Abstract

This paper examines the cyclogenesis of the “Perfect Storms” of late October and early November 1991 over the North Atlantic and focuses on the influence of Hurricane Grace (HG) toward their development. The two storms considered are the “Perfect Storm” (PS) that underwent a warm seclusion process and an extratropical cyclone (EC1) with two development phases. HG, which initially formed via tropical transition (TT), influenced the first phase of EC1 via reduced atmospheric static stability and enhanced low-level baroclinicity. As a result, deep moist convection and latent heat release produced maxima in midtropospheric diabatic heating and lower-tropospheric potential vorticity (PV) that aided the development of EC1. Backward air parcel trajectories and large diabatic contributions to eddy available potential energy (APE) generation suggests that EC1 developed as a diabatic Rossby vortex (DRV)-like feature.

The second and explosively deepening phase of EC1 occurred as the cyclone coupled with an upper-tropospheric PV disturbance (PVD) over the eastern North Atlantic. Backward air parcel trajectories demonstrate the explosive deepening of EC1 involved airstreams originating from east of HG and from over the Labrador Sea. Parcel trajectories and a large baroclinic contribution to eddy APE generation further suggests that the two-phase development of EC1 may have involved a DRV-like feature.

The subsequent recurvature and extratropical transition (ET) of HG occurred in the warm sector of the PS downstream of a second upper-tropospheric PVD over the western North Atlantic. Reduced atmospheric static stability, enhanced warm air advection, and strong latent heat release during the recurvature and ET of HG contributed to the development of a strong, zonally oriented warm front and the warm seclusion of the PS. Parcel trajectory analysis demonstrates that the PS warm seclusion involved the isolation of air parcels by a bent-back warm front that were warmed via sensible heating from the underlying Gulf Stream.

1. Introduction

a. Background and motivation

The “Perfect Storms” (PSs) of late October and early November 1991 were a series of three extratropical cyclones (ECs) that impacted the North Atlantic and North America with large waves, coastal flooding, heavy snow, and accumulating ice (e.g., Cordeira and Bosart 2010). The PSs included the so-called Perfect Storm (PS; Cardone et al. 1996; Junger 1997) and two additional ECs over the North Atlantic and central United States (EC1 and EC2, respectively). EC1 was an intense cyclone that developed in two distinct phases that resulted from interactions with Hurricane Grace (HG), the PS, and an upper-tropospheric potential vorticity (PV) disturbance (PVD) of arctic origin. EC2 is colloquially known as the 1991 Halloween Blizzard in the Midwest United States and will not be discussed further in this paper. The purpose of this paper is twofold: 1) to investigate interactions between HG, the PS, and EC1, and 2) document their subsequent evolutions over the North Atlantic.

Previous research on the PS by Cordeira and Bosart (2010) showed that downstream baroclinic development originated over the North Pacific, progressed across North America, and produced conditions favorable for the cyclogenesis of the PS and EC1 over the western North Atlantic. These favorable conditions for cyclogenesis, shown by high values of the Eady baroclinic growth rate, developed in conjunction with the equatorward displacement of upper-tropospheric and lower-stratospheric PVDs in proximity to enhanced low-level baroclinicity [cf. Fig. 16 from Cordeira and Bosart (2010) and adapted here as Fig. 1]. Figure 1 additionally depicts the 6-h locations of HG, the PS, and EC1, and illustrates that the PS and EC1 developed poleward of the region where HG underwent TT, recurvature, and extratropical transition (ET).

Fig. 1.

Six-hour positions of HG, the PS, and EC1 between 25 Oct and 3 Nov 1991 with the maximum grid-point values of the Eady baroclinic growth rate parameter (σE) at 700 hPa (day−1; shaded according to scale) and the time-mean 300-hPa geopotential height (solid contours every 12 dam) between 28 Oct and 1 Nov 1991. Here, σE = 0.31fUzN−1, f = Coriolis parameter, Uz = vertical derivative of the horizontal wind, N = Brunt–Väisälä frequency and was calculated over the 800–600-hPa layer with the ERA–Interim. Tropical symbols denote categorical intensity and characteristics of HG and the PS with day numbers given at 0000 UTC. Inset: SLP derived from the National Hurricane Center (NHC) Best Track data, ERA–Interim analyses, and National Meteorological Center (NMC) Final Analyses.

Fig. 1.

Six-hour positions of HG, the PS, and EC1 between 25 Oct and 3 Nov 1991 with the maximum grid-point values of the Eady baroclinic growth rate parameter (σE) at 700 hPa (day−1; shaded according to scale) and the time-mean 300-hPa geopotential height (solid contours every 12 dam) between 28 Oct and 1 Nov 1991. Here, σE = 0.31fUzN−1, f = Coriolis parameter, Uz = vertical derivative of the horizontal wind, N = Brunt–Väisälä frequency and was calculated over the 800–600-hPa layer with the ERA–Interim. Tropical symbols denote categorical intensity and characteristics of HG and the PS with day numbers given at 0000 UTC. Inset: SLP derived from the National Hurricane Center (NHC) Best Track data, ERA–Interim analyses, and National Meteorological Center (NMC) Final Analyses.

HG underwent TT following a weak extratropical cyclone (WEC) precursor pathway on 26–27 October over the subtropical North Atlantic (Davis and Bosart 2003, 2004). EC1 subsequently developed poleward of HG on 28–29 October during its first development phase, and later collocated with the PS warm front along the southern periphery of the Eady baroclinic growth rate parameter maximum on 29 October (Fig. 1). EC1 rapidly traversed the North Atlantic at >25 m s−1 and “crossed” the Eady baroclinic growth rate parameter maximum and the time–mean 300-hPa geopotential height gradient during its second development phase on 30 October. The second development phase was marked by explosive deepening of 49 hPa (24 h)−1 between 1800 UTC 29 October and 1800 UTC 30 October.1 Meanwhile, the PS retrograded toward the northeast United States, underwent a warm seclusion process (e.g., Shapiro and Keyser 1990) and subsequent TT, and became a weak unnamed hurricane on 1 November (Pasch 1991; Pasch and Avila 1992).

b. Development of the PS and EC1

Observational and numerical evidence suggest that extratropical cyclogenesis is influenced by 1) the configuration of the planetary- and synoptic-scale flow regime, 2) physical processes contributing to enhanced baroclinicity, and 3) latent heating via precipitation processes (e.g., Sanders and Gyakum 1980; Reed and Albright 1986; Sanders 1986; Reed et al. 1992; Bosart et al. 1996; Lackmann et al. 1996; Dickinson et al. 1997; Deveson et al. 2002; Roebber 2009). Given that Cordeira and Bosart (2010) focused on the antecedent planetary- and synoptic-scale flow regime, herein we investigate the influence of enhanced baroclinicity and latent heating via precipitation process during the life cycle of HG on the development of the PS and EC1. The following hypotheses are considered:

  1. From a PV perspective, Davis and Bosart (2004) suggest that WEC TT events typically occur when mid- to upper-tropospheric PVDs encounter lower-tropospheric baroclinicity in the subtropics. For example, baroclinicity was a critical ingredient in producing a region of quasi-balanced upward motion that focused deep moist convection (DMC) northeast of the developing low-level cyclonic circulation associated with TC Diana (1984) and to the east of a mid- and upper-tropospheric PVD. This region of DMC subsequently allowed for a vertical redistribution of mass and PV that reduced vertical wind shear and produced an environment conducive to tropical cyclogenesis (Davis and Bosart 2001). Hulme and Martin (2009) also show that a region of DMC may remain poleward of a TC that has undergone TT. This region of DMC is situated along a zone of enhanced baroclinicity and is typically characterized by frontogenesis, significant latent heat release, and diabatic PV generation (see their Figs. 3–4). Therefore, we hypothesize that a region of enhanced baroclinicity poleward of HG during TT may have influenced the development of the PS and EC1.

  2. Climatologies of TCs undergoing ET in the western Pacific Ocean note a maximum in scalar frontogenesis to the north and east of poleward-moving TC remnants (Harr and Elsberry 2000; Sinclair 2002) with a resultant deformation pattern consistent with warm frontogenesis (Keyser et al. 1988; Schultz et al. 1998; Jones et al. 2003). Therefore, we hypothesize that regions of enhanced baroclinicity and warm frontogenesis poleward of HG during recurvature and ET may have influenced the development of the PS and EC1.

  3. The location of a TC downstream of an upstream trough (Harr and Elsberry 2000) is similar in structure to Petterssen and Smebye (1971) type-B cyclogenesis events (Klein et al. 2002). The region of reduced atmospheric static stability, enhanced warm air advection, and likely strong latent heat release induced by the TC is in phase with the extratropical forcing for ascent ahead of the upper-level trough, and provides a focusing mechanism for explosive cyclone intensification consistent with quasigeostrophic theory and PV thinking (e.g., Agustí-Panareda et al. 2005; Agustí-Panareda 2008). Therefore, we hypothesize that a reduction in atmospheric static stability, enhanced warm air advection, and strong latent heat release induced by HG downstream of the time–mean upper-level trough seen in Fig. 1 may have influenced the development of the PS and EC1.

c. Evolution of the PS and EC1

The warm seclusion of the PS and the two-phase development of EC1 over the North Atlantic are subsequently investigated. Emphasis will be given to the two-phase development of EC1. The two-phase development process is reminiscent of cyclogenesis events with delayed intensification due to a period of unfavorable coupling between upper- and lower-tropospheric PVDs (e.g., Buzzi and Tibaldi 1978; Farrell 1984; Gyakum et al. 1992; Wernli et al. 2002; Moore et al. 2008). The first development phase is often characterized by diabatic low-level PV generation beneath a region of deep moist ascent along a low-level baroclinic zone. The resulting lower-tropospheric interior PV anomaly represents a diabatic Rossby vortex (DRV) that can be self-maintained via continuous diabatic heating in conjunction with the transport of warm, moist air poleward and normal to a baroclinic zone downstream by the cyclonic circulation induced by the PV anomaly (e.g., Raymond and Jiang 1990; Parker and Thorpe 1995; Moore and Montgomery 2004, 2005). The second development phase is often characterized by explosive cyclogenesis in conjunction with a period of mutual intensification between upper- and lower-tropospheric PVDs as in Hoskins et al. (1985). This “DRV cyclogenesis pathway” has been investigated by Wernli et al. (2002) for extreme winter storm Lothar (1999) over Western Europe and by Moore et al. (2008) for a 2005 East Coast U.S. snowstorm. A majority of this investigation focuses on EC1 given, to our knowledge, an analysis of DRV formation and subsequent explosive extratropical cyclogenesis downstream (poleward) of a region where a TC is undergoing ET has yet to be undertaken.

d. Paper organization

This paper will be organized as follows. Section 2 outlines the data and methods used for the investigation. Sections 35 are an overview the TT of HG, the development and warm seclusion of the PS, and the two-phase development of EC1, with an emphasis on diagnostic Lagrangian and energetic analyses. Sections 6 and 7 contain a discussion of the results and conclusions.

2. Data and methods

Gridded datasets of the European Centre for Medium-Range Weather Forecasts (ECMWF) Interim reanalysis (ERA-Interim; Simmons et al. 2007), with ~1.5° horizontal grid spacing, 23 vertical pressure levels between 1000 and 50 hPa, and 6-h temporal resolution constituted the primary data source for analysis. The ERA-Interim analyses are complemented by infrared satellite (IR) imagery from the National Climatic Data Center (NCDC) through their Global International Satellite Cloud Climatology Project (ISCCP) B1 Browse System (GIBBS) archive (online at http://www.ncdc.noaa.gov/gibbs/) and as gridded brightness temperature (Tb) data from the Cloud Archive User Service (CLAUS; Hodges et al. 2000) with 0.5° horizontal grid spacing and 3-h temporal resolution.

A number of diagnostic quantities are utilized to analyze the development and evolution of the PS and EC1. The energetics analysis in section 5 is based upon equations presented in Moore and Montgomery (2005) and Conzemius et al. (2007), as originally developed and reformulated by Lorenz (1955), Muench (1965), and Norquist et al. (1977). The time rate of change of eddy available potential energy (APE) is given by

 
formula

where AE is the eddy APE, CA is the conversion from basic-state APE to eddy APE (e.g., baroclinic generation of eddy APE), CE is the conversion of eddy APE to eddy kinetic energy, and GE is the diabatic generation of eddy APE. The following definitions of these terms, following Moore and Montgomery (2005) and Conzemius et al. (2007), contain area mean quantities (denoted by ), zonally averaged values (denoted by [ ]), deviations from the area mean (denoted by *), and deviations from the zonal mean (denoted by primed values):

 
formula
 
formula
 
formula
 
formula

Terms in (2)(5) have their usual meteorological meaning except the mean static stability parameter () is defined following Norquist et al. (1977):

 
formula

The energetics of EC1 are calculated over a ~500 km × ~500 km feature-following domain with vertical integration limits of pb = 925 hPa and pt = 250 hPa. The lower limit of integration was chosen to eliminate uncertainties introduced by SLP values <1000 hPa over the North Atlantic.

The Lagrangian derivative of potential temperature (PT) /dt is used to approximate diabatic heating as

 
formula

The Lagrangian derivative of PT is computed from the ERA-Interim using centered 12-h air parcel trajectories (i.e., 6-h forward and 6-h backward trajectories) with a 60-min time step for 1° latitude and longitude increments at each pressure level over the North Atlantic. This method for calculating diabatic heating from a dataset with 6-h temporal resolution follows McTaggart-Cowan et al. (2007) and Galarneau et al. (2009). The diabatic heating rate () is presented as Q(cp)−1. The Lagrangian derivative of PT used in this study is a useful diagnostic tool for approximating diabatic heating rates. Limitations in the calculation are noted given the 6-h interval between analysis times and inconsistencies that may result from a lack of temporal continuity, and the interpolation scheme used to locate air parcels between analysis times. From the three-dimensional structure in diabatic heating, it was subsequently possible to calculate diabatic PV tendencies. However, given the aforementioned limitations in the calculation of diabatic heating, we express only estimates of PV tendencies from the Lagrangian analyses or from the vertical gradient in the diabatic heating rate as defined following Martin (2006, p. 293):

 
formula

This paper also considers the near-surface contribution to diabatic temperature changes via sensible heat (SH) flux and is calculated as the local rate in temperature change via SH–flux convergence over a volume using the bulk aerodynamic method for a hydrostatic atmosphere following Bluestein (1992, p. 311):

 
formula

where SH is absorbed evenly over a lower-tropospheric depth of Δp. The SH flux is provided from the ERA-Interim analyses at 0000 and 1200 UTC.

Finally, the Rossby radius of deformation (R0) is estimated from the ERA-Interim analyses to provide evidence for the influence of the ageostrophic adjustment process associated with upper-tropospheric PVDs on the development of EC1 in section 5c. The R0 value is a spatially averaged value for a ~500 km × ~500 km domain centered on the location of each PVD on the 2-PV unit [PVU (1 PVU = 10−6 K m2 kg−1 s−1)] surface, and is estimated as NH(f0)−1, where N is the Brunt–Väisälä frequency calculated between 900 and 200 hPa, H is the height of the 2-PVU surface, and f0 is the Coriolis parameter at 45°N. The R0 value is compared to the horizontal separation distance between the center locations of the upper-tropospheric PVDs and the lower-tropospheric maximum in PV associated with EC1.

3. Tropical transition of HG: 25–27 October

The TT of HG followed a similar path by TCs that form as a result of WEC precursors (Davis and Bosart 2003, 2004). Accordingly, images of PV, wind, and pressure on the 330-K PT surface depict an elongated streamer of high PV air that stretched across the western North Atlantic between 25 and 28 October (Fig. 2). The antecedent 850-hPa relative vorticity maximum associated with pre-HG was located near the “tail” of the PV streamer near 25°N at 0000 UTC 25 October (Fig. 2a). Deep-layer wind shear values over pre-HG were ~14 m s−1 with convection and values of Tb < 240 K located primarily downstream. The convection was located in a region of likely isentropic upper-level ascent as suggested by a component of the wind parallel to the isentropic pressure gradient (Fig. 2a).

Fig. 2.

The 330-K PV (PVU; shaded blue), wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1), and pressure (hPa, contoured red), with 850-hPa relative vorticity (solid black contours every 2.0 × 10−5 s−1 starting at 2.0 × 10−5 s−1) and CLAUS brightness temperatures (K; shaded red) at 0000 UTC 25–28 Oct. The symbols “G” and “1” represent the locations of the HG and EC1, respectively. The black dashed line marks the location of the PV streamer discussed in section 3. Yellow box illustrates domain for area averaged deep-layer wind shear over Grace (defined as the 850-hPa wind subtracted from the 330-K wind) displayed in lower left. Inset: Corresponding IR satellite imagery with PV streamer location.

Fig. 2.

The 330-K PV (PVU; shaded blue), wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1), and pressure (hPa, contoured red), with 850-hPa relative vorticity (solid black contours every 2.0 × 10−5 s−1 starting at 2.0 × 10−5 s−1) and CLAUS brightness temperatures (K; shaded red) at 0000 UTC 25–28 Oct. The symbols “G” and “1” represent the locations of the HG and EC1, respectively. The black dashed line marks the location of the PV streamer discussed in section 3. Yellow box illustrates domain for area averaged deep-layer wind shear over Grace (defined as the 850-hPa wind subtracted from the 330-K wind) displayed in lower left. Inset: Corresponding IR satellite imagery with PV streamer location.

The PV streamer thinned on the 330-K PT surface poleward of pre-HG between 25 and 26 October (Fig. 2b). At 850-hPa, relative vorticity increased monotonically within pre-HG from 2.0 × 10−5 s−1 to 6.0 × 10−5 s−1 between 0000 UTC 25 October and 0000 UTC 26 October (Figs. 2a,b), and increased rapidly to ~12.0 × 10−5 s−1 by 1200 UTC 26 October (not shown). The rapid increase in low-level relative vorticity was associated with an increase in pre-HG’s classification strength to a subtropical storm at 0600 UTC 26 October. The thinning of the PV streamer was likely influenced by a negative PV tendency above the maximum in midtropospheric diabatic heating associated with convection to the northeast of pre-HG on 27 October (Fig. 2c). Additionally, low-level PV generation beneath the maximum in midtropospheric diabatic heating likely influenced an increase in low-level relative vorticity to >18.0 × 10−5 s−1 (Fig. 2c). Deep-layer wind shear values decreased to ~10 m s−1 on 27 October and to ~7 m s−1 on 28 October (Figs. 2c,d). Observations from a reconnaissance flight by the Air Force prompted the National Hurricane Center to upgrade pre-HG to a tropical storm at 1800 UTC 27 October and to a hurricane by 0000 UTC 28 October (Fig. 2d).

4. The “Perfect Storm”

a. Development: 28–29 October

The PS appears as a frontal wave over New England, located between a >1040-hPa surface anticyclone over eastern Canada (“H”) and HG over the western Atlantic at 0000 UTC 28 October (Fig. 3a). A low-level 850-hPa PV maximum was located over Quebec at 0000 UTC 28 October, coincident with a low-level baroclinic zone (Fig. 3a), and to the east of an upper-level trough illustrated by dynamic tropopause (DT) pressure maximum XP (i.e., PVD XP; Fig. 4a). The amplifying-wave phase of the PS ensued as the warm, moist air mass associated with HG moved poleward in the PS warm sector on 29 October (Figs. 3c,d) and resulted in a ~20% reduction in 1000–700-hPa static stability near 40°N, 55°W [as determined from (6); not shown]. The thermodynamic structure of the environment over the western North Atlantic in the region of the PS warm front changed significantly between 1200 UTC 28 October and 1200 UTC 29 October (Fig. 5). This 24-h period featured deep warm air advection downstream of PVD XP that contributed to 8°–10°C of warming between 1000 and 700 hPa and contributed to the nearby 20% reduction in the 1000–700-hPa static stability. The environment supported deep moist convection (DMC) in conjunction with vertical wind shear values >30 m s−1, strong vertical motion <−10 μb s−1, and CAPE values (calculated from the level of free convection) >250 J kg−1 between 0000 and 1200 UTC 29 October.

Fig. 3.

The 850-hPa potential temperature (dashed blue contours every 3 dam) and potential temperature gradient [K (100 km)−1; shaded according to red scale], wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1; plotted for magnitudes >5 m s−1), sea level pressure (solid black contours every 4 hPa), and precipitable water (mm, shaded according to grayscale) at 0000 and 1200 UTC 28 and 29 Oct. The symbols “H,” “P,” “G,” and “1” represent the locations of the anticyclone discussed in section 3, the PS, HG, and EC1, respectively.

Fig. 3.

The 850-hPa potential temperature (dashed blue contours every 3 dam) and potential temperature gradient [K (100 km)−1; shaded according to red scale], wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1; plotted for magnitudes >5 m s−1), sea level pressure (solid black contours every 4 hPa), and precipitable water (mm, shaded according to grayscale) at 0000 and 1200 UTC 28 and 29 Oct. The symbols “H,” “P,” “G,” and “1” represent the locations of the anticyclone discussed in section 3, the PS, HG, and EC1, respectively.

Fig. 4.

Dynamic tropopause (DT = 2 PVU surface) pressure (shaded according to blue scale in hPa) and wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1 starting at 40 m s−1), 850-hPa potential vorticity (contoured at 0.75, 1.0, 1.25, and 1.5 PVU with every 0.5 PVU contoured thereafter), and CLAUS brightness temperature (K, shaded according to red scale) 0000 UTC 28 Oct and 1200 UTC 29 Oct. The symbols “P,” “G,” and “1” represent the locations of the PS, HG, and EC1, respectively. The symbols XP and XA represent the locations of DT pressure maxima in proximity to the PS and EC1, respectively. (a)–(c) Inset IR satellite imagery to assist in locating EC1. Cross-section lines shown for Fig. 12.

Fig. 4.

Dynamic tropopause (DT = 2 PVU surface) pressure (shaded according to blue scale in hPa) and wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1 starting at 40 m s−1), 850-hPa potential vorticity (contoured at 0.75, 1.0, 1.25, and 1.5 PVU with every 0.5 PVU contoured thereafter), and CLAUS brightness temperature (K, shaded according to red scale) 0000 UTC 28 Oct and 1200 UTC 29 Oct. The symbols “P,” “G,” and “1” represent the locations of the PS, HG, and EC1, respectively. The symbols XP and XA represent the locations of DT pressure maxima in proximity to the PS and EC1, respectively. (a)–(c) Inset IR satellite imagery to assist in locating EC1. Cross-section lines shown for Fig. 12.

Fig. 5.

Skew T–logp diagram of air temperature (°C), dewpoint (°C), and wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1) with the vertical profile of vertical velocity (μb s−1) at 1200 UTC 28 Oct (blue), 0000 UTC 29 Oct (red), and 1200 UTC 29 Oct (black) at 45.0°N, 49.0°W. Locations of soundings coincide with the location of EC1 at 1200 UTC 29 Oct. Inset: CAPE (calculated from the level of free convection), surface-to-500-hPa shear magnitude, and precipitable water (PW) values.

Fig. 5.

Skew T–logp diagram of air temperature (°C), dewpoint (°C), and wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1) with the vertical profile of vertical velocity (μb s−1) at 1200 UTC 28 Oct (blue), 0000 UTC 29 Oct (red), and 1200 UTC 29 Oct (black) at 45.0°N, 49.0°W. Locations of soundings coincide with the location of EC1 at 1200 UTC 29 Oct. Inset: CAPE (calculated from the level of free convection), surface-to-500-hPa shear magnitude, and precipitable water (PW) values.

The ensuing amplifying-wave phase of the PS was associated with the amalgamation of three low-level PV maxima associated with HG, EC1 (discussed in section 5), and the PS to create “frontal-like” bands of PV between 0000 and 1200 UTC 29 October (Figs. 4c,d). During this period, the minimum SLP of the PS decreased from 999 to 990 hPa (Fig. 1). The period between 1200 UTC 28 October and 1200 UTC 29 October featured an increase in southerly low-level winds from ~15 to >25 m s−1, the formation of DMC along and poleward of the frontogenetic PS warm front (suggested by an increase in the magnitude of the 850-hPa PT gradient), and a downstream intensification of the winds on the DT (i.e., the North Atlantic jet stream) from ~60 to >70 m s−1 (Figs. 4c,d). The intensification of the North Atlantic jet stream is consistent with a redistribution of PV along the PS warm front such that DMC contributed to negative PV tendencies above a midtropospheric maximum in diabatic heating. As a result, the upper-level PV and DT pressure gradients were likely enhanced, contributing to an intensification of the upper-level westerly flow. The intensification of the North Atlantic jet stream is revisited from PV and Lagrangian perspectives in sections 5c and 5d.

b. Warm seclusion: 30–31 October

The warm seclusion phase of the PS resembled the Shapiro and Keyser (1990) cyclone model and initiated with the development of a “T-bone” frontal structure between 0000 and 1200 UTC 30 October (Figs. 6a,b). The seclusion is later shown by a bent-back warm front enclosing 850-hPa PT > 294 K at 1200 UTC 30 October and 0000 UTC 31 October (Figs. 6b,c). The bent-back warm front at 1200 UTC 30 October was characterized by a strong 850-hPa PT gradient to the north and west of the PS (Fig. 6b). The bent-back warm front represented a region of strong frontogenesis and upward vertical motion (not shown), DMC, and likely strong latent heat release to the northwest of the PS. Wernli (1997), Bosart (1999), and Posselt and Martin (2004) suggest that midtropospheric diabatic heating along the bent-back warm front is responsible for the upper-level redistribution of PV during the development of a PV hook (e.g., the treble-clef-like feature seen in Fig. 7 on the DT). A population of 36-h backward air parcel trajectories ending near the PV hook, summarized by population–mean trajectories, suggest two air parcel source regions: 1) air parcels comprising the region of high PV air within the PV hook at 0000 UTC 30 October that descended from over Canada (Fig. 7a), and 2) air parcels comprising the region of low PV air wrapping cyclonically around and into the PV hook at 1200 UTC 30 October that ascended to the east of HG and along the PS warm front (Fig. 7b). A visual comparison between regions of Tb < 220 K at 0000 UTC 30 October in Fig. 7a and the location of air parcels at 0000 UTC 30 October that ultimately comprised the region of low PV air in Fig. 7b suggest that the ascent of these air parcels occurred in conjunction with DMC and negative upper-level PV tendencies within the PV hook, and contributed to the ultimate fracture of PVD XP on 31 October (Figs. 7c,d).

Fig. 6.

As in Fig. 3, but at 0000 and 1200 UTC 30 and 31 Oct with (a) sensible heat flux contours every 100 W m−2 starting at 200 W m−2; and (b)–(d), the 294-K 850-hPa potential contour emphasized as a solid blue contour.

Fig. 6.

As in Fig. 3, but at 0000 and 1200 UTC 30 and 31 Oct with (a) sensible heat flux contours every 100 W m−2 starting at 200 W m−2; and (b)–(d), the 294-K 850-hPa potential contour emphasized as a solid blue contour.

Fig. 7.

As in Fig. 4, but at 0000 30 Oct and 1200 UTC 31 Oct with 36-h backward air parcel trajectories ending at (a) 550 hPa [(b) 350 hPa] and at 0000 (1200) UTC on 30 Oct. Air parcel pressure is given in hPa every 12 h at the location of the white-filled circle. These 36-h trajectories are population-mean trajectories for air parcels ending at nearby locations. Note 850-hPa potential vorticity is contoured at 1.0, 1.25, and 1.5 PVU with every 0.5 PVU contoured thereafter.

Fig. 7.

As in Fig. 4, but at 0000 30 Oct and 1200 UTC 31 Oct with 36-h backward air parcel trajectories ending at (a) 550 hPa [(b) 350 hPa] and at 0000 (1200) UTC on 30 Oct. Air parcel pressure is given in hPa every 12 h at the location of the white-filled circle. These 36-h trajectories are population-mean trajectories for air parcels ending at nearby locations. Note 850-hPa potential vorticity is contoured at 1.0, 1.25, and 1.5 PVU with every 0.5 PVU contoured thereafter.

The warm seclusion process of the PS progressed with weakening lower-tropospheric baroclinicity (Fig. 6d) and a decrease in DT pressure within the PV hook that defined the location of PVD XP over the western North Atlantic on 31 October (Fig. 7d). The bent-back warm front moved toward eastern New England and was associated with >100 mm of total rainfall [cf. Table 1 of Cordeira and Bosart (2010)]. The low-level secluded warm core of the PS subsequently transitioned into a subtropical storm at 1800 UTC 31 October, aided by localized convection over the Gulf Stream (Fig. 8a). The warm core of the PS underwent TT via the strong extratropical precursor pathway discussed by Davis and Bosart (2003, 2004) and developed into the unnamed hurricane on 1 November (Fig. 8b; e.g., Pasch and Avila 1992).

Fig. 8.

GOES-7 visible satellite imagery at (a) 1801 UTC 31 Oct and (b) 1801 UTC 1 Nov. Image adapted from the University of Wisconsin.

Fig. 8.

GOES-7 visible satellite imagery at (a) 1801 UTC 31 Oct and (b) 1801 UTC 1 Nov. Image adapted from the University of Wisconsin.

c. Lagrangian diagnostic

Air parcel trajectories were generated from regions where the 850-hPa PT exceeded 295 K during the warm seclusion phase of the PS at 1200 UTC 30 October (Fig. 9a). The Lagrangian diagnostic is used to ascertain the origin of air parcels that constituted the region of warm air secluded within the PS and to assess whether or not these air parcels originated in proximity to HG. Backward trajectories for 26 air parcels with 850-hPa PT > 295 K at 1200 UTC 30 October within the warm seclusion of the PS indicate that parcels originated poleward of the PS at 0000 UTC 29 October, comprised an airstream consistent with a cold conveyor belt (e.g., Carlson 1980), and encircled the bent-back warm front between 0000 and 1200 UTC 30 October (Fig. 9a). These 26 air parcels are separated into two populations: (i) 9 early-ascending air parcels that ascended between 1200 UTC 29 October and 0000 UTC 30 October, descended between 0000 and ~0600 UTC 30 October, and re-ascended to 850 hPa by 1200 UTC 30 October (illustrated in red; Fig. 9b); and (ii) 17 later-ascending air parcels that remained near the surface between 0000 UTC 29 October and 0000 UTC 30 October, but then ascended abruptly to 850 hPa between 0000 and 1200 UTC 30 October (illustrated in blue; Fig. 9b). The backward trajectories indicate that the early-ascending trajectories were encircled and “overtaken” by the later-ascending trajectories. The computed PT change following the two populations suggest that the early-ascending trajectories (red) warmed significantly as they first traveled at low levels poleward of the PS and to the northwest of the bent-back warm front between 1200 UTC 29 October and 0000 UTC 30 October, whereas the later-ascending trajectories (blue) warmed significantly as they traveled along the bent-back warm front between 0000 and 0600 UTC 30 October (Fig. 9c). The warming of both air parcel populations via SH flux is considered in section 6c, given that both populations appeared to warm significantly as they traveled over the Gulf Stream.

Fig. 9.

(a) Backward air parcel trajectories for the 36-h period ending at 1200 UTC 30 Oct atop the 850-hPa potential temperature (K, shaded) at 1200 UTC 30 Oct. Air parcels end at 850 hPa within a region of 850-hPa potential temperature (θ850) >295 K. The trajectories are labeled sequentially at their “endpoints” and contain white-filled circles every 12 h. The trajectories are colored red (blue) if they ascended before (after) 0000 UTC 29 Oct. Select sea surface temperature (SST) contours from the week ending on 1 Nov 1991 are shown subjectively drawn in green. The (b) pressure and (c) potential temperature values for these trajectories given with shading as in (a).

Fig. 9.

(a) Backward air parcel trajectories for the 36-h period ending at 1200 UTC 30 Oct atop the 850-hPa potential temperature (K, shaded) at 1200 UTC 30 Oct. Air parcels end at 850 hPa within a region of 850-hPa potential temperature (θ850) >295 K. The trajectories are labeled sequentially at their “endpoints” and contain white-filled circles every 12 h. The trajectories are colored red (blue) if they ascended before (after) 0000 UTC 29 Oct. Select sea surface temperature (SST) contours from the week ending on 1 Nov 1991 are shown subjectively drawn in green. The (b) pressure and (c) potential temperature values for these trajectories given with shading as in (a).

5. EC1

a. Development (phase one): 28–29 October

The locations of EC1 were traced backward in time and its center was located to the northeast of HG at 0000 UTC 28 October. EC1 was located along a weak baroclinic zone, embedded within precipitable water values >50 mm (Fig. 3a), and collocated with an isolated region of DMC as suggested by values of Tb < 210 K (Fig. 4a). EC1 subsequently moved poleward in ~20 m s−1 850-hPa southerly flow within the PS warm sector and downstream of PVD XP between 1200 UTC 28 October and 1200 UTC 29 October (Figs. 3 and 4). The above environment in which EC1 developed suggests that weak slantwise ascent in the presence of tropical moisture associated with HG and weak low-level baroclinicity contributed to diabatic low-level PV generation and the EC1 cyclogenesis process. This first development phase of EC1, defined as the initial period of decreasing SLP values in Fig. 1, was complete by 1200 UTC 29 October as the minimum SLP decreased to 994 hPa and lower-tropospheric PV at 850-hPa increased to >2 PVU along the PS warm front (Fig. 4d). The location of EC1 at 1200 UTC 29 October, while embedded within the vorticity and PV of the PS warm front, is evident from the distinct local maximum in low-level PV ~750 km to the northeast of the low-level circulation and SLP minimum of the PS.

b. Explosive deepening (phase two): 30–31 October

The explosive deepening and second development phase of EC1 began poleward of remnant tropical moisture associated with HG at 0000 UTC 30 October (Fig. 10a). EC1 weakened in the 12-h period ending at 0000 UTC 30 October, but was likely maintained by enhanced surface SH fluxes > (50–100) W m−2 over the North Atlantic (Fig. 10a). The SH fluxes were enhanced in conjunction with low-level cold air residing over relatively warmer water in the wake of a large antecedent cyclone over the central North Atlantic at 0000 UTC 30 October. The enhanced SH fluxes contributed to a ~20% reduction in the low-level atmospheric static stability (not shown). As a result, weak slantwise ascent in the presence of reduced static stability was likely sufficient in maintaining (or fending off the dissipation of) EC1 between 1200 UTC 29 October and 0000 UTC 30 October.

Fig. 10.

As in Fig. 6, but over the northeast Atlantic with sensible heat flux contoured every 50 W m−2 starting at (a) 50 W m−2 and shading for the 850-hPa potential temperature gradient beginning at 2.0 K (100 km)−1.

Fig. 10.

As in Fig. 6, but over the northeast Atlantic with sensible heat flux contoured every 50 W m−2 starting at (a) 50 W m−2 and shading for the 850-hPa potential temperature gradient beginning at 2.0 K (100 km)−1.

The low-level EC1 PV maximum at 0000 UTC 30 October was located on the anticyclonic shear-side of the North Atlantic jet stream and equatorward of an upper-tropospheric PVD that originated over the Arctic (i.e., PVD XA; Fig. 11a). EC1 rapidly developed characteristics of a Norwegian-type cyclone within the exit region of the North Atlantic jet stream as the low-level thermal ridge rapidly narrowed and the minimum SLP decreased 23 hPa from 983 to 960 hPa between 0000 and 1200 UTC 30 October (Fig. 10b). The minimum SLP of EC1 decreased an additional 8 to 952 hPa by 0000 UTC 31 October (Fig. 10c). The second development phase of EC1 occurred as EC1 “crossed” the axis of the North Atlantic jet stream downstream of PVD XA and the DT “lowered” adjacent to EC1 between 0000 UTC 30 October and 0000 UTC 31 October (Figs. 11c,d). The lowering of the DT, suggested by increasing DT pressure to the west of EC1 (i.e., PVD X1), is consistent with a vertical and lateral coupling of EC1 with PVD XA. The coupling of EC1 with PVD XA occurred in conjunction with an increase in low-level PV from ~1.5 to >2.5 PVU.

Fig. 11.

As in Fig. 7, but over the northeast Atlantic and with 850-hPa PV contoured every 0.5 PVU starting at 1.0 PVU. Cross section lines shown for Fig. 12.

Fig. 11.

As in Fig. 7, but over the northeast Atlantic and with 850-hPa PV contoured every 0.5 PVU starting at 1.0 PVU. Cross section lines shown for Fig. 12.

c. Vertical evolution

The first development phase of EC1 was associated with midtropospheric estimates of > 20 K day−1, evident from cross sections through PVD XP and EC1 at 0000 UTC 29 October, and PVD XA and EC1 at 1200 UTC 29 October (Figs. 12a,b, respectively). Lower tropospheric PV values >2.0 PVU, located in regions of upward-increasing diabatic heating rates, suggested that latent heating associated with DMC contributed to low-level PV generation within EC1 during the first phase of development. Low-level diabatic PV tendencies, estimated from (8) during the first phase of development, exceeded 2 PVU day−1. Oppositely, upward-decreasing values of diabatic heating (above ~500 hPa) contributed to a relative minimum in middle- and upper-tropospheric PV (Figs. 12a,b). Upper-level estimates of the diabatic PV tendency were ~−0.5 PVU day−1. The minimum in upper-tropospheric PV and negative diabatic PV tendency subsequently enhanced the north–south upper-tropospheric PV gradient, contributed to the intensification of the North Atlantic jet stream, and likely played a role in negative upper-level PV tendencies proximate PVD XP discussed in section 4b.

Fig. 12.

The EC1 cross-section analysis of potential vorticity (shaded according to scale in PV units), Lagrangian estimate of diabatic heating (white contours every 5 K day−1 with negative values dashed), potential temperature (black contours every 4 K), and horizontal wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1) at 0000 29 Oct and 1200 UTC 30 Oct. Cross-section lines are shown in Figs. 4 and 10. (b),(c) The colored arrows represent the approximate location of select airstreams flowing orthogonal to the cross section discussed in text.

Fig. 12.

The EC1 cross-section analysis of potential vorticity (shaded according to scale in PV units), Lagrangian estimate of diabatic heating (white contours every 5 K day−1 with negative values dashed), potential temperature (black contours every 4 K), and horizontal wind (half barb = 2.5 m s−1; full barb = 5.0 m s−1; pennant = 25.0 m s−1) at 0000 29 Oct and 1200 UTC 30 Oct. Cross-section lines are shown in Figs. 4 and 10. (b),(c) The colored arrows represent the approximate location of select airstreams flowing orthogonal to the cross section discussed in text.

EC1 rapidly crossed the North Atlantic with a translation speed >25 m s−1 and was located equatorward of PVD XA at 0000 UTC 30 October (Fig. 12c). The explosive deepening of EC1 during the second development phase occurred as 1) EC1 became coupled with PVD XA between 0000 and 1200 UTC 30 October (Figs. 12c,d), and 2) midtropospheric estimates of the diabatic heating rate exceeded 25 K day−1 over EC1. The low-level EC1 PV maximum was located in a region of upward-increasing diabatic heating, suggesting that latent heating associated with DMC contributed to low-level PV generation within EC1. Low-level estimates of the diabatic PV tendency exceeded 2.5 PVU day−1.

As discussed above and observed in Fig. 12, low-level PV generation associated with the two-phase development of EC1 occurred downstream of three distinct upper-tropospheric PVDs on 28–29 October (XP) and on 30–31 October (XA and X1). The Rossby radius of deformation (R0) associated with the upper-tropospheric PVDs and the horizontal separation distance between the upper-tropospheric PVDs and EC1 is estimated for select times between 28 and 31 October (Table 1). The horizontal separation distance between these upper-tropospheric PVDs and EC1 slowly decreased from ~550 km on 28 October to <100 km on 31 October as R0 remained ~(400–500) km. The horizontal separation distances and R0 values suggest a possible, but likely small, dynamical influence of PVD XP on the first phase of development, and a more significant dynamical influence of PVD XA and X1 on the second phase of development.

Table 1.

Comparison of the separation distance D between PVDs XP, XA, and X1 and EC1 and the approximate Rossby radius of deformation (R0) for select times.

Comparison of the separation distance D between PVDs XP, XA, and X1 and EC1 and the approximate Rossby radius of deformation (R0) for select times.
Comparison of the separation distance D between PVDs XP, XA, and X1 and EC1 and the approximate Rossby radius of deformation (R0) for select times.

d. Lagrangian diagnostic

Air parcel trajectories were generated for 36-h periods ending (or beginning) in regions where 600-hPa () exceeded various thresholds during the evolution of EC1 at 1200 UTC 28 October and 1200 UTC 30 October (Figs. 1315, respectively). The Lagrangian diagnostic is used to ascertain the origin of air parcels that likely contributed to maxima in midtropospheric diabatic heating estimated in Fig. 12. As a result, the Lagrangian analysis assesses the origin of air parcels that produced increasing diabatic heating rates with height and low-level diabatic PV generation within EC1.

Backward trajectories for 25 air parcels with >10 K day−1 at 1200 UTC 28 October during the first development phase of EC1 indicate that parcels originated in the lower troposphere to the southeast of HG at 0000 UTC 27 October (Fig. 13a). Air parcels ascended to the east of EC1 and contained increasing PT and PV values between 0000 and 1200 UTC 28 October, consistent with slantwise ascent and ultimately a maximum in midtropospheric diabatic heating (Figs. 13a–c). The increasing PT and PV values along these air parcel trajectories likely provide a downstream source for low-level PV generation within EC1. Forward trajectories for this population show that these air parcels comprised an airstream consistent with a warm conveyor belt (e.g., Carlson 1980) as they ascended to above 300 hPa and rapidly crossed the North Atlantic at ~50 m s−1 on the anticyclonic shear side of the North Atlantic jet stream (Fig. 14). Most of these forward air parcels were associated with a net decrease in PV of ~0.5 PVU as they ascended above the region of maximum diabatic heating between 1200 UTC 28 October and 0600 UTC 29 October (Fig. 14b). A few air parcels were associated with large increases in PV as they passed very near or through PVD XA between 0000 and 1200 UTC 29 October. The locations of these trajectories are likely incorrect and the result of the interpolation scheme used to locate air parcels between six-hour analysis times as discussed in section 2.

Fig. 13.

(a) Backward air parcel trajectories for the 36-h period ending at 1200 UTC 28 Oct shaded according to their pressure elevation (hPa, according to color-bar scale). Air parcels end at 600 hPa within the region of estimated diabatic heating rates () >10 K day−1 illustrated by the thin dashed contour. Diabatic heating rates are shaded every 5 K day−1 according to gray-bar scale. The trajectories are labeled sequentially at their “endpoints” and contain white-filled circles every 12 h. The locations of the PS, HG, and EC1 at times A through D are given by the subscripted markers P, G, and 1, respectively. Population–mean trajectory shown in black. (b),(c) The potential temperature and potential vorticity values for trajectories drawn in (a) with the population–mean values denoted in red.

Fig. 13.

(a) Backward air parcel trajectories for the 36-h period ending at 1200 UTC 28 Oct shaded according to their pressure elevation (hPa, according to color-bar scale). Air parcels end at 600 hPa within the region of estimated diabatic heating rates () >10 K day−1 illustrated by the thin dashed contour. Diabatic heating rates are shaded every 5 K day−1 according to gray-bar scale. The trajectories are labeled sequentially at their “endpoints” and contain white-filled circles every 12 h. The locations of the PS, HG, and EC1 at times A through D are given by the subscripted markers P, G, and 1, respectively. Population–mean trajectory shown in black. (b),(c) The potential temperature and potential vorticity values for trajectories drawn in (a) with the population–mean values denoted in red.

Fig. 14.

(a) As in Fig. 13a, but for forward air parcel trajectories for the 36-h beginning at 1200 UTC 28 Oct atop the magnitude of the 300-hPa wind speed (contoured in black every 10 m s−1 starting at 50 m s−1 at 0600 UTC 29 Oct). The (b) potential vorticity and (c) zonal wind speed values for these trajectories, with the population–mean values drawn in red. The population mean excludes all trajectories drawn in blue. Letters A–D correspond to times inset in (a).

Fig. 14.

(a) As in Fig. 13a, but for forward air parcel trajectories for the 36-h beginning at 1200 UTC 28 Oct atop the magnitude of the 300-hPa wind speed (contoured in black every 10 m s−1 starting at 50 m s−1 at 0600 UTC 29 Oct). The (b) potential vorticity and (c) zonal wind speed values for these trajectories, with the population–mean values drawn in red. The population mean excludes all trajectories drawn in blue. Letters A–D correspond to times inset in (a).

Backward and forward trajectories for 21 air parcels with > 25 K day−1 at 1200 UTC 29 October indicate that air parcels originated in the same low-level subtropical environment to the southeast of HG at 0000 UTC 28 October and terminated at upper levels over the eastern North Atlantic at 0000 UTC 31 October (not shown). The backward and forward trajectories suggest that air parcels contributing to maxima in midtropospheric diabatic heating downstream of EC1 were subsequently associated with decreases in upper-level PV on the anticyclonic shear side of the North Atlantic jet stream, increases in the meridional PV gradient, and an intensification of the upper-level westerly flow (Fig. 14c). Backward trajectories were also generated from regions of upper-tropospheric low PV air on the anticyclonic shear-side of the North Atlantic jet stream at 1200 UTC 29 October and 0000 UTC 30 October (not shown). These two time periods coincided with the intensification of the North Atlantic jet stream and were discussed previously in association with negative upper-tropospheric PV tendencies over the North Atlantic. A large majority >90% of these air parcels originated proximate to the lower troposphere Hurricane Grace and experienced large increases (decreases) in PT (PV) as they ascended and accelerated across the North Atlantic, supporting a strong diabatic influence on the intensification of the North Atlantic jet stream.

Backward trajectories for 40 air parcels with > 20 K day−1 at 1200 UTC 30 October during the second development phase of EC1 indicate that a majority of these parcels originated in the lower troposphere well to the east of HG at 0000 UTC 29 October and comprised a warm conveyor belt (Fig. 15a). The approximate location of these air parcels can be observed equatorward of EC1 at 0000 UTC 30 October as depicted by the red arrow in Fig. 12c. However, a portion of the total air parcel population originated at upper levels over the Labrador Sea at 0000 UTC 29 October. The approximate location of these air parcels can be observed equatorward of PVD XA at 1200 UTC 29 October and 0000 UTC 30 October as depicted by blue arrows in Figs. 12b,c, respectively. The backward trajectories illustrate that air parcels following the warm conveyor belt abruptly rose from ~950 to 600 hPa and merged with descending air parcels with source regions over the Labrador Sea. The merger of these two airstreams, containing notably different characteristics in PT (Fig. 15c), was associated with an increase in the tropospheric PT and PV gradients proximate PVD XA and EC1 at 1200 UTC 30 October. As a result, the increased PT and PV gradients locally enhanced dynamical forcing for ascent downstream of PVD XA and over EC1 during the period of coupling, contributing to large diabatic heating rates and low-level PV generation within EC1 (Fig. 15b).

Fig. 15.

As in Fig. 13, but with air parcel trajectories ending at 1200 UTC 30 Oct and > 20 K day−1. The (b) potential temperature and (c) potential vorticity values for these trajectories, and are colored by their source region: North Atlantic (red) and Labrador Sea (blue). Population–mean values are denoted by the heavy solid contours.

Fig. 15.

As in Fig. 13, but with air parcel trajectories ending at 1200 UTC 30 Oct and > 20 K day−1. The (b) potential temperature and (c) potential vorticity values for these trajectories, and are colored by their source region: North Atlantic (red) and Labrador Sea (blue). Population–mean values are denoted by the heavy solid contours.

e. Energetics

The quasi-Lagrangian evolution of the diabatic generation of eddy APE (GE) and the baroclinic generation of eddy APE (CA) was calculated to identify dynamical contributions to the development of EC1 (Fig. 16). The magnitude of the diabatic generation GE compared to the baroclinic CA generation of eddy APE is a useful means to differentiate between the dynamics associated with baroclinic Rossby waves and diabatic Rossby waves (i.e., DRVs; Parker and Thorpe 1995; Moore and Montgomery 2004, 2005; Moore et al. 2008). Typically, synoptic-scale baroclinic waves are dominated by baroclinic generation CA in association with large contributions to eddy APE generation via thermal advection, whereas mesoscale DRVs are dominated by diabatic generation GE in association with large contributions to eddy APE generation via latent heat release and diabatic heating. Figure 16 illustrates that diabatic generation GE > 0.5 kg s−3 was the dominant source for eddy APE generation during the first development phase of EC1 between 1200 UTC 28 October and 0600 UTC 29 October. The first development phase coincided with a SLP decrease from 1008 to 994 hPa and an increase in 850-hPa PV from 0.9 to 2.3 PVU. As compared to Moore and Montgomery (2004, 2005) and Moore et al. (2008), the first development phase of EC1 appears to be energetically consistent with a DRV-like feature (Fig. 16).

Fig. 16.

Lagrangian time series of the minimum sea level pressure (hPa; short-dashed contour), baroclinic conversion of basic-state APE to eddy APE (CA; kg s−3; long-dashed contour), diabatic generation of eddy APE (GE; kg s−3; solid contour), and the 850-hPa maximum PV (PVU; inset bar chart) computed for a ~500 km × ~500 km box centered on EC1.

Fig. 16.

Lagrangian time series of the minimum sea level pressure (hPa; short-dashed contour), baroclinic conversion of basic-state APE to eddy APE (CA; kg s−3; long-dashed contour), diabatic generation of eddy APE (GE; kg s−3; solid contour), and the 850-hPa maximum PV (PVU; inset bar chart) computed for a ~500 km × ~500 km box centered on EC1.

Diabatic generation GE decreased from >1.0 to <−0.3 kg s−3 during the transition period from the first to second development phases between 1200 and 1800 UTC 29 October (Fig. 16). This transition period was associated with EC1 minimum SLP values increasing above 1000- and 850-hPa PV values decreasing below 2.0 PVU. Baroclinic generation CA increased from ~0.2 to ~0.4 kg s−3 and became the dominant source for eddy APE generation during the second development phase on 30 October. However, subsequent brief maxima in diabatic generation GE were observed at 1200 UTC 30 October and 0000 UTC 31 October during the period of coupling between PVD XA and EC1. The maximum in baroclinic generation CA (and brief maxima in diabatic generation GE) occurred in conjunction with a decrease in the EC1 minimum SLP to 952 hPa and an increase in 850-hPa PV to 3.4 PVU at 1800 UTC 30 October. The baroclinic generation CA and diabatic generation GE subsequently decreased to ~0.0 kg s−3 as EC1 slowly filled on 31 October. The second development phase of EC1 appears to be more energetically consistent with a synoptic-scale baroclinic wave, influenced by brief diabatic contributions. The two-phase development process in which phase 1 (2) was dominated by the diabatic generation GE (baroclinic generation CA) of eddy APE suggests the evolution of EC1 followed a process similar to the DRV cyclogenesis pathway described in section 1c.

6. Discussion

The synoptic, Lagrangian, and energetic perspectives on the development and evolution of the PS and EC1 illustrate that the structure of HG following TT and during its recurvature and ET may have created conditions favorable for subsequent cyclogenesis over the western North Atlantic. This discussion focuses on how HG likely contributed to the development of EC1 and the PS, and is followed by comments on the subsequent evolution of EC1 as a DRV-like feature and the warm seclusion of the PS.

a. Influence of HG on the development of EC1 and the PS

The TT of HG occurred following the weak extratropical cyclone precursor pathway in which an incipient PV streamer focused and organized convection. Convection to the east of pre-HG slowly weakened the vertical wind shear and produced a favorable environment for TT [compare Fig. 2 to Fig. 3 of Davis and Bosart (2004)]. Recall that Hulme and Martin (2009) show that a region of convection poleward of the TC undergoing TT is situated along a zone of enhanced baroclinicity and is typically characterized by frontogenesis, significant latent heat release, and diabatic PV generation. In the present case, this region of enhanced baroclinicity to the northeast of HG was a favorable region for air parcels to ascend to their lifted condensation level and induce latent heat release and diabatic PV generation below the level of maximum diabatic heating. The path of air parcels ending at 1200 UTC 28 October, derived from the Lagrangian analysis (Fig. 13a), support a hypothesis suggesting that weak slantwise ascent near the region of enhanced baroclinicity associated with the TT of HG contributed to low-level PV generation during the first development phase of EC1. The dominant contribution of the diabatic generation GE of eddy APE during the first development phase of EC1 (Fig. 16) additionally suggests that this region of enhanced baroclinicity contributed to the development of EC1 as a DRV-like feature on 28 October.

As EC1 exited the western North Atlantic along the PS warm front, the structure of HG during its recurvature and ET on 29 October likely influenced the coincident development of the PS. The region of enhanced baroclinicity poleward of HG, characterized by latent heat release and diabatic PV generation, aided the development of a strong warm front to the east of the PS in a manner similar to poleward-moving TC remnants discussed in section 1b. The process of warm frontal development to the north and east of poleward-moving TC remnants, however, was likely exacerbated in the present case given the a priori enhanced baroclinic zone associated with the TT of HG.

A trajectory analysis suggests that air parcels comprising the midtropospheric maximum in diabatic heating rates over the PS warm front (and downstream of EC1) originated to the east of HG, ascended, and subsequently contributed to decreases in upper-level PV on the anticyclonic shear side of the North Atlantic jet stream. The decrease in upper-level PV facilitated an increase in the meridional PV gradient downstream of PVD XP and an intensification of the North Atlantic jet stream (Figs. 12a and 14, respectively). The negative upper-tropospheric PV tendencies and ascent of these air parcels are likely an indication that latent heat release and DMC associated with the ascent of tropical air modified the environment downstream of PVD XP and produced conditions favorable for the rapid cyclogenesis of the PS. The effect of latent heat release and diabatic heating downstream of PVD XP and over the PS compares favorably with the self-development cyclogenesis paradigm (e.g., Sutcliffe and Forsdyke 1950). As hypothesized, the region of reduced atmospheric static stability, enhanced warm air advection, and latent heat release induced by the recurvature and ET of HG was likely in phase with the extratropical forcing for ascent ahead of PVD XP, and provides a focusing mechanism for rapid cyclone intensification consistent with quasigeostrophic theory and PV thinking (e.g., Agustí-Panareda et al. 2005; Agustí-Panareda 2008).

The subsequent warm seclusion of the PS may have additionally been influenced by the increase in the meridional PV gradient downstream of PVD XP and the intensification of the North Atlantic jet stream. Schultz et al. (1998) demonstrate that cyclones tend to undergo the warm seclusion process when large-scale confluent flow is found downstream (i.e., within the entrance region of the North Atlantic jet stream). A large-scale confluent flow pattern generally acts to orient axes of dilatation zonally over the cyclone, favoring warm frontogenesis and a strong zonally oriented warm front (cf. Fig. 10b in Schultz et al. 1998). The localized effect of latent heat release and diabatic heating downstream of PVD XP, however, suggests that these processes only contributed to or accelerated a predefined evolution of the large-scale flow over the North Atlantic, likely more influenced by the large-scale antecedent conditions upstream (e.g., Cordeira and Bosart 2010). This hypothesis could be further assessed using numerical modeling to quantify the influence of latent heat release downstream of PVD XP on the subsequent evolution of the PS, but is beyond the scope of the current investigation.

b. Comments on the evolution of EC1

Conclusive results on the development of EC1 as a DRV on 28 October are complicated by the proximity of the upstream upper-level trough (PVD XP) over the northwest Atlantic (Figs. 4c and 12a). The ~650 km horizontal separation between these two features was slightly larger than the estimated ~450 km Rossby radius of deformation (Table 1). The marginal separation distance between these two features as compared to the Rossby radius of deformation suggests that convection during the development of EC1 may have been influenced by dynamical forcing for ascent and convective destabilization in the presence of tropical moisture downstream of PVD XP on 28–29 October. For example, convective destabilization in the PS warm sector was observed in the thermodynamic profile over the North Atlantic between 1200 UTC 28 October and 1200 UTC 29 October (Fig. 5). The modification of these profiles was likely associated with isentropic ascent and enhanced warm-air advection downstream of PVD XP, subsequent saturation and cooling, and attendant changes to the convective environment (e.g., Juckes and Smith 2000). Although it is quantitatively difficult to determine the influence of the PVD on the development of EC1 as a DRV without a more rigorous diagnostic such as PV inversion, the Lagrangian and energetic diagnostics presented suggest that lower-tropospheric PV generation during the first development phase of EC1 occurred mostly independent of PVD XP.

Further evidence for the DRV-like development of EC1 is garnered from the apparent two-phase evolution of EC1 characterized by two periods of decreasing minimum SLP values and increasing 850-hPa PV values, the rapid translation speed >25 m s−1 of EC1 across the North Atlantic, and the transition from diabatic generation GE to baroclinic generation CA of eddy APE (Fig. 16). The second and explosive deepening phase of EC1 occurred with the interaction between, and coupling of, PVD XA with EC1 as suggested by the merger of two differing airstreams comprising the region of midtropospheric diabatic heating at 1200 UTC 30 October (Fig. 15). This second development phase closely resembled an upper-level-induced cyclogenesis pathway (Petterssen 1955; Petterssen and Smebye 1971) and the coupling of upper- and lower-tropospheric PVDs (Hoskins et al. 1985). The characteristics of EC1 during the two-phase development process are remarkably similar to DRV-like developments documented by Wernli et al. (2002) and Moore et al. (2008).

The second development phase of EC1 was additionally associated with EC1 crossing the axis of the North Atlantic jet stream (Fig. 17). EC1 moved along the anticyclonic shear side of the North Atlantic jet stream with > 20 K day−1 poleward of the cyclone and equatorward of the jet axis between 0000 UTC 29 October and 0000 UTC 30 October (Fig. 17a). The midtropospheric maximum in diabatic heating poleward of EC1 likely influenced negative upper-tropospheric PV tendencies and a steepening of the isentropic surfaces between EC1 and PVD XA. As the horizontal separation distance between PVD XA and EC1 decreased to within the Rossby radius of deformation, the two features became laterally and vertically coupled by ~0600 UTC 30 October (Figs. 9 and 11). The EC1 crossed the jet stream axis as suggested by a maximum in deep-layer shear >15 m s−1 over EC1 by 0600 UTC 30 October (Fig. 17b) and became collocated with PVD X1 and a region with > 30 K day−1 by 1200 UTC 30 October.

Fig. 17.

(a) Six-hour locations of EC1 and PVDs XA and X1 with the maximum grid-point values of the 300-hPa wind speed (m s−1; shaded according to scale) between 0000 UTC 29 Oct and 1800 UTC 30 Oct. The estimated diabatic heating rate is color contoured every 5 K day−1 starting at 25 K day−1. (b) Time series of the deep-layer wind shear [m s−1; solid (defined as the magnitude of the 850-hPa wind subtracted from 200-hPa wind)] and the vertically averaged 925–200-hPa local domain-maximum meridional PV gradient [PVU (1000 km)−1; dashed] averaged over a ~500 km × ~500 km domain centered on EC1.

Fig. 17.

(a) Six-hour locations of EC1 and PVDs XA and X1 with the maximum grid-point values of the 300-hPa wind speed (m s−1; shaded according to scale) between 0000 UTC 29 Oct and 1800 UTC 30 Oct. The estimated diabatic heating rate is color contoured every 5 K day−1 starting at 25 K day−1. (b) Time series of the deep-layer wind shear [m s−1; solid (defined as the magnitude of the 850-hPa wind subtracted from 200-hPa wind)] and the vertically averaged 925–200-hPa local domain-maximum meridional PV gradient [PVU (1000 km)−1; dashed] averaged over a ~500 km × ~500 km domain centered on EC1.

The propensity for strong cyclones to have a cross-jet or poleward component to their movement in the NH is well known (e.g., Simmons and Hoskins 1978; Hoskins and West 1979; Wallace et al. 1988; Schär and Wernli 1993; Takayabu 1991; Plu and Arbogast 2005), and was frequently observed during the Fronts and Atlantic Storm Track Experiment (FASTEX; Baehr et al. 1999). Takayabu (1991) and Rivière (2008) suggest that a nonlinear interaction between upper- and lower-tropospheric PVDs may lead to the poleward displacement of low-level cyclones. Shapiro (1992) and Gilet et al. (2009) demonstrate that the poleward displacement of a cyclone occurs in association with an enhanced vertically averaged meridional PV gradient. The physical process modulating the strength of the vertically averaged meridional PV gradient during the coupling of upper- and lower-tropospheric PVDs occurs in association with an enhancement of the nonlinear stretching term that contributes to the low-level relative vorticity tendency (Gilet et al. 2009). As a result, the lower-tropospheric PVD experiences a poleward shift relative to the upper-tropospheric PVD that is consistent with an enhancement of the vertically averaged PV gradient. In the present case, the vertically averaged meridional PV gradient averaged over EC1 was likely enhanced in association with large midtropospheric diabatic heating rates, and reached a maximum during the jet crossing at 0600 UTC 30 October (Fig. 17b).

The explosive cyclogenesis and jet crossing observed by EC1 occurred during the nonlinear interaction between EC1 and PVD XA. Given a DRV-like feature such as EC1, however, explosive cyclogenesis and the jet-crossing process appear to be modulated by the presence of an intense upper-tropospheric PVD and the maintenance of the DRV-like feature until it can interact with the upper-tropospheric PVD. These factors likely contribute to results that identify ~15% (~5%) of DRVs develop into intense cyclones with minimum SLP values below 990 (970) hPa (Kenzelmann 2005). In the present case, the equatorward displacement (or extraction) of PVD XA and the maintenance of EC1 by enhanced SH fluxes over the North Atlantic were likely crucial ingredients in the subsequent explosive cyclogenesis, jet crossing, and second development phase of EC1. These processes support initial findings by Cordeira and Bosart (2010) that suggested the antecedent large-scale flow regime produced conditions favorable for the development of the PS and EC1.

c. Comments on the evolution of the PS

The Lagrangian analysis performed for the warm seclusion phase of the PS illustrated that air parcels encircled the bent-back warm front, warmed, and became secluded as a warm core (Fig. 9). The two air parcel populations described in section 4c showed that the air inside the warm core originated poleward of the PS (e.g., the early-ascending population) and that this air was encircled by more rapidly moving air originating farther poleward (e.g., the later-ascending population), and corroborates similar findings by Shapiro and Keyser (1990) and Kuo et al. (1992). The early-ascending population ascended from 0000 UTC 29 October to 0000 UTC 30 October, descended from 0000 to 0600 UTC 30 October, and subsequently ascended to their termination pressure at 850-hPa by 1200 UTC 30 October (Figs. 9a,b). The PT of these air parcels increased ~18 K in the 24-h period prior to making the cyclonic turn around the bent-back warm front over the Gulf Stream at ~0000 UTC 30 October. The later-ascending population lagged the early-ascending parcels by ~6 h (Fig. 9a). These air parcels ascended from ~950 hPa to their termination pressure at 850-hPa between 0000 and 1200 UTC 30 October. On average, the PT of the later-ascending parcels increased ~21 K in the 24-h period prior to making the cyclonic turn around the bent-back warm front over the Gulf Stream at ~0600 UTC 30 October.

The diabatic temperature tendency via SH-flux convergence from (9) is used to explore the large lower-tropospheric increases in PT computed during the warm seclusion of the PS. Diabatic temperature tendencies exceeded 15 K day−1 for the early-ascending air parcels at ~1200 UTC 29 October and for the later-ascending air parcels at ~0000 UTC 30 October if SH was absorbed evenly over a depth of ~100 hPa between the surface and air parcel pressures (Fig. 18). To first order, the calculated diabatic temperature tendencies balance the thermodynamic equation locally when the magnitude of cold air advection over the western North Atlantic is considered (not shown). Corresponding SH flux values along the paths of these air parcels are ~200 W m−2 for both early- and later-ascending air parcels at 1200 UTC 29 October and 0000 UTC 30 October, respectively (Fig. 18; also contoured in Fig. 5a at 0000 UTC 30 October). The SH flux values of ~200 W m−2 are commonly observed during cold air outbreaks over the Gulf Stream (e.g., Bunker 1976). Warming via sensible heating of ~(15–20) K day−1 is approximately equal to the computed diabatic increases in PT following the air parcel clusters between 1200 UTC 29 October and 1200 UTC 30 October. These results suggest that the warm seclusion of the PS involved the seclusion of air parcels by the bent-back warm front that were warmed in conjunction with sensible heating from the underlying Gulf Stream.

Fig. 18.

Time series of SH flux (W m−2) and diabatic temperature tendency (K day−1) for the 26 air parcel trajectories in Fig. 8. Trajectories are divided into early-ascending parcels (dashed; N = 9) and later-ascending parcels (solid; N = 17) as defined in section 3. The diabatic temperature tendency is calculated from (8) with Δp = 100 hPa.

Fig. 18.

Time series of SH flux (W m−2) and diabatic temperature tendency (K day−1) for the 26 air parcel trajectories in Fig. 8. Trajectories are divided into early-ascending parcels (dashed; N = 9) and later-ascending parcels (solid; N = 17) as defined in section 3. The diabatic temperature tendency is calculated from (8) with Δp = 100 hPa.

7. Conclusions

This investigation focused on interactions between HG, the PS, and EC1 over the North Atlantic in late October and early November 1991. The TT, recurvature, and ET of HG occurred within the warm sector of the PS on 25–29 October and likely enhanced the development of the PS by reducing atmospheric static stability, enhancing warm air advection, and inducing strong latent heat release downstream of upper-tropospheric PVD XP. The evolution of the PS closely resembled the Shapiro and Keyser (1990) cyclone model, with strong warm frontogenesis located within the entrance region of the North Atlantic jet stream and the subsequent development of a bent-back warm front over the western North Atlantic. A Lagrangian trajectory analysis of the PS at 1200 UTC 30 October demonstrated that the warm seclusion process involved the isolation of air parcels by the bent-back warm front that were likely warmed in conjunction with large sensible heat fluxes from the underlying Gulf Stream.

The evolution of EC1 contained two distinct phases of development, denoted by two periods of decreasing SLP and increasing lower-tropospheric PV. The first phase of development occurred along a region of enhanced baroclinicity and convection poleward of HG that formed during the TT of HG, whereas the second phase of development occurred as EC1 rapidly cross the North Atlantic and interacted with upper-tropospheric PVD XA. A Lagrangian trajectory analysis demonstrated that air parcels to the east of HG, in conjunction with tropical moisture, contributed to diabatic lower-tropospheric PV production between 1200 UTC 28 October and 29 October as they ascended to the east of EC1 during the first phase of development. An analysis of the eddy APE tendency equation confirmed that the first phase of development was dominated by the diabatic generation (GE) of eddy APE. Subsequently, the Lagrangian trajectory analysis demonstrated that the merger of two distinct airstreams on 30 October from the region to the east of HG and from over the Labrador Sea contributed to the second phase of development. The eddy APE tendency equation showed that the baroclinic generation (CA) of eddy APE dominated this latter phase of development as EC1 crossed the axis of the North Atlantic jet stream and explosively deepened on 30 October. The temporal evolution of the Lagrangian and energetics analyses suggested EC1 initially evolved as a DRV-like feature and followed the DRV-cyclogenesis pathway observed by Wernli et al. (2002) and Moore et al. (2008).

The likely contributions of HG to the development and evolution of the PS and EC1 were associated with reduced atmospheric static stability, enhanced warm air advection and strong latent heat release in the presence of tropical moisture and enhanced baroclinicity over the North Atlantic. It is difficult, however, to quantitatively determine the dynamical influence of HG on the subsequent warm seclusion of the PS and the DRV-like development of EC1. Forthcoming work will incorporate the use of the Advanced Research core of the Weather Research and Forecasting Model (WRF-ARW) to simulate and explore the influence of HG on the development and evolution of the PS and EC1.

Acknowledgments

This research was supported by National Science Foundation (NSF) Grants ATM-0304254 and ATM-0553017. Discussions with Thomas Galarneau, Jr. and Alan Srock (University at Albany, SUNY) contributed to the manuscript. The authors also thank Dr. Anantha Aiyyer (North Carolina State University) for developing software to calculate air parcel trajectories. This manuscript was greatly improved by comments and suggestions from two anonymous reviewers. Data was provided by the National Climatic Data Center, the ECMWF, and the Data Support Section of the Computational and Information Systems Laboratory at NCAR.

REFERENCES

REFERENCES
Agustí-Panareda
,
A.
,
2008
:
The contribution of ex–Tropical Cyclone Gert (1999) toward the weakening of a midlatitude cyclogenesis event
.
Mon. Wea. Rev.
,
136
,
2091
2111
.
Agustí-Panareda
,
A.
,
S. L.
Gray
,
G. C.
Craig
, and
C. D.
Thorncroft
,
2005
:
The extratropical transition of Tropical Cyclone Lili (1996) and its crucial contribution to a moderate extratropical development
.
Mon. Wea. Rev.
,
133
,
1562
1573
.
Baehr
,
C.
,
B.
Pouponneau
,
F.
Ayrault
, and
A.
Joly
,
1999
:
Dynamical characterization of the FASTEX cyclogenesis cases
.
Quart. J. Roy. Meteor. Soc.
,
125
,
3469
3494
.
Bluestein
,
H.
,
1992
:
Synoptic-Dynamic Meteorology in Midlatitudes
.
Vol. I, Principles of Kinematics and Dynamics, Oxford University Press, 431 pp
.
Bosart
,
L. F.
,
1999
:
Observed cyclone life cycles
.
The Life Cycles of Extratropical Cyclones, M. A. Shapiro and S. Grønås, Eds., Amer. Meteor. Soc., 187–213
.
Bosart
,
L. F.
,
G. J.
Hakim
,
K. R.
Tyle
,
M. A.
Bedrick
,
W. E.
Bracken
,
M. J.
Dickinson
, and
D. M.
Schultz
,
1996
:
Large-scale antecedent conditions associated with the 12–14 March 1993 cyclone (“Superstorm ’93”) over eastern North America
.
Mon. Wea. Rev.
,
124
,
1865
1891
.
Bunker
,
A. F.
,
1976
:
Computations of surface energy flux and annual air–sea interaction cycles of the North Atlantic Ocean
.
Mon. Wea. Rev.
,
104
,
1122
1140
.
Buzzi
,
A.
, and
S.
Tibaldi
,
1978
:
Cyclogenesis in the lee of the Alps: A case study
.
Quart. J. Roy. Meteor. Soc.
,
104
,
271
287
.
Cardone
,
V. J.
,
R. E.
Jensen
,
D. T.
Resio
,
V. R.
Swail
, and
A. T.
Cox
,
1996
:
Evaluation of contemporary ocean wave models in rare extreme events: the “Halloween Storm” of October 1991 and the “Storm of the Century” of March 1993
.
J. Atmos. Oceanic Technol.
,
13
,
198
230
.
Carlson
,
T. N.
,
1980
:
Airflow through mid-latitude cyclones and the comma cloud pattern
.
Mon. Wea. Rev.
,
108
,
1498
1509
.
Conzemius
,
R. J.
,
R. W.
Moore
,
M. T.
Montgomery
, and
C. A.
Davis
,
2007
:
Mesoscale convective vortex formation in a weakly sheared moist neutral environment
.
J. Atmos. Sci.
,
64
,
1443
1466
.
Cordeira
,
J. M.
, and
L. F.
Bosart
,
2010
:
The antecedent large-scale conditions of the “Perfect Storms” of late October and early November 1991
.
Mon. Wea. Rev.
,
138
,
2546
2569
.
Davis
,
C. A.
, and
L. F.
Bosart
,
2001
:
Numerical simulations of the genesis of Hurricane Diana. Part I: Control simulation
.
Mon. Wea. Rev.
,
129
,
1859
1881
.
Davis
,
C. A.
, and
L. F.
Bosart
,
2003
:
Baroclinically induced tropical cyclogenesis
.
Mon. Wea. Rev.
,
131
,
2730
2747
.
Davis
,
C. A.
, and
L. F.
Bosart
,
2004
:
The TT problem: Forecast the tropical transition of cyclones
.
Bull. Amer. Meteor. Soc.
,
85
,
1657
1662
.
Deveson
,
A. C. L.
,
K. A.
Browning
, and
T. D.
Hewson
,
2002
:
A classification of FASTEX cyclones using a height-attributable quasi-geostrophic vertical-motion diagnostic
.
Quart. J. Roy. Meteor. Soc.
,
128
,
93
117
.
Dickinson
,
M. J.
,
L. F.
Bosart
,
W. E.
Bracken
,
G. J.
Hakim
,
D. M.
Schultz
,
M. A.
Bedrick
, and
K. R.
Tyle
,
1997
:
The March 1993 superstorm cyclogenesis: Incipient phase synoptic- and convective-scale flow interaction and model performance
.
Mon. Wea. Rev.
,
125
,
3041
3072
.
Farrell
,
B.
,
1984
:
Modal and nonmodal baroclinic waves
.
J. Atmos. Sci.
,
41
,
668
673
.
Galarneau
,
T. J.
, Jr.
,
L. F.
Bosart
,
C. A.
Davis
, and
R.
McTaggart-Cowan
,
2009
:
Baroclinic transition of a long-lived mesoscale convective vortex
.
Mon. Wea. Rev.
,
137
,
562
584
.
Gilet
,
J.-B.
,
M.
Plu
, and
G.
Rivière
,
2009
:
Nonlinear baroclinic dynamics of surface cyclones crossing a zonal jet
.
J. Atmos. Sci.
,
66
,
3021
3041
.
Gyakum
,
J. R.
,
P. J.
Roebber
, and
T. A.
Bullock
,
1992
:
The role of antecedent surface vorticity development as a conditioning process in explosive cyclone intensification
.
Mon. Wea. Rev.
,
120
,
1465
1489
.
Harr
,
P. A.
, and
R. L.
Elsberry
,
2000
:
Extratropical transition of tropical cyclones over the western North Pacific. Part I: Evolution of structural characteristics during the transition process
.
Mon. Wea. Rev.
,
128
,
2613
2633
.
Hodges
,
K. I.
,
D. W.
Chappell
,
G. J.
Robinson
, and
G.
Yang
,
2000
:
An improved algorithm for generating global window brightness temperatures from multiple satellite infrared imagery
.
J. Atmos. Oceanic Technol.
,
17
,
1296
1312
.
Hoskins
,
B. J.
, and
N. V.
West
,
1979
:
Baroclinic waves and frontogenesis. Part II: Uniform potential vorticity jet flows—Cold and warm fronts
.
J. Atmos. Sci.
,
36
,
1663
1680
.
Hoskins
,
B. J.
,
M. E.
McIntyre
, and
A. W.
Robertson
,
1985
:
On the use and significance of the isentropic potential vorticity maps
.
Quart. J. Roy. Meteor. Soc.
,
111
,
877
946
.
Hulme
,
A. L.
, and
J. E.
Martin
,
2009
:
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
.
Jones
,
S. C.
, and
Coauthors
,
2003
:
The extratropical transition of tropical cyclones: Forecast challenges, current understanding, and future directions
.
Wea. Forecasting
,
18
,
1052
1092
.
Juckes
,
M.
, and
R. K.
Smith
,
2000
:
Convective destabilization by upper-level troughs
.
Quart. J. Roy. Meteor. Soc.
,
126
,
111
123
.
Junger
,
S.
,
1997
:
The Perfect Storm
.
W. W. Norton & Company, 227 pp
.
Kenzelmann
,
P.
,
2005
:
Dynamics and Climatology of Diabatic Rossby Waves
.
Diplomarbeit, University of Mainz, 90 pp
.
Keyser
,
D.
,
M. J.
Reeder
, and
R. J.
Reed
,
1988
:
A generalization of Petterssen’s frontogenesis function and its relation to the forcing of vertical motion
.
Mon. Wea. Rev.
,
116
,
762
780
.
Klein
,
P. M.
,
P. A.
Harr
, and
R. L.
Elsberry
,
2002
:
Extratropical transition of western North Pacific tropical cyclones: Midlatitude and tropical cyclone contributions to reintensification
.
Mon. Wea. Rev.
,
130
,
2240
2259
.
Kuo
,
Y.-H.
,
R. J.
Reed
, and
S.
Low-Nam
,
1992
:
Thermal structure and airflow in a model simulation of an occluded marine cyclone
.
Mon. Wea. Rev.
,
120
,
2280
2297
.
Lackmann
,
G. M.
,
L. F.
Bosart
, and
D.
Keyser
,
1996
:
Planetary- and synoptic-scale characteristics of explosive wintertime cyclogenesis over the western North Atlantic Ocean
.
Mon. Wea. Rev.
,
124
,
2729
2758
.
Lorenz
,
E. N.
,
1955
:
Available potential energy and the maintenance of the general circulation
.
Tellus
,
7
,
157
167
.
Martin
,
J. E.
,
2006
:
Mid-Latitude Atmospheric Dynamics: A First Course
.
John Wiley & Sons, 324 pp
.
McTaggart-Cowan
,
R.
,
L. F.
Bosart
,
J. R.
Gyakum
, and
E. H.
Atallah
,
2007
:
Hurricane Katrina (2005). Part II: Evolution and hemispheric impacts of a diabatically generated warm pool
.
Mon. Wea. Rev.
,
135
,
3927
3949
.
Moore
,
R. W.
, and
M. T.
Montgomery
,
2004
:
Reexamining the dynamics of short-scale, diabatic Rossby waves and their role in midlatitude moist cyclogenesis
.
J. Atmos. Sci.
,
61
,
754
768
.
Moore
,
R. W.
, and
M. T.
Montgomery
,
2005
:
Analysis of an idealized, three-dimensional diabatic Rossby vortex: A coherent structure of the moist baroclinic atmosphere
.
J. Atmos. Sci.
,
62
,
2703
2725
.
Moore
,
R. W.
,
M. T.
Montgomery
, and
H. C.
Davies
,
2008
:
The integral role of a diabatic Rossby vortex in a heavy snowfall event
.
Mon. Wea. Rev.
,
136
,
1878
1897
.
Muench
,
H. S.
,
1965
:
On the dynamics of the wintertime stratospheric circulation
.
J. Atmos. Sci.
,
22
,
349
360
.
Norquist
,
D. C.
,
E. E.
Recker
, and
R. J.
Reed
,
1977
:
The energetics of African wave disturbances as observed during Phase III of GATE
.
Mon. Wea. Rev.
,
105
,
334
342
.
Parker
,
D. J.
, and
A. J.
Thorpe
,
1995
:
Conditional convective heating in a baroclinic atmosphere: A model of convective frontogenesis
.
J. Atmos. Sci.
,
52
,
1699
1711
.
Pasch
,
R. J.
,
1991
:
Preliminary report: Unnamed Hurricane 28 October– 3 November 1991
.
National Hurricane Center, 7 pp. [Available online at http://www.nhc.noaa.gov/.]
Pasch
,
R. J.
, and
L. A.
Avila
,
1992
:
Atlantic hurricane season of 1991
.
Mon. Wea. Rev.
,
120
,
2688
2696
.
Petterssen
,
S.
,
1955
:
A general survey of factors influencing development at sea level
.
J. Meteor.
,
12
,
36
42
.
Petterssen
,
S.
, and
S. J.
Smebye
,
1971
:
On the development of extratropical cyclones
.
Quart. J. Roy. Meteor. Soc.
,
97
,
457
482
.
Plu
,
M.
, and
P.
Arbogast
,
2005
:
A cyclogenesis evolving into two distinct scenarios and its implications for short-term ensemble forecasting
.
Mon. Wea. Rev.
,
133
,
2016
2029
.
Posselt
,
D. J.
, and
J. E.
Martin
,
2004
:
The effect of latent heat release on the evolution of a warm occluded thermal structure
.
Mon. Wea. Rev.
,
132
,
578
599
.
Raymond
,
D. J.
, and
H.
Jiang
,
1990
:
A theory for long-lived mesoscale convective systems
.
J. Atmos. Sci.
,
47
,
3067
3077
.
Reed
,
R. J.
, and
M. D.
Albright
,
1986
:
A case study of explosive cyclogenesis in the eastern Pacific
.
Mon. Wea. Rev.
,
114
,
2297
2319
.
Reed
,
R. J.
,
M. T.
Stoelinga
, and
Y.-H.
Kuo
,
1992
:
A model-aided study of the origin and evolution of the anomalously high potential vorticity in the inner region of a rapidly deepening marine cyclone
.
Mon. Wea. Rev.
,
120
,
893
913
.
Rivière
,
G.
,
2008
:
Barotropic regeneration of upper-level synoptic disturbances in different configurations of the zonal weather regime
.
J. Atmos. Sci.
,
65
,
3159
3178
.
Roebber
,
P. J.
,
2009
:
Planetary waves, cyclogenesis, and the irregular breakdown of zonal motion over the North Atlantic
.
Mon. Wea. Rev.
,
137
,
3907
3917
.
Sanders
,
F.
,
1986
:
Explosive cyclogenesis in the west-central North Atlantic Ocean, 1981–1984. Part I: Composite structure and mean behavior
.
Mon. Wea. Rev.
,
114
,
1781
1794
.
Sanders
,
F.
, and
J. R.
Gyakum
,
1980
:
Synoptic-dynamic climatology of the “Bomb.”
.
Mon. Wea. Rev.
,
108
,
1589
1606
.
Schär
,
C.
, and
H.
Wernli
,
1993
:
Structure and evolution of an isolated semi-geostrophic cyclone
.
Quart. J. Roy. Meteor. Soc.
,
119
,
57
90
.
Schultz
,
D. M.
,
D.
Keyser
, and
L. F.
Bosart
,
1998
:
The effect of large-scale flow on low-level frontal structure and evolution in midlatitude cyclones
.
Mon. Wea. Rev.
,
126
,
1767
1791
.
Shapiro
,
L.
,
1992
:
Hurricane vortex motion and evolution in a three-layer model
.
J. Atmos. Sci.
,
49
,
140
154
.
Shapiro
,
M. A.
, and
D.
Keyser
,
1990
:
Fronts, jet streams and the tropopause
.
Extratropical Cyclones: The Erik Palmén Memorial Volume, C. W. Newton and E. O Holopainen, Eds., Amer. Meteor. Soc., 167–191
.
Simmons
,
A. J.
, and
B. J.
Hoskins
,
1978
:
The life cycles of some nonlinear baroclinic waves
.
J. Atmos. Sci.
,
35
,
414
432
.
Simmons
,
A. J.
,
S.
Uppala
,
D.
Dee
, and
S.
Kobayashi
,
2007
:
ERA-Interim: New ECMWF reanalysis products from 1989 onwards
.
ECMWF Newsletter, No. 110, ECMWF, Reading, United Kingdom, 25–35
.
Sinclair
,
M. R.
,
2002
:
Extratropical transition of southwest Pacific tropical cyclones. Part I: Climatology and mean structure changes
.
Mon. Wea. Rev.
,
130
,
590
609
.
Sutcliffe
,
R. C.
, and
A. G.
Forsdyke
,
1950
:
The theory and use of upper air thickness patterns in forecasting
.
Quart. J. Roy. Meteor. Soc.
,
76
,
189
217
.
Takayabu
,
I.
,
1991
:
“Coupling development”: An efficient mechanism for the development of extratropical cyclones
.
J. Meteor. Soc. Japan
,
69
,
837
841
.
Wallace
,
J. M.
,
G.-H.
Lim
, and
M. L.
Blackmon
,
1988
:
Relationship between cyclone tracks, anticyclone tracks and baroclinic waveguides
.
J. Atmos. Sci.
,
45
,
439
462
.
Wernli
,
H.
,
1997
:
A Lagrangian-based analysis of extratropical cyclones. II: A detailed case study
.
Quart. J. Roy. Meteor. Soc.
,
123
,
1677
1706
.
Wernli
,
H.
,
S.
Dirren
,
M. A.
Liniger
, and
M.
Zillig
,
2002
:
Dynamical aspects of the life cycle of the winter storm “Lothar” (24–26 December 1999)
.
Quart. J. Roy. Meteor. Soc.
,
128
,
405
429
.

Footnotes

1

The minimum SLP and deepening rate of EC1 is given as 949 and 50 hPa (24 h)−1 in the 40-yr ECMWF Re-Analysis (ERA-40) between 0000 UTC 30 October and 0000 UTC 31 October (cf. Cordeira and Bosart 2010), and 952 and 49 hPa (24 h)−1 in the ERA-Interim between 1800 UTC 29 October and 1800 UTC 30 October.