Abstract

Tropical cyclone (TC) development near upper-level potential vorticity (PV) streamers in the North Atlantic is studied from synoptic climatology, composite, and case study perspectives. Midlatitude anticyclonic wave breaking is instrumental in driving PV streamers into subtropical and tropical latitudes, in particular near the time-mean midocean trough identified previously as the tropical upper-tropospheric trough. Twelve TCs developed within one Rossby radius of PV streamers in the North Atlantic from June through November 2004–08. This study uses composite analysis in the disturbance-relative framework to compare the structural and thermodynamic evolution for developing and nondeveloping cases.

The results show that incipient tropical disturbances are embedded in an environment characterized by 850–200-hPa westerly vertical wind shear and mid- and upper-level quasigeostrophic ascent associated with the PV streamer, with minor differences between developing and nondeveloping cases. The key difference in synoptic-scale flow between developing and nondeveloping cases is the strength of the anticyclone north of the incipient tropical disturbance. The developing cases are marked by a stronger near-surface pressure gradient and attendant easterly flow north of the vortex, which drives enhanced surface latent heat fluxes and westward (upshear) water vapor transport. This evolution in water vapor facilitates an upshear propagation of convection, and the diabatically influenced divergent outflow erodes the PV streamer aloft by negative advection of PV by the divergent wind. This result suggests that the PV streamer plays a secondary role in TC development, with the structure and intensity of the synoptic-scale anticyclone north of the incipient vortex playing a primary role.

1. Introduction

Tropical cyclone (TC) development in the North Atlantic occurs in conjunction with a rich array of precursor disturbances. These disturbances range from easterly waves that originate over Africa (e.g., Kiladis et al. 2006; Burpee 1972) to baroclinic cyclones that originate in the midlatitude westerly flow (e.g., Davis and Bosart 2003). The environmental factors that are favorable for TC development are well documented and include warm sea surface temperatures (SSTs), moist conditions in midlevels, strong divergence in upper levels, weak vertical wind shear, and preexisting low-level cyclonic vorticity (e.g., Gray 1968; DeMaria et al. 2001). It is known, however, that TCs also develop in environments with moderate to strong deep-layer vertical wind shear (e.g., Bracken and Bosart 2000; Davis and Bosart 2001, 2002, 2003, 2004, 2006) and over relatively cool SSTs (e.g., McTaggart-Cowan et al. 2006). In fact, previous research has demonstrated that half of all TC developments over the North Atlantic occur in environments with moderate to strong baroclinicity and attendant vertical wind shear (e.g., McTaggart-Cowan et al. 2008, 2013; Hess et al. 1995; Elsner et al. 1996). These baroclinically influenced TC developments encompass subtropical cyclones that undergo tropical transition (TT; e.g., Davis and Bosart 2003, 2004; Evans and Guishard 2009; Guishard et al. 2009; Hulme and Martin 2009a,b) and precursor low-level cyclonic vorticity centers that interact with upper-tropospheric troughs (e.g., Sadler 1976, 1978; Lander 1994; Molinari et al. 1995; Chen et al. 2008).

In this study, we will examine North Atlantic TC development in conjunction with precursor low-level cyclonic vorticity centers located within one Rossby radius of upper-level troughs referred to as potential vorticity (PV) streamers (Wernli and Sprenger 2007). Potential vorticity streamers extend southwestward into lower latitudes in response to increased stretching deformation on the periphery of the upstream anticyclone (Martius et al. 2008) during anticyclonic wave breaking [AWB; termed “baroclinic life cycle 1” events; Thorncroft et al. (1993)]. The intrusion of PV streamers into lower latitudes occurs most frequently near time-mean midocean trough axes, commonly referred to as the tropical upper-tropospheric troughs (TUTTs; Sadler 1967). TUTTs have been described as dynamically active regions of AWB and preferred areas for the intrusion of cyclonic PV into lower latitudes (e.g., Postel and Hitchman 1999; Ndarana and Waugh 2011).

Previous studies that examined the influence of TUTTs, and upper-level cutoff lows that fracture from TUTTs, in TC development focused primarily on TCs over the western North Pacific. Riehl (1948) examined synoptic charts at lower- and upper-tropospheric levels for July–September 1945–47 and found that upper-level disturbances were quite common throughout the tropical and subtropical Pacific. TC development was more favorable when tropical disturbances embedded in the lower-tropospheric easterlies moved into regions of near-surface convergence driven by these upper-level disturbances. Sadler (1976, 1978) suggested that these upper-level cutoff lows (referred to as TUTT cells), which are embedded in the southern end of TUTTs, promote TC development by enhancing the upper-level divergent outflow and establishing an upper-level outflow channel near the low-level disturbance. Kelley and Mock (1982), in a compositing study of TUTT cells, similarly documented near-surface convergence in conjunction with deep-layer quasigeostrophic (QG) ascent on the southeastern flank of the upper-level disturbance [see also Yu and Kwon (2005) and references therein]. The influence of upper-tropospheric troughs on TC development has also been documented in the southern Indian Ocean (Payne and Methven 2012) and near Australia (McBride and Keenan 1982).

While previous research has shown that TUTT cells provide a relatively efficient genesis pathway for producing TCs over the western North Pacific and North Atlantic basins (McTaggart-Cowan et al. 2013), the drier midlevel air mass and stronger vertical wind shear that accompany upper-tropospheric troughs, referred to here as PV streamers, can be detrimental to TC development (Knaff 1997). Additionally, Chen et al. (2008) suggested that low-level tropical disturbances associated with PV streamers rarely develop and account for only 1.9% of all TC developments over the western North Pacific from 1979 to 2002. However, upper-level PV streamers that fracture into cutoff lows account for 13% of monsoon gyre formation events, which can thereafter contribute to TC development (Lander 1994). Still, the body of work on TC development in conjunction with upper-level PV features prompts questions about what factors dictate whether TC development will occur in conjunction with an upper-level PV streamer.

An important goal of recent TC-related field programs1 was to sample tropical disturbances early in their life cycles through their development into named TCs or their dissipation. Analyses of dropsonde data from the Pre-Depression Investigation of Cloud-systems in the Tropics (PREDICT) field program have shown relatively moist conditions within the “pouch” [defined as the region of recirculation in the disturbance-relative framework (Dunkerton et al. 2009)] for developing TCs compared to nondeveloping disturbances (Davis and Ahijevych 2012; Smith and Montgomery 2012; Komaromi 2013). The importance of vertical alignment of the developing TC vortex through a deep layer was emphasized by Davis and Ahijevych (2012) as being instrumental for reducing the infiltration of dry air into the pouch, thus facilitating moistening through a deep layer (see also Nolan 2007; Raymond et al. 2011; Rappin and Nolan 2012; Dunkerton et al. 2009). Davis and Ahijevych (2013) added that deep-layer moistening and enhanced static stability near the center of the pouch facilitates a vertical mass flux profile that contributes to low-level convergence and spin up of cyclonic vorticity (e.g., Bister and Mapes 2004; Raymond and Sessions 2007). To date, however, the analysis of cases from recent field programs primarily involves easterly wave disturbances over the central and western tropical North Atlantic, making it unclear if the results are applicable to tropical disturbances that interact with upper-level PV streamers.

Figure 1 shows a schematic of TC development near an upper-level PV streamer during AWB {represented by the 200-hPa isobar on the dynamic tropopause [DT; defined as the 2.0 PV unit (PVU) surface; 1.0 PVU = 1.0 × 10−6 m2 s−1 K kg−1]} and describes the type of event that will be examined in this study. TC development is preferred along the southeastern flank of the PV streamer as it progresses southeastward. This study will use National Centers for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR; Saha et al. 2010) data to examine TC development near upper-level PV streamers in the North Atlantic during June–November 2004–08. We will use composite and statistical analysis to examine the structure and thermodynamic differences between developing and nondeveloping tropical disturbances interacting with upper-level PV streamers and infer how these differences may relate to TC development. At issue is whether the upper-level PV streamer plays an integral role in TC development, or whether the structure of the precursor vortex and nearby synoptic-scale environment in the lower and midtroposphere has a dominant influence in determining the outcome of the interaction with the upper-level disturbance. The focus of this study is thus necessarily split between a diagnosis of the evolution of potentially important PV streamers at low latitudes and an analysis of the morphology of potential TC precursors in the lower and midtroposphere. The connection of these two elements of the investigation is what will allow us to further our understanding of the impact of cyclonic upper-level circulations on the TC development process.

Fig. 1.

Schematic illustration of AWB on the DT. The 200-hPa isobar on the DT is marked by a black solid contour at t − 0 h. Circulations associated with pressure anomalies are indicated by dotted circles with arrows. Positive pressure anomalies (+p) are equivalent to positive PV anomalies, and negative pressure anomalies (−p) correspond to negative PV anomalies. The preferred location for TC development is marked by the red hurricane symbol.

Fig. 1.

Schematic illustration of AWB on the DT. The 200-hPa isobar on the DT is marked by a black solid contour at t − 0 h. Circulations associated with pressure anomalies are indicated by dotted circles with arrows. Positive pressure anomalies (+p) are equivalent to positive PV anomalies, and negative pressure anomalies (−p) correspond to negative PV anomalies. The preferred location for TC development is marked by the red hurricane symbol.

This paper is organized as follows. Section 2 describes the data and analysis methods employed throughout the investigation. A brief case study of the development of TC Otto (2010) is presented in section 3. Section 4 will provide a climatology of TC development and AWB during 2004–08, and section 5 will present the composite results. The concluding discussion will be presented in section 6.

2. Data and methods

a. Datasets

The NCEP CFSR is available four times daily at 0.5° × 0.5° horizontal grid spacing on isobaric levels at 50-hPa increments (25-hPa increments below 750 hPa) and constitutes the primary data source for this study. A recent investigation by Schenkel and Hart (2012) examined the validity of using atmospheric reanalysis datasets in TC-related studies and found that the CFSR [and the Japanese 25-year Reanalysis (Onogi et al. 2007)] had the smallest position errors relative to observations for TCs compared to other reanalysis datasets, likely because of the use of vortex relocation. Schenkel and Hart (2012) also identified an error in the vortex relocation code that created discontinuities in the mass field for a nontrivial fraction of TCs in the CFSR, particularly for TCs near complex terrain. These discontinuities in the mass field were not apparent near the TCs examined in the present study, however. Therefore, we can have some confidence in the ability of the CFSR to represent synoptic-scale and near-storm subsynoptic-scale environment and flow features. The CFSR and NOAA Optimum Interpolated SST fields (Reynolds and Smith 1994) were obtained from the National Center for Atmospheric Research (NCAR) Research Data Archive and the NOAA/Earth System Research Laboratory, respectively. Satellite-derived products used in the case study were obtained from the Naval Research Laboratory.

b. Identification of AWB events and vortex tracking methodology

Anticyclonic wave breaking events that occurred from North America to western Europe2 during June–November 2004–08 were manually identified by examining 6-hourly DT analyses. A schematic of the structure and key features of an AWB event are shown in Fig. 1. Time t − 0 h for an AWB event was identified when the following criteria were first satisfied as illustrated in Fig. 1: i) the midlatitude anticyclone exhibited a closed circulation, and ii) an upper-level PV streamer [as defined by Wernli and Sprenger (2007), their section 2b and Fig. 1] was located directly south of the center of the anticyclone. While a latitude restriction was not employed, the PV streamer was required to be connected to the higher-latitude PV reservoir on the eastern flank of the anticyclone.

To identify low-level disturbances located within one Rossby radius3 (typically about 1000 km) of the upper-level PV streamers, an objective vortex-tracking algorithm was used following the methodology of Davis et al. (2008). All vorticity centers that formed over the North Atlantic basin south of 50°N and exceeded a maximum 450 × 450 km2 area-averaged relative vorticity value of 1.5 × 10−5 s−1 at 900 hPa for 48 h or more were identified and tracked. Any tracks that corresponded to TCs that reached tropical storm intensity in the Hurricane Best Track Database (HURDAT; Landsea et al. 2004) were categorized as developing TCs, and all remaining tracks were categorized as nondeveloping disturbances. The 900-hPa vorticity centers that were located within one Rossby radius of a PV streamer on its southeastern flank at t − 0 h during their development stage (defined as a 48-h period where the 900-hPa vorticity was increasing) were then identified. It is these identified vorticity centers that are considered in this study as tropical disturbances that occurred near an upper-level PV streamer.

c. Compositing methodology

The composite analysis was performed in disturbance-relative coordinates based on the position of the 900-hPa cyclonic vorticity center, as determined by the objective vortex tracking algorithm. For the composites, CFSR data interpolated to a Lambert conformal grid with 50-km grid spacing and a 61 × 61 grid point box centered on the 900-hPa vorticity center was used for averaging. The composites were computed at time t − 72, t − 48, t − 24, t − 12, t − 0, and t + 24 h. Although the composite times are relative to AWB life cycle, with t − 0 h as defined earlier, the times also coincide with the life cycle of the tropical disturbance itself, with nondeveloping disturbances peaking in intensity on average at time t − 0 h and subsequently weakening, while developing TCs continued to intensify after t − 0 h. The disturbance-relative composite analysis facilitates the examination of the relevant near-storm synoptic- and subsynoptic-scale kinematic and thermodynamic structures of developing and nondeveloping disturbances. Statistical significance was assessed using the nonparametric Wilcoxon–Mann–Whitney rank sum test (Wilks 1995, 138–143), unless otherwise noted, with significance defined at the 95% level. We used a nonparametric statistical test to account for skewness in the data distributions.

In addition to diagnostic calculations described in the results section, we solved for the QG vertical motion using the routine available in the Read/Interpolate/Plot software package (users’ guide available online at http://www2.mmm.ucar.edu/wrf/users/docs/ripug.htm), which inverts (using overrelaxation) the Q vector form of the QG omega equation, defined as [Bluestein 1992, see his Eq. (5.7.56)]

 
formula

where Q was defined as [Bluestein 1992, see his Eq. (5.7.55)]

 
formula

with air pressure p, air temperature T, geostrophic wind , gradient on a pressure surface , and gas constant for dry air R. This calculation uses a domain-average Coriolis parameter f and vertical static stability (σ) profile, with the domain defined as the North Atlantic basin, and it is performed on isobaric levels.

Other diagnostic calculations used herein include the (i) moist static energy (MSE), which is defined as , where Cp is the heat capacity, T is the temperature, Φ is the geopotential, Lυ is the latent heat of vaporization, and q is the water vapor mixing ratio; and (ii) vertically integrated water vapor transport (IVT), calculated as , where q is the water vapor mixing ratio, V is the horizontal wind, p0 is 900 hPa, p is 700 hPa, and g is the acceleration due to gravity (see also Moore et al. 2012).

3. Brief analysis of TC Otto (2010)

Within 1 week of the conclusion of the field phase of PREDICT, TC Otto developed in the North Atlantic basin. The development of Otto is noteworthy, because it occurred as a low-level tropical disturbance interacted with the southern end of an upper-level PV streamer. The track and satellite depiction of Otto during its entire life cycle is shown in Fig. 2. Otto began as a westward-moving easterly wave disturbance that is first trackable in the eastern North Atlantic basin at 0000 UTC 1 October (Table 1). Convection was located on the eastern flank of Otto as it moved westward across the North Atlantic basin in an environment characterized by westerly vertical wind shear through 5 October (Figs. 2 and 3a–c). By 0000 UTC 5 October, the pre-Otto disturbance is located in the eastern Caribbean Sea on the southern flank of an upper-level cyclonic circulation (Fig. 3c). The upper-level feature originated on the southern end of a PV streamer that was located over the subtropical North Atlantic during the first week of October (Figs. 3a–c). The PV streamer was reinforced over the subtropics by two AWB events, the first of which (A1) was underway at 0000 UTC 3 October and the second (A2) at 0000 UTC 4 October (Figs. 3a,b). The pre-Otto disturbance was located at 17.8°N, 62.3°W on the southeastern flank of the PV streamer at 0000 UTC 4 October, which corresponds to time t − 0 h for A2, as used for the composite analysis shown later (Fig. 3b). The PV streamer fractured by 0000 UTC 5 October and moved westward with the pre-Otto disturbance (Fig. 3c).

Fig. 2.

Microwave images of polarized corrected temperature (shaded according to the color bar; K) from the Tropical Rainfall Measuring Mission (TRMM) and the Special Sensor Microwave Imager (SSM/I) superposed on visible geostationary satellite images for selected times during the life cycle of TC Otto (2010). The satellite imagery is overlaid on the storm track, comprised of positions from NOAA/NWS/NHC Automated Tropical Cyclone Forecast (ATCF) fix file (f-decks) observations (dashed line) and from HURDAT (solid line). The red circles denote the position of TC Otto at the synoptic time closest to the time of the satellite image.

Fig. 2.

Microwave images of polarized corrected temperature (shaded according to the color bar; K) from the Tropical Rainfall Measuring Mission (TRMM) and the Special Sensor Microwave Imager (SSM/I) superposed on visible geostationary satellite images for selected times during the life cycle of TC Otto (2010). The satellite imagery is overlaid on the storm track, comprised of positions from NOAA/NWS/NHC Automated Tropical Cyclone Forecast (ATCF) fix file (f-decks) observations (dashed line) and from HURDAT (solid line). The red circles denote the position of TC Otto at the synoptic time closest to the time of the satellite image.

Table 1.

HURDAT (Landsea et al. 2004) and NOAA/NWS/NHC ATCF fix file (f-decks) observations for TC Otto (Oct 2010). All times are UTC. Abbreviations for the state of the system are low (LO), extratropical (EX), subtropical depression (SD), subtropical storm (ST), tropical storm (TS), and hurricane (HU). Position information prior to 0600 UTC 6 Oct 2010 was obtained from the f-decks.

HURDAT (Landsea et al. 2004) and NOAA/NWS/NHC ATCF fix file (f-decks) observations for TC Otto (Oct 2010). All times are UTC. Abbreviations for the state of the system are low (LO), extratropical (EX), subtropical depression (SD), subtropical storm (ST), tropical storm (TS), and hurricane (HU). Position information prior to 0600 UTC 6 Oct 2010 was obtained from the f-decks.
HURDAT (Landsea et al. 2004) and NOAA/NWS/NHC ATCF fix file (f-decks) observations for TC Otto (Oct 2010). All times are UTC. Abbreviations for the state of the system are low (LO), extratropical (EX), subtropical depression (SD), subtropical storm (ST), tropical storm (TS), and hurricane (HU). Position information prior to 0600 UTC 6 Oct 2010 was obtained from the f-decks.
Fig. 3.

Sea level pressure (white contours every 2 hPa), 600–400-hPa layer-mean ascent (red contours every 0.3 Pa s−1 starting at −0.3 Pa s−1), 850–200-hPa vertical wind shear (vectors; m s−1), PV in the 300–200-hPa layer (black contours every 0.5 PVU starting at 0.5 PVU), and surface latent heat flux (shaded; W m−2) at 0000 UTC (a) 3, (b) 4, (c) 5, (d) 6, (e) 7, and (f) 8 Oct 2010. The positions of TC Otto were obtained from the NOAA/NWS/NHC ATCF fix file (f-decks) observations and HURDAT and are indicated by the white-filled red circles. The centers of the two AWB events are marked A1 and A2. The axis and southern end of the PV streamer are marked by the dashed magenta line and ×, respectively.

Fig. 3.

Sea level pressure (white contours every 2 hPa), 600–400-hPa layer-mean ascent (red contours every 0.3 Pa s−1 starting at −0.3 Pa s−1), 850–200-hPa vertical wind shear (vectors; m s−1), PV in the 300–200-hPa layer (black contours every 0.5 PVU starting at 0.5 PVU), and surface latent heat flux (shaded; W m−2) at 0000 UTC (a) 3, (b) 4, (c) 5, (d) 6, (e) 7, and (f) 8 Oct 2010. The positions of TC Otto were obtained from the NOAA/NWS/NHC ATCF fix file (f-decks) observations and HURDAT and are indicated by the white-filled red circles. The centers of the two AWB events are marked A1 and A2. The axis and southern end of the PV streamer are marked by the dashed magenta line and ×, respectively.

The distribution of convection relative to the center of Otto markedly evolved on 5–6 October. On 5 October, dry air west of Otto wrapped cyclonically around the western and southern flank, while moist air and convection wrapped cyclonically around the northern side (Figs. 2,,3c,,d, and 4a,b). The upper-level PV anomaly began to weaken by 0000 UTC 6 October because of increased straining associated with the approach of a strong midlatitude trough from the northwest and the building anticyclonic outflow north of the developing Otto (Fig. 3d). Convection moved to the northwestern (upshear) flank of Otto on 6 October (Fig. 2), and Otto became more circular (Fig. 4c) with the development of a deep warm-core structure, as shown by the anticyclonically circulating 850–200-hPa thermal wind vectors by 0000 UTC 7 October (Fig. 3e). The National Hurricane Center (NHC) designated Otto as a subtropical storm at 1200 UTC 6 October (Table 1). On 7–8 October, the upper-level PV anomaly dissipated, and Otto developed into a category 1 hurricane on the Saffir–Simpson scale (Simpson 1974) as it moved northeastward (Table 1) ahead of the midlatitude trough over the eastern United States (Figs. 2 and 3e–f).

Fig. 4.

The 700–500-hPa layer-mean relative humidity (shaded according to the color bar; %), 900-hPa relative vorticity (black contours every 2.0 × 10−5 s−1), and 700–500-hPa layer-mean wind (arrows; m s−1) at 0000 UTC (a) 5, (b) 6, and (c) 7 Oct 2010. The white-filled red circle marks the HURDAT position of Otto.

Fig. 4.

The 700–500-hPa layer-mean relative humidity (shaded according to the color bar; %), 900-hPa relative vorticity (black contours every 2.0 × 10−5 s−1), and 700–500-hPa layer-mean wind (arrows; m s−1) at 0000 UTC (a) 5, (b) 6, and (c) 7 Oct 2010. The white-filled red circle marks the HURDAT position of Otto.

The formation of Otto is an example of how TC development can evolve in the presence of a PV streamer, but it raises questions about the processes involved. Did the PV streamer play an integral role in TC development? What processes govern the evolution of the PV streamer that interacted with Otto? Is the upshear propagation of convection linked to the enhanced surface latent heat fluxes north of Otto (Figs. 3c–e)? Climatology and composite analyses will allow us to begin to address these questions in a general sense.

4. Climatology for the 2004–08 TC seasons

In this section, we provide a preliminary climatological overview of AWB and TC development during the 2004–08 TC seasons (defined here as June–November) to provide context for the types of TC developments studied herein. Figure 5 shows the geographical distribution of AWB events, where the AWB is defined at the location of the center of anticyclonic circulation on the DT at time t − 0 h (“−p” in Fig. 1). The climatology shows that AWB events occur across the North Atlantic basin north of 35°N. Over the western North Atlantic (defined as west of 45°W), 12 AWB events associated with TC development were identified (Fig. 5a). These cases are summarized in Table 2. Note that most of the TCs remained relatively weak, reaching tropical storm or category 1 intensity, except for Rita (2005) and Gordon (2006), both of which reached at least category 3 intensity. Of the 12 developing TCs, 6 made landfall over the U.S. mainland, making this formation category particularly relevant from a socioeconomic perspective.

Fig. 5.

Time t − 0 h locations (black dots) of AWB events associated with (a) a developing TC, (b) a trackable nondeveloping 900-hPa cyclonic vorticity center, and (c) no trackable 900-hPa cyclonic vorticity center during June–November 2004–08. The locations indicated mark the center of anticyclonic circulation on the DT, and the 900-hPa cyclonic vorticity centers are required to be within one Rossby radius of the PV streamer on the leading flank of the anticyclone on the DT.

Fig. 5.

Time t − 0 h locations (black dots) of AWB events associated with (a) a developing TC, (b) a trackable nondeveloping 900-hPa cyclonic vorticity center, and (c) no trackable 900-hPa cyclonic vorticity center during June–November 2004–08. The locations indicated mark the center of anticyclonic circulation on the DT, and the 900-hPa cyclonic vorticity centers are required to be within one Rossby radius of the PV streamer on the leading flank of the anticyclone on the DT.

Table 2.

List of North Atlantic tropical cyclones that developed in conjunction with AWB during June–November 2004–08. The genesis date and maximum intensity are indicated. The two rightmost columns indicate whether the developing TC made landfall in the United States and was initially classified as a subtropical storm. These data were obtained from HURDAT. The statistics for Ivan follow its redevelopment as a tropical disturbance after reemerging in the Gulf of Mexico on 21 Sep 2004.

List of North Atlantic tropical cyclones that developed in conjunction with AWB during June–November 2004–08. The genesis date and maximum intensity are indicated. The two rightmost columns indicate whether the developing TC made landfall in the United States and was initially classified as a subtropical storm. These data were obtained from HURDAT. The statistics for Ivan follow its redevelopment as a tropical disturbance after reemerging in the Gulf of Mexico on 21 Sep 2004.
List of North Atlantic tropical cyclones that developed in conjunction with AWB during June–November 2004–08. The genesis date and maximum intensity are indicated. The two rightmost columns indicate whether the developing TC made landfall in the United States and was initially classified as a subtropical storm. These data were obtained from HURDAT. The statistics for Ivan follow its redevelopment as a tropical disturbance after reemerging in the Gulf of Mexico on 21 Sep 2004.

Throughout the North Atlantic basin, 44 AWB events associated with a trackable nondeveloping tropical disturbance were identified, with 19 and 25 events identified over the western and eastern regions, respectively (Fig. 5b). Anticyclonic wave breaking events not associated with any trackable low-level disturbance also occur across the basin, with 123 cases identified between 2004 and 2008 (Fig. 5c). The geographical distribution of AWB events during the TC season bears similarity with the summertime PV streamer climatology constructed by Wernli and Sprenger (2007). They show that PV streamers on the 340-K potential temperature surface occur preferentially over the eastern North Atlantic, while PV streamers on the 350-K potential temperature surface occur preferentially over the western North Atlantic (see their Fig. 6).

To determine the isentropic level at which the AWB signature was strongest, the area-averaged (10° × 10° box) relative vorticity centered on the anticyclonic circulation on the DT at time t − 0 h was computed. The level with minimum vorticity was then selected as the isentropic level for the AWB event, resulting in the distribution shown in Fig. 6. The distribution of isentropic levels for western and eastern North Atlantic AWB events is statistically different, with western North Atlantic events having a normal distribution and eastern North Atlantic events having a negatively skewed distribution with a reduced frequency of higher potential temperature values (Fig. 6a and Table 3). When comparing AWB events that are accompanied by a trackable 900-hPa disturbance, AWB events associated with developing TCs are statistically different from AWB events with nondeveloping disturbances (Fig. 6b and Table 3). These results are consistent with Wernli and Sprenger (2007) and suggest that western North Atlantic AWB events occur preferentially at higher isentropic levels compared to eastern North Atlantic AWB events. From the relative distributions of developing and nondeveloping disturbances, AWB events that occur at lower isentropic levels appear to be more detrimental for TC development.

Fig. 6.

Histogram of AWB isentropic level for (a) all North Atlantic AWB events and (b) AWB events with a trackable 900-hPa cyclonic vorticity center. Events over the western and eastern North Atlantic are shown in blue and red, respectively, while those events with developing TCs are shown in black. The isentropic level is defined as the level of minimum area-averaged (10° × 10° box) relative vorticity centered on the anticyclonic circulation of the AWB.

Fig. 6.

Histogram of AWB isentropic level for (a) all North Atlantic AWB events and (b) AWB events with a trackable 900-hPa cyclonic vorticity center. Events over the western and eastern North Atlantic are shown in blue and red, respectively, while those events with developing TCs are shown in black. The isentropic level is defined as the level of minimum area-averaged (10° × 10° box) relative vorticity centered on the anticyclonic circulation of the AWB.

Table 3.

Average and median isentropic levels (K) for AWB events. The isentropic level is defined as the level of minimum area-averaged relative vorticity centered on the anticyclonic circulation of the AWB. The distribution of isentropic levels is shown in Fig. 6.

Average and median isentropic levels (K) for AWB events. The isentropic level is defined as the level of minimum area-averaged relative vorticity centered on the anticyclonic circulation of the AWB. The distribution of isentropic levels is shown in Fig. 6.
Average and median isentropic levels (K) for AWB events. The isentropic level is defined as the level of minimum area-averaged relative vorticity centered on the anticyclonic circulation of the AWB. The distribution of isentropic levels is shown in Fig. 6.

The intraseasonal frequency of all North Atlantic AWB events shows a maximum value in September and a minimum value in November (Fig. 7a). The temporal distribution of AWB events associated with a trackable low-level tropical disturbance is not statistically different from the overall climatology but occupies a greater relative frequency in mid- to late season. A statistical summary is shown in Table 4. The statistics indicate that 31% of all North Atlantic AWB events are accompanied by a trackable low-level disturbance, with 7% accompanied by a developing TC. Considering the western North Atlantic only, the frequency of AWB events peaks in June, with a secondary peak in September (Fig. 7c). The proportion of AWB events that are accompanied by a trackable low-level disturbance (or a developing TC) peaks in the late season. Overall, 34% of these events are accompanied by a trackable low-level disturbance, with 13% accompanied by a developing TC. In contrast to western North Atlantic events, eastern North Atlantic AWB events occur most frequently in August–November and are maximized in September (Fig. 7d). In the context of all North Atlantic TCs during 2004–08, 15% developed under the influence of AWB, with a higher relative frequency in October and November (Fig. 7b and Table 4). In all, the statistics show that, on average, over two TCs per year develop in conjunction with AWB over the North Atlantic, half of which made landfall on the U.S. coastline during 2004–08. This result demonstrates that this type of TC development is not a rare event over the North Atlantic.

Fig. 7.

Intraseasonal frequency of (a) North Atlantic AWB, (b) North Atlantic TCs, (c) western North Atlantic AWB, and (d) eastern North Atlantic AWB during June–November 2004–08. In (a),(c),(d), all events are in black, events with a trackable 900-hPa cyclonic vorticity center are in red, and events in which the trackable 900-hPa cyclonic vorticity center developed into a TC are in blue. In (b), all TCs are in black and TCs associated with AWB are in blue. Individual years are shown by the small dots.

Fig. 7.

Intraseasonal frequency of (a) North Atlantic AWB, (b) North Atlantic TCs, (c) western North Atlantic AWB, and (d) eastern North Atlantic AWB during June–November 2004–08. In (a),(c),(d), all events are in black, events with a trackable 900-hPa cyclonic vorticity center are in red, and events in which the trackable 900-hPa cyclonic vorticity center developed into a TC are in blue. In (b), all TCs are in black and TCs associated with AWB are in blue. Individual years are shown by the small dots.

Table 4.

Statistical summary of North Atlantic AWB and TC activity during June–November 2004–08. The west, east, and total columns represent the western North Atlantic, eastern North Atlantic, and entire North Atlantic, respectively. All values are percentages.

Statistical summary of North Atlantic AWB and TC activity during June–November 2004–08. The west, east, and total columns represent the western North Atlantic, eastern North Atlantic, and entire North Atlantic, respectively. All values are percentages.
Statistical summary of North Atlantic AWB and TC activity during June–November 2004–08. The west, east, and total columns represent the western North Atlantic, eastern North Atlantic, and entire North Atlantic, respectively. All values are percentages.

Figure 8 shows all of the trackable low-level tropical disturbances in latitude and SST phase space at time t − 0 h. Nondeveloping disturbances generally occur at higher latitudes and over cooler SSTs compared to developing TCs, a result that is statistically significant. Additionally, the SST for nondeveloping disturbances associated with AWB over the western North Atlantic is significantly warmer than for AWB over the eastern North Atlantic. The gray box encloses the phase-space region for developing TCs, with the lower bound of SST and upper bound of latitude defined by the Mahalanobis distance (Mahalanobis 1936) for the developing TC cases. The Mahalanobis distance is defined as the distance of individual data points from the mean, scaled by the covariance (cf. van Lier-Walqui et al. 2014), and is used here to define the maximum distance that encloses all of the developing TC data points. The gray box is based on the threshold SST and latitude values that enclose the full zone of the developing TCs, as defined by the Mahalanobis distance (i.e., the green dashed area), but also includes lower-latitude systems and those with very warm SSTs, since there is no physical rationale for excluding them from the dataset. Only the developing TCs and nondeveloping disturbances associated with AWB over the western North Atlantic that have SST ≥ 24.1°C and occur at latitudes ≤ 34.7°N will be used in the composite analysis presented section 5. It is these disturbances that are considered candidates for possible TC development. The developing TCs and nondeveloping disturbances associated with AWB over the western North Atlantic were used in the composite analyses in order to compare developing and nondeveloping disturbances that occurred within a similar geographical region in conjunction with AWB at higher isentropic surfaces. The disturbances associated with AWB over the eastern North Atlantic, none of which developed into TCs, were not included, since they appear to occur in conjunction with AWB at lower isentropic surfaces and over cooler SSTs.

Fig. 8.

Distribution of 900-hPa cyclonic vorticity centers at time t − 0 h in the latitude (°N) and SST (°C) domain. Nondeveloping vorticity centers over the eastern (western) North Atlantic are indicated by the blue (red) circles, and the developing TCs are indicated by the black circles. The SST value is the area-averaged SST over a 5.0° × 5.0° latitude–longitude box centered on the 900-hPa cyclonic vorticity center. The mean latitude–SST value for developing TCs is indicated by the green square. The Mahalanobis distance (Mahalanobis 1936) for the developing TCs is shown by the green dashed ellipse. The gray shading marks the region of phase space considered favorable for TC development, as defined by the Mahalanobis distance.

Fig. 8.

Distribution of 900-hPa cyclonic vorticity centers at time t − 0 h in the latitude (°N) and SST (°C) domain. Nondeveloping vorticity centers over the eastern (western) North Atlantic are indicated by the blue (red) circles, and the developing TCs are indicated by the black circles. The SST value is the area-averaged SST over a 5.0° × 5.0° latitude–longitude box centered on the 900-hPa cyclonic vorticity center. The mean latitude–SST value for developing TCs is indicated by the green square. The Mahalanobis distance (Mahalanobis 1936) for the developing TCs is shown by the green dashed ellipse. The gray shading marks the region of phase space considered favorable for TC development, as defined by the Mahalanobis distance.

5. Composite analysis

This section presents a statistical and spatial disturbance-relative composite analysis of several quantities to highlight differences between developing and nondeveloping tropical disturbances that interact with a PV streamer. Over the western North Atlantic, 12 developing and 14 nondeveloping tropical disturbances were identified, as described at the end of section 4 (see also Fig. 8). Although the tropical disturbances are composited at times relative to AWB life cycle, the time of genesis for the developing tropical disturbances was defined as the first occurrence of tropical or subtropical storm status in the HURDAT and occurred, on average, 15 h after time t − 0 h. Thus, the period from time t − 72 through t + 24 h serves as the development, or genesis, phase of the tropical disturbances analyzed herein.

a. Structural evolution

The composite structure and evolution of the upper-level PV streamer and low-level tropical disturbance for the developing and nondeveloping cases are shown in Figs. 9 and 10, respectively. For both developing and nondeveloping cases, the PV streamer moves southeastward to near the incipient tropical disturbance by t − 24 h (Figs. 9a–c and 10a–c). Convection, as represented by CFSR mean ascent in the 700–500-hPa layer, develops on the northeastern flank of the low-level disturbance in both composites. The surface anticyclone (labeled “H”) west of the PV streamer axis is stronger for the developing cases, with an increased sea level pressure gradient and attendant easterly geostrophic flow north of the incipient tropical disturbance. From t − 24 to t − 0 h, the PV streamer begins to thin in response to enhanced diabatically driven divergent outflow north and west of the developing TC (Figs. 9d,e). The divergent outflow impedes the southeastward progression of the PV streamer through negative PV advection by the divergent wind and is enhanced as convection moves westward to the northwestern side of the low-level disturbance. The nondeveloping composite is marked by reduced diabatic outflow interacting with the PV streamer as convection remained on the northeastern flank of the low-level disturbance (Figs. 10d,e). As a consequence, the PV streamer continues southeastward to over the low-level disturbance by t + 24 h (Fig. 10f).

Fig. 9.

Disturbance-relative composite-mean PV in the 250–150-hPa layer (black contours every 0.5 PVU starting at 1.0 PVU), sea level pressure (dashed blue contours every 2 hPa), 200-hPa divergent wind (arrows; m s−1), 700–500-hPa layer-mean ascent (solid red contours every 0.1 Pa s−1 starting at −0.2 Pa s−1), and surface latent heat flux (shaded according to the color bar; W m−2) for the developing cases at time (a) t − 72, (b) t − 48, (c) t − 24, (d) t − 12, (e) t − 0, and (f) t + 24 h. The surface anticyclone is marked “H,” and the PV streamer axis is marked with a dashed red line. In (f), the boxes cover regions where area-averaged quantities were computed in Fig. 22. Box 1 marks the surface anticyclone region, box 2 marks the area north of the vortex in which the surface latent heat fluxes were computed, and box 3 marks the mean sea level pressure of the disturbance.

Fig. 9.

Disturbance-relative composite-mean PV in the 250–150-hPa layer (black contours every 0.5 PVU starting at 1.0 PVU), sea level pressure (dashed blue contours every 2 hPa), 200-hPa divergent wind (arrows; m s−1), 700–500-hPa layer-mean ascent (solid red contours every 0.1 Pa s−1 starting at −0.2 Pa s−1), and surface latent heat flux (shaded according to the color bar; W m−2) for the developing cases at time (a) t − 72, (b) t − 48, (c) t − 24, (d) t − 12, (e) t − 0, and (f) t + 24 h. The surface anticyclone is marked “H,” and the PV streamer axis is marked with a dashed red line. In (f), the boxes cover regions where area-averaged quantities were computed in Fig. 22. Box 1 marks the surface anticyclone region, box 2 marks the area north of the vortex in which the surface latent heat fluxes were computed, and box 3 marks the mean sea level pressure of the disturbance.

Fig. 10.

As in Fig. 9, but for the nondeveloping cases.

Fig. 10.

As in Fig. 9, but for the nondeveloping cases.

The disturbance-relative position of PV streamers for the developing and nondeveloping cases is summarized in Fig. 11. The developing cases are located, on average, 750 km northeast of the southern end of the PV streamer (Figs. 11a,c). The nondeveloping cases are located, on average, 1100 km northeast of the southern end of the PV streamer, a distance that is significantly larger than for the developing cases (Figs. 11b,c), and interact with wider PV streamers on average (Fig. 11d). Despite the differences in PV streamer structure and evolution, the developing and nondeveloping disturbances are embedded in comparable westerly vertical wind shear, as shown in a contour frequency by altitude diagram (CFAD; Yuter and Houze 1995) of zonal wind at t − 0 h (Fig. 12). The developing cases have slightly stronger vertical wind shear and are skewed toward easterly flow (Fig. 12a). The nondeveloping cases are skewed toward westerly flow (Fig. 12b). The importance of deeper and stronger easterly flow and its linkage to the anticyclone north of the low-level disturbance, enhanced surface fluxes, and upshear movement of convection will be discussed in more detail in the next subsection.

Fig. 11.

(top) Disturbance-relative positions of the axis of PV streamers at t − 0 h for (a) developing and (b) nondeveloping cases. The filled circles mark the southern end of the PV streamer. (bottom) Box-and-whisker diagrams of (c) disturbance distance (km) from the southern end of the PV streamer and (d) width (km) of the PV streamer at t − 0 h. The lower and upper bounds of the box mark the 25th and 75th percentiles, respectively. The black line within the box represents the median and the red diamond represents the mean. The whiskers mark the maximum and minimum values. The width of the PV streamer is defined as in Fig. 1 in Wernli and Sprenger (2007), except that the distance is bounded by the 1.0-PVU contour in the 250–150-hPa layer.

Fig. 11.

(top) Disturbance-relative positions of the axis of PV streamers at t − 0 h for (a) developing and (b) nondeveloping cases. The filled circles mark the southern end of the PV streamer. (bottom) Box-and-whisker diagrams of (c) disturbance distance (km) from the southern end of the PV streamer and (d) width (km) of the PV streamer at t − 0 h. The lower and upper bounds of the box mark the 25th and 75th percentiles, respectively. The black line within the box represents the median and the red diamond represents the mean. The whiskers mark the maximum and minimum values. The width of the PV streamer is defined as in Fig. 1 in Wernli and Sprenger (2007), except that the distance is bounded by the 1.0-PVU contour in the 250–150-hPa layer.

Fig. 12.

CFAD plot (shaded according to the color bar; %) and composite-mean zonal environment wind (m s−1; developing TC solid; nondeveloping disturbance dashed) for (a) developing and (b) nondeveloping cases at time t − 0 h. The bin size is 1.0 m s−1. The environment wind was computed by taking the area average over a 400–1000-km annulus centered on the 900-hPa cyclonic vorticity center. The mean profiles are plotted on both panels, and the zero line is plotted in magenta.

Fig. 12.

CFAD plot (shaded according to the color bar; %) and composite-mean zonal environment wind (m s−1; developing TC solid; nondeveloping disturbance dashed) for (a) developing and (b) nondeveloping cases at time t − 0 h. The bin size is 1.0 m s−1. The environment wind was computed by taking the area average over a 400–1000-km annulus centered on the 900-hPa cyclonic vorticity center. The mean profiles are plotted on both panels, and the zero line is plotted in magenta.

The composite vertical structure of PV is shown from t − 48 to t − 0 h in Fig. 13. The south–north cross sections at t − 48 h show the presence of a westerly upper-level wind maximum above the low-level disturbance for both the developing and nondeveloping cases (Figs. 13a,d). The westerly wind maximum is associated with southwesterly flow ahead of the PV streamer. The developing cases have deeper easterly flow north of the low-level disturbance, and both composites have isentropes that slope upward from south to north, indicative of a baroclinic environment. The upper-level westerly wind maximum moves south of the strengthening PV tower in the developing cases by t − 24 h, indicative of the southwestward disturbance-relative movement of the southern end of the PV streamer (Fig. 9c). By t − 0 h, the PV tower for both the developing and nondeveloping cases has intensified, but the developing cases have a well-defined PV tower that extends to 400 hPa and have a better-defined warm-core structure in the 700–300-hPa layer (Figs. 13c,f and 14a). The nondeveloping cases have a significantly lower 700–300-hPa thickness anomaly at t − 0 h because of the continued presence of the PV streamer aloft (Fig. 13f). While the developing cases are marked by a stronger low-level disturbance throughout the period from t − 72 to t + 24 h, the difference in 850–700-hPa layer-mean relative vorticity becomes significant at t − 0 h (Fig. 14b). The intensification of the developing cases occurs as convection wraps cyclonically to the northwestern (upshear) side of the low-level disturbance and divergent outflow impedes the southeastward movement of the PV streamer (cf. Fig. 9). The differences in PV streamer–divergent outflow interaction shown in Figs. 9 and 10 are consistent with Davis and Bosart (2004, their Fig. 3), who emphasized the importance of convectively generated divergent outflow in reducing positive PV aloft over the low-level disturbance during TT.

Fig. 13.

Disturbance-relative composite-mean south–north vertical cross section of PV (shaded according to the grayscale; PVU), potential temperature (red contours every 5 K), and zonal wind (black contours every 5.0 m s−1; easterly dashed) for developing cases at time (a) t − 48, (b) t − 24, and (c) t − 0 h, and for nondeveloping disturbances at (d) t − 48, (e) t − 24, and (f) t − 0 h.

Fig. 13.

Disturbance-relative composite-mean south–north vertical cross section of PV (shaded according to the grayscale; PVU), potential temperature (red contours every 5 K), and zonal wind (black contours every 5.0 m s−1; easterly dashed) for developing cases at time (a) t − 48, (b) t − 24, and (c) t − 0 h, and for nondeveloping disturbances at (d) t − 48, (e) t − 24, and (f) t − 0 h.

Fig. 14.

Quasi-Lagrangian time series of area-averaged (a) 700–300-hPa thickness anomaly (m) and (b) 850–700-hPa layer-averaged relative vorticity (×10−5 s−1) from time t − 72 to t + 24 h. The developing TCs are shown in blue and nondeveloping disturbances are shown in red. The composite-mean value is shown with error bars representing the 90% confidence interval. Statistical significance is indicated by a black dot. The thickness anomaly is computed as the area average at radius ≤ 200 km minus the area average over the 200–1000-km radial band.

Fig. 14.

Quasi-Lagrangian time series of area-averaged (a) 700–300-hPa thickness anomaly (m) and (b) 850–700-hPa layer-averaged relative vorticity (×10−5 s−1) from time t − 72 to t + 24 h. The developing TCs are shown in blue and nondeveloping disturbances are shown in red. The composite-mean value is shown with error bars representing the 90% confidence interval. Statistical significance is indicated by a black dot. The thickness anomaly is computed as the area average at radius ≤ 200 km minus the area average over the 200–1000-km radial band.

Figure 15 shows zonal vertical cross sections of relative humidity, relative vorticity, and QG ascent from t − 48 to t − 0 h. The relative vorticity structure of the low-level disturbance is stronger for the developing cases throughout (see also Fig. 14b). The QG vertical motion in the middle and upper troposphere is similar for both developing and nondeveloping cases over the low-level disturbance but is stronger for the nondeveloping cases east of the low-level disturbance. The vertical axis of enhanced relative humidity is collocated with the QG ascent maximum for the nondeveloping cases from t − 48 to t − 0 h (Figs. 15d–f). While the developing cases show a similar relative humidity pattern at t − 48 and t − 24 h, the moist axis shifts from the QG ascent maximum to the cyclonic vorticity tower of the developing cases by t − 0 h (Figs. 15a–c). This shift in the moist axis is indicative of the transition of a baroclinic system dominated by QG processes to a convective disturbance driven by diabatic processes and occurs as the convection moves to the northwestern side of the incipient disturbance (cf. Fig. 9e).

Fig. 15.

Disturbance-relative composite-mean west–east vertical cross section of relative humidity (solid red contours every 10%, starting at 50%), relative vorticity (black contours every 2.0 × 10−5 s−1; zero contour omitted; positive solid; negative dashed), and QG vertical motion (shaded according to the color bar; ×10−1 Pa s−1) for developing cases at time (a) t − 48, (b) t − 24, and (c) t − 0 h and for nondeveloping disturbances at (d) t − 48, (e) t − 24, and (f) t − 0 h. The vertical axis of maximum relative humidity and cyclonic relative vorticity are indicated by a dashed magenta and a dashed black line, respectively.

Fig. 15.

Disturbance-relative composite-mean west–east vertical cross section of relative humidity (solid red contours every 10%, starting at 50%), relative vorticity (black contours every 2.0 × 10−5 s−1; zero contour omitted; positive solid; negative dashed), and QG vertical motion (shaded according to the color bar; ×10−1 Pa s−1) for developing cases at time (a) t − 48, (b) t − 24, and (c) t − 0 h and for nondeveloping disturbances at (d) t − 48, (e) t − 24, and (f) t − 0 h. The vertical axis of maximum relative humidity and cyclonic relative vorticity are indicated by a dashed magenta and a dashed black line, respectively.

In addition to the PV streamer evolution that is more favorable for development, the developing cases have a stronger and better-defined warm-core vorticity tower throughout the development period, suggesting that stronger incipient tropical disturbances are better candidates to survive interaction with a PV streamer and undergo genesis, a result consistent with previous studies (e.g., DeMaria et al. 2001; Davis et al. 2008; Dunkerton et al. 2009). Additionally, the developing cases are characterized by a stronger anticyclone and deeper easterly flow north of the incipient low-level disturbance, suggesting that while the PV streamer itself may play a secondary role in development, the structure of the AWB pattern may be the key synoptic-scale precursor to TC development. The next section will examine the thermodynamic evolution differences between the developing and nondeveloping cases and how these distinctions link to the documented differences in synoptic-scale flow.

b. Thermodynamics and role of synoptic-scale flow in moisture evolution

The CFAD is utilized in this section to examine the vertical structure of distributions of thermodynamic variables for the developing and nondeveloping cases. The profile of virtual temperature anomaly4 shows differences in evolution for the developing and nondeveloping cases from t − 48 to t + 24 h (Fig. 16). The developing TC profile shows a warm anomaly from t − 48 to t + 24 h (Figs. 16a–d) throughout the troposphere, particularly in the 500–300-hPa layer, where the warm anomaly reaches +1–2 K by t − 24 h (Fig. 16c). The virtual temperature anomaly is negatively skewed in the 500–300-hPa layer at t − 24 and t + 24 h, which reduces the mean value. The profile for the nondeveloping cases shows a positively skewed neutral thermal profile, except for a +0.5-K warm anomaly in the 500–300-hPa layer at t − 0 h (Figs. 16e–f). By t + 24 h, the thermal profile for the nondeveloping cases is positively skewed and neutral again, as the cold upper-level trough is aligned above the low-level disturbance (see also Fig. 14a). In all, the largest differences in the virtual temperature profile between the developing and nondeveloping cases reside in the middle and upper levels, where a warm anomaly is present for the developing cases. The virtual temperature profile for the developing cases is consistent with a positive warm-core PV anomaly maximized in the midtroposphere, as shown in the PV cross sections (Fig. 13c), while the profile for the nondeveloping cases is consistent with a weaker low-level PV anomaly and a stronger upper-level cold-core PV anomaly—a signature of the PV streamer aloft (see also Figs. 10e and 14a).

Fig. 16.

Virtual temperature anomaly CFAD plots (shaded according to the color bar; %) and composite-mean (K; developing TC solid; nondeveloping disturbance dashed) for developing TCs at (a) t − 48, (b) t − 24, (c) t − 0, and (d) t + 24 h and nondeveloping disturbances at (e) t − 48, (f) t − 24, (g) t − 0, and (h) t + 24 h. The anomaly is computed as the area average at radius ≤ 200 km minus the area average over the 200–1000-km radial band. The bin size is 0.5 K. The mean profiles are plotted on all panels, and the zero line is plotted in magenta.

Fig. 16.

Virtual temperature anomaly CFAD plots (shaded according to the color bar; %) and composite-mean (K; developing TC solid; nondeveloping disturbance dashed) for developing TCs at (a) t − 48, (b) t − 24, (c) t − 0, and (d) t + 24 h and nondeveloping disturbances at (e) t − 48, (f) t − 24, (g) t − 0, and (h) t + 24 h. The anomaly is computed as the area average at radius ≤ 200 km minus the area average over the 200–1000-km radial band. The bin size is 0.5 K. The mean profiles are plotted on all panels, and the zero line is plotted in magenta.

Composite profiles of MSE anomaly show a positive anomaly in the lower troposphere for both developing and nondeveloping cases at t − 48 h (Figs. 17a,e). An abrupt increase and narrowing of the distribution occurs in the 700–500-hPa layer for developing cases at t − 24 h (Fig. 17b). Positive MSE anomalies remain larger for developing cases compared to nondeveloping cases through t + 24 h (Figs. 17c,d,g,h). The increase in the positive MSE anomaly centered on 700 hPa is due to an increase in water vapor mixing ratios close to the disturbance center (radius ≤ 200 km) for the developing cases and leads to a larger vertical gradient of the perturbation MSE in the 900–700-hPa layer for the developing cases from t − 24 through t + 24 h (Figs. 17b–d,f–h).

Fig. 17.

As in Fig. 16, except showing moist static energy anomaly (kJ kg−1). The bin size is 0.5 kJ kg−1.

Fig. 17.

As in Fig. 16, except showing moist static energy anomaly (kJ kg−1). The bin size is 0.5 kJ kg−1.

Disturbance-relative composites of 800-hPa geopotential height and 900–700-hPa layer-mean water vapor mixing ratio and IVT for developing and nondeveloping cases are shown in Figs. 18 and 19, respectively. The developing cases show a marked increase in water vapor mixing ratio near the disturbance center, exceeding 11 g kg−1 by t − 0 h (Fig. 18e), which is consistent with the increase in MSE in the lower and midtroposphere (see also Fig. 17). Conversely, the nondeveloping cases are characterized by water vapor mixing ratios that remain near 10 g kg−1 near the disturbance center and are slightly asymmetric with the center of the moisture plume located downshear (Fig. 19). The 700–400-hPa layer-mean relative humidity at t − 0 h shows a similar pattern as the water vapor mixing ratio (Fig. 20). Both the developing and nondeveloping cases are characterized by a dry environment north and west of the low-level disturbance in conjunction with the PV streamer. The developing cases have a larger closed disturbance-relative circulation, or pouch region (Dunkerton et al. 2009) and have higher relative humidity values, particularly on the northwestern side of the disturbance (Fig. 20a). The nondeveloping cases have a more asymmetric distribution of moisture near the disturbance center, with much lower relative humidity values on the northwestern side (Fig. 20b).

Fig. 18.

Disturbance-relative composite-mean 800-hPa geopotential height (solid contours every 10 m), 900–700-hPa layer-mean water vapor mixing ratio (dashed contours every 1 g kg−1), and 900–700-hPa IVT (arrows with magnitude shaded according to the color bar; kg m−1 s−1) for the developing cases at time (a) t − 72, (b) t − 48, (c) t − 24, (d) t − 12, (e) t − 0, and (f) t + 24 h.

Fig. 18.

Disturbance-relative composite-mean 800-hPa geopotential height (solid contours every 10 m), 900–700-hPa layer-mean water vapor mixing ratio (dashed contours every 1 g kg−1), and 900–700-hPa IVT (arrows with magnitude shaded according to the color bar; kg m−1 s−1) for the developing cases at time (a) t − 72, (b) t − 48, (c) t − 24, (d) t − 12, (e) t − 0, and (f) t + 24 h.

Fig. 19.

As in Fig. 18, but for the nondeveloping cases.

Fig. 19.

As in Fig. 18, but for the nondeveloping cases.

Fig. 20.

Disturbance-relative composite-mean 700–400-hPa layer-mean relative humidity (shaded according to the color bar; %), relative vorticity (red contours every 2.0 × 10−5 s−1 starting at 2.0 × 10−5 s−1), and disturbance-relative streamlines (black lines with arrow heads) at t − 0 h for (a) developing and (b) nondeveloping cases. The boxes mark the regions used for averaging relative humidity and vertical velocity shown in Fig. 21.

Fig. 20.

Disturbance-relative composite-mean 700–400-hPa layer-mean relative humidity (shaded according to the color bar; %), relative vorticity (red contours every 2.0 × 10−5 s−1 starting at 2.0 × 10−5 s−1), and disturbance-relative streamlines (black lines with arrow heads) at t − 0 h for (a) developing and (b) nondeveloping cases. The boxes mark the regions used for averaging relative humidity and vertical velocity shown in Fig. 21.

The evolution of relative humidity and CFSR vertical velocity is shown in Fig. 21. The time series show that 700–500-hPa layer-mean relative humidity on the northwestern side of the low-level disturbance increases for the developing cases through t − 0 h, becoming significantly larger than the nondeveloping cases at t − 0 and t + 24 h (Fig. 21a). Likewise, the 900–700-hPa layer-mean relative humidity increases through t − 0 h, becoming significantly larger than the nondeveloping cases at t − 12 h (Fig. 21a). Conversely, 700–500- and 900–700-hPa layer-mean relative humidity on the northeastern side of the low-level disturbance increases for both the developing and nondeveloping cases through t − 0 h, with no significant differences between the two composite classes (Fig. 21b). This result is consistent with the spatial distribution of relative humidity shown in Fig. 20, where both developing and nondeveloping cases are relatively moist on the northeastern flank (downshear side) of the low-level disturbance, but the moisture wraps cyclonically westward to the northwestern side for the developing cases. A similar evolution is shown in the time series of 700–500-hPa layer-mean vertical velocity, which shows how ascent becomes significantly stronger on the northwestern side of the developing cases by t − 0 h (Fig. 21c) and is not significantly different on the northeastern side of the low-level disturbance until t + 24 h (Fig. 21d). In all, the increase in moisture and convection on the northwestern side of the developing cases indicates that TC development is occurring, as convection on the upshear side of the low-level disturbance can readily weaken and impede the southeastward movement of the PV streamer aloft, as previously discussed.

Fig. 21.

Time series of area-averaged 900–700- and 700–500-hPa layer-mean relative humidity (%) (a) northwest and (b) northeast of the low-level disturbance, and 700–500-hPa vertical velocity (Pa s−1) (c) northwest and (d) northeast of the low-level disturbance from time t − 72 to t + 24 h. The regions used for area averaging are indicated in Fig. 20. The composite-mean value is shown with error bars representing the 90% confidence interval. Statistical significance is indicated by a black dot. The developing TCs are shown in blue and nondeveloping disturbances are shown in red.

Fig. 21.

Time series of area-averaged 900–700- and 700–500-hPa layer-mean relative humidity (%) (a) northwest and (b) northeast of the low-level disturbance, and 700–500-hPa vertical velocity (Pa s−1) (c) northwest and (d) northeast of the low-level disturbance from time t − 72 to t + 24 h. The regions used for area averaging are indicated in Fig. 20. The composite-mean value is shown with error bars representing the 90% confidence interval. Statistical significance is indicated by a black dot. The developing TCs are shown in blue and nondeveloping disturbances are shown in red.

The increase in relative humidity on the northwestern side of the developing cases appears to be driven by two processes, both linked to the structure and strength of the anticyclone north of the low-level disturbance. First, well-defined westward IVT in the 900–700-hPa layer is apparent for the developing cases, becoming established as the anticyclone develops and strengthens north of the low-level disturbance (Fig. 18). Westward IVT increases through t − 0 h, with well-defined moisture flux convergence located on the northwestern flank of the low-level disturbance (Fig. 18e). For the nondeveloping cases, IVT is weak on the northern side of the low-level vortex in response to the absence of the anticyclone to the north (Fig. 19). Well-defined northward IVT develops on the eastern side of the nondeveloping cases by t − 24 h as the anticyclone becomes established to the east, placing the moisture flux convergence on the northeastern side of the vortex during t − 24 to t + 24 h (Figs. 19c–f). Second, the strength of the near-surface anticyclone for the developing cases contributes to enhanced surface latent heat fluxes on the northern side of the low-level disturbance, becoming significantly larger than the nondeveloping cases by t − 24 h (Fig. 22a; see also Figs. 9 and 10). The enhanced low-level geostrophic easterlies and attendant latent heat fluxes are associated with a sea level pressure gradient that is significantly stronger for the developing cases from t − 24 to t + 24 h (Fig. 22b), associated with a significantly stronger surface anticyclone north of the low-level disturbance prior to t − 0 h (Fig. 22c) and a significantly lower sea level pressure minimum associated with the developing disturbance after t − 0 h (Fig. 22d).

Fig. 22.

Time series of area-averaged (a) surface latent heat flux (W m−2), (b) meridional sea level pressure gradient [hPa (100 km)−1], (c) sea level pressure near anticyclone (hPa), and (d) sea level pressure near incipient tropical disturbance (hPa) from time t − 72 to t + 24 h. The composite-mean value is shown with error bars representing the 90% confidence interval. Statistical significance is indicated by a black dot. The regions used for area averaging are indicated in Fig. 9f. Box 1 marks the surface anticyclone region, box 2 marks the area north of the vortex in which the surface latent heat fluxes were computed, and box 3 marks the mean sea level pressure of the disturbance.

Fig. 22.

Time series of area-averaged (a) surface latent heat flux (W m−2), (b) meridional sea level pressure gradient [hPa (100 km)−1], (c) sea level pressure near anticyclone (hPa), and (d) sea level pressure near incipient tropical disturbance (hPa) from time t − 72 to t + 24 h. The composite-mean value is shown with error bars representing the 90% confidence interval. Statistical significance is indicated by a black dot. The regions used for area averaging are indicated in Fig. 9f. Box 1 marks the surface anticyclone region, box 2 marks the area north of the vortex in which the surface latent heat fluxes were computed, and box 3 marks the mean sea level pressure of the disturbance.

6. Conclusions

Tropical cyclone development near upper-level PV streamers is examined from synoptic climatology, composite, and case study perspectives for the 2004–08 North Atlantic seasons. The development of TCs near PV streamers is a specific type of baroclinically influenced development. These PV streamers are driven into the subtropical and tropical latitudes by straining on the leading edge of AWB events (Martius et al. 2008). The climatology shows that for 2004–08, 15% of all TC development occurred in conjunction with PV streamer interaction during AWB. The frequency and spatial distribution of AWB is consistent with previous AWB and PV streamer climatologies for the North Atlantic (e.g., Wernli and Sprenger 2007; Martius et al. 2008; Postel and Hitchman 1999). Incipient tropical disturbances over the western North Atlantic that were located within one Rossby radius of a PV streamer were considered candidate disturbances for TC development. Disturbances in the eastern North Atlantic were not included in the analysis, since they were associated with AWB at lower isentropic surfaces over cooler SSTs, while candidate disturbances in the western North Atlantic were associated with AWB at higher isentropic surfaces. The latter appears to be more favorable for TC development. Of all AWB events over the western North Atlantic during 2004–08, 39% had a tropical disturbance within one Rossby radius of the PV streamer. Furthermore, 13% of all AWB events were associated with TC development: two to three events per year, of which half made landfall in the United States.

During the 2004–08 North Atlantic TC seasons, 12 developing and 14 nondeveloping tropical disturbances that interacted with a PV streamer were identified and included in composite analyses. The structural evolution of PV streamers during AWB was different for developing and nondeveloping cases. In both composites, an upper-level PV streamer approached the low-level incipient vortex from the northwest. In the developing cases, upper-level convectively generated divergent outflow over and northwest of the low-level tropical disturbance contributed to negative PV advection by the divergent wind on the southeastern flank of the PV streamer, halting its southeastward progression [Molinari et al. (1995); see also Archambault et al. (2013) for discussion of the impact of divergent outflow on upper-level PV gradients]. For the nondeveloping cases, the upper-level outflow and convection was much weaker and located on the northeastern flank of the low-level disturbance and, hence, had a negligible impact on the PV streamer. This allowed the PV streamer to move southeastward to over the low-level disturbance, a relative position that was unfavorable for the development of a deep warm-core structure.

The thermodynamic structure of the developing and nondeveloping cases was in general agreement with recent observational studies from the PREDICT field program (e.g., Davis and Ahijevych 2013; Smith and Montgomery 2012; Komaromi 2013; Davis and Ahijevych 2012). Findings include static stabilization below 700 hPa within 200 km of the center and moistening at midlevels during development. Moist and stable conditions in developing disturbances have been documented to favor a vertical mass flux profile characterized by maximum ascent in the low to midtroposphere, helping to increase low-level convergence and thus cyclonic vorticity by stretching (e.g., Schumacher et al. 2007; Raymond and Sessions 2007; Davis and Galarneau 2009; Gjorgjievska and Raymond 2014).

A recent study by Rappin and Nolan (2012) addressed TC genesis and vertical wind shear orientation. Using idealized numerical experiments, their results showed that TC genesis occurred more readily when the low-level wind and vertical wind shear were counteraligned (defined as low-level easterly flow in an environment with westerly vertical wind shear). In this scenario, boundary layer air was moistened in the downshear flank by convection. This convection then propagated in the upshear direction on the northern side of the disturbance as the boundary layer was moistened by enhanced surface latent heat fluxes. The propagation of convection to the upshear side helped to reduce vortex tilt and to promote more rapid development of a TC. While our study used coarser-resolution reanalysis data (and composite analysis that is inherently smoothed) that was likely unable to resolve the evolution of finescale vortex structure such as tilt, the results presented herein suggest that a similar propagation of convection in the upshear direction may be promoting TC development near PV streamers.

The developing cases were characterized by a stronger anticyclone to the north of the low-level vortex, which helped to moisten the upshear side of the disturbance by low-level westward IVT and moistening by enhanced surface latent heat fluxes. These processes were also documented in the development of TC Otto (2010) as described in section 3, are similar to the evolution of TC Diana (1984) documented by Bosart and Bartlo (1991), and contributed to an upshear propagation of convection that has been documented to be important for TC development (Rappin and Nolan 2012). The strong surface anticyclone north of the developing cases and its attendant enhanced sea level pressure gradient and surface latent heat fluxes was the key linkage between the synoptic-scale flow associated with the AWB event and the near-storm thermodynamic conditions associated with TC development. For the nondevelopers, the weaker anticyclone north of the vortex may, in part, be related to the position of the low-level disturbance along the southeastern side of the PV streamer. A position farther northeast along the PV streamer would place the low-level disturbance farther eastward away from the core of the AWB and its attendant anticyclone. It is proposed that the PV streamer itself does not play a beneficial role in TC development; rather, the synoptic-scale flow associated with the AWB event may promote a horizontal distribution of moisture and convection that promotes the destruction of the PV streamer and subsequent TC development.

The high frequency of occurrence of tropical disturbances in association with AWB events suggests that the connections between them are physically relevant. The fact that 15% of TC development in the North Atlantic basin takes place in conjunction with AWB, and that half of these storms make landfall, demonstrates that an improved understanding of these events could yield important socioeconomic benefits. The purpose of this study was to identify the linkages between PV streamers and developing versus nondeveloping tropical disturbances in order to lay a foundation for future investigations that focus on attribution. In particular, cloud-resolving modeling studies are required to assess the relative impact of the beneficial and detrimental elements of the PV streamer’s presence and evolution throughout the predevelopment phase of the disturbance life cycle. Armed with such additional insight, improvements in our ability to predict the likelihood of this complex subtype of TC development will hopefully have an important impact on TC predictions across the region.

Acknowledgments

Partial support was provided by NSF grants ATM-0553017, ATM-0849491, and AGS-0935830 and an NCAR Advanced Study Program Graduate Student Visitor Fellowship awarded to Galarneau for the 2009 spring semester. McTaggart-Cowan’s collaborative visit to NCAR in April 2012 was supported by the MMM division visitor funds. The authors thank James Done (NCAR) for his valuable comments on the manuscript and John Molinari, Dan Keyser, and Chris Thorncroft (all University at Albany) for helpful discussions during the early stages of this research. Two anonymous reviewers and Dave Raymond (New Mexico Tech) are thanked for their insightful comments and suggestions that helped to improve the paper.

REFERENCES

REFERENCES
Archambault
,
H. M.
,
L. F.
Bosart
,
D.
Keyser
, and
J. M.
Cordeira
,
2013
:
A climatological analysis of the extratropical flow response to recurving western North Pacific tropical cyclones
.
Mon. Wea. Rev.
,
141
,
2325
2346
, doi:.
Bister
,
M.
, and
B. E.
Mapes
,
2004
:
Effect of vertical dipole temperature anomalies on convection in a cloud model
.
J. Atmos. Sci.
,
61
,
2092
2100
, doi:.
Bluestein
,
H. B.
,
1992
: Synoptic-Dynamic Meteorology in Midlatitudes: Volume 1. Principles of Kinematics and Dynamics. Oxford University Press, 431 pp.
Bosart
,
L. F.
, and
J. A.
Bartlo
,
1991
:
Tropical storm formation in a baroclinic environment
.
Mon. Wea. Rev.
,
119
,
1979
2013
, doi:.
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
,
353
376
, doi:.
Braun
,
S. A.
, and Coauthors
,
2013
:
NASA’s Genesis and Rapid Intensification Processes (GRIP) field experiment
.
Bull. Amer. Meteor. Soc.
,
94
,
345
363
, doi:.
Burpee
,
R. W.
,
1972
:
The origin and structure of easterly waves in the lower troposphere of North Africa
.
J. Atmos. Sci.
,
29
,
77
90
, doi:.
Chen
,
T.-C.
,
S.-Y.
Wang
,
M.-C.
Yen
, and
A. J.
Clark
,
2008
:
Are tropical cyclones less effectively formed by easterly waves in the western North Pacific than in the North Atlantic?
Mon. Wea. Rev.
,
136
,
4527
4540
, doi:.
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
,
1859
1881
, doi:.
Davis
,
C. A.
, and
L. F.
Bosart
,
2002
:
Numerical simulations of the genesis of Hurricane Diana (1984). Part II: Sensitivity of track and intensity prediction
.
Mon. Wea. Rev.
,
130
,
1100
1124
, doi:.
Davis
,
C. A.
, and
L. F.
Bosart
,
2003
:
Baroclinically induced tropical cyclogenesis
.
Mon. Wea. Rev.
,
131
,
2730
2747
, doi:.
Davis
,
C. A.
, and
L. F.
Bosart
,
2004
:
The TT problem: Forecasting the tropical transition of cyclones
.
Bull. Amer. Meteor. Soc.
,
85
,
1657
1662
, doi:.
Davis
,
C. A.
, and
L. F.
Bosart
,
2006
:
The formation of hurricane Humberto (2001): The importance of extra-tropical precursors
.
Quart. J. Roy. Meteor. Soc.
,
132
,
2055
2085
, doi:.
Davis
,
C. A.
, and
T. J.
Galarneau
Jr.
,
2009
:
The vertical structure of mesoscale convective vortices
.
J. Atmos. Sci.
,
66
,
686
704
, doi:.
Davis
,
C. A.
, and
D. A.
Ahijevych
,
2012
:
Mesoscale structural evolution of three tropical weather systems observed during PREDICT
.
J. Atmos. Sci.
,
69
,
1284
1305
, doi:.
Davis
,
C. A.
, and
D. A.
Ahijevych
,
2013
:
Thermodynamic environments of deep convection in Atlantic tropical disturbances
.
J. Atmos. Sci.
,
70
,
1912
1928
, doi:.
Davis
,
C. A.
,
C.
Snyder
, and
A. C.
Didlake
Jr.
,
2008
:
A vortex-based perspective of eastern Pacific tropical cyclone formation
.
Mon. Wea. Rev.
,
136
,
2461
2477
, doi:.
DeMaria
,
M.
,
J. A.
Knaff
, and
B. H.
Connell
,
2001
:
A tropical cyclone genesis parameter for the tropical Atlantic
.
Wea. Forecasting
,
16
,
219
233
, doi:.
Dunkerton
,
T. J.
,
M. T.
Montgomery
, and
Z.
Wang
,
2009
:
Tropical cyclogenesis in a tropical wave critical layer: Easterly waves
.
Atmos. Chem. Phys.
,
9
,
5587
5646
, doi:.
Elsberry
,
R. L.
, and
P. A.
Harr
,
2008
:
Tropical Cyclone Structure (TCS08) field experiment science basis, observational platforms, and strategy
.
Asia-Pac. J. Atmos. Sci.
,
44
,
209
231
.
Elsner
,
J. B.
,
G. S.
Lehmiller
, and
T. B.
Kimberlain
,
1996
:
Objective classification of Atlantic hurricanes
.
J. Climate
,
9
,
2880
2889
, doi:.
Evans
,
J. L.
, and
M. P.
Guishard
,
2009
:
Atlantic subtropical storms. Part I: Diagnostic criteria and composite analysis
.
Mon. Wea. Rev.
,
137
,
2065
2080
, doi:.
Gjorgjievska
,
S.
, and
D. J.
Raymond
,
2014
:
Interaction between dynamics and thermodynamics during tropical cyclogenesis
.
Atmos. Chem. Phys.
,
14
,
3065
3082
, doi:.
Gray
,
W. M.
,
1968
:
Global view of the origin of tropical disturbances and storms
.
Mon. Wea. Rev.
,
96
,
669
700
, doi:.
Guishard
,
M. P.
,
J. L.
Evans
, and
R. E.
Hart
,
2009
:
Atlantic subtropical storms. Part II: Climatology
.
J. Climate
,
22
,
3574
3594
, doi:.
Hess
,
J. C.
,
J. B.
Elsner
, and
N. E.
LaSeur
,
1995
:
Improving seasonal hurricane predictions for the Atlantic basin
.
Wea. Forecasting
,
10
,
425
432
, doi:.
Hulme
,
A. L.
, and
J. E.
Martin
,
2009a
:
Synoptic- and frontal-scale influences on tropical transition events in the Atlantic basin. Part I: A six-case survey
.
Mon. Wea. Rev.
,
137
,
3605
3625
, doi:.
Hulme
,
A. L.
, and
J. E.
Martin
,
2009b
:
Synoptic- and frontal-scale influences on tropical transition events in the Atlantic basin. Part II: Tropical transition of Hurricane Karen
.
Mon. Wea. Rev.
,
137
,
3626
3650
, doi:.
Kelley
,
W. E.
, and
D. R.
Mock
,
1982
:
A diagnostic study of upper tropospheric cold lows over the western North Pacific
.
Mon. Wea. Rev.
,
110
,
471
480
, doi:.
Kiladis
,
G. N.
,
C. D.
Thorncroft
, and
N. M. J.
Hall
,
2006
:
Three-dimensional structure and dynamics of African easterly waves. Part I: Observations
.
J. Atmos. Sci.
,
63
,
2212
2230
, doi:.
Knaff
,
J. A.
,
1997
:
Implications of summertime sea level pressure anomalies in the tropical Atlantic region
.
J. Climate
,
10
,
789
804
, doi:.
Komaromi
,
W. A.
,
2013
:
An investigation of composite dropsonde profiles for developing and nondeveloping tropical waves during the 2010 PREDICT field campaign
.
J. Atmos. Sci.
,
70
,
542
558
, doi:.
Lander
,
M. A.
,
1994
:
Description of a monsoon gyre and its effects on the tropical cyclones in the western North Pacific during August 1991
.
Wea. Forecasting
,
9
,
640
654
, doi:.
Landsea
,
C.
, and Coauthors
,
2004
: The Atlantic hurricane database re-analysis project: Documentation for 1851–1910 alterations and additions to the HURDAT database. Hurricanes and Typhoons: Past, Present, and Future, R. J. Murnane and K.-B. Liu, Eds., Columbia University Press, 177–221.
Mahalanobis
,
P. C.
,
1936
:
On the generalized distance in statistics
.
Proc. Natl. Inst. Sci. India
,
2
,
49
55
.
Martius
,
O.
,
C.
Schwierz
, and
M.
Sprenger
,
2008
:
Dynamical tropopause variability and potential vorticity streamers in the Northern Hemisphere—A climatological analysis
.
Adv. Atmos. Sci.
,
25
,
367
380
, doi:.
McBride
,
J. L.
, and
T. D.
Keenan
,
1982
:
Climatology of tropical cyclone genesis in the Australian region
.
J. Climatol.
,
2
,
13
33
, doi:.
McTaggart-Cowan
,
R.
,
L. F.
Bosart
,
C. A.
Davis
,
E. H.
Atallah
,
J. R.
Gyakum
, and
K. A.
Emanuel
,
2006
:
Analysis of Hurricane Catarina (2004)
.
Mon. Wea. Rev.
,
134
,
3029
3053
, doi:.
McTaggart-Cowan
,
R.
,
G. D.
Deane
,
L. F.
Bosart
,
C. A.
Davis
, and
T. J.
Galarneau
Jr.
,
2008
:
Climatology of tropical cyclogenesis in the North Atlantic (1948–2004)
.
Mon. Wea. Rev.
,
136
,
1284
1304
, doi:.
McTaggart-Cowan
,
R.
,
T. J.
Galarneau
Jr.
,
L. F.
Bosart
,
R. W.
Moore
, and
O.
Martius
,
2013
:
A global climatology of baroclinically influenced tropical cyclogenesis
.
Mon. Wea. Rev.
,
141
,
1963
1989
, doi:.
Molinari
,
J.
,
S.
Skubis
, and
D.
Vollaro
,
1995
:
External influences on hurricane intensity. Part III: Potential vorticity structure
.
J. Atmos. Sci.
,
52
,
3593
3606
, doi:.
Montgomery
,
M. T.
, and Coauthors
,
2012
:
The Pre-Depression Investigation of Cloud-Systems in the Tropics (PREDICT) experiment: Scientific basis, new analysis tools, and some first results
.
Bull. Amer. Meteor. Soc.
,
93
,
153
172
, doi:.
Moore
,
B. J.
,
P. J.
Neiman
,
F. M.
Ralph
, and
F. E.
Barthold
,
2012
:
Physical processes associated with heavy flooding rainfall in Nashville, Tennessee, and vicinity during 1–2 May 2010: The role of an atmospheric river and mesoscale convective systems
.
Mon. Wea. Rev.
,
140
,
358
378
, doi:.
Ndarana
,
T.
, and
D. W.
Waugh
,
2011
:
A climatology of Rossby wave breaking on the Southern Hemisphere tropopause
.
J. Atmos. Sci.
,
68
,
798
811
, doi:.
Nolan
,
D.
,
2007
:
What is the trigger for tropical cyclogenesis?
Aust. Meteor. Mag.
,
56
,
241
266
.
Onogi
,
K.
, and Coauthors
,
2007
:
The JRA-25 Reanalysis
.
J. Meteor. Soc. Japan
,
85
,
369
432
, doi:.
Payne
,
B.
, and
J.
Methven
,
2012
:
The role of baroclinic waves in the initiation of tropical cyclones across the southern Indian Ocean
.
Atmos. Sci. Lett.
,
13
,
88
94
, doi:.
Postel
,
G. A.
, and
M. H.
Hitchman
,
1999
:
A climatology of Rossby wave breaking along the subtropical tropopause
.
J. Atmos. Sci.
,
56
,
359
373
, doi:.
Rappin
,
E. D.
, and
D. S.
Nolan
,
2012
:
The effect of vertical shear orientation on tropical cyclogenesis
.
Quart. J. Roy. Meteor. Soc.
,
138
,
1035
1054
, doi:.
Raymond
,
D. J.
, and
S.
Sessions
,
2007
:
Evolution of convection during tropical cyclogenesis
.
Geophys. Res. Lett.
,
34
,
L06811
, doi:.
Raymond
,
D. J.
,
S. L.
Sessions
, and
C.
López Carrillo
,
2011
:
Thermodynamics of tropical cyclogenesis in the northwest Pacific
.
J. Geophys. Res.
,
116
,
D18101
, doi:.
Reynolds
,
R. W.
, and
T. M.
Smith
,
1994
:
Improved global sea surface temperature analyses using optimum interpolation
.
J. Climate
,
7
,
929
948
, doi:.
Riehl
,
H.
,
1948
:
On the formation of typhoons
.
J. Meteor.
,
5
,
247
265
, doi:.
Rogers
,
R.
, and Coauthors
,
2006
:
The intensity forecasting experiment: A NOAA multiyear field program for improving tropical cyclone intensity forecasts
.
Bull. Amer. Meteor. Soc.
,
87
,
1523
1537
, doi:.
Sadler
,
J. C.
,
1967
: The tropical upper tropospheric trough as a secondary source of typhoons and a primary source of trade wind disturbances. Tech. Rep. No. 67-12, Hawaii Institute of Geophysics, University of Hawaii, Honolulu, HI 96822, 44 pp.
Sadler
,
J. C.
,
1976
:
A role of the tropical upper tropospheric trough in early season typhoon development
.
Mon. Wea. Rev.
,
104
,
1266
1278
, doi:.
Sadler
,
J. C.
,
1978
:
Mid-season typhoon development and intensity changes and the tropical upper tropospheric trough
.
Mon. Wea. Rev.
,
106
,
1137
1152
, doi:.
Saha
,
S.
, and Coauthors
,
2010
:
The NCEP Climate Forecast System Reanalysis
.
Bull. Amer. Meteor. Soc.
,
91
,
1015
1057
, doi:.
Schenkel
,
B. A.
, and
R. E.
Hart
,
2012
:
An examination of tropical cyclone position, intensity, and intensity life cycle within atmospheric reanalysis datasets
.
J. Climate
,
25
,
3453
3475
, doi:.
Schumacher
,
C.
,
M. H.
Zhang
, and
P. E.
Ciesielski
,
2007
:
Heating structures of the TRMM field campaigns
.
J. Atmos. Sci.
,
64
,
2593
2610
, doi:.
Simpson
,
R. H.
,
1974
:
The hurricane disaster–potential scale
.
Weatherwise
,
27
,
169
186
, doi:.
Smith
,
R. K.
, and
M. T.
Montgomery
,
2012
:
Observations of the convective environment in developing and non-developing tropical disturbances
.
Quart. J. Roy. Meteor. Soc.
,
138
,
1721
1739
, doi:.
Thorncroft
,
C. D.
,
B. J.
Hoskins
, and
M. E.
McIntyre
,
1993
:
Two paradigms of baroclinic-wave life-cycle behavior
.
Quart. J. Roy. Meteor. Soc.
,
119
,
17
55
, doi:.
van Lier-Walqui
,
M.
,
T.
Vukicevic
, and
D. J.
Posselt
,
2014
:
Linearization of microphysical parameterization uncertainty using multiplicative process perturbation parameters
.
Mon. Wea. Rev.
,
142
,
401
413
, doi:.
Wernli
,
H.
, and
M.
Sprenger
,
2007
:
Identification and ERA-15 climatology of potential vorticity streamers and cutoffs near the extratropical tropopause
.
J. Atmos. Sci.
,
64
,
1569
1586
, doi:.
Wilks
,
D. M.
,
1995
: Statistical Methods in the Atmospheric Sciences: An Introduction. Academic Press, 467 pp.
Yu
,
H.
, and
H. J.
Kwon
,
2005
:
Effect of TC–trough interaction on the intensity change of two typhoons
.
Wea. Forecasting
,
20
,
199
211
, doi:.
Yuter
,
S. E.
, and
R. A.
Houze
Jr.
,
1995
:
Three-dimensional kinematic and microphysical evolution of Florida cumulonimbus. Part II: Frequency distributions of vertical velocity, reflectivity, and differential reflectivity
.
Mon. Wea. Rev.
,
123
,
1941
1963
, doi:.

Footnotes

*

The National Center for Atmospheric Research is sponsored by the National Science Foundation.

1

The Tropical Cyclone Structure-2008 (TCS-08; Elsberry and Harr 2008), PREDICT (Montgomery et al. 2012), National Aeronautics and Space Administration (NASA) Genesis and Rapid Intensification Processes (GRIP; Braun et al. 2013), and the NOAA Intensity Forecasting Experiment (IFEX; Rogers et al. 2006) field programs.

2

AWB events that occurred between 100°W and 30°E were considered, because their attendant PV streamers tend to be located over the subtropical and tropical North Atlantic based on our subjective investigation of DT maps over the 2004–08 North Atlantic TC seasons.

3

The Rossby radius of deformation is defined as , where H is the scale height, N is the Brunt–Väisälä frequency, and f0 is the Coriolis constant 1.0 × 10−4 s−1.

4

The anomaly represents the departure of the mean within 200 km from the 200–1000-km radial and azimuthal average.