Buoyancy of Convective Vertical Motions in the Inner Core of Intense Hurricanes. Part II: Case Studies

Matthew D. Eastin Department of Atmospheric Science, Colorado State University, Fort Collins, Colorado

Search for other papers by Matthew D. Eastin in
Current site
Google Scholar
PubMed
Close
,
William M. Gray Department of Atmospheric Science, Colorado State University, Fort Collins, Colorado

Search for other papers by William M. Gray in
Current site
Google Scholar
PubMed
Close
, and
Peter G. Black Hurricane Research Division, NOAA/AOML, Miami, Florida

Search for other papers by Peter G. Black in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

This is the second of two papers on the buoyancy of convective vertical motions in the inner core of intense hurricanes. This paper uses extensive airborne radar, dropwindsonde, and flight-level observations in Hurricanes Guillermo (1997) and Georges (1998) to illustrate typical azimuthal distribution of buoyant convection and demonstrate that the low-level eye can be an important source region for buoyant eyewall convection.

In both hurricanes, eyewall vertical velocity and radar reflectivity are asymmetric and exhibit persistent relationships with the direction of the environmental vertical wind shear. Mesoscale vertical motions exhibit a wavenumber-1 structure with maximum ascent downshear and weak descent upshear. The mesoscale reflectivity maxima are located left-of-shear. Buoyant eyewall updraft cores and transient convective-scale reflectivity cells are predominantly downshear and left-of-shear. Most eyewall downdraft cores that transport significant mass downward are located upshear. Negative buoyancy was most common in left-of-shear downdrafts, with positive buoyancy dominant in upshear downdrafts. Inward-spiraling rainbands located outside the eyewall exhibit upband/downband asymmetries. Upband segments contain more convective reflectivity cells and buoyant updraft cores than the more stratiform downband segments. Equal numbers of downdraft cores are found upband and downband, but the majority exhibit negative buoyancy.

Several buoyant updraft cores encountered in the midlevel eyewall exhibit equivalent potential temperatures (θe) much higher than the θe observed in the low-level eyewall, but equivalent to the θe observed in the low-level eye. Asymmetric low-wavenumber circulations appear responsible for exporting the high-θe eye air into the relatively low-θe eyewall and generating the locally buoyant updraft cores.

Implications of these results upon conceptual models of hurricane structure are discussed. Three mechanisms, whereby an ensemble of asymmetric buoyant convection could contribute to hurricane evolution, are also discussed.

Corresponding author address: Matthew D. Eastin, Department of Math and Computer Science, Central College, 812 University, Pella, IA 50219. Email: eastinm@central.edu

Abstract

This is the second of two papers on the buoyancy of convective vertical motions in the inner core of intense hurricanes. This paper uses extensive airborne radar, dropwindsonde, and flight-level observations in Hurricanes Guillermo (1997) and Georges (1998) to illustrate typical azimuthal distribution of buoyant convection and demonstrate that the low-level eye can be an important source region for buoyant eyewall convection.

In both hurricanes, eyewall vertical velocity and radar reflectivity are asymmetric and exhibit persistent relationships with the direction of the environmental vertical wind shear. Mesoscale vertical motions exhibit a wavenumber-1 structure with maximum ascent downshear and weak descent upshear. The mesoscale reflectivity maxima are located left-of-shear. Buoyant eyewall updraft cores and transient convective-scale reflectivity cells are predominantly downshear and left-of-shear. Most eyewall downdraft cores that transport significant mass downward are located upshear. Negative buoyancy was most common in left-of-shear downdrafts, with positive buoyancy dominant in upshear downdrafts. Inward-spiraling rainbands located outside the eyewall exhibit upband/downband asymmetries. Upband segments contain more convective reflectivity cells and buoyant updraft cores than the more stratiform downband segments. Equal numbers of downdraft cores are found upband and downband, but the majority exhibit negative buoyancy.

Several buoyant updraft cores encountered in the midlevel eyewall exhibit equivalent potential temperatures (θe) much higher than the θe observed in the low-level eyewall, but equivalent to the θe observed in the low-level eye. Asymmetric low-wavenumber circulations appear responsible for exporting the high-θe eye air into the relatively low-θe eyewall and generating the locally buoyant updraft cores.

Implications of these results upon conceptual models of hurricane structure are discussed. Three mechanisms, whereby an ensemble of asymmetric buoyant convection could contribute to hurricane evolution, are also discussed.

Corresponding author address: Matthew D. Eastin, Department of Math and Computer Science, Central College, 812 University, Pella, IA 50219. Email: eastinm@central.edu

1. Introduction

An understanding of how hurricane convection behaves, organizes, and influences storm evolution is important. Convection directly provides the latent heat release and vertical transport of mass and energy required to maintain the warm core. The organization of this convection is influenced by internal dynamics as well as large-scale environmental forcings and ocean–atmosphere interactions. Any organizational change can feed back upon the local environment to influence subsequent convection and storm evolution. These processes must be better understood before significant advancements in hurricane intensity prediction can be achieved (Elsberry et al. 1992). Numerous observational studies have documented the structure and organization of hurricane convection (e.g., Barnes et al. 1983; Jorgensen 1984a,b; Willoughby 1990; Marks et al. 1992; Black et al. 2002), but many fundamental properties, such as buoyancy, remain undocumented.

This paper is the second of two papers discussing the buoyancy of convective-scale vertical motions in the inner core of intense hurricanes as observed from research aircraft. In the first paper (Eastin et al. 2005, hereafter Part I), a statistical summary of total buoyancy was presented for convective vertical velocity events called cores. To review, flight-level data from 14 intense hurricanes were initially filtered to separate the slowly evolving (or balanced) mesoscale structure from any transient (or unbalanced) convective structure. Updraft and downdraft cores were then defined from the convective vertical velocity (|wc| > 1.0 m s−1 for at least 0.5 km). The total buoyancy (B) was defined from the convective virtual potential temperature, pressure, and liquid water content using the mesoscale structure as the reference state. Most updraft cores were weak (<2 m s−1) and small (<2 km in radial extent). The mean eyewall (rainband) updraft exhibited a small positive B below 4 km (between 2 and 5 km) that was more than adequate to explain the strongest of core magnitudes (5–7 m s−1). Buoyant updraft cores occupied <5% of the total inner-core area (within 150 km of the storm center), but accomplished ∼40% of the total upward mass transport. Most downdraft cores transporting mass downward (i.e., not superimposed upon stronger mesoscale updrafts) were similarly weak and small with positive mean B, suggesting that downdraft cores are more transient than updraft cores. In this paper, the buoyancy characteristics of two hurricanes are examined in greater detail to illustrate the typical azimuthal distribution of buoyant convection, demonstrate that the low-level eye can be an important source region for buoyant eyewall convection and provide evidence of three physical links between buoyant convection and storm evolution. Part I should be consulted for a more detailed discussion of the dataset, analysis techniques, and convective core statistics.

2. Selected cases

The case studies examined here in greater detail are eastern Pacific Hurricane Guillermo on 2 and 3 August 1997 and Atlantic Hurricane Georges on 19 September 1998. Table 1 presents intensity, motion, and large-scale environmental characteristics for each hurricane during the inner-core observation periods. The cases represent typical intense hurricanes embedded within environments generally favorable for maintenance or intensification. Despite such similarities, our findings are markedly robust and representative of a larger group of hurricanes. See Eastin (2003) for analyses of additional cases.

a. Hurricane Guillermo

The inner core of Hurricane Guillermo was first observed by the two National Oceanic and Atmospheric Administration (NOAA) WP-3D aircraft between 1800 UTC on 2 August and 0100 UTC on 3 August 1997. Lawrence (1999) gives a detailed account of the history of Guillermo. During this period, Guillermo was located 850 km south-southwest of Puerto Vallarta, Mexico, moving at 4.5 m s−1 toward the west. Sea surface temperatures (SSTs) were above 29°C, and the moderate vertical shear was oriented from north to south. Guillermo was rapidly deepening with an average drop in minimum sea level pressure (MSLP) of 2.4 mb h−1. Maximum observed flight-level winds were 61 m s−1 at 3.0-km altitude, and the average MSLP was 950 mb. The lower aircraft (NOAA-42 or H) made 10 eye penetrations (20 radial legs) at 3.0-km altitude, and the upper aircraft (NOAA-43 or I) made 6 eye penetrations at 5.5-km altitude. All quadrants were well sampled by each aircraft. Radiometric temperatures were only available from the upper aircraft, and thus analysis regarding convective core buoyancy was limited to the 12 radial legs at 5.5 km.

The two NOAA WP-3D aircraft returned to Guillermo on the following day and observed the inner core between 1900 UTC on 3 August and 0100 UTC on 4 August. During this period, Guillermo continued to move westward at 5.5 m s−1 over SSTs > 29°C and under the influence of moderate northerly vertical shear. Storm intensity was relatively steady with maximum observed flight-level winds of 71 m s−1 at 3.0-km altitude and a MSLP of 923 mb. The aircraft flew flight patterns nearly identical to those of the first day at the same altitudes. Analysis of convective core buoyancy was again limited to the 12 radial legs at 5.5 km.

b. Hurricane Georges

The inner core of Hurricane Georges was observed by the two NOAA WP-3D aircraft between 1900 UTC on 19 September and 0100 UTC on 20 September 1998. Pasch et al. (2001) provides a detailed account of the storm history. During this period, Georges was located 1000 km east-northeast of St. Martin, moving westward at 7.0 m s−1 over 28°–29°C SSTs, within a region of moderate west-northwesterly vertical shear. Maximum observed flight-level winds were 69 m s−1 at 4.2-km altitude, and the MSLP was 939 mb. Georges was intensifying with an average drop in MSLP of 0.8 mb h−1. Both aircraft made eye penetrations at ∼4.2-km altitude. The first aircraft (I) made three penetrations between 1900 and 2100 UTC, while the second aircraft (H) made two penetrations between 2300 and 0100 UTC. All quadrants were sampled at least once by each aircraft. Analysis of convective core buoyancy was limited to the four radial legs flown by the second aircraft.

3. Azimuthal distribution of buoyant convection

Numerous studies have investigated the impact of environmental vertical wind shear on the azimuthal distribution of eyewall convection. Using idealized numerical models with only balanced dry dynamics, Jones (1995) and DeMaria (1996) showed that the tilting of a vortex by vertical shear will induce a compensating transverse circulation in an attempt to bring the vortex back toward a vertical orientation and maintain balanced flow. This circulation, which comprises downshear upward motion and upshear downward motion, creates a cold (warm) potential temperature anomaly downshear (upshear) of the storm center. Subsequent flow along the distorted isentropes will induce enhanced upward motion in the downshear-right quadrant. In full-physics numerical simulations, however, Frank and Ritchie (1999, 2001 showed that condensational heating effectively removes the downshear cold anomaly, and the updraft maximum shifts to the downshear-left quadrant.

Observational studies of hurricane eyewalls embedded within moderate to strong vertical shear (>5 m s−1) show a relatively consistent picture. The mesoscale vertical motion and composite radar reflectivity exhibit a pronounced wavenumber-1 asymmetry with the ascent (decent) maximum downshear (upshear) and the reflectivity maximum left-of-shear (Marks et al. 1992; Franklin et al. 1993; Reasor et al. 2000; Black et al. 2002). Convective updrafts tend to develop downshear at low levels, advect cyclonically around the eyewall at speeds slower than the azimuthal mean wind, and approach the tropopause upshear. Convective downdrafts are most common upshear. Black et al. (2002) attributed updraft (downdraft) accelerations to buoyancy (water loading), but no direct estimates of buoyancy were presented. Here, we examine the buoyancy and vertical velocity characteristics of each case’s eyewall with respect to the environmental vertical wind shear vector.

Observational studies of hurricane rainbands have noted distinct asymmetries. In particular, the upwind (or upband) segments tend to be more cellular in radar structure with many strong convective updrafts, while the downwind (or downband) segments are more stratiform with relatively few strong updrafts (Barnes et al. 1983; Powell 1990; Barnes et al. 1991). Barnes et al. (1991) attributed these asymmetries to downband sounding stabilization [i.e., a reduction in convective available potential energy (CAPE)] induced by the melting and evaporation of hydrometeors transported from upband regions. Furthermore, numerous studies have argued that cool, dry (i.e., negatively buoyant) convective downdrafts within rainbands were responsible for an overall thermodynamic modification of the boundary layer and the subsequent suppression of deep convection at inner radii, including eyewall convection (Cione et al. 2000, and references therein). Here, we also examine the buoyancy characteristics of each case’s rainband cores within the context of these previous rainband studies.

a. Hurricane Guillermo on 2 August 1997

Figure 1 shows horizontal single scans of radar reflectivity obtained from the upper aircraft lower-fuselage (LF) radar during each eye penetration. The radar data were mapped into a storm-relative coordinate system using an objectively constructed track from flight-level winds (Willoughby and Chelmow 1982). Superimposed upon the reflectivity fields are the radial legs that compose each penetration. Initially at 1859 UTC (Fig. 1a), the eyewall was elliptical and asymmetric with enhanced convection (>30 dBZ) left-of-shear. By 1935 UTC (Fig. 1b) the primary eyewall band had rotated cyclonically to the upshear-left quadrant and propagated radially outward. Furthermore, several isolated convective cells had developed in each quadrant as the reflectivity interface between the eye and eyewall acquired a higher-wavenumber appearance. Recent studies (e.g., Schubert et al. 1999; Reasor et al. 2000; Kossin and Schubert 2001) have suggested that such reflectivity structure and evolution is evidence of vortex Rossby waves and mesovortices embedded within the eyewall. Upon the third eye penetration at 2115 UTC (Fig. 1c), the eyewall consisted of an asymmetric inward-spiraling band with enhanced convection left-of-shear. The eyewall maintained this wavenumber-1 structure and general orientation throughout the remaining 3 h of inner-core observation. Surrounding the eyewall were several convective and stratiform rainbands.

Inspection of animated radar reflectivity fields revealed that the eyewall contained multiple distinct convective cells (>30 dBZ, with diameters <10 km) that predominantly formed on the downshear side and moved cyclonically around to the upshear side. Likewise, most rainbands were composed of many isolated convective cells embedded within a stratiform rain area (e.g., the rainband located ∼60 km east of the center in Fig. 1f). Numerous convective cells were tracked1 to provide a description of their life cycle. Most eyewall cells were trackable for ∼120° from formation in the downshear quadrants to dissipation in the upshear quadrants. Cell lifetimes were typically 10–20 min, which translates to orbital velocities of ∼47 m s−1 (at 25-km radius) or ∼85% of the azimuthal mean horizontal wind in the 3.0–5.5-km layer. Rainband cells were trackable for ∼30°–60°over 10–30 min, and roughly moved with the local azimuthal mean layer wind (25–35 m s−1). These cell characteristics are consistent with previous observations (Parrish et al. 1984; Black et al. 2002). While individual convective cells can not be uniquely associated with individual updraft cores, their spatial similarity suggests that such cells may be manifestations of the cores.

It is interesting to note that during the first three eye penetrations a few eyewall reflectivity cells >10 km in diameter were trackable for ∼270° over 30–60 min. These larger cells moved much slower, at ∼65% of the azimuthal mean layer wind, which is consistent with expectations for convectively excited vortex Rossby waves (Montgomery and Kallenbach 1997).

Figure 2 shows representative flight-level data from the fifth eye penetration between 2316 and 2339 UTC (see Fig. 1e). The vertical lines denote convective-scale updraft cores. In the downshear-left eyewall, the aircraft encountered a ∼1.5 m s−1 mesoscale updraft upon which two updraft cores and one downdraft core was superimposed. The innermost updraft core exhibited average convective vertical velocity (wc) > 3 m s−1 and average total buoyancy (B) > 0.5 K, which exceed the respective upper 10% values of each distribution discussed in Part I. The outer updraft core was weaker (wc < 1.5 m s−1) and negatively buoyant (B ∼ −0.70 K), as was the downdraft core (B ∼ −0.15 K). In contrast, in the upshear-right eyewall the aircraft encountered minimal mesoscale vertical motion, no convective-scale cores, minimal buoyancy (positive or negative), and little cloud or precipitation water content. Interestingly, the equivalent potential temperature (θe) and relative vertical vorticity (ζ)2 were much less asymmetric; both radial legs exhibited θe and ζ maxima in the eyewall region with depressed values inside the eye. Note that the strong buoyant updraft core in the downshear-left eyewall is collocated with both θe and ζ maxima.

The rainband encountered 90–110 km southeast of the storm center (Fig. 2) contained a ∼2 m s−1 mesoscale updraft upon which three updraft cores and two downdraft cores were superimposed. Each updraft core was positively buoyant (B ∼ 0.3 K),and the downdraft cores were negatively buoyant (B ∼ −0.10 K). Elevated values of θe and ζ were also located within the rainband. Similar characteristics were observed during other rainband penetrations.

Figure 3 summarizes the vertical velocity and core buoyancy characteristics along all 12 radial legs flown at 5.5 km. The figure’s construction and utility are described here in some detail because similar summary figures are presented for each case. In Figs. 3a–c, radial leg data were superimposed upon a single-scan horizontal reflectivity field from a representative eye penetration. Figure 3a highlights sections of each radial leg in which the mesoscale vertical velocity (wm) exceeded certain appreciable magnitudes (3, 1, and −1 m s−1). Figure 3b depicts the locations of all identified updraft cores stratified by their wc and B characteristics. The stratification thresholds (wc > 2 m s−1 and B > 0.25 K) were designed to isolate updraft cores with appreciable vertical motion or positive buoyancy. Within the context of the distributions presented in Part I, the thresholds isolate cores from the upper ∼33% of each distribution. Figure 3c denotes the locations of downdraft cores with average total vertical velocity (w) < −1 m s−1 (i.e., stronger than 1 m s−1 downward) stratified by their B characteristics (either positive or negative). Downdraft cores with w > −1 m s−1 were excluded since they do not contribute significantly to downward mass transport (see section 3c in Part I for further discussion). The figures are utilized to depict gross azimuthal relationships between vertical velocity, buoyancy, and radar reflectivity. Convective-scale features are not directly comparable since the radial leg data were collected over the course of several hours. However, the quasi-steadiness in the mesoscale structure during each inner-core observation period should permit comparison of gross features.

Mesoscale updrafts (Fig. 3a) in excess of 1 m s−1 dominated the downshear and left-of-shear regions of the eyewall. Mesoscale descent <−1 m s−1 dominated the upshear eyewall. The wavenumber-1 eyewall reflectivity signature was rotated ∼60°–90° downwind of the mesoscale updraft band, with the mesoscale descent downwind of the reflectivity maximum. Previous studies (Marks et al. 1992; Reasor et al. 2000; Black et al. 2002) have shown similar azimuthal offsets that are consistent with the cyclonic advection of hydrometeors generated by the mesoscale (and convective-scale) updrafts. Appreciable mesoscale ascent was also encountered during several rainband penetrations.

A total of 49 updraft cores (Fig. 3b) were encountered at 5.5-km altitude within 120 km of the storm center. The 32 updraft cores in the eyewall region were evenly distributed about the center with 2–5 cores along each leg. However, the six strongest and most buoyant updraft cores (both wc > 2 m s−1 and B > 0.25 K) were superimposed upon mesoscale ascent and collocated with the higher radar reflectivities in the downshear and left-of-shear quadrants. Roughly half of the rainband updraft cores (8 of 17) exhibited B > 0.25 K. All such cores were encountered within rainband segments containing isolated convective cells. Clearly, the azimuthal distribution of buoyant updraft cores was highly asymmetric, yet well correlated with regions of enhanced convection.

Table 2 summarizes the percent area occupied and the percent total (convective and mesoscale) upward mass transport accomplished by updraft cores as a function of region and core characteristics. The six strongest and most buoyant eyewall cores occupied ∼2% of the eyewall region but accomplished ∼25% of the total transport. The most buoyant rainband updraft cores occupied ∼3% of the rainband region but accomplished ∼30% of the total transport. Clearly, these few buoyant cores significantly contribute to the total upward mass transport.

A total of 21 downdraft cores with w < −1 m s−1 were encountered within 120 km of the storm center (Fig. 3c). The majority of such eyewall downdraft cores (10 of 12) were in the left-of-shear and upshear quadrants either collocated with or downwind of reflectivity maxima. Eyewall cores in the upshear-left (upshear right) quadrant tended to exhibit positive (negative) total buoyancy, while most rainband downdraft cores (7 of 9) were negatively buoyant. Of all the downdrafts with B < 0.0 K, most (8 of 12) exhibited average thermal buoyancy (TB) < −0.25 K and average water loading (WL) >−0.10 K in the presence of liquid water, suggesting that the negative B were supported primarily by hydrometeor evaporation.

b. Hurricane Guillermo on 3 August 1997

Guillermo’s radar reflectivity structure on 3 August was strikingly similar to the structure observed on 2 August. The single-scan LF radar reflectivity fields obtained during each eye penetration at 5.5 km (Fig. 4) depict an asymmetric eyewall with enhanced convection in the downshear and left-of-shear quadrants. This wavenumber-1 pattern of high reflectivity remained relatively fixed with respect to the storm center during the 6-h observation period. The convective-scale reflectivity structure, however, exhibited much variability from penetration to penetration. Surrounding the eyewall were several inward-spiraling rainbands. Isolated convective cells tended to dominate the upwind portion of each rainband, while downwind portions exhibited more stratiform precipitation. Convective cell tracking in animated radar imagery yielded results similar to those of the day before.

Figure 5 shows flight-level data from the first eye penetration between 1855 and 1923 UTC (see Fig. 4a). In the downshear eyewall, the aircraft encountered a strong mesoscale updraft >3 m s−1 and two convective-scale updraft cores. The inner updraft core exhibited wc > 4 m s−1, B > 0.4 K, and was superimposed upon the mesoscale updraft maximum. The outer core (at 36-km radius) was weaker (wc < 1.5 m s−1) but equally buoyant (B ∼ 0.5 K). In the upshear eyewall, the aircraft encountered minimal mesoscale vertical motion, little cloud or precipitation water content, two strong downdraft cores (wc < −2 m s−1), and four weak updraft cores (wc < 2 m s−1). The inner downdraft core (at 25 km) contained ∼0.15 g m−3 of liquid water and was negatively buoyant, while the outer downdraft core (at 38 km) contained <0.05 g m−3 of liquid water and was positively buoyant. Of the updraft cores, only the inner two exhibited positive buoyancy (B < 0.3 K). The θe and ζ structures were, again, much less asymmetric with maxima in the eyewall and depressed values inside the eye. Thus, the upshear–downshear radial structure observed during the first eye penetration was qualitatively similar to the upshear–downshear structure observed ∼20 h prior on 2 August at the same altitude (see Fig. 2).

This general wavenumber-1 pattern in eyewall reflectivity and vertical velocity persisted during the 6-h observation period, but the θe and ζ radial profiles underwent significant structural changes. Figure 6 shows the flight-level data from the fifth eye penetration between 2333 and 2359 UTC (see Fig. 4e). As before, the aircraft encountered a strong mesoscale updraft and two buoyant updraft cores in the downshear-left eyewall, while a mesoscale downdraft and several relatively weak and near-neutral (|B| < 0.1 K) convective cores were encountered on the upshear-right side. However, in contrast to the first eye penetration, the θe and ζ profiles exhibited pronounced mesoscale maxima in the eye, rather than in the eyewall. Such dramatic transitions in the θe and ζ radial structure have been attributed to horizontal mixing associated with vortex Rossby waves and mesovortices (Kossin and Eastin 2001).

Figure 7 summarizes the vertical velocity and core buoyancy characteristics for all 12 radial legs at 5.5 km. Many characteristics evident on 3 August are strikingly similar to those observed on 2 August. Mesoscale updrafts (Fig. 7a) in excess of 3 m s−1 dominated the eyewall’s downshear quadrants, while mesoscale downdrafts (of which a few exceeded −1 m s−1) were predominant upshear. Again, the strong mesoscale updrafts were located ∼90° upwind of the reflectivity maximum (in the left-of-shear quadrants), which was upwind of the mesoscale downdrafts. No appreciable mesoscale downdrafts were encountered during rainband penetrations, but several mesoscale updrafts were.

A total of 71 updraft cores (Fig. 7b) and 33 downdraft cores with w <  −1 m s−1 (Fig. 7c) were encountered within 120 km of the storm center. The 41 eyewall updraft cores were again evenly distributed about the center, but the 10 strongest and most buoyant updraft cores were limited to the downshear and left-of-shear quadrants. Roughly 45% of rainband updraft cores (13 of 30) exhibited B > 0.25 K and were encountered within more cellular rainband segments. The most buoyant updraft cores of each region occupied <10% of the total area but accomplished ∼30%–40% of the total upward mass transport (see Table 2). The majority of eyewall downdraft cores (18 of 26) were located upshear, downwind of the reflectivity maxima. Of the 20 downdraft cores with negative B, over 70% exhibited TB < −0.25 K and WL > −0.10 K in the presence of liquid water.

c. Hurricane Georges on 19 September 1998

In contrast to Guillermo, the single-scan LF reflectivity fields from the five eye penetrations of Georges at 4.2-km altitude (not shown) exhibited a more closed and nearly symmetric eyewall (i.e., continuous ring of reflectivity >30 dBZ) at ∼25-km radius. However, consistent with Guillermo, the vertical velocity and core buoyancy characteristics (Fig. 8) were highly asymmetric. Mesoscale updrafts >1 m s−1 dominated the downshear and left-of-shear eyewall, while weaker mesoscale ascent (and descent) was predominant upshear and right-of-shear. A total of six updrafts cores and five downdraft cores with w < −1 m s−1 were encountered in the eyewall region by the second aircraft. While azimuthal relationships are difficult to ascertain from such little data, two noteworthy characteristics are evident. First, only two of the five strongest updraft cores were positively buoyant. However, both cores exhibited B > 0.25 K and were located left-of-shear along the inner eyewall edge. Second, the four downdraft cores located just outside the eyewall reflectivity maximum exhibited negative B with significant contributions from both TB and WL.

Multiple inward-spiraling convective and stratiform rainbands surrounded the eyewall. As in Guillermo, isolated convective cells tended to dominate the upwind portion of each rainband with stratiform precipitation dominant downwind. No appreciable mesoscale downdrafts were observed in rainbands, but several mesoscale updrafts were. A total of 11 updrafts cores and 2 downdraft cores with w < −1 m s−1 were encountered by the second aircraft. Roughly half of the updraft cores (6 of 11) exhibited B > 0.25 K. Both downdraft cores were negatively buoyant.

d. Synthesis

The three cases exhibited several striking similarities in azimuthal eyewall organization consistent with the conceptual model of Black et al. (2002; Fig. 9). Mesoscale vertical motions and gross radar reflectivity exhibited persistent wavenumber-1 asymmetries with maximum ascent (decent) downshear (upshear) and the precipitation maximum left-of-shear. Moreover, the buoyancy characteristics of convective cores exhibited consistent asymmetric relationships with respect to the vertical shear vector. In order to quantify these relationships, the radial legs in each case were rotated around the storm center so that the shear vector for the case was pointing due north. The eyewall area was divided into four regions with respect to the vertical shear vector: downshear right and left, and upshear right and left. The number of updraft and downdraft cores per 100 km of total eyewall region leg length was then determined for each quadrant. The normalization by total leg length was employed to account for inequalities in leg numbers per shear-rotated quadrant.

Figure 10 clearly shows that the strongest and most buoyant eyewall updraft cores were most often encountered in the downshear-left quadrant, but were rarely encountered in the upshear-right quadrant despite a nearly equal number of total updraft cores. Over 60% of downshear-left updrafts exhibited wc > 2 m s−1 and ∼55% contained B > 0.25 K. In contrast, less than 15% of upshear-right updrafts exhibited either B > 0.25 K or wc > 2 m s−1. Also note that the majority of the most buoyant updraft cores (∼65%) were located in the downshear quadrants, while most of the strongest updrafts (∼75%) were encountered in the left-of-shear quadrants. This cyclonic shift in the preferred quadrants is consistent with the typical updraft core evolution envisioned by Black et al. (2002): buoyant updraft cores originate at low levels in the downshear quadrants and begin to accelerate upward while being advected cyclonically around to the left-of-shear quadrants by the primary circulation. The convective cell activity observed in the Guillermo LF radar animations further supports this notion.

Figure 11 shows the normalized number of downdraft cores with w < −1 m s−1 encountered in each quadrant. Over 70% of downdrafts were located in the two upshear quadrants. When the cores are stratified by B, over 55% (85%) of negatively (positively) buoyant downdrafts were encountered in the left-of-shear (upshear) quadrants. Again, the cyclonic shift in preferred quadrants may be indicative of typical downdraft core evolution. Consider a negatively buoyant downdraft core (resulting from localized evaporative cooling and/or water loading) that originates at upper levels left-of-shear. As the core accelerates downward and is advected cyclonically around to the upshear quadrants, adiabatic warming will begin to overcome evaporational cooling (as the liquid water content decreases), and the downdraft core will become positively buoyant. The observed decreases in gross radar reflectivity as one moves cyclonically around the eyewalls from the left-of-shear quadrants to the upshear quadrants (see Figs. 3 and 7) further support this notion.

The buoyancy characteristics of rainband cores were also consistent with the aforementioned studies of hurricane rainbands. Buoyant updraft cores were more frequently encountered during upband penetrations (e.g., legs 48 and 49 in Fig 3; legs 61, 55, 58, and 64 in Fig. 7) than during downband penetrations (e.g., legs 46, 52, and 41 in Fig. 3; legs 53, 59, and 62 in Fig. 7). Upband segments also contained more isolated reflectivity cells than the more stratiform downband segments. Furthermore, of the 18 downdraft cores with w < −1 m s−1 encountered in rainbands (at altitudes above 4 km), 10 exhibited TB < −0.25 K in the presence of liquid water. Assuming these downdrafts can maintain their negative TB through hydrometeor evaporation and penetrate into the boundary layer (below ∼0.5-km altitude), our results support the notion that rainband downdraft cores can cool the low-level inflow. Such locally cool downdrafts were encountered equally in both upband and downband segments.

4. An unconventional source region for buoyant eyewall updrafts

Conventional wisdom states that the vast majority of air ascending in the eyewall originates in the large-scale environment and is driven inward by frictional dissipation. As the air flows inward, surface heat and moisture fluxes increase the mean boundary layer θe to values well above the large-scale environment. Upon approaching the eyewall, the air converges and is lifted. If, at some point during this forced ascent, an air parcel’s θe exceeds the local environmental θe, then the parcel will become locally buoyant and accelerate upward until the two θe values reequilibrate (i.e., the parcel consumes the local CAPE). Observations, however, indicate that appreciable CAPE is rarely observed in and near the eyewall (e.g., Bogner et al. 2000). Furthermore, as discussed previously, convective downdrafts within rainbands can inject low-θe air into the boundary layer and significantly lower the mean θe of the inflowing air. If adequate thermodynamic recovery of the inflow is not achieved prior to reaching the eyewall, subsequent eyewall convection may be suppressed (Powell 1990). In light of such obstacles, one might expect a buoyant eyewall updraft to constitute a rare event. However, the results presented here and in Part I indicate that buoyant updrafts are not so rare (∼33% of all eyewall updraft cores exhibited B > 0.25 K). How then do so many eyewall updraft cores acquire appreciable buoyancy?

Observations (Hawkins and Imbembo 1976; Jorgensen 1984b; Willoughby 1998; Kossin and Eastin 2001) and numerical simulations (Liu et al. 1997; Braun 2002; Persing and Montgomery 2003) indicate that θe in the low-level eye can, at times, exceed eyewall θe by 5–10 K. Braun (2002) and Persing and Montgomery (2003) have recently suggested that the outward advection of this high-θe eye air into the eyewall can generate locally buoyant updrafts. Here, we present observational evidence that several buoyant eyewall updraft cores in both Guillermo and Georges originated in the low-level eye and were generated by asymmetric outflow associated with low-wavenumber mesovortices embedded within the eye and eyewall.

In order to diagnose the source region of buoyant eyewall updraft cores observed at flight level, core e values were compared to vertical profiles of θe obtained below flight level in both the eye and eyewall by GPS dropwindsondes (Hock and Franklin 1999). On 3 August 1997, the two aircraft dropped three sondes in Guillermo’s eye and six in the eyewall. Multiple sondes were also deployed in the eye (4 sondes) and eyewall (14 sondes) of Georges on 19 September 1998. The sonde data for each case were obtained from the NOAA Hurricane Research Division quality-controlled data archive and further scrutinized following Bogner et al. (2000) to correct any mixed-layer humidity-sensor wetting errors.3 Vertical profiles of θe were then computed following Bolton (1980).

Figure 12a shows the drop locations and descent trajectories for the nine sondes deployed in Guillermo on 3 August. Each of the eye sondes were dropped well away from the eyewall in close proximity to the circulation center. Four of the eyewall sondes were deployed along the inner eyewall edge, while the other two were dropped within the high radar reflectivity region. Figure 12b shows the sonde-derived θe profiles. Eyewall θe ranged from 358 to 368 K but never exceeded 370 K, and the individual profiles exhibited little altitude dependence. In the eye, however, two profiles exhibited a rapid decrease in θe from ∼375 K (below 2 km) to ∼360 K (above 2 km). Therefore, below 5.5 km, air parcels with θe > 370 K were only observed in the low-level eye. Comparison of the θe profiles with the θe observed in the eyewall updraft cores at 5.5-km altitude (Fig. 12b) reveals that six cores contained θe > 370 K, of which three exhibited B > 0.25 K. Assuming the eyewall updrafts originated below 2 km, the low-level eye was the apparent source region for these six cores, not the boundary layer inflow below the eyewall.

It is interesting to note that the eyewall updraft cores could be roughly divided into two groups: cores with θe > 365 K and cores with θe < 365 K. Given that lower-θe air (∼360 K) was located both inside and outside the eyewall above 2-km altitude (see Figs. 5, 6, 12b), one might expect entrainment mixing to reduce the updraft core θe during their ascent from 2 to 5.5 km. Thus, it seems plausible that several of the cores with θe > 365 K may have also originated in the low-level eye. Further note that most cores with B > 0.25 K (12 of 17) and wc > 2.0 m s−1 (10 of 15) contained θe > 365 K. Therefore, the low-level eye may have been a primary source region for the strongest and most buoyant eyewall updrafts in Guillermo on 3 August.

Figure 13a shows the drop locations and descent trajectories for the 18 sondes deployed in Georges on 19 September. As in Guillermo, the eye sondes were dropped away from the eyewall in close proximity to the circulation center, and the eyewall sondes were dropped both within and along the inner edge of the eyewall. Most eyewall θe profiles (Fig. 13b) exhibited little altitude dependence as typical θe values ranged between 355 and 365 K. A few eyewall profiles decreased from ∼360 K near the surface to ∼350 K at 4-km altitude, indicative of appreciable positive CAPE. Each eye sonde exhibited a pronounced decrease in θe from ∼370 K (below 1 km) to ∼360 K (above 1 km). Therefore, below 4.2 km, air parcels with θe > 365 K were only observed in the low-level eye. Comparison with eyewall updraft core θe at 4.2-km altitude (Fig. 14b) reveals that two cores contained θe > 365 K, and both exhibited B > 0.25 K and wc > 2 m s−1. Again, assuming the updrafts originated below ∼1 km, the lowlevel eye was the apparent source region for both cores.

The results from Guillermo and Georges clearly suggest that the high-θe low-level eye was an important source region for the buoyant eyewall updrafts. An important remaining question is, what dynamical mechanism(s) transported the high-θe eye air outward into the eyewall? Potential mechanisms include thermally and dynamically driven transverse circulations (e.g., Shapiro and Willoughby 1982); local supergradient accelerations (e.g., Gray and Shea 1973); “through flows” induced by storm translation (e.g., Weatherford 1989; Peng and Williams 1990); and asymmetric low-wavenumber circulations (e.g., Marks et al. 1992; Reasor et al. 2000). Here, we focus on the latter. Schubert et al. (1999) demonstrated that a ζ structure composed of an annular ring of enhanced ζ in the eyewall with relatively weak ζ in the eye and outside the eyewall can support the growth of barotropic instabilities (i.e., vortex Rossby waves) that lead to vigorous asymmetric mixing between the eye and eyewall. During the mixing process, the annular ring breaks down into low-wavenumber mesovortices that initially distort the local flow into polygonal patterns (Kossin and Schubert 2001) and then actively exchange mass between the eye and eyewall (Kossin and Eastin 2001). A number of observational studies have documented mesovortices in the eye and eyewall of intense hurricanes (Kossin et al. 2002, and references therein).

Several observations from both hurricanes suggest that such mesovortices may have been responsible for the outward advection of high-θe air into the eyewall. First, the flight-level ζ profiles in Guillermo (Figs. 2 and 5) and Georges (not shown) satisfy the criteria for barotropic instability (Schubert et al. 1999). Second, the evolution of Guillermo’s ζ profiles (Figs. 5 and 6) is consistent with expectations for vigorous asymmetric mixing between the eye and eyewall (Kossin and Eastin 2001). Third, the polygonal (or wavelike) pattern in radar reflectivity along the inner edge of Guillermo’s eyewall at 3.0-km altitude (Fig. 12a) may be indicative of asymmetric hydrometeor advection induced by mesovortical flow (Kossin and Schubert 2001). Fourth, the θe evolution in the low-level eyes of Guillermo and Georges is suggestive of substantial rapid mixing between the high-θe eye and relatively low-θe eyewall. For example, in Guillermo (Fig. 12b) the average θe observed near the circulation center below 2 km changed from ∼375 K at 1948 UTC to a typical eyewall value of ∼365 K at 2125 UTC, and then returned to ∼375 K at 2351 UTC. Finally, video imagery obtained from Georges’ eye (Fig. 14) reveals striking coherent mesovortical structures in the low-level eye clouds adjacent to the eyewall.

5. Discussion

Both Guillermo and Georges contained several buoyant convective updraft cores during periods of significant intensification. The question remains as to what roles asymmetric buoyant convection play in hurricane evolution. Three specific roles (or mechanisms) are discussed here. For each mechanism, circumstantial support is provided by the cases as well as previous observational and numerical studies. The discussions focus on how an ensemble of buoyant convection can contribute to the maintenance of a hurricane by inducing a net spinup of the mean vortex equal to the net spindown. Extension to intensification (weakening) simply requires the buoyant convection to induce a net spinup greater than (less than) the net spindown. It should be emphasized that we do not contend any single mechanism produced the observed intensifications. Rather, hurricane evolution is envisioned to occur through a combination of each mechanism as well as a contribution from the symmetric dynamics.

First, buoyant inner-core convection can drive a net deep-layer radial inflow above the frictional boundary layer that, through absolute angular momentum conservation, can spinup and maintain the mean vortex against frictional dissipation. Since the upward acceleration of an air parcel will induce lateral entrainment and a local mass convergence, it seems plausible that an ensemble of buoyant updraft cores can drive such a net radial inflow. Figure 15 shows that each case exhibited significant radial inflow outside the eyewall above the boundary layer. Such inflow is not unique to these cases. Numerous observational analyses of intense hurricanes also depict significant deep-layer radial inflow (e.g., Riehl and Malkus 1961; Jorgensen 1984b; Marks and Houze 1987; Dodge et al. 1999). This mechanism is consistent with Ooyama’s (1969, 1982 conceptual model of hurricane evolution.

It is interesting to note that the steady-state fields predicted by the nonhydrostatic model of Rotunno and Emanuel (1987) also show a deep-layer radial inflow and an accelerating eyewall updraft in the lowest 10 km (see their Fig. 5). Such structure is not consistent with strictly moist-neutral ascent above the frictional boundary layer, as contended by Emanuel (1986). Rather, the accelerating updraft and midlevel inflow suggest that some buoyancy is present in the steady state, and thus required for hurricane maintenance.

Second, buoyant convection can contribute to hurricane evolution through the generation of adiabatic heating from mass-compensating subsidence. Since the upward acceleration of an air parcel will induce a local overturning circulation, it seems plausible that an ensemble of buoyant convection will drive mesoscale subsidence. Portions of this subsidence may be sustained and locally enhanced by hydrometeor evaporation, generating an ensemble of convective downdrafts. In a nearly moist-neutral environment (i.e., the inner core), any given subsidence or downdraft will become an adiabatic heat source (and positively buoyant) once cooling induced by hydrometeor evaporation is overcome. Recall that the eyewall and rainbands of each case contained numerous positively buoyant downdraft cores in close proximity to buoyant updraft cores (Figs. 3, 7 and 8). Several studies of intense hurricanes have also documented locally warm downdrafts adjacent to deep convection (Jorgensen 1984a; Black et al. 1994; Heymsfield et al. 2001). Any downdraft heating will tend to become locally trapped due to the high vorticity and inertial stability in the inner core (Shapiro and Willoughby 1982; Hack and Schubert 1986). Moreover, a heat source will induce a local warming in the vertical temperature profile, which, through hydrostatics, will enhance the local pressure gradient and lower the surface pressure. Therefore, hurricane maintenance can be achieved if an ensemble of buoyant convection can generate an enhancement of the mean radial pressure gradient and a spinup of the mean vortex equal to the net dissipation. This mechanism is consistent with the Heymsfield et al. (2001) conceptual model of how deep convective bursts influence hurricane evolution.

Third, buoyant convection can contribute to hurricane evolution by the generation of localized ζ anomalies and subsequent upscale dynamical interactions. Montgomery and Enagonio (1998) demonstrated that positive potential vorticity (or ζ) anomalies superimposed upon a strong cyclonic vortex will become axisymmetrized and spinup the vortex mean flow. Since the upward acceleration of an air parcel will converge and stretch the background ζ, creating a local ζ anomaly, it seems plausible that buoyant updrafts generate such ζ anomalies. Recall that several of the most buoyant updraft cores in Guillermo were collocated with localized positive ζ anomalies (Figs. 2 and 6). Such collocations were also observed in Georges. Enagonio and Montgomery (2001) further showed that the axisymmetrization of a single anomaly produced only a small intensification of the mean vortex (∼0.5 m s−1 over ∼3 days), but the combined effects of several “pulsed” anomalies produced a more appreciable intensification (∼10 m s−1 over ∼3 days). Again, recall that Guillermo’s buoyant updraft cores were frequently located in regions where many isolated convective-scale reflectivity cells were repeatedly observed to develop, mature, and dissipate over the course of 10–20 min (i.e., pulse). Evidence of the axisymmetrization process, as described in Kossin and Eastin (2001), was also observed in the flight-level profiles of Guillermo (Figs. 5 and 6) and the video imagery of Georges (Fig. 14). Previous studies of intense hurricanes have also documented pulsating convection (Heymsfield et al. 2001; Black et al. 2002) and evidence of the axisymmetrization process (Reasor et al. 2000). Therefore, hurricane maintenance can be achieved if an ensemble of buoyant convection can generate, through such upscale dynamical interactions, a spinup of the mean vortex equal to the net dissipation. This mechanism is consistent with Braun’s (2002) envisionment of how “hot towers” influence hurricane evolution.

6. Summary and conclusions

Extensive aircraft data collected in Hurricanes Guillermo on 2 and 3 August 1997 and Georges on 19 September 1998 were used to document several aspects of buoyant inner-core convection. For each case, airborne radar and in situ flight-level data were combined to elucidate gross azimuthal relationships between the precipitation and vertical velocity structure, with particular emphasis on the buoyancy characteristics of convective-scale vertical velocity cores. Additional observations from GPS dropwindsondes were used to investigate the source of buoyant updraft cores in the eyewall. The analyses revealed multiple striking similarities in convective structure and organization. Our significant findings are as follows:

  1. Each eyewall exhibited a quasi-stationary wavenumber-1 asymmetry in mesoscale vertical velocity and precipitation. The orientation of these asymmetries with the large-scale vertical shear vector was consistent with previous observations, theoretical expectations, and numerical simulations. Mesoscale ascent maxima were located downshear with relatively weaker mesoscale descent upshear. Mesoscale precipitation maxima were left-of-shear.

  2. Convective updraft cores were evenly distributed around the eyewalls. However, updraft cores with either appreciable average total buoyancy (B > 0.25 K) or strong average convective vertical velocity (wc > 2 m s−1) exhibited consistent asymmetric relationships with the vertical shear vector orientation. The majority of these strongest (most buoyant) updrafts were located left-of-shear (downshear). In each case, the updraft cores with B > 0.25 K occupied <10% of the total eyewall area but accomplished ∼25%–40% of the total upward mass transport.

  3. Numerous isolated convective precipitation cells (>30 dBZ with diameters <10 km) formed in the downshear eyewall and moved at ∼85% of the azimuthal mean wind around to the left-of-shear or upshear eyewall before dissipation. Cells lifetimes were typically 10–20 min. The spatial similarity and azimuthal collocation of the convective cells and the strongest and most buoyant updrafts cores suggests that the transient cells may be manifestations of these updraft cores.

  4. Convective downdraft cores that transport significant mass downward (w < −1 m s−1) were asymmetrically distributed in the eyewall. Over 70% were located upshear. Negative (positive) buoyancy was most common in left-of-shear (upshear) cores.

  5. Inward-spiraling rainbands exhibited an upband/downband asymmetry. Upband segments tended to contain more convective cells and buoyant (B > 0.25 K) updraft cores, while downband segments were more stratiform in nature with fewer buoyant updrafts. No pronounced asymmetry was evident for downdraft cores with w < −1 m s−1, but ∼55% exhibited appreciable negative buoyancy (B < −0.25 K).

  6. The low-level eye was an important source region for buoyant eyewall convection. In both hurricanes, multiple GPS dropwindsonde profiles revealed that θe in the low-level eye (below 1–2 km) exceeded eyewall θe by 5–10 K, and comparisons with θe in eyewall updraft cores at midlevels indicated that several of the most buoyant cores likely originated in the low-level eye. Extensive circumstantial evidence suggests that asymmetric low-wavenumber circulations generated the buoyant updrafts by exporting high-θe air from the low-level eye into the relatively low-θe eyewall.

  7. Three physical mechanisms whereby an ensemble of asymmetric buoyant convection could contribute to hurricane maintenance and evolution were discussed. Circumstantial evidence supporting each mechanism was observed.

Acknowledgments

The authors are grateful to Mark DeMaria for providing the SHIPS environmental predictor database; James Franklin, Mike Black, and Steve Feuer for providing the GPS dropsonde data; Hugh Willoughby for providing Fig. 9; Nancy Griffin for creating the radar animations; and Neal Dorst for providing the aircraft video imagery. We wish to thank Jim Kossin, John Gamache, and John Persing for helpful discussions regarding this work. Constructive comments by Mike Montgomery, Wayne Schubert, Mike Black, and one anonymous reviewer were also very beneficial. This research was supported by NSF Grants ATM-0071369 and ATM-9616818.

REFERENCES

  • Barnes, G. M., E. J. Zipser, D. Jorgensen, and F. Marks Jr., 1983: Mesoscale and convective structure of a hurricane rainband. J. Atmos. Sci., 40 , 21272137.

    • Search Google Scholar
    • Export Citation
  • Barnes, G. M., J. F. Gamache, M. A. LeMone, and G. J. Stossmeister, 1991: A convective cell in a hurricane rainband. Mon. Wea. Rev., 119 , 776794.

    • Search Google Scholar
    • Export Citation
  • Black, M. L., J. F. Gamache, F. D. Marks Jr., C. E. Samsury, and H. E. Willoughby, 2002: Eastern Pacific Hurricanes Jimena of 1991 and Olivia of 1994: The effect of vertical shear on structure and intensity. Mon. Wea. Rev., 130 , 22912312.

    • Search Google Scholar
    • Export Citation
  • Black, R. A., H. B. Bluestein, and M. L. Black, 1994: Unusually strong vertical motions in a Caribbean hurricane. Mon. Wea. Rev., 122 , 27222739.

    • Search Google Scholar
    • Export Citation
  • Bogner, P. B., G. M. Barnes, and J. L. Franklin, 2000: Conditional instability and shear for six hurricanes over the Atlantic ocean. Wea. Forecasting, 15 , 192207.

    • Search Google Scholar
    • Export Citation
  • Bolton, D., 1980: The computation of equivalent potential temperature. Mon. Wea. Rev., 108 , 10461053.

  • Braun, S. A., 2002: A cloud-resolving simulation of Hurricane Bob (1991): Storm structure and eyewall buoyancy. Mon. Wea. Rev., 130 , 15731592.

    • Search Google Scholar
    • Export Citation
  • Cione, J. J., P. G. Black, and S. H. Houston, 2000: Surface observations within the hurricane environment. Mon. Wea. Rev., 128 , 15501561.

    • Search Google Scholar
    • Export Citation
  • DeMaria, M., 1996: The effect of vertical shear on tropical cyclone intensity change. J. Atmos. Sci., 53 , 20762087.

  • DeMaria, M., and J. Kaplan, 1994a: Sea surface temperature and the maximum intensity of Atlantic tropical cyclones. J. Climate, 7 , 13241334.

    • Search Google Scholar
    • Export Citation
  • DeMaria, M., and J. Kaplan, 1994b: A statistical hurricane intensity prediction scheme (SHIPS) for the Atlantic basin. Wea. Forecasting, 9 , 209220.

    • Search Google Scholar
    • Export Citation
  • DeMaria, M., and J. Kaplan, 1999: An updated statistical hurricane intensity prediction scheme (SHIPS) for the Atlantic and eastern North Pacific basin. Wea. Forecasting, 14 , 326337.

    • Search Google Scholar
    • Export Citation
  • Dodge, P., R. W. Burpee, and F. D. Marks Jr., 1999: The kinematic structure of a hurricane with sea level pressure less than 900 mb. Mon. Wea. Rev., 127 , 9871004.

    • Search Google Scholar
    • Export Citation
  • Eastin, M. D., 2003: Buoyancy of convective vertical motions in the inner core of intense hurricanes. Ph.D. dissertation, Colorado State University, 152 pp. [Available from Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523.].

  • Eastin, M. D., W. M. Gray, and P. G. Black, 2005: Buoyancy of convective vertical motions in the inner core of intense hurricanes. Part I: General statistics. Mon. Wea. Rev., 133 , 188208.

    • Search Google Scholar
    • Export Citation
  • Elsberry, R. L., G. J. Holland, H. Garrish, M. DeMaria, and C. P. Gaurd, 1992: Is there any hope for tropical cyclone intensity change prediction?—A panel discussion. Bull. Amer. Meteor. Soc., 73 , 264275.

    • Search Google Scholar
    • Export Citation
  • Emanuel, K. A., 1986: An air–sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. J. Atmos. Sci., 43 , 585604.

    • Search Google Scholar
    • Export Citation
  • Enagonio, J., and M. T. Montgomery, 2001: Tropical cyclogenesis via convectively forced vortex Rossby waves in a shallow water primitive equation model. J. Atmos. Sci., 58 , 685705.

    • 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
  • Franklin, J. L., S. J. Lord, S. E. Feuer, and F. D. Marks, 1993: The kinematic structure of Hurricane Gloria (1985) determined from nested analyses of dropwindsonde and Doppler radar data. Mon. Wea. Rev., 121 , 24332451.

    • Search Google Scholar
    • Export Citation
  • Gray, W. M., and D. J. Shea, 1973: The hurricane’s inner core region. II: Thermal stability and dynamic characteristics. J. Atmos. Sci., 30 , 15651576.

    • Search Google Scholar
    • Export Citation
  • Hack, J. J., and W. H. Schubert, 1986: Nonlinear response of atmospheric vortices to heating by organized cumulus convection. J. Atmos. Sci., 43 , 15591573.

    • Search Google Scholar
    • Export Citation
  • Hawkins, H. F., and S. M. Imbembo, 1976: The structure of a small, intense hurricane—Inez, 1966. Mon. Wea. Rev., 104 , 418442.

  • Heymsfield, G. M., J. B. Halverson, J. Simpson, L. Tian, and T. P. Bui, 2001: ER-2 Doppler radar investigations of the eyewall of Hurricane Bonnie during the Convection and Moisture Experiment-3. J. Appl. Meteor., 40 , 13101330.

    • Search Google Scholar
    • Export Citation
  • Hock, T. F., and J. L. Franklin, 1999: The NCAR GPS dropwindsonde. Bull. Amer. Meteor. Soc., 80 , 407420.

  • Jones, S. C., 1995: The evolution of vortices in vertical shear. Part II: Initially barotropic vortices. Quart. J. Roy. Meteor. Soc., 121 , 821851.

    • Search Google Scholar
    • Export Citation
  • Jorgensen, D. P., 1984a: Mesoscale and convective-scale characteristics of mature hurricanes. Part I: General observations by aircraft. J. Atmos. Sci., 41 , 12681285.

    • Search Google Scholar
    • Export Citation
  • Jorgensen, D. P., 1984b: Mesoscale and convective-scale characteristics of mature hurricanes. Part II: Inner core structure of Hurricane Allen (1980). J. Atmos. Sci., 41 , 12871311.

    • Search Google Scholar
    • Export Citation
  • Kossin, J. P., and M. D. Eastin, 2001: Two distinct regimes in the kinematic and thermodynamic structure of the hurricane eye and eyewall. J. Atmos. Sci., 58 , 10791090.

    • Search Google Scholar
    • Export Citation
  • Kossin, J. P., and W. H. Schubert, 2001: Mesovortices, polygonal flow patterns, and rapid pressure falls in hurricane-like vortices. J. Atmos. Sci., 58 , 21962209.

    • Search Google Scholar
    • Export Citation
  • Kossin, J. P., B. D. McNoldy, and W. H. Schubert, 2002: Vortical swirls in hurricane eye clouds. Mon. Wea. Rev., 130 , 31443149.

  • Kraft, R. H., 1961: The hurricane’s central pressure and highest wind. Mar. Wea. Log, 5 , 155.

  • Lawrence, M. B., 1999: Eastern North Pacific hurricane season of 1997. Mon. Wea. Rev., 127 , 24402454.

  • Liu, Y., D-L. Zhang, and M. K. Yau, 1997: Multiscale numerical study of Hurricane Andrew (1992). Part I: An explicit simulation. Mon. Wea. Rev., 125 , 25972616.

    • Search Google Scholar
    • Export Citation
  • Marks Jr., F. D., and R. A. Houze Jr., 1987: Inner core structure of Hurricane Alicia from airborne Doppler radar observations. J. Atmos. Sci., 44 , 12961317.

    • Search Google Scholar
    • Export Citation
  • Marks Jr., F. D., R. A. Houze Jr., and J. F. Gamache, 1992: Dual-aircraft investigation of the inner core of Hurricane Norbert (1984). Part I: Kinematic structure. J. Atmos. Sci., 49 , 919942.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., and R. J. 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
  • Neumann, C. J., B. J. Jarvinen, C. J. McAdie, and G. R. Hammer, 1999: Tropical cyclones of the North Atlantic Ocean, 1871–1998. Historical Climatology Series Paper 6-2, National Climatic Data Center in Cooperation with the National Hurricane Center, 206 pp.

  • Ooyama, K., 1969: Numerical simulation of the life cycle of tropical cyclones. J. Atmos. Sci., 26 , 340.

  • Ooyama, K., 1982: Conceptual evolution of the theory and modeling of the tropical cyclone. J. Meteor. Soc. Japan, 60 , 369380.

  • Parrish, J. R., R. W. Burpee, F. D. Marks, and C. W. Landsea, 1984: Mesoscale and convective-scale characteristics of Hurricane Frederic during landfall. Preprints,. 15th Conf. on Hurricanes and Tropical Meteorology, Miami, FL, Amer. Meteor. Soc., 415–420.

    • Search Google Scholar
    • Export Citation
  • Pasch, R. J., L. A. Avila, and J. L. Guiney, 2001: Atlantic hurricane season of 1998. Mon. Wea. Rev., 129 , 30863123.

  • Peng, M. S., and R. T. Williams, 1990: Dynamics of vortex asymmetries and their influence on vortex motion on a β-plane. J. Atmos. Sci., 47 , 19872003.

    • Search Google Scholar
    • Export Citation
  • Persing, J., and M. T. Montgomery, 2003: Hurricane superintensity. J. Atmos. Sci., 60 , 23492371.

  • Powell, M. D., 1990: Boundary layer structure and dynamics in outer hurricane rainbands. Part II: Downdraft modification and mixed layer recovery. Mon. Wea. Rev., 118 , 918938.

    • 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
  • Reynolds, R. W., and T. M. Smith, 1993: An improved real-time global sea surface temperature analysis. J. Climate, 6 , 114119.

  • Riehl, H., and J. Malkus, 1961: Some aspects of Hurricane Daisy (1958). Tellus, 13 , 181213.

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

    • Search Google Scholar
    • Export Citation
  • Schubert, W. H., M. T. Montgomery, R. K. Taft, T. A. Guinn, S. R. Fulton, J. P. Kossin, and J. P. Edwards, 1999: Polygonal eyewalls, asymmetric eyewall contraction, and potential vorticity mixing in hurricanes. J. Atmos. Sci., 56 , 11971223.

    • Search Google Scholar
    • Export Citation
  • Shapiro, L. J., and H. E. Willoughby, 1982: The response of balanced hurricanes to local sources of heat and momentum. J. Atmos. Sci., 39 , 378394.

    • Search Google Scholar
    • Export Citation
  • Wang, J., H. L. Cole, D. J. Carlson, E. R. Miller, K. Beierle, A. Paukkunen, and T. K. Laine, 2002: Corrections of humidity measurement errors from the Vaisala RS80 radiosonde—Application to TOGA COARE data. J. Atmos. Oceanic Technol., 19 , 9811002.

    • Search Google Scholar
    • Export Citation
  • Weatherford, C. L., 1989: The structural evolution of typhoons. Atmospheric Science Paper 446, Department of Atmospheric Science, Colorado State University, 198 pp. [Available from Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523.].

  • Willoughby, H. E., 1990: Temporal changes of the primary circulation in tropical cyclones. J. Atmos. Sci., 47 , 242264.

  • Willoughby, H. E., 1998: Tropical cyclone eye thermodynamics. Mon. Wea. Rev., 126 , 30533067.

  • Willoughby, H. E., and M. B. Chelmow, 1982: Objective determination of hurricane tracks from aircraft observations. Mon. Wea. Rev., 110 , 12981305.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Storm-relative radar reflectivity at ∼5.5-km altitude for Hurricane Guillermo on 2 Aug 1997 during each eye penetration at (a) 1859, (b) 1935, (c) 2115, (d) 2155, (e) 2333, and (f) 0003 UTC on 3 Aug. The domain of (a)–(f) is 240 km × 240 km, and tick marks are shown every 24 km. Solid lines denote the radial legs flown during each eye penetration. Leg number increments denote aircraft heading during penetrations. The vertical wind shear vector is shown in the lower-left corner. Note that the eyewall was composed of a persistent wavenumber-1 reflectivity field and many transient smaller-scale features throughout much of the flight.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 2.
Fig. 2.

Storm-relative radial profiles of (a) total vertical velocity w, (b) total liquid water content qc + qr, (c) total buoyancy B, (d) equivalent potential temperature θe, and (e) relative vertical vorticity ζ for Hurricane Guillermo during the fifth eye penetration at ∼5.5-km altitude between 2316 and 2339 UTC on 2 Aug 1997. The bold line in (a) denotes the mesoscale vertical velocity wm estimated through application of a 20-km running Bartlett filter to the w data. Differences between w and wm define the convective vertical velocity wc from which updraft and downdraft cores were defined (|wc| > 1 m s−1 for at least 0.5 km). Vertical lines denoted identified convective updraft cores.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 3
Fig. 3

Summary of flight-level vertical velocity and core buoyancy for Hurricane Guillermo between 1800 UTC on 2 Aug and 0100 UTC on 3 Aug 1997 at ∼5.5-km altitude. (a) Sections of each radial leg in which the mesoscale vertical velocity wm exceeded certain magnitudes. (b) Locations of convective updraft cores stratified by their average convective vertical velocity wc and average total buoyancy B characteristics. (c) Locations of convective downdraft cores with average total vertical velocity w < −1 m s−1 stratified by their B characteristics. The radial leg data in (a)–(c) is superimposed upon a representative storm-relative radar reflectivity field observed at 2333 UTC during the fifth eye penetration. The domain of (a)–(c) is 240 km – 240 km, and tick marks are shown every 24 km. The vertical wind shear vector is shown in the upper-right corner. Dashed circles denote the approximate separation radius between the eyewall and rainband regions.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 4.
Fig. 4.

As in Fig. 1, but on 3 Aug at (a) 1912, (b) 1949, (c) 2125, (d) 2204, (e) 2351, and on 4 Aug of (f) 0029UTC.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 5.
Fig. 5.

As in Fig. 2, but during the first eye penetration between 1855 and 1923 UTC on 3 Aug.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 6.
Fig. 6.

As in Fig. 2, but during the fifth eye penetration between 2333 and 2359 UTC on 3 Aug.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 7.
Fig. 7.

As in Fig. 3, but between 1900 UTC on 3 Aug and 0100 UTC on 4 Aug. The representative storm-relative radar reflectivity field was observed at 2204 UTC during the fourth eye penetration.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 8.
Fig. 8.

As in Fig. 3, but for Hurricane Georges between 1900 UTC on 19 Sep and 0100 UTC on 20 Sep 1998 at ∼4.2-km altitude. The representative storm-relative radar reflectivity field was observed at 1955 UTC during the second eye penetration by the first aircraft. Note that core buoyancy characteristics could only be determined along the four radial legs flown by the second aircraft between 2300 and 0100 UTC.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 9.
Fig. 9.

Schematic illustration of the shear-induced convective asymmetry in hurricane eyewalls. The low-level environmental flow is denoted by the large gray arrow. Upper-level flow is indicated by the three blue arrows. Buoyant updrafts form somewhat upwind of the downshear side of the eyewall. They accelerate upward and advect cyclonically around the eyewall, ascending to upper levels on the upshear side. Downdrafts driven by evaporative cooling and water loading dominate the upshear eyewall. Adapted from Black et al. (2002).

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 10.
Fig. 10.

Shear-rotated quadrant plots showing the number of updraft cores encountered per 100 km of flight through the eyewall regions of the three cases. Cores are stratified by their average convective vertical velocity wc and average total buoyancy B characteristics. The center of each box represents the storm center, and the upper two quadrants of each box represent the downshear direction.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 11.
Fig. 11.

Shear-rotated quadrant plots showing the number of downdraft cores with average total vertical velocity w < −1 m s−1 encountered per 100 km of flight through the eyewall regions of the three cases. Cores are stratified by their average total buoyancy B characteristics. The center of each box represents the storm center, and the upper two quadrants of each box represent the downshear direction.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 12.
Fig. 12.

(a) Storm-relative radar reflectivity observed at ∼3.0 km in Hurricane Guillermo at 2054 UTC on 3 Aug 1997. The domain is 120 km × 120 km, and tick marks are shown every 12 km. Superimposed are GPS dropwindsonde launch locations (filled circles) and their storm-relative trajectories (lines). The dashed inner box, enlarged in the upper right-hand corner, shows the trajectories for the GPS sondes launched in the eye. The enlarged inner box domain is 10 km × 10 km, with tick marks every 1 km, and the small cross indicates the circulation center. (b) Vertical profiles of equivalent potential temperature θe obtained by the GPS sondes deployed in the eye (solid) and eyewall (dashed) of Hurricane Guillermo between 1900 UTC on 3 Aug and 0100 UTC on 4 Aug. Also shown at the top of (b) are core average θe values for all eyewall updraft cores (filled circles) observed at ∼5.5-km altitude during the same period, as well as the subset of updraft cores with average convective vertical velocity wc > 2 m s−1 (filled squares) and the subset of updraft cores with average total buoyancy B > 0.25 K (filled triangles).

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 13.
Fig. 13.

As in Fig. 12 but (a) at ∼4.2 km in Hurricane Georges at 0041 UTC on 20 Sep 1998, and (b) for GPS sondes deployed in Hurricane Georges between 1900 UTC on 19 Sep and 0100 UTC on 20 Sep. Core average equivalent potential temperature θe values are only shown for eyewall updraft cores encountered at ∼4.2-km altitude by the second aircraft between 2300 and 0100 UTC.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 14.
Fig. 14.

Video imagery of clouds in the low-level eye and along the inner edge of Georges’ eyewall during the first flight between 1900 and 2100 UTC on 19 Sep 1998. Images were obtained by the aircraft’s nose video camera (the gust probe arm is visible in the lower right-hand corner). Views are toward (a) the south at 1954 UTC, (b) the south at 1955 UTC, and (c) the north at 2019 UTC. Note the cyclonic swirls in the low-level eye clouds and the cumuliform towers in the eyewall adjacent to the cyclonic swirls.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Fig. 15.
Fig. 15.

Azimuthal mean radial velocity u at (a) ∼3.0-km altitude in Hurricane Guillermo between 1800 UTC on 2 Aug and 0100 UTC on 3 Aug 1997, (b) ∼5.5-km altitude in Hurricane Guillermo between 1900 UTC on 3 Aug and 0100 UTC on 4 Aug 1997, and (c) ∼4.2-km altitude in Hurricane Georges between 1900 UTC on 19 Sep and 0100 UTC on 20 Sep 1998. Azimuthal means were computed for each 0.5-km radial bin in which data were available from at least half of the radial legs flown. Vertical dashed lines denote the approximate separation radius between the eyewall and rainband regions. Note that each profile is dominated by radial inflow beyond ∼30 km.

Citation: Monthly Weather Review 133, 1; 10.1175/MWR-2849.1

Table 1.

Summary of intensity, motion, and large-scale environmental characteristics for the hurricane cases examined in this study.

Table 1.
Table 2.

Summary of radial legs and updraft core statistics for each case study examined here and all intense hurricanes examined in Part I. Updraft cores are stratified by region, average convective vertical velocity (wc), and average total buoyancy (B). Core statistics include number (N), percentage of total area occupied, and percentage of total upward mass transport (MT) accomplished.

Table 2.

1

Convective cell identification and location (radius and azimuth) was determined using the sequential storm-relative LF reflectivity fields available every 30 s.

2

Equivalent potential temperatures were computed following Bolton (1980), and relative vertical vorticity was estimated from the tangential wind profiles using a simple difference approximation of ζ(r) = ∂(rυ)/r∂r.

3

Humidity-sensor wetting errors may exist above the mixed layer. The errors will result in an overestimate of relative humidity, and thus an overestimate of θe. Eastin (2003) argued that realistic humidity errors of 5%–10% (θe errors of 3–5 K) may be present in the eyewall profiles and the low-level portions of the eye profiles. Such errors, however, do not alter our conclusions and are probably somewhat offset by a known dry bias (Wang et al. 2002).

Save
  • Barnes, G. M., E. J. Zipser, D. Jorgensen, and F. Marks Jr., 1983: Mesoscale and convective structure of a hurricane rainband. J. Atmos. Sci., 40 , 21272137.

    • Search Google Scholar
    • Export Citation
  • Barnes, G. M., J. F. Gamache, M. A. LeMone, and G. J. Stossmeister, 1991: A convective cell in a hurricane rainband. Mon. Wea. Rev., 119 , 776794.

    • Search Google Scholar
    • Export Citation
  • Black, M. L., J. F. Gamache, F. D. Marks Jr., C. E. Samsury, and H. E. Willoughby, 2002: Eastern Pacific Hurricanes Jimena of 1991 and Olivia of 1994: The effect of vertical shear on structure and intensity. Mon. Wea. Rev., 130 , 22912312.

    • Search Google Scholar
    • Export Citation
  • Black, R. A., H. B. Bluestein, and M. L. Black, 1994: Unusually strong vertical motions in a Caribbean hurricane. Mon. Wea. Rev., 122 , 27222739.

    • Search Google Scholar
    • Export Citation
  • Bogner, P. B., G. M. Barnes, and J. L. Franklin, 2000: Conditional instability and shear for six hurricanes over the Atlantic ocean. Wea. Forecasting, 15 , 192207.

    • Search Google Scholar
    • Export Citation
  • Bolton, D., 1980: The computation of equivalent potential temperature. Mon. Wea. Rev., 108 , 10461053.

  • Braun, S. A., 2002: A cloud-resolving simulation of Hurricane Bob (1991): Storm structure and eyewall buoyancy. Mon. Wea. Rev., 130 , 15731592.

    • Search Google Scholar
    • Export Citation
  • Cione, J. J., P. G. Black, and S. H. Houston, 2000: Surface observations within the hurricane environment. Mon. Wea. Rev., 128 , 15501561.

    • Search Google Scholar
    • Export Citation
  • DeMaria, M., 1996: The effect of vertical shear on tropical cyclone intensity change. J. Atmos. Sci., 53 , 20762087.

  • DeMaria, M., and J. Kaplan, 1994a: Sea surface temperature and the maximum intensity of Atlantic tropical cyclones. J. Climate, 7 , 13241334.

    • Search Google Scholar
    • Export Citation
  • DeMaria, M., and J. Kaplan, 1994b: A statistical hurricane intensity prediction scheme (SHIPS) for the Atlantic basin. Wea. Forecasting, 9 , 209220.

    • Search Google Scholar
    • Export Citation
  • DeMaria, M., and J. Kaplan, 1999: An updated statistical hurricane intensity prediction scheme (SHIPS) for the Atlantic and eastern North Pacific basin. Wea. Forecasting, 14 , 326337.

    • Search Google Scholar
    • Export Citation
  • Dodge, P., R. W. Burpee, and F. D. Marks Jr., 1999: The kinematic structure of a hurricane with sea level pressure less than 900 mb. Mon. Wea. Rev., 127 , 9871004.

    • Search Google Scholar
    • Export Citation
  • Eastin, M. D., 2003: Buoyancy of convective vertical motions in the inner core of intense hurricanes. Ph.D. dissertation, Colorado State University, 152 pp. [Available from Department of Atmospheric Science, Colorado State University, Fort Collins, CO 80523.].

  • Eastin, M. D., W. M. Gray, and P. G. Black, 2005: Buoyancy of convective vertical motions in the inner core of intense hurricanes. Part I: General statistics. Mon. Wea. Rev., 133 , 188208.

    • Search Google Scholar
    • Export Citation
  • Elsberry, R. L., G. J. Holland, H. Garrish, M. DeMaria, and C. P. Gaurd, 1992: Is there any hope for tropical cyclone intensity change prediction?—A panel discussion. Bull. Amer. Meteor. Soc., 73 , 264275.

    • Search Google Scholar
    • Export Citation
  • Emanuel, K. A., 1986: An air–sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. J. Atmos. Sci., 43 , 585604.

    • Search Google Scholar
    • Export Citation
  • Enagonio, J., and M. T. Montgomery, 2001: Tropical cyclogenesis via convectively forced vortex Rossby waves in a shallow water primitive equation model. J. Atmos. Sci., 58 , 685705.

    • 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
  • Franklin, J. L., S. J. Lord, S. E. Feuer, and F. D. Marks, 1993: The kinematic structure of Hurricane Gloria (1985) determined from nested analyses of dropwindsonde and Doppler radar data. Mon. Wea. Rev., 121 , 24332451.

    • Search Google Scholar
    • Export Citation
  • Gray, W. M., and D. J. Shea, 1973: The hurricane’s inner core region. II: Thermal stability and dynamic characteristics. J. Atmos. Sci., 30 , 15651576.

    • Search Google Scholar
    • Export Citation
  • Hack, J. J., and W. H. Schubert, 1986: Nonlinear response of atmospheric vortices to heating by organized cumulus convection. J. Atmos. Sci., 43 , 15591573.

    • Search Google Scholar
    • Export Citation
  • Hawkins, H. F., and S. M. Imbembo, 1976: The structure of a small, intense hurricane—Inez, 1966. Mon. Wea. Rev., 104 , 418442.

  • Heymsfield, G. M., J. B. Halverson, J. Simpson, L. Tian, and T. P. Bui, 2001: ER-2 Doppler radar investigations of the eyewall of Hurricane Bonnie during the Convection and Moisture Experiment-3. J. Appl. Meteor., 40 , 13101330.

    • Search Google Scholar
    • Export Citation
  • Hock, T. F., and J. L. Franklin, 1999: The NCAR GPS dropwindsonde. Bull. Amer. Meteor. Soc., 80 , 407420.

  • Jones, S. C., 1995: The evolution of vortices in vertical shear. Part II: Initially barotropic vortices. Quart. J. Roy. Meteor. Soc., 121 , 821851.

    • Search Google Scholar
    • Export Citation
  • Jorgensen, D. P., 1984a: Mesoscale and convective-scale characteristics of mature hurricanes. Part I: General observations by aircraft. J. Atmos. Sci., 41 , 12681285.

    • Search Google Scholar
    • Export Citation
  • Jorgensen, D. P., 1984b: Mesoscale and convective-scale characteristics of mature hurricanes. Part II: Inner core structure of Hurricane Allen (1980). J. Atmos. Sci., 41 , 12871311.

    • Search Google Scholar
    • Export Citation
  • Kossin, J. P., and M. D. Eastin, 2001: Two distinct regimes in the kinematic and thermodynamic structure of the hurricane eye and eyewall. J. Atmos. Sci., 58 , 10791090.

    • Search Google Scholar
    • Export Citation
  • Kossin, J. P., and W. H. Schubert, 2001: Mesovortices, polygonal flow patterns, and rapid pressure falls in hurricane-like vortices. J. Atmos. Sci., 58 , 21962209.

    • Search Google Scholar
    • Export Citation
  • Kossin, J. P., B. D. McNoldy, and W. H. Schubert, 2002: Vortical swirls in hurricane eye clouds. Mon. Wea. Rev., 130 , 31443149.

  • Kraft, R. H., 1961: The hurricane’s central pressure and highest wind. Mar. Wea. Log, 5 , 155.

  • Lawrence, M. B., 1999: Eastern North Pacific hurricane season of 1997. Mon. Wea. Rev., 127 , 24402454.

  • Liu, Y., D-L. Zhang, and M. K. Yau, 1997: Multiscale numerical study of Hurricane Andrew (1992). Part I: An explicit simulation. Mon. Wea. Rev., 125 , 25972616.

    • Search Google Scholar
    • Export Citation
  • Marks Jr., F. D., and R. A. Houze Jr., 1987: Inner core structure of Hurricane Alicia from airborne Doppler radar observations. J. Atmos. Sci., 44 , 12961317.

    • Search Google Scholar
    • Export Citation
  • Marks Jr., F. D., R. A. Houze Jr., and J. F. Gamache, 1992: Dual-aircraft investigation of the inner core of Hurricane Norbert (1984). Part I: Kinematic structure. J. Atmos. Sci., 49 , 919942.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., and R. J. 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
  • Neumann, C. J., B. J. Jarvinen, C. J. McAdie, and G. R. Hammer, 1999: Tropical cyclones of the North Atlantic Ocean, 1871–1998. Historical Climatology Series Paper 6-2, National Climatic Data Center in Cooperation with the National Hurricane Center, 206 pp.

  • Ooyama, K., 1969: Numerical simulation of the life cycle of tropical cyclones. J. Atmos. Sci., 26 , 340.

  • Ooyama, K., 1982: Conceptual evolution of the theory and modeling of the tropical cyclone. J. Meteor. Soc. Japan, 60 , 369380.

  • Parrish, J. R., R. W. Burpee, F. D. Marks, and C. W. Landsea, 1984: Mesoscale and convective-scale characteristics of Hurricane Frederic during landfall. Preprints,. 15th Conf. on Hurricanes and Tropical Meteorology, Miami, FL, Amer. Meteor. Soc., 415–420.

    • Search Google Scholar
    • Export Citation
  • Pasch, R. J., L. A. Avila, and J. L. Guiney, 2001: Atlantic hurricane season of 1998. Mon. Wea. Rev., 129 , 30863123.

  • Peng, M. S., and R. T. Williams, 1990: Dynamics of vortex asymmetries and their influence on vortex motion on a β-plane. J. Atmos. Sci., 47 , 19872003.

    • Search Google Scholar
    • Export Citation
  • Persing, J., and M. T. Montgomery, 2003: Hurricane superintensity. J. Atmos. Sci., 60 , 23492371.

  • Powell, M. D., 1990: Boundary layer structure and dynamics in outer hurricane rainbands. Part II: Downdraft modification and mixed layer recovery. Mon. Wea. Rev., 118 , 918938.

    • 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
  • Reynolds, R. W., and T. M. Smith, 1993: An improved real-time global sea surface temperature analysis. J. Climate, 6 , 114119.