations generated by a series of deep cumulonimbus convective events, like the one documented here, within an environment of elevated cyclonic vorticity was instrumental to the formation of the Dolly surface vortex.

Corresponding author address: Dr. Paul D. Reasor, Dept. of Meteorology, The Florida State University, Tallahassee, FL 32306. Email: reasor@met.fsu.edu

Abstract

Recent numerical studies of tropical cyclone genesis suggest a new paradigm for how the surface vortex is established based on a highly nonaxisymmetric mechanism involving the interaction of low-level cyclonic circulations generated by deep cumulonimbus convection. A reexamination of mesoscale observations during the genesis of Hurricane Guillermo (1991) confirms the presence of multiple cyclonic circulations. More recently, airborne Doppler radar wind observations during the genesis of Atlantic Hurricane Dolly (1996) also reveal multiple lower-to-middle-tropospheric mesoscale cyclonic circulations during sequential 15–20-min compositing periods.

A particularly well-organized, but initially weak (mean tangential wind of 7 m s−1), low-level cyclonic vortex embedded within the pre-Dolly tropical disturbance is observed coincident with deep, vertically penetrating cumulonimbus convection. The earliest observations of this vortex show the peak circulation near 2-km height with a mean diameter of 30–40 km. The circulation undergoes a slight intensification over a 2-h period, with the maximum tangential winds ultimately peaking below 1-km height. Approximately 18 h after these observations Dolly is classified as a hurricane by the National Hurricane Center.

A synthesis of observations during the early development of Dolly supports a stochastic view of tropical cyclone genesis in which multiple lower-to-middle-tropospheric mesoscale cyclonic circulations are involved in building the surface cyclonic circulation. It is suggested that, in particular, the interaction of low-level circulations generated by a series of deep cumulonimbus convective events, like the one documented here, within an environment of elevated cyclonic vorticity was instrumental to the formation of the Dolly surface vortex.

Corresponding author address: Dr. Paul D. Reasor, Dept. of Meteorology, The Florida State University, Tallahassee, FL 32306. Email: reasor@met.fsu.edu

1. Introduction

Tropical cyclone (TC) genesis requires the transformation of a tropical disturbance, generally characterized by a mesoscale region of convection, and in some cases a midlevel cyclonic flow, into a warm-core TC vortex with maximum winds near the surface. Globally, the majority of tropical disturbances do not develop into TCs. The relative scarcity of TC genesis events may in part be a consequence of the complex multiscale nature of the phenomenon, involving a cooperation of the synoptic scale (e.g., Bracken and Bosart 2000), mesoscale (e.g., Ritchie and Holland 1997, hereafter RH97), and convective scale (Simpson et al. 1997; Montgomery and Enagonio 1998; Hendricks et al. 2004; Montgomery et al. 2005, manuscript submitted to J. Atmos. Sci., hereafter M05). Here we examine the mesoscale and convective-scale aspects of the problem utilizing a unique airborne in situ and Doppler radar dataset, described in detail by Bracken (1999, hereafter B99) that documents the three-dimensional wind structure of TC Dolly (1996) during the genesis stage.

Based on recent observations and numerical studies of the mesoscale and convective-scale aspects of TC genesis, a growing consensus exists for separating the early development of the TC into two stages. The latter stage of development occurs when air–sea interaction processes begin to dominate the growth of a finite-amplitude surface circulation in a manner consistent with the wind-induced surface heat exchange (WISHE) theory of Rotunno and Emanuel (1987). The onset of WISHE is characterized by increases in surface fluxes with wind speed, and is generally accompanied by increases in the surface equivalent potential temperature (θe) above ambient values. Molinari et al. (2004) term the initial genesis stage before significant increases of surface θe as pre-WISHE. During the pre-WISHE period mesoscale and convective-scale vortical interactions take place to build the seedling surface circulation. It is the pre-WISHE stage that we explore here.

We believe that the establishment of the surface vortex during the transformation of a tropical disturbance is central to the problem of TC genesis. Using a synthesis of flight-level aircraft measurements and composited airborne Doppler radar winds, Bister and Emanuel (1997, hereafter BE97) proposed that the surface vortex is established in part through vertical advection of vorticity associated with midlevel mesoscale convective vortices (MCVs) by downdrafts in the stratiform rain region. RH97 used flight-level measurements, omega dropwindsondes, and satellite imagery to demonstrate the interaction between lower-tropospheric synoptic-scale circulation and MCVs during the genesis process. Simpson et al. (1997) also emphasized the role of MCVs using a dataset similar to RH97. They argued that the stochastic interaction of midlevel MCVs in an environment of elevated ambient cyclonic vorticity (e.g., a monsoon trough) builds the circulation down to the surface.

Montgomery and Enagonio (1998) argued that the surface vortex need not originate at midlevels. Motivated by Zehr’s (1992) observations of convective pulsing during the early genesis stage, they focused on the dynamical role of convectively generated potential vorticity anomalies within MCV environments. The limited spatial and temporal resolution of the observations available at the time made numerical simulation the only viable option for studying the smaller-scale processes. Using a three-dimensional quasigeostrophic model, they demonstrated how low-level vorticity generated by convective bursts within a preexisting midlevel MCV can spin up the surface circulation beneath the MCV through horizontal axisymmetrization. Reasor and Montgomery (2001) clarified the vertical alignment dynamics of the preexisting vortex and convectively generated vorticity anomaly. More recently, Davis and Bosart (2001) found in a numerical case study of Hurricane Diana (1984) using parameterized convection that the surface vortex can arise through multiple mergers of low-level cyclonic vorticity. Hendricks et al. (2004) used cloud-resolving numerical simulations to demonstrate the importance of “vortical” hot towers (VHTs) in the generation of low-level vorticity during Diana’s genesis phase. The VHTs differ from the deep, vertically penetrating cumulonimbus convection discussed by Riehl and Malkus (1958) in that the tilting and stretching of elevated preexisting vorticity by the intense convection yields a strong rotational component of the flow. M05 clarified the dynamics of VHTs and their role in the formation of the TC surface circulation through a series of idealized simulations of convection initiated within a preexisting MCV.

In the present study we examine data from National Oceanic and Atmospheric Administration (NOAA) research flights into Hurricane Dolly (1996) during the genesis phase. The Dolly case compliments prior observational studies in the sense that the flight-level and satellite data confirm the important role of midlevel MCVs during the development of the TC from an incipient tropical disturbance. The Dolly case is unique because of the detailed look at the wind field on the cloud scale to mesoscale provided by the airborne Doppler radar during the period when the surface vortex began to form. We believe this to be the first TC genesis dataset to clearly document the wind evolution of a low-level cyclonic circulation on the VHT scale (10–20 km) in apparent association with deep cumulonimbus convection.

The outline of the paper is as follows. Section 2 reviews the use of Doppler radar observations during the Tropical Experiment in Mexico (TEXMEX) in 1991 (BE97; Raymond et al. 1998), which aside from Dolly (1996) is the only known source of high temporal resolution three-dimensional wind measurements of TC genesis to date. The aircraft flight-level data, Doppler radar data, and satellite measurements used in the present study of Dolly are described in section 3. In section 4, the genesis of Hurricane Dolly is presented through a synthesis of the data. A discussion of the Dolly results in the context of recent cloud-resolving numerical simulations is presented in section 5.

2. TEXMEX genesis field experiment

The first organized field campaign designed specifically to examine the mesoscale structure of tropical disturbances in the genesis phase was TEXMEX in 1991. TEXMEX focused on the eastern Pacific, climatologically the most active region of TC development on earth (based on the density of genesis cases per unit area per unit time). The experiment ran from July to August and captured six tropical disturbances in various stages of development, four of which ultimately became TCs. One of the central goals of TEXMEX, as discussed by BE97, was to test the hypothesis that the elevation of middle-tropospheric θe (i.e., moistening) above a near-surface vorticity maximum is a sufficient condition for TC genesis to occur.

The only case during TEXMEX to capture the transformation from tropical disturbance to TC was Hurricane Guillermo. Guillermo developed from an easterly wave that propagated across the Atlantic into the eastern Pacific. Figure 1 shows Guillermo’s track after the easterly wave disturbance was upgraded to a TC by the National Hurricane Center (NHC). Aircraft in situ and Doppler radar measurements obtained during two flights through the developing disturbance prior to tropical storm classification (1800 UTC 4 August) were used to test the TEXMEX hypothesis. The observational domains relative to the storm track are also indicated in Fig. 1.

An analysis of Guillermo’s genesis is provided by BE97 and B99. Here we summarize what the Doppler observations revealed about the mesoscale structures within the developing disturbance. The first flight made by the NOAA WP-3D (P3) aircraft into the pre-Guillermo disturbance began at 2300 UTC 2 August and ended at 0700 UTC 3 August. From the Doppler radar analyses B99 identified two distinct low-level (1–3 km) and midlevel (5–7 km) cyclonic circulation pairs. The locations and observation times of each vortex, identified by the center of circulation, are shown in Fig. 2. Approximately 24 h later, the developing disturbance was once again observed by Doppler radar. BE97 suggested that an elongated low-level vortex first identified at 0412 UTC transformed into Tropical Storm Guillermo by 1800 UTC.

The lack of time continuity in this case limits our ability to clearly connect the vortex observed on 4 August with the circulations on the previous day. A comparison of composited radar and flight-level data from the two flights by BE97 suggests that a moderate increase in middle-tropospheric θe was not immediately followed by the intensification of the surface circulation. To explain the observed evolution of Guillermo, BE97 proposed an axisymmetric model of TC genesis in which the advection of elevated vorticity associated with a midlevel MCV by downdrafts in the stratiform rain region serves to initiate the surface vortex.

In B99’s analysis of Guillermo the true three-dimensional asymmetric structure of the disturbance flow was highlighted. Multiple elevated vorticity centers (20–40-km diameter) were documented, some of which are associated with the distinct vortex circulations identified in Fig. 2. In particular, a cyclonic vorticity anomaly 25–50 km to the north of the circulation identified by BE97 on 4 August was argued by B99 to develop into Tropical Storm Guillermo. The circulations observed by Doppler radar and identified by BE97 and B99 during the pre-WISHE stage of development are likely only a small sample of a greater population of localized cyclonic circulations generated by intense convection, or other means (Hendricks et al. 2004; M05). If one takes the view put forth by Ooyama (1982) that genesis is a stochastic process involving the nonlinear interaction of various scales, and considers the mesoscale alone, the Doppler observations suggest that no one mesoscale circulation may be directly responsible for the transformation of the disturbance into a TC. According to the numerical studies of Hendricks et al. (2004) and M05, a number of convective scale to mesoscale circulations actually act in concert to form the system-scale cyclonic surface vortex. This paradigm is addressed further in the next sections through a detailed analysis of another storm of easterly wave origin, Hurricane Dolly (1996).

3. Data and method

a. Synoptic background

According to Rappaport (1996), Hurricane Dolly (1996) developed from an easterly wave that moved off the coast of Africa on 9 August. As the wave propagated westward through the Caribbean Sea, NHC upgraded the associated tropical disturbance to a tropical depression at 0600 UTC 19 August near 17.3°N, 80.2°W. An analysis of the pre-Dolly large-scale environment by B99 showed that warm SSTs (>27°C), weak vertical wind shear (<10 m s−1 in the 850–200-hPa layer), and synoptic forcing of vertical motion over a broad lower-tropospheric cyclonic vorticity maximum favored TC genesis. Approximately 36 h after being declared a tropical depression, Dolly was upgraded to a hurricane. Figure 3 shows Dolly’s track prior to making landfall at 1800 UTC 23 August.

b. Aircraft measurements

Mesoscale observations of the pre-Dolly tropical disturbance were made by P3 aircraft during the 1996 Hurricane Field Program of NOAA’s Hurricane Research Division (HRD). Flights on 15, 17, and 19 August document the evolution of the disturbance utilizing a flight pattern that maximizes temporal resolution of the wind field during a single investigation period, providing a unique dataset for use in studying the convective scale to mesoscale aspects of the genesis phenomenon. Figure 4 shows an example of the P3 aircraft flight tracks at lower-to-middle-tropospheric levels on 15 and 17 August. The repeated penetrations of the aircraft into the chosen core of the disturbance was a key factor in improving the temporal resolution of the Doppler observations over those in the TEXMEX Guillermo (1991) case, where a stair step flight pattern was employed to achieve the greatest spatial coverage.

On 15 and 17 August two Doppler—radar-equipped P3 aircraft flew through the pre-Dolly disturbance. Due to an aircraft problem, only one aircraft was available on 19 August. The 48-h time separation between flights, combined with the more focused flight patterns, precludes a comprehensive study of the broadscale (>100 km) circulation evolution, as was done for TEXMEX storms by Raymond et al. (1998). The 1-s flight-level thermodynamic and wind measurements provided by HRD are used here primarily to supplement the mesoscale Doppler wind analyses.

c. Doppler radar measurements

The repeated penetrations through the estimated center of the disturbance, especially on 19 August, permit multiple looks at the wind field with Doppler radar. The 3-cm wavelength tail Doppler radar of the P3s operated in a mode consistent with the fore–aft scanning technique (FAST). In FAST mode the radar alternately scans in a cone up to 25° fore and aft of the flight track perpendicular. Within an approximately 120-km-wide swath centered on the flight track the fore and aft scans provide dual looks at the flow. When the two wind measurements are supplemented with the continuity equation, estimates of the three components of the vector wind field are then obtained.

In standard iterative techniques the two Doppler measurements of velocity (Vr1 and Vr2) at a location are expressed in terms of the Cartesian wind components (U, V, W) (Jorgensen et al. 1983),
i1520-0469-62-9-3151-e1
where β1,2 are the antenna pointing angles relative to north and θ1,2 are the elevation angles from vertical. The system of equations is solved for U and V, given the hydrometeor fall speed VT and an initial guess for W. The continuity equation is then integrated vertically downward (assuming W = 0 at echo top) to obtain an updated estimate of the vertical velocity. The equations are solved iteratively until, in theory, convergence is reached. Here, we solve the projection equations and continuity equation simultaneously using the variational technique of Gamache (1998). The method is more stable than the standard iterative technique for elevation angles greater than 45°, and has demonstrated success in previous applications to hurricanes (e.g., Reasor et al. 2000).

The geometry of FAST requires that the time between the two looks at the wind field increase with distance away from the flight track. If the compositing domain is restricted to 120 km on a side (which is dictated largely by the 2° vertical beamwidth of the P3 tail radar), the time between measurements at a point is no more than a few minutes. The total compositing time for a 120-km-long swath is 15–20 min. Because the antenna must alternate between fore and aft orientations, the horizontal resolution of the observations parallel to the flight track is less than that of fixed antenna scanning techniques. The horizontal resolution in FAST mode is approximately 2 km (Gamache et al. 1995). The Doppler observations shown in the analyses here are mapped to a Cartesian coordinate system with horizontal and vertical spacings of 1.5 and 1 km, respectively.

Although the Dolly genesis missions were designed to take advantage of the two Doppler-equipped P3 aircraft, a number of setbacks limited the amount of useful data. On 15 August convection was too sparse, and the scatterers too few, to provide an extensive mapping of the wind field from Doppler radar. On 17 August the radar data tape from one plane was found to be defective upon post processing. A cracked windshield grounded one of the planes on 19 August, leaving only one P3 available to participate in the mission. The approximate location and time of each flight segment used to construct the Doppler analyses shown in the next section are listed in Table 1. The flight segments will be referred to by the number given in the table (e.g., FS1 for flight segment 1).

d. Satellite measurements

Satellite visible and infrared imagery are utilized to characterize the general structure of the pre-Dolly disturbance and to enhance our interpretation of the Doppler wind fields. The Geostationary Operational Environmental Satellite-8 (GOES-8) data were retrieved from archives at the Cooperative Institute for Research in the Atmosphere at Colorado State University.

4. The genesis of Hurricane Dolly (1996)

a. Early disturbance observations

The rapid transformation of the pre-Dolly disturbance into a TC began on 19 August. During the preceding days, convection became increasingly concentrated within the synoptic-scale cyclonic circulation of the easterly wave (Rappaport 1996). According to the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) 2.5° Re-Analysis (ERA-40) of the 700–500-hPa absolute vorticity (not shown), the wave axis moved from approximately 66°W at 1200 UTC 15 August to 73°W at 1200 UTC 17 August, at an average speed of 4 to 5 m s−1. During this period deep convection developed, but failed to organize into a concentrated pattern. Following a period of convective reinvigoration and horizontal concentration late on 17 August, the ERA-40 analyses indicate an increase in the synoptic-scale middle-tropospheric cyclonic vorticity through the end of the field observation period (0000 UTC 20 August).

Flight-level wind measurements during the early observation period from 15–17 August are shown in Fig. 4. Winds sampled over the 5° box confirm the presence of a broadscale cyclonic curvature of the flow. Cyclonic curvature is also evident on the mesoscale in the flight-level observations on 15 and 17 August. In Fig. 4, and subsequent figures, mesoscale regions of pronounced flight-level cyclonic flow curvature are denoted by “V” if coincident with well-defined cyclonic circulation and elevated relative vorticity (>5 × 10−4 s−1) in the Doppler radar analyses (e.g., a closed earth-relative circulation). We identify by “C” those regions of flight-level cyclonic flow curvature that either fall outside the Doppler radar domain or are identified with a simple cyclonic wind shift in the Doppler analyses.

On 15 August near 13.75°N, 67.5°W the wind shift along the flight track at 570 hPa indicates a possible midlevel cyclonic circulation. Precipitation scatterers were too sparse on 15 August, however, to confirm a circulation with the Doppler radar. On 17 August a cyclonic wind shift in the 640-hPa flight-level data near 15.4°N, 75°W (henceforth referred to as V1) did fall within a region of adequate precipitation to permit a three-dimensional look at the flow. The vertical structure of a lower-tropospheric vortex is shown in Fig. 5 based on Doppler radar measurements along FS1 at 1914 UTC. Although present from 1 to 5 km height, the vortex circulation is most clearly defined in the analysis at 3 km. Vortex V1 is characterized by a relative vorticity maximum of approximately 30 × 10−4 s−1 with horizontal diameter of the vorticity core of 15–30 km. Unfortunately, subsequent observations of V1 were not made—the next flights did not occur until 1830 UTC 19 August, almost 48 h later. In the meantime convection remained horizontally concentrated on the mesoscale, tracking to the west-northwest.

b. Disturbance transformation

1) Overview of observations

On 19 August the pre-Dolly tropical disturbance transformed into a TC. Based on a synthesis of flight-level, satellite, and buoy observations, NHC upgraded the pre-Dolly disturbance to a tropical depression at 0600 UTC, and then to a tropical storm by 1800 UTC at 18.2°N, 83°W (Rappaport 1996). While Air Force C-130 near-surface observations late on 19 August, shown in Fig. 6a, do indicate a region of enhanced southeasterly flow in excess of 15–20 m s−1 east of 84°W and south of 19°N, a well-defined accompanying system-scale cyclonic vortex is not easily found. Several mesoscale cyclonic wind shifts are evident, however, at locations denoted along the P3 and C-130 flight tracks. Identification of a center at this time is therefore not a well-defined task.

As sometimes happens in genesis cases, the pre-Dolly disturbance organized rapidly, spending only a brief period of time in the depression stage (R. Pasch 2004, personal communication). In such a situation, precise definitions of when and where genesis occurs can be somewhat elusive. In the upcoming analysis we adopt the position that the stage of development captured by the Doppler radar measurements late on 19 August, whether it be predepression or postdepression, is ambiguous. Dolly’s status on 19 August will henceforth be referred to as one of early development.

Visible satellite imagery beginning at 1200 UTC 19 August shows a broadscale (∼300–400 km) cyclonic rotation of the midlevel cloud field associated with the pre-Dolly disturbance (not shown). The 480-hPa flight-level observations beginning at 1830 UTC 19 August (Fig. 6b) confirm the presence of westerlies south of 19°N. Infrared imagery shown in Fig. 7 from 2145 to 2345 UTC 19 August, and then at 1745 UTC the next day, indicates that deep convection was focused within two regions of the disturbance, north and south of 19°N, approximately 150 km apart. Although most clearly defined in the satellite animations, the two convective regions appear to make a partial cyclonic orbit about a westward moving centroid. Such orbiting of mesoscale regions of convection is reminiscent of the midlevel MCV interactions documented by RH97 and Simpson et al. (1997) in Pacific genesis cases. The two convective regions ultimately merge toward a common center, and at 1800 UTC 20 August Hurricane Dolly emerges.

2) Observed circulations: Origin

Doppler radar measurements were made within both convective regions less than 24 h before hurricane formation, between 1900 and 2400 UTC 19 August. Figure 8 shows the Doppler-derived wind fields along FS2 and FS3 at the respective 1 and 6 km heights in the southern convective region. At 1956 UTC an elongated midlevel vortex V2 is located near 18.7°N, 83.1°W. The characteristic relative vorticity of V2 is 5–15 × 10−4 s−1, well above ambient planetary vorticity values at this latitude. Two and a half hours later, at 2230 UTC, another midlevel cyclonic circulation (denoted C) appears in the early stages of development southwest of V2 at 18.5°N, 83.6°W along a boundary between southwesterly and northwesterly flow. Although some cyclonic curvature of the flow is evident beneath V2, no well-defined mesoscale cyclonic circulations are observed at 1-km height within this region.

We cannot be certain of the origin of these middle-tropospheric cyclonic circulations, but their presence is consistent with previous observations of MCVs during TC genesis (e.g., BE97; RH97). The MCV is thought to originate within the stratiform precipitation region of a larger mesoscale convective system (e.g., Zhang and Fritsch 1987; Brandes 1990; Johnson and Bartels 1992). It has been argued that in some cases the MCV may form through the vertical gradient of diabatic heating in the stratiform region, resulting in positive midlevel potential vorticity generation (Hertenstein and Schubert 1991). Figure 9 shows an east–west cross section of reflectivity and wind through midlevel vortex V2 at 1956 UTC. The radar brightband, which is typically indicative of stratiform precipitation and roughly denotes the layer where melting of frozen hydrometeors occurs (hence the elevated reflectivity), is only faintly seen here, but is consistently observed between 3.5- and 4-km height elsewhere in the Doppler domain. Cooling at and below the melting layer and heating above should place the maximum positive potential vorticity generation somewhere between 3- and 5-km height. The observed primary vortex circulation of V2 peaks near 5-km height at 1956 UTC. The circulation shows little downward vertical extent at this time; thus we surmise that it is in a decaying stage, having possibly formed prior to the observation period within the stratiform precipitation region.

Figure 10a shows that within the northern convective region (FS4–FS6) another midlevel vortex V3 was present during a 2-h period overlapping with the observation period in the southern region. Similar to V2, the core relative vorticity of V3 is 5–15 × 10−4 s−1, but the circulation of V3 is much more circular in appearance. The horizontal scale of V3 is difficult to determine due to the limited observational domain, but the cyclonic wind field is at least 100 km across. When the flows from V2 and V3 (and perhaps other midlevel vortices not documented here) are superposed, it becomes clear that the broadscale cyclonic rotation of the cloud field previously noted in the satellite imagery is likely a consequence of these nearby midlevel vortices embedded within the synoptic-scale cyclonic flow of the easterly wave.

The relative vorticity at 1-km height in the northern convective region is shown in Fig. 10b. Midlevel vortex V3 differs from V2 in that a prominent low-level cyclonic vortex V4 exists nearby. Vortex V4, first observed at 2128 UTC near 19.5°N, 84.4°W, has a peak relative vorticity value of 35 × 10−4 s−1. The horizontal diameter of the high vorticity core is 20–30 km. The initial horizontal separation between the centers of V3 and V4 is 25–30 km, with V3 to the southeast of low-level vortex V4.

It is conceivable that V3 and V4 began as a single vertically coherent vortex. Midlevel vortex V3 would then likely originate through vertical shearing, rather than diabatic heating in the stratiform precipitation region, as suggested for V2. To determine whether this scenario is possible the 850–500-hPa vertical shear is computed separately using National Centers for Environmental Prediction (NCEP) reanalysis data (provided by the NOAA–Cooperative Institute for Research in Environmental Sciences (CIRES) Climate Diagnostics Center, Boulder, Colorado, from their Website (available online at http://www.cdc.noaa.gov) and 2.5° ERA-40 analyses. A 5° box centered on the vortex location at 2128 UTC is used to compute the horizontally averaged flow. Both analyses indicate a prominent northerly component to the weak vertical shear in the 850–500-mb layer, consistent with an observed upper-level anticyclone to the northwest of the pre-Dolly disturbance and east-southeasterly flow at low levels (B99). Local estimates of vertical shear through the same depth using area-averaged winds within each Doppler wind composite also indicate a northerly component (not shown). Thus, there is a consistency between the vertical shear direction and the presumed southeasterly vortex tilt with height.

The question remains as to whether the observed vertical shear could tilt a vortex of the magnitude and size of V3/V4. Reasor et al. (2004, hereafter RMG04) showed that the resiliency of a vortex in vertical shear can be determined through knowledge of the vortex Rossby number, ratio of horizontal vortex scale to Rossby deformation radius, and vertical shear magnitude. For a broadly distributed initial vortex (e.g., a Gaussian radial profile of vorticity), the resiliency will depend on the relative magnitudes of the intrinsic precession frequency of the tilted vortex and the differential advection rate associated with the vertical shear. From Fig. 10 we estimate the maximum tangential winds of the hypothetical initially aligned vortex to be 5 m s−1 at 50-km radius. According to RMG04, a vortex similar to V3/V4, in the absence of convective heating, would precess at such a slow rate that even 5 m s−1 vertical shear could tear it apart (see Fig. 7 of RMG04). It is therefore plausible that V3 and V4 at one time was a single vortex. Without a longer observation window it is difficult to determine how and when this vortex may have originated. In the upper northwest portion of the 1-km height observational domain at 1956 UTC in Fig. 8 (i.e., to the southeast of the low-level vortex at 2128 UTC) there is some hint of a cyclonic curvature of the flow, possibly the southern extent of a cyclonic vortex. The presence of a vortex, however, could not be independently confirmed.

It is also possible that midlevel vortex V3 arose through diabatic heating in the stratiform precipitation region, as suggested for the formation of V2. In this scenario it is unlikely that low-level vorticity of the magnitude depicted in Fig. 10b at 2128 UTC would exist as part of the downward extension of the midlevel MCV. One explanation for the formation of V4 lies within a closer examination of the infrared imagery in Fig. 7. The imagery at 2145 UTC shows an isolated cold cloud top almost directly above the low-level vortex V4. M05 have shown that deep, penetrating cumulonimbus convection, or hot towers, can attain a strong vortical component when generated within environments of elevated near-surface vorticity. In the numerical simulations of M05 these vortical hot towers, or VHTs, emerge within an initial MCV environment. The origin of low-level relative vorticity associated with the VHTs can be understood by recalling the equation for the material rate of change of vertical vorticity,
i1520-0469-62-9-3151-e2
where ζ is the vertical component of relative vorticity, u, υ, and w are the x, y, and z components of the velocity, respectively; ρ is the density; p is the pressure; and Fx,y represent frictional dissipation. Contributions to the material rate of change of vertical vorticity include (from left to right) stretching, tilting, solenoidal production, and diffusion. The latter two contributions are not believed to be fundamental to the VHT evolution, so we focus here on the stretching and tilting terms. M05 found that when deep cumulus convection is initiated, the ζ tendency is at first dominated by tilting of ambient horizontal vorticity filaments into the vertical by the horizontally sheared updraft, similar to the development of rotation in a supercell thunderstorm. A vertical vorticity dipole emerges. The stretching of ambient and newly generated vertical vorticity then dominates the ζ tendency, leading to an amplification of the positive vertical vorticity. M05 suggest that the VHTs may then behave as nonentraining plumes, but note that higher resolution numerical simulations are required to represent explicitly the lateral mixing outside the VHT core. It is plausible that V4 was a direct result of one of these VHT events. A further examination of this connection between the observations and the recent genesis theory of M05 is reserved for section 5.

In a case study of the genesis of Hurricane Danny (1997), Molinari et al. (2004) present indirect evidence for the generation of low-level vertical vorticity, and in one instance a coherent vortex, within the periphery of a preexisting MCV during the pre-WISHE stage. It is well known that convection is often initiated downshear of a midlevel MCV embedded within a vertically sheared environment (Raymond and Jiang 1990). Molinari et al. argue that the merger and axisymmetrization of this downshear convectively generated vorticity aided the development of Danny, as demonstrated in idealized numerical simulations by M05. Based on this previous work, the interaction of midlevel vortex V3 with the northerly vertical shear would yield a convective asymmetry, and potential VHTs, south-southeast of the vortex. Low-level vortex V4, which is found to the northwest of V3, thus may not necessarily be tied directly to the interaction of MCV and large-scale vertical shear at 2128 UTC. The possibility that V4 was initiated through such an interaction prior to the observation period cannot be ruled out.

Vertical shear may still be playing a role in the early development of V4, but perhaps a role different from that envisioned above. Figure 11 shows an east–west cross section of vertical velocity through V4 at 2128 UTC. The lower-to-middle-tropospheric updraft maximum is 3.5 m s−1 to the east of the 1-km relative vorticity maximum.1 This location of the updraft maximum is downshear left of the large-scale vertical shear vector, consistent with studies of convective vortices embedded in vertical shear flow (e.g., Wang and Holland 1996; Frank and Ritchie 1999, 2001). If V4 were initiated prior to the observation period through, for example, a VHT event, the subsequent interaction of that small-scale vortex with vertical shear could excite new nearby convection and enhanced vorticity generation through stretching of vorticity filaments associated with the preexisting vortex. While it is possible that V4 itself was initiated through the deep convective event identified in the infrared imagery of Fig. 7, the available observations seem to point to this convective event as being forced by an already existing small-scale low-level vortex. The impact of this presumed secondary convective event on the vorticity structure of V4 is evident in a three-dimensional representation of the relative vorticity shown in Fig. 12. The 15 × 10−4 s−1 relative vorticity iso-surface tilts from west to east with height at 2128 UTC. The upward extension of vorticity east of the 1-km circulation coincides with the location of the vigorous updraft and appears to be a new addition to the low-level vorticity core.

A simple calculation of vorticity tendency based on the stretching of ambient vorticity alone is done to test whether the above convective mechanisms for the formation and modification of V4 are plausible. The low-level horizontal convergence (estimated from the Doppler analysis) within the convective cell is approximately 1 × 10−3 s−1. Given this value of local convergence and an initial low-level value of absolute vorticity of 2.5 × 10−4 s−1 associated with the presumed downward extension of V3 (an east–west cross section of relative vorticity through V3, not shown, indicates values of 1–3 (×10−4 s−1) between 1- and 2-km height), Eq. (2) predicts an exponential ζ increase of 12.5 × 10−4 s−1 can be achieved through vertical stretching in a 30-min period. This estimate is a conservative one given that tilting of horizontal vorticity by the updraft will add to the initial absolute vertical vorticity available for stretching. Thus, V4 could have been initiated within the environment of V3. If we assume, instead, that the updraft in Fig. 11 represents a secondary convective event on a preexisting VHT with conservative initial peak absolute vorticity of 15 × 10−4 s−1, an exponential ζ increase of 75 × 10−4 s−1 is possible in a 30-min period. This generation rate easily accounts for the low-level vorticity values observed at 2128 UTC in association with V4.

3) Observed circulations: Evolution

Midlevel vortex V3 moves away from V4 in time to the south-southwest. In Fig. 10b the position of V3 relative to V4 is shown. The absence of a cyclonic precession of the vortices about their centroid suggests little interaction between the two circulations. The basic structure of V3 is revealed by azimuthally averaging about the 5 km height vortex center (defined for simplicity as the center of circulation). The radius–height structure of symmetric tangential wind is shown in Fig. 13 at 2128, 2245, and 2345 UTC. Early in the observation period the symmetric tangential wind peaks between 4 and 5 km height and has a maximum value of at least 8 m s−1 beyond 30 km radius. The strong downward extension of the symmetric circulation to 1-km height at 2128 UTC may not be representative of the initial MCV (assuming such an isolated vortex did exist) and is most likely a consequence of the developing nearby low-level vortex V4. As V3 moves away from V4, the mean tangential wind diminishes throughout the depth, and the more vertically confined nature of the vortex becomes increasingly evident.

Figure 14 shows the evolution of the local azimuthal-mean structure of V4 using the 1-km height vortex center. In addition to the tangential wind, the symmetric radial wind and relative vertical vorticity are shown. At 2128 UTC the symmetric tangential wind of V4 peaks near 2 km height with a maximum value of 7 m s−1 at 18 km radius. Peak vorticity values of 25 × 10−4 s−1 fall along the axis of rotation of the vortex and are confined to the lower troposphere. The symmetric radial flow is convergent inside the radius of maximum tangential wind (RMW) below 4 km height and is divergent above 4 km height. This pattern of convergence is consistent with the convective cell, identified in Fig. 11, that formed near the center of V4.

As expected, the enhanced low-level symmetric convergence leads to an intensification of the low-level symmetric vortex by 2245 UTC. A double maximum in tangential wind emerges, one near 10-km radius and another beyond 20-km radius. Both maxima appear to peak at or below 1-km height. The maximum vorticity increases to 35 × 10−4 s−1 and becomes more vertically coherent. Overall, the transverse circulation weakens, and the midlevel divergence extends through a greater depth. By 2345 UTC a well-defined symmetric vortex has formed with maximum tangential winds of 8 m s−1 at 15-km radius. The maximum winds still peak at or below 1-km height but now extend through a greater vertical depth. The symmetric relative vorticity also increases in vertical extent (the decrease in peak value from 2245 UTC is a consequence of using the circulation center rather than vorticity center). Indicative of a period of rapid growth, the radial flow near the RMW has become convergent throughout the observed depth.

The increase in the symmetric relative vorticity of V4 is consistent with the increase in total relative vorticity at 1-km height shown in Fig. 10b. The evolution of the vorticity, however, is highly asymmetric. The rate of change of the symmetric vorticity in cylindrical coordinates, derived from Eq. (2) (neglecting tilting, solenoidal, and diffusion terms), has contributions from both symmetric and asymmetric processes:
i1520-0469-62-9-3151-e3
where D represents the symmetric material derivative and is the symmetric convergence. The first term represents the symmetric stretching of symmetric absolute vorticity. The last two terms represent the radial eddy vorticity flux convergence and eddy vertical vorticity advection, respectively. Montgomery and Kallenbach (1997) have shown that the axisymmetrization of vorticity anomalies on a symmetric vortex can contribute significantly to the symmetric vortex evolution through the eddy flux convergence term. Asymmetric bands of elevated vorticity, typical of the axisymmetrization process, emanate from the high vorticity core of V4 at all times (e.g., the band to the southeast of V4 at 2345 UTC in Fig. 10b). The three-dimensional perspective of relative vorticity in Fig. 12 also highlights the development of an asymmetric band, referred to here as a feeder band, on the east side of V4. Although this band does not connect directly to the vorticity core of V4, it becomes increasingly organized and moves closer to the vortex core over the 2-h observation period. This feeder band is also evident in the reflectivity field. Figure 15 shows the 30-dBZ iso-surface of reflectivity. The band of reflectivity, and hence precipitation, is coincident with the band of vorticity. The generation of vorticity and elevated θe within the convective band may feed the developing surface circulation, as discussed in the next section.

5. Discussion

Low-level mesoscale cyclonic vortices are likely common features within tropical disturbances during the pre-WISHE stage of TC development. Direct evidence of two such circulations within the pre-Dolly disturbance was found on 17 and 19 August. A simple vorticity budget on 19 August suggested that deep cumulonimbus convection could converge the elevated low-level absolute vorticity of an observed downward-penetrating midlevel cyclonic vortex (V3) to yield near-hurricane-strength vorticity values on very small scales (10–20 km). The resulting low-level vortex (V4) on 19 August was observed over a 2-h period and increased in peak strength (in an azimuthal-mean sense) from 7 m s−1 at 2-km height to 8 m s−1 at or below 1-km height. Additional low-level mesoscale cyclonic wind shifts identified in the 975-hPa flight-level winds of the Air Force C-130 aircraft on 19 August could not be independently linked to actual cyclonic circulations with the Doppler radar owing to spatial and temporal limitations of the dataset. Based on animations of the infrared satellite imagery, however, deep convective events appear to be ubiquitous during the observation period. Isolated cold cloud tops appear and disappear in a pulsating fashion. Where these convective events are able to tilt and converge existing low-level absolute vorticity (e.g., in the vicinity of preexisting midlevel vortices), we expect that low-level closed or enhanced cyclonic circulations like the ones observed will result (Hendricks et al. 2004; M05).

Sufficient evidence was found by NHC to upgrade Dolly to a tropical storm just prior to the Doppler observation period on 19 August documented here. However, the apparent absence of a system-scale cyclonic flow at low levels during the Doppler observation period on 19 August suggests that Dolly may have been at an earlier stage of development than tropical storm stage. In the idealized study of the predepression stage of TC genesis by M05, a warm bubble, representing a convective cell, was initiated near the radius of maximum tangential wind of a midlevel cyclonic MCV. Within the convectively unstable and vorticity rich environment of the MCV, rapidly rotating VHTs were observed to form. The individual VHTs are associated with strong vertical vorticity through tilting and stretching of ambient elevated vorticity by the convergent flow of the buoyant updraft. The local enhancement of rotation within these VHTs makes the conversion of latent heat energy into rotational kinetic energy more efficient (Schubert and Hack 1982). M05 showed that the net effect of diabatic heating by a number of these VHT events is to drive a larger-scale toroidal circulation that concentrates the low-level angular momentum of the initial MCV and the subsequently produced vorticity anomalies, leading to the spinup of a tropical depression strength, system-scale surface vortex.

Figure 16 shows an example of a typical VHT event from the idealized Regional Atmospheric Modeling System (RAMS) simulations of M05. Actually, multiple VHT events are present, identified by the localized regions of elevated absolute vorticity. The primary anomaly is our focus here. The low-level absolute vertical vorticity maximum is approximately 50 × 10−4 s−1, similar to that shown in Fig. 10b for the low-level vortex (V4) within the Dolly disturbance. The horizontal diameter of the high vorticity core of the VHT is 10–20 km, compared to 20–30 km for V4. In the example shown in Fig. 16, the initial convective event has already locally tilted and stretched the absolute vorticity of the MCV. The predominant vertically penetrating cyclonic vorticity anomaly results from subsequent mergers with nearby VHT vorticity. The band of elevated absolute vorticity to the southwest of the primary anomaly is one example of such a merger.

A similar merger of vorticity was indicated at 2128 UTC in the early evolution of V4. A convective cell to the east of the low-level circulation appeared to stretch the absolute vorticity of the small-scale vortex, leading to a deeper vertical penetration of the cyclonic circulation. It is possible that the mesoscale vortex V4 observed at 2128 UTC in the Dolly case is also a consequence of the diabatic merger of several VHTs, which M05 have shown increases the horizontal scale of the circulation. Additional merger events were suggested during the observed lifetime of V4 in the form of vorticity bands adjacent to the vortex core (e.g., the band to the southeast of V4 at 2345 UTC in Fig. 10b is remarkably similar to the simulated merger event in Fig. 16) and through the development of a larger-scale feeder band displaced from the vortex core, but spiraling in with time (see Figs. 12 and 15).

The origin of the feeder band is unknown. Such banded structures were argued by Montgomery and Kallenbach (1997) to be manifestations of the vortex axisymmetrization process. Weak baroclinic zones in the pre-TC environment might also have contributed to the band formation through frontogenetic processes. Limited thermodynamic measurements in the vicinity of V4 prevent a comprehensive study of the thermodynamic structure of the bands. Flight segment FS6 at 480 hPa on 19 August, however, did transect both the spiral vorticity band to the southeast of V4 at 2345 and the feeder band farther to the east (for reference, FS6 bisects the 2345 UTC Doppler domain at 19.2°N in Fig. 10). Figure 17 shows a time series of θe along FS6 with 1-s time resolution. Elevated θe is first encountered during the transect of the spiral band near V4, with values exceeding 350 K. Outside of this band values fall to 342–343 K. Another region of slightly elevated θe of 345.5 K is encountered during passage through the feeder band between 83.9° and 84.1°W. Both bands may play a role in organizing convection and vorticity generation. May and Holland (1999) presented evidence for the generation of substantial positive potential vorticity within the stratiform precipitation region of hurricane spiral bands, and argued that the inward flux of this high potential vorticity air into the vortex core could positively impact intensity. Whether these bands are capable of producing similar intensification of mesoscale vortices like V4 is the subject of future investigation.

No further Doppler observations of the Dolly disturbance were made beyond 19 August. Therefore, we can only speculate as to the role of the observed low-level VHT-scale vortex V4. The intensifying vortex V4 may very well have become the center of circulation of TC Dolly. Another possibility is that subsequent convective bursts within the northern and southern convective regions identified in Fig. 7 continued to produce low-level cyclonic circulations like the one observed. In support of the latter is the observation from infrared satellite imagery that the northern and southern convective regions appeared to merge in time. One might be inclined to argue that this merger of the mesoscale regions of convection is evidence for the midlevel merger of MCVs and downward building of the surface cyclonic circulation (RH97; Simpson et al. 1997). The Doppler radar observations did indicate two midlevel MCVs in close proximity, although the one in the southern convective region appeared to decay during the observation period. The Doppler composites do not unambiguously clarify the role of the MCVs in this particular case, but they do confirm that the VHTs seem to thrive in their presence. The Dolly observations suggest that ample vorticity associated with a small-scale VHT-like circulation already existed at low levels. The simulations of M05 with one midlevel MCV demonstrate that a series of nearby VHT events drives an increasing system-scale low-level inflow. An increasing system-scale low-level inflow will tend to organize convection closer to the system center, as observed. Thus, while the observed low-level vortex may not have been solely responsible for Dolly’s surface circulation, it was likely instrumental to its development.

The overall evolution of the Dolly disturbance described here supports a stochastic view of TC genesis. While large-scale conditions were favorable for TC genesis and elevated cyclonic vorticity on the synoptic scale was present at midlevels associated with the easterly wave, it was the interplay between midlevel mesoscale circulations and deep convection that appeared to initiate the near-surface circulation development. Yet on 17 August a fairly vigorous small-scale lower-tropospheric cyclonic vortex V1 was observed, but genesis did not occur at that time. In the simulations of M05, VHTs form and dissipate over time, but in doing so elevate both the low-level pool of absolute vorticity and mid-to-upper-level moisture for future, more resilient, VHT events. Whether V1, and other vortices not sampled during the period leading up to the observations on 19 August, played this role is unclear. Improved temporal continuity of the data (e.g., missions every 12 h) is essential for future investigations of TC genesis so that the changes on the mesoscale described here can be related to the development of the broadscale circulation in a quantitative way.

On 19 August the stochastic nature of the genesis process is manifested most clearly through the apparent vorticity mergers during the development of V4, which itself occurs within an environment of interacting midlevel vortices. Future observational investigations would greatly benefit from having two radar platforms sampling nearby regions of the tropical disturbance (e.g., the northern and southern convective regions here) simultaneously to assess the presence of multiple small-scale (10–30 km) vortices, and at the same time document with sufficient detail the vorticity mergers taking place. However, this should not be done at the expense of temporal continuity, if only two radar-equipped aircraft are available. We strongly advocate a future TC genesis field experiment to continue the observational exploration of ideas presented here.

Acknowledgments

This work was supported in part by National Science Foundation Grant ATM-0101781, Office of Naval Research Grant N00014-02-1-0474, and Colorado State University. The first author was supported in part by the National Research Council while a postdoctoral research associate at HRD, and by The Florida State University. Lance Bosart was supported by National Science Foundation Grants ATM-9413012 and ATM-9612485. We foremost acknowledge HRD, whose 1996 tropical cyclogenesis mission made this dataset possible. In particular we thank Frank Marks, John Gamache, and Pete Black for their willing assistance with the data and many helpful discussions. We wish to express our gratitude to Ray Zehr and John Knaff of CIRA for providing the GOES-8 imagery used in this study. We also thank Matt Eastin for assistance with the flight-level data, and Steve Feuer and Mike Black for providing ODW data for the Dolly case. Finally, this paper greatly benefited from the constructive comments of Dave Raymond and two anonymous reviewers.

REFERENCES

  • Bister, M., , and K. A. Emanuel, 1997: The genesis of Hurricane Guillermo: TEXMEX analyses and a modeling study. Mon. Wea. Rev., 125 , 26622682.

    • Search Google Scholar
    • Export Citation
  • Bracken, W. E., 1999: A multiscale examination of tropical cyclogenesis. Ph.D. dissertation, The University at Albany, State University of New York, 340 pp. [Available from Dept. of Atmospheric Science, The University at Albany, State University of New York, Albany, NY 12222.].

  • Bracken, W. E., , and L. F. Bosart, 2000: The role of synoptic-scale flow during tropical cyclogenesis over the North Atlantic Ocean. Mon. Wea. Rev., 128 , 353376.

    • Search Google Scholar
    • Export Citation
  • Brandes, E. A., 1990: Evolution and structure of the 6–7 May 1985 mesoscale convective system and associated vortex. Mon. Wea. Rev., 118 , 109127.

    • Search Google Scholar
    • Export Citation
  • Davis, C. A., , and L. F. Bosart, 2001: Numerical simulations of the genesis of Hurricane Diana (1984). Part I: Control simulation. Mon. Wea. Rev., 129 , 18591881.

    • Search Google Scholar
    • Export Citation
  • Frank, W. M., , and E. A. Ritchie, 1999: Effects of environmental flow upon tropical cyclone structure. Mon. Wea. Rev., 127 , 20442061.

  • Frank, W. M., , and E. A. Ritchie, 2001: Effects of vertical wind shear on the intensity and structure of numerically simulated hurricanes. Mon. Wea. Rev., 129 , 22492269.

    • Search Google Scholar
    • Export Citation
  • Gamache, J. F., 1998: Evaluation of a fully three-dimensional variational Doppler analysis technique. Preprints, 28th Conf. on Radar Meteorology, Austin, TX, Amer. Meteor. Soc., 422–423.

  • Gamache, J. F., , F. D. Marks, , and F. Roux, 1995: Comparison of three airborne Doppler sampling techniques with airborne in situ wind observations in Hurricane Gustav (1990). J. Atmos. Oceanic Technol., 12 , 171181.

    • Search Google Scholar
    • Export Citation
  • Gerrish, H., 1991: Preliminary report: Hurricane Guillermo 4–10 August 1991. National Hurricane Center Rep., 7 pp. [Available online at ftp://ftp.nhc.noaa.gov/pub/storm_archives/pacific/prelimep/ep1991/guillerm.].

  • Hendricks, E. A., , M. T. Montgomery, , and C. A. Davis, 2004: On the role of “vortical” hot towers in tropical cyclone formation. J. Atmos. Sci., 61 , 12091232.

    • Search Google Scholar
    • Export Citation
  • Hertenstein, R., , and W. H. Schubert, 1991: Potential vorticity anomalies associated with squall lines. Mon. Wea. Rev., 119 , 16631672.

    • Search Google Scholar
    • Export Citation
  • Johnson, R. H., , and D. L. Bartels, 1992: Circulations associated with a mature-to-decaying midlatitude mesoscale convective system. Part II: Upper-level features. Mon. Wea. Rev., 120 , 13011321.

    • Search Google Scholar
    • Export Citation
  • Jorgensen, D. P., , P. H. Hildebrand, , and C. L. Frush, 1983: Feasibility test of an airborne pulse-Doppler meteorological radar. J. Appl. Meteor., 22 , 744757.

    • Search Google Scholar
    • Export Citation
  • Mapes, B. E., , and R. A. Houze, 1995: Diabatic divergence profiles in Western Pacific mesoscale convective systems. J. Atmos. Sci., 52 , 18071828.

    • Search Google Scholar
    • Export Citation
  • May, P. T., , and G. J. Holland, 1999: The role of potential vorticity generation in tropical cyclone rainbands. J. Atmos. Sci., 56 , 12241228.

    • Search Google Scholar
    • Export Citation
  • Molinari, J., , D. Vollaro, , and K. L. Corbosiero, 2004: Tropical cyclone formation in a sheared environment: A case study. J. Atmos. Sci., 61 , 24932509.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , and R. Kallenbach, 1997: A theory for vortex Rossby-waves and its application to spiral bands and intensity changes in hurricanes. Quart. J. Roy. Meteor. Soc., 123 , 435465.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , and J. Enagonio, 1998: Tropical cyclogenesis via convectively forced vortex Rossby waves in a three-dimensional quasigeostrophic model. J. Atmos. Sci., 55 , 31763207.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , M. E. Nicholls, , T. A. Cram, , and A. B. Saunders, 2005: A vortical hot tower route to tropical cyclogenesis. J. Atmos. Sci., , in press.

    • Search Google Scholar
    • Export Citation
  • Ooyama, K. V., 1982: Conceptual evolution of the theory and modeling of the tropical cyclone. J. Meteor. Soc. Japan, 60 , 369379.

  • Rappaport, E. N., 1996: Preliminary report: Hurricane Dolly 19–25 August 1996. National Hurricane Center Rep., 7 pp. [Available online at http://www.nhc.noaa.gov/1996dolly.html.].

  • Raymond, D. J., , and H. Jiang, 1990: A theory for long-lived mesoscale convective systems. J. Atmos. Sci., 47 , 30673077.

  • Raymond, D. J., , C. Lopez-Carrillo, , and L. Lopez Cavazos, 1998: Case-studies of developing east Pacific easterly waves. Quart. J. Roy. Meteor. Soc., 124 , 20052034.

    • Search Google Scholar
    • Export Citation
  • Reasor, P. D., , and M. T. Montgomery, 2001: Three-dimensional alignment and corotation of weak, TC-like vortices via linear vortex Rossby waves. J. Atmos. Sci., 58 , 23062330.

    • Search Google Scholar
    • Export Citation
  • Reasor, P. D., , M. T. Montgomery, , F. D. Marks Jr., , and J. F. Gamache, 2000: Low-wavenumber structure and evolution of the hurricane inner core observed by airborne dual-Doppler radar. Mon. Wea. Rev., 128 , 16531680.

    • Search Google Scholar
    • Export Citation
  • Reasor, P. D., , M. T. Montgomery, , and L. D. Grasso, 2004: A new look at the problem of tropical cyclones in vertical shear flow: Vortex resiliency. J. Atmos. Sci., 61 , 322.

    • Search Google Scholar
    • Export Citation
  • Riehl, H., , and J. Malkus, 1958: On the heat balance in the equatorial trough zone. Geophysica, 6 , 503538.

  • Ritchie, E. A., , and G. J. Holland, 1997: Scale interactions during the formation of Typhoon Irving. Mon. Wea. Rev., 125 , 13771396.

  • Rotunno, R., , and K. A. Emanuel, 1987: An air–sea interaction theory for tropical cyclones. Part II: Evolutionary study using a nonhydrostatic axisymmetric numerical model. J. Atmos. Sci., 44 , 542561.

    • Search Google Scholar
    • Export Citation
  • Schubert, W. H., , and J. J. Hack, 1982: Inertial stability and tropical cyclone development. J. Atmos. Sci., 39 , 16881697.

  • Simpson, J., , E. Ritchie, , G. J. Holland, , J. Halverson, , and S. Stewart, 1997: Mesoscale interactions in tropical cyclone genesis. Mon. Wea. Rev., 125 , 26432661.

    • Search Google Scholar
    • Export Citation
  • Wang, Y., , and G. J. Holland, 1996: Tropical cyclone motion and evolution in vertical shear. J. Atmos. Sci., 53 , 33133332.

  • Zehr, R., 1992: Tropical cyclogenesis in the western North Pacific. NOAA Tech. Rep. NESDIS 61, 181 pp.

  • Zhang, D-L., , and J. M. Fritsch, 1987: Numerical simulation of the meso-β scale structure and evolution of the 1977 Johnstown flood. Part II: Inertially stable warm-core vortex and the mesoscale convective complex. J. Atmos. Sci., 44 , 25932612.

    • Search Google Scholar
    • Export Citation
Fig. 1.
Fig. 1.

Best track for Tropical Cyclone Guillermo (1991) from NHC (Gerrish 1991). The TS and H labels denote the locations where Guillermo was upgraded to a Tropical Storm and hurricane, respectively. Also shown are the TEXMEX observational domains on 3 and 4 Aug.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 2.
Fig. 2.

Location of all mesoscale cyclonic circulations observed with airborne Doppler radar inside the TEXMEX domains shown in Fig. 1. The times indicated are in UTC, and M denotes the 5–7-km midlevel layer and L denotes the 1–3-km low-level layer. The center of circulation is used to define the vortex locations.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 3.
Fig. 3.

Best track for Tropical Cyclone Dolly (1996) from NHC (Rappaport 1996). The TS and H labels denote the locations where Dolly was upgraded to a tropical storm and hurricane, respectively. Also shown are the observational domains on 15, 17, and 19 Aug.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 4.
Fig. 4.

Flight tracks and flight-level winds within the pre-Dolly disturbance from the NOAA P3 aircraft at 570 hPa on 15 Aug and at 640 hPa on 17 Aug (see Fig. 3 for the location of the observations relative to Dolly’s storm track). In all figures a barb equals 2.5 m s−1 and a flag equals 10 m s−1. Labels C and V1 denote cyclonic circulations of interest. See text for further details.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 5.
Fig. 5.

Vertical structure of winds and relative vorticity associated with low-level vortex V1 at 1914 UTC 17 Aug (FS1). Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 6.
Fig. 6.

Flight tracks and flight-level winds within the pre-Dolly disturbance from (a) the Air Force C-130 at 975 hPa and (b) the NOAA P3 aircraft at 480 hPa on 19 Aug. Labels C and V2–4 denote cyclonic circulations of interest. Barbs equal 2.5 m s−1 and flags equal 10 m s−1. See text for further details.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 7.
Fig. 7.

GOES infrared satellite images of the Dolly disturbance on 19 Aug at (top left) 2145, (top right) 2245, and (bottom left) 2345 UTC. Two regions of deep convection are observed north and south of 19°N. The convective regions propagate westward, merging to form Hurricane Dolly 18 h later around (bottom right) 1745 UTC 20 Aug. An emerging convective cell and vortex V4 are identified in the northern region to the northwest of midlevel vortex V3. In the southern region, a midlevel cyclonic circulation C is identified. See text for further details. The airborne Doppler observation domain is contained within the heavy box.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 8.
Fig. 8.

Wind and relative vorticity at (left) 1- and (right) 6-km height at (top) 1956 (FS2) and (bottom) 2230 (FS3) UTC 19 Aug within the southern convective region identified in Fig. 7. No low-level cyclonic circulations are evident, but two midlevel circulations, V2 and C, are present. The heavy solid line denotes the cross section through V2 in Fig. 9, and X indicates the location of V2 and C on the 1-km wind fields. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 9.
Fig. 9.

East–west cross section (see Fig. 8) of reflectivity through V2 at 1956 UTC 19 Aug. The peak tangential circulation is located near 5-km height at V2 within a stratiform region. The xz flow within the cross section is also shown. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 10.
Fig. 10.

Wind and relative vorticity (1- and 5-km height) at 2128 (FS4), 2245 (FS5), and 2345 (FS6) UTC 19 Aug within the northern convective region identified in Fig. 7. (a) Midlevel vortex V3 moves to the south-southwest, and (b) the evolution of a low-level cyclonic vortex V4 is observed. The heavy solid line denotes the cross section through V4 displayed in Fig. 11, and X denotes the location of V3 on the 1-km wind fields. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 10.
Fig. 10.

(Continued)

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 11.
Fig. 11.

East–west cross section (see Fig. 10b) of vertical velocity at 2128 UTC 19 Aug associated with the convective cell identified in infrared imagery in Fig. 7. The 1-km location of vortex V4 is identified. Contour interval is 1 m s−1. Negative values are dashed and the zero contour is not shown. The xz flow within the cross section is also shown. Barbs equal 2.5 m s−1 and flags equal 10 m s−1.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 12.
Fig. 12.

Three-dimensional iso-surface of the vertical component of relative vorticity (15 × 10−4 s−1) showing the development of the low-level vortex at 2128, 2245, and 2345 UTC (clockwise from top). The depth shown is from 1 to 7 km. The feeder band about vortex V4 intensifies and spirals in toward the vortex core.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 13.
Fig. 13.

Radius–height structure of azimuthal mean (using the estimated center of circulation at 5-km height) tangential wind (V, contour interval 1 m s−1) for vortex V3 in the northern convective region on 19 Aug. The tangential wind decreases in magnitude with time indicating a decay of the vortex circulation as it moves away from low-level vortex V4.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 14.
Fig. 14.

As in Fig. 13 but for low-level vortex V4 centered at 1-km height. In addition to tangential wind, radial wind (U, contour interval 0.2 m s−1 and inward flow dashed) and the vertical component of relative vorticity (ζ, contour interval 0.5 × 10−3 s−1) are shown. The peak tangential wind increases from 7 to 8 m s−1 and moves down from a height near 2 km to below 1 km. The peak vorticity takes on a more vertically coherent structure, extending from near surface to above 7 km by the end of the observation period. The radial flow becomes increasingly dominated by inflow as time progresses.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 15.
Fig. 15.

As in Fig. 12 but for reflectivity (30 dBZ). Shown are the primary convective cell near the center of V4 and the development of a feeder band coincident with the band of vorticity in Fig. 12.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 16.
Fig. 16.

Absolute vertical vorticity from the RAMS TC genesis control simulation of M05. A number of VHTs are observed at the time shown, but one in particular stands out having converged elevated vorticity from previous convective events. The horizontal diameter of the dominant VHT core at 1-km height is 10–20 km with a peak absolute vorticity value of 50 × 10−4 s−1.

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Fig. 17.
Fig. 17.

Time series of 1-s P3 flight-level θe during FS6 (480 hPa) shown as a function of longitude. The elevated regions of θe correspond to transects through the spiral band near V4 and the feeder band to its east (see the 2345 UTC panel of Fig. 10 near 19.2°N for reference).

Citation: Journal of the Atmospheric Sciences 62, 9; 10.1175/JAS3540.1

Table 1.

Central time and location of flight segments (see Figs. 4 and 6) used to construct the Doppler wind analyses of the pre-Dolly disturbance Aug 1996.

Table 1.
1

The kinematically determined vertical velocity derived from Doppler radar measurements is subject to errors resulting from the imposition of an echo-top boundary condition of W = 0 and the contamination of Doppler winds below 1-km height by sea clutter. Regarding the former, Mapes and Houze (1995) provide evidence that Doppler radar likely undersamples strong divergences aloft (at and above echo top), in some cases leading to a significant underestimation of vertical velocity.

Save