Observations of the Convective Environment in Developing and Non-Developing Tropical Disturbances

QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/qj.000

Observations of the convective environment in developing and non-developing tropical disturbances

Roger K. Smitha1and Michael T. Montgomeryb a Meteorological Institute, University of Munich, Munich, Germany b Dept. of Meteorology, Naval Postgraduate School, Monterey, CA & NOAA’s Hurricane Research Division, Miami, FL, USA.

Abstract: Analyses of thermodynamic data gathered from airborne dropwindsondes released from the upper troposphere during the Pre- Depression Investigation of Cloud Systems in the Tropics (PREDICT) experiment are presented. The main focus is on two systems that finally became Hurricanes Karl and Matthew and one system that attained Tropical Storm status (Gaston), but subsequently weakened and never redeveloped during five days of monitoring. Data for all events show that the largest values of total precipitable water are collocated with the surface trough and with some high values of convective available potential energy, which coincide roughly with low values of convective inhibition. Vertical profiles of virtual potential temperature show little variability between soundings on a particular day, but the system means from day to day show a slight warming. In contrast, vertical profiles of pseudo- equivalent potential temperature, θe, show much more variability between soundings on a particular day on account of the variability in moisture. In all systems, there is a tendency for the lower troposphere to moisten, but in the system that did not develop, the middle and upper troposphere became progressively drier during the five missions. In the developing systems, the upper levels moistened. The most prominent difference between the non-developing system and the two systems that developed was the much larger reduction of θe between the surface and a height of 3 km, typically 25 K in the non-developing system, compared with only 17 K in the systems that developed. Conventional wisdom would suggest that, for this reason, the convective downdraughts would be stronger in the non-developing system and would thereby act to suppress the development. Here we invoke an alternative hypothesis that the drier mid-level air weakens the convective updraughts and thereby weakens the amplification of system relative vorticity necessary for development. Copyright c 2011 Royal Meteorological Society

KEY WORDS Tropical cyclone, hurricane, typhoon, spin-up, intensiﬁcation

Received June 23, 2011; Revised ; Accepted

1 Introduction Atmospheric Administration (NOAA) and National Sci- ence Foundation (NSF) to investigate tropical cyclogenesis in the Caribbean and West Atlantic and the sub- Understanding the dynamics of tropical cyclogenesis remains one of the great unsolved problems in tropical sequent intensification of storms in these regions. While meteorology. One reason for the lack of understanding is two of the experiments1 included intensification in their undoubtedly the fact that genesis occurs over the trop- portfolio of objectives, the Pre-Depression Investigation ical oceans where conventional observational data such of Cloud Systems in the Tropics (PREDICT) experiment as radar data, radiosonde soundings and surface data are was designed exclusively to study genesis. Priority was relatively sparse. While there have been a few field exper- given to developing storms prior to their classification as iments directed at documenting cyclogenesis (Bister and tropical depressions as defined by forecasters2, even when Emanuel 1997, Ritchie and Holland 1999, Elsberry and Harr 2008) and a few serendipitous sets of measurements (Reasor et al. 2005, Houze et al. 2009), many questions remain about the processes involved and their relative 1The Genesis and Rapid Intensification Processes (GRIP) project of importance. A recent review of work over the past few NASA and the Intensify Forecast Experiment (IFEX) of the NOAA. years is given by Montgomery and Smith (2010). 2The glossary on NOAAs Hurricane Research Divisions website uses “tropical cyclone as the generic term for a nonfrontal synoptic-scale In the late summer of 2010, a trio of field exper- low-pressure system over tropical or sub-tropical waters with organized iments was conducted by the National Aeronautics and convection (i.e. thunderstorm activity) and a definite cyclonic surface wind circulation. Notably, this definition does not invoke any wind Space Administration (NASA), National Oceanic and threshold. The same glossary defines a “tropical depression as a tropical cyclone with maximum sustained surface winds of less than 17 m s−1 (34 kt, 39 mph) and, in the Atlantic and Eastern Pacific Basins, a tropical 1Correspondence to: Prof. Roger K. Smith, Meteorological Institute, storm as a tropical cyclone with surface winds between 17 m s−1 and 33 Ludwig-Maximilians University of Munich, Theresienstr. 37, 80333 ms−1. In this study we will define genesis as the formation of a tropical Munich, Germany. E-mail: [email protected] depression and we impose no formal threshold on wind speed.

Copyright c 2011 Royal Meteorological Society Prepared using qjrms3.cls [Version: 2007/01/05 v1.00] 2 R. K. SMITH AND M. T. MONTGOMERY mature storms were present nearby. The primary mea- dropwindsonde data and methods of analysis. Section 4 surement platform of PREDICT was the National Cen- presents the results of the analyses for each storms. The ter for Atmospheric Research (NCAR) GV research air- results are discussed in section 5 and the conclusions are craft, equipped with dropsondes and onboard sensors for given in section 6. meteorological variables and ice microphysics. The range and speed of the GV, and the high altitude (≈13-14 km) from which it could release dropsondes, were exploited to 2 Events discussed sample storm formation from Central America to the mid- The storms sampled by PREDICT aircraft missions are Atlantic (roughly 40◦W) operating out of St. Croix in the summarized by Montgomery et al. (2011, see their Table U.S. Virgin Islands. Some measurements were made also 1). These storms were fairly evenly split between devel- by the NASA DC8 research aircraft, which, has the capa- oping and non-developing systems. While some cases had bility to release dropsondes from moderately high in the a high potential for genesis from the time of the first troposphere (≈10-11 km). GV flight into them and subsequently developed, one sys- The PREDICT field experiment aimed to gather data tem departed from this pattern. Gaston was a surpris- on developing and non-developing tropical disturbances ing instance where development occurred initially, but and to test the recently proposed marsupial model of following a period of decline (sampled by the GV) re- tropical cyclogenesis in association with tropical easterly development did not occur. Matthew surprisingly formed waves (Dunkerton et al. 2009). The overarching hypoth- in only three days from a weak disturbance in the mid- esis was that tropical depression formation is greatly Atlantic Inter-Tropical-Convergence-Zone (ITCZ). favoured in the critical-layer region of the synoptic-scale, The main focus of this study is on the three storms: pre-depression wave or subtropical disturbance. A sum- ex-Tropical Storm Gaston, pre-Hurricanes Karl and pre- mary of the scientific basis for the experiment as well Tropical Storm Matthew, but some data will be presented as some highlights of the data obtained is described by for pre-Hurricane Nichol. We contrast these data with with Montgomery et al. (2011). The experiment collected, a brief discussion of those of obtained during the mature inter alia, unprecedented3 dropwindsonde data from two stage of Hurricane Earl. high-altitude flying aircraft (the NCAR GV and NASA In the analyses presented, the negative ocean influ- DC8). The flights were focused on the so-called “wave ence of mixing up of cold water beneath the thermocline pouch” (Dunkerton et al. 2009) and its immediate vicinity. to the surface is neglected because the development and The wave pouch is defined as the region of recirculating non-development of tropical disturbances occurs at rela- streamlines depicted in the frame of reference at a given tively low surface winds of approximately 30 knots or less altitude moving with the tropical disturbance in the lower where the impact of wind-induced mixing of cold water is troposphere and the centre of circulation is referred to as small (Shay, 2010). the “sweet spot”, the location at which, according to the marsupial paradigm, a tropical cyclone is most likely to form. In the PREDICT experiment, the 700 hPa level was 3 Dropwinsonde data typically used to track the pouch of a candidate disturbance and to delineate its boundary with environmental Between 20 and 25 dropwindsondes were made during air. These data provide a unique opportunity to examine each GV flight mission. Diurnal cycle effects were largely the evolution of the thermodynamic and kinematic struc- filtered by the fact that the GV flights were carried out at ture of the wave pouches identified for both developing approximately the same time of day. After quality control and non-developing systems. using the NCAR Aspen software and further manual In this study we will examine thermodynamic aspects inspection to eliminate sondes that did not record all the of the sounding data acquired during the experiment, way to the surface, the precise number of usable sondes focusing on the day-to-day changes in the mean profiles from the GV is summarized here. For ex-Tropical Storm of potential temperature and equivalent potential tempera- Gaston: 20 on 2 September, 21 on 3 and 5 September, ture. While these analyses cannot directly offer an under- and 22 on 6 and 7 September. For pre-Hurricane Karl: standing of the spin up process on the mesoscale, they 21 for the two flights on 10 September and the flights nonetheless offer quantitative information on the changes on 11 and 14 September, 22 on 12 September and 20 in the thermodynamic mean states during development or on 13 September. For pre-Tropical Storm Matthew: 21 non-development, such as the stabilization or otherwise of on 20, 21 and 24 September and 16 on 22 September. the mean sounding and the moistening of the mean sound- For pre-Hurricane Nichol: 21 on 27 and 28 September. ing during the evolution of the disturbance. Many of these soundings were released from altitudes The outline of this paper is as follows. Section 2 between 13 and 14 km giving an unprecedented glimpse details the events discussed and section 3 describes the of the structure of the much of the tropical troposphere in the systems studied. Radiosonde soundings from island 3In the TCS08 experiment, the US Air Force C130 aircraft was flown at stations at similar latitudes show that the tropopause lies a relatively high altitude of approximately 9 km in conjunction with the around 15-16 km in the region of interest. Naval Research Laboratory P3 with the ELDORA radar and Doppler wind lidar on board. However during the earlier experiments, flight Soundings were made also by the DC8 during the last altitudes were limited to a few kilometres. two days of ex-Tropical Storm Gaston (06-07 September)

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj CONVECTIVE ENVIRONMENT IN DEVELOPING AND NON-DEVELOPING TROPICAL DISTURBANCES 3 and for the last three days of during pre-Hurricane Karl paper by Rutherford and Montgomery (2011). That work (12-14 September). documents the import of dry air into the pouch in the mid- The data from all of these soundings were inter- dle troposphere and the severing of the “umbilical chord” polated to levels at intervals of 100 m above the sur- of moisture from the ITCZ. For reasons articulated below, face. These data were then used to derive various quan- these conditions do not favour development. tities of thermodynamic interest such as the total pre- The presence of dry air is confirmed by the TPW- cipitable water (TPW), Convective Available Potential analyses and especially on 5 September, dry air is seen Energy (CAPE) and Convective Inhibition (CIN). The to be wrapping around the southern side of the depres- definition that we use for CAPE and CIN and the method sion. Nevertheless the core, or pouch region, maintains for calculating these quantities are given briefly in the relatively high TPW values (> 50 kg m−2) throughout appendix. The DC8 sondes were released mostly from the period of study. Indeed, we present evidence below to altitudes of around 10 km, several kilometres lower than show that the lower troposphere actually moistened during from the GV and often well below the level of neutral the five aircraft missions, although the middle and upper buoyancy (LNB) for air parcels lifted from near the sur- troposphere did not (see Figure 7 below). It will be noted face. For this reason, CAPE and CIN values were calcu- that the highest TPW values lie broadly within the region lated only for the GV sondes. of lowest surface pressures on all days and encompass the sweet spot. Figure 2 shows frequency diagrams of TPW- 4 Results values from the five flights. Significantly, the percentage of relatively high TPW-values (≥ 55 kg m−2) on the five 4.1 Ex-Gaston: 2-7 September days and the number of usable sondes (in brackets) were 55% (20), 67% (21), 48% (21), 33% (22) and 45% (22), Tropical Storm Gaston developed from an African wave and was named by the National Hurricane Center (NHC) respectively. It is interesting to compare these percentages at 15 UTC 1 September. However, dropsondes from the with the corresponding ones for pre-Hurricane Karl given GV deployed during its first mission into Gaston on in section 4.2, which are considerably higher. 2 September indicated that the storm had weakened to Figure 3 shows the vertical structure of virtual poten- a tropical depression, whereupon Gaston was officially tial temperature (θv), pseudo-equivalent potential tem- 4 downgraded to an INVEST, i.e. an area of disturbed perature (θe), and saturation pseudo-equivalent poten- tropical weather lacking organization of deep convection. tial temperature (θes) for four selected soundings on 5 The disturbance maintained an identity that could be September, two in the region of highest TPW and two in tracked across the Caribbean although the convective the dry air to the south of this region. The vertical lines in activity weakened considerably after 7 September. Five the figure show the θe values for air parcels at the surface GV-flights were made into the disturbance on each day and at a height of 100 m above the surface. Since θe is from 2 to 7 September, except on 4 September. conserved for undilute ascent with or without condensa- Figure 1 shows objective analyses of the surface tion, these lines represent the θe of moist air parcels lifted pressure for each of the five flights. The flight tracks, from these levels. Moreover, the distance between the ver- drop points and drop times are shown also. In general, tical line and the θes-curve at a given height is roughly the fields show coherent patterns on a scale larger than proportional to the buoyancy of the lifted air parcel at this the spacing between sondes, a feature that we interpret to height, with the buoyancy being positive when the par- be affirmation of the integrity of the data. One surface cel line is to the right of the θes-curve. Thus, assuming pressure reading at 1745 UTC on 2 September seemed undilute ascent, the first intersection of the vertical line suspiciously high and was replaced by an average of with the θes-curve is the approximate level of free con- that from neighbouring points at about the same latitude. vection (LFC) and the final intersection is approximately Interestingly, the pressure pattern on 2 September has the equal to the LNB for the particular air parcel (see Ooyama appearance of a trough, whereas on subsequent days there 1969, Emanuel 1994, Holton 2004). Furthermore, the area were some closed isobars (at 0.5 mb spacing), suggesting between these lines and the θes-curve in the range between that Gaston was trying to redevelop. As the main focus the LFC and LNB is roughly proportional to the CAPE of this paper is on the thermodynamical aspects of the and that between the surface and the LFC is roughly pro- systems, the relationship of the surface pressure field to portional to the CIN. The following features may be noted: the observed flow field will be discussed in a separate publication. The right panels in Figure 1 show the correspond- • The height from which the soundings were made ing distribution of TPW derived from the sondes. Satel- was mostly close to the LNB allowing reasonably lite imagery indicated that Gaston was surrounded by accurate estimates of CAPE5. relatively dry air during its westwards transit across the Atlantic and it was speculated by forecasters that this was 4The pseudo-equivalent potential temperature was calculated using the reason for Gaston’s failure to redevelop. A Lagrangian Bolton’s formula (Bolton 1980). 5In cases where the LNB was above the sounding, the sounding was perspective on the interaction between the dry environ- extended, guided by that in the European Centre for Medium Range ment and the moist pouch will be presented in a separate Weather Forecasts analyses in the storm location.

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 4 R. K. SMITH AND M. T. MONTGOMERY

Figure 1. Objective analyses of surface pressure (left panels) and total precipitable water (right panels) for ex-Tropical Storm Gaston on 02, 03 and 05 September 2010. Contour interval 0.5 mb for pressure, 10 kg m−2 for TPW. The aircraft track is marked with the location of soundings along the track (small circles). The data are given below the circle and time of the sounding in UTC is given above the circle in the TPW plots. The original surface pressure from the sounding at 1745 UTC on 2 September was suspiciously high (1015.6 mb) and was replaced by the average of the two adjacent values at 1733 UTC and 1843 UTC. The approximate track of the pouch centre during the ﬂight with starting and ending times indicated is marked by the thick green line. Figure continued on the next page.

• The vertical structure of θv shows the expected acquire little or no buoyancy below a height of about monotonic increase with height and varies little 4 km. A striking feature of these sounding is the between soundings. low values of θe in the low to middle troposphere, • The LFC of the two soundings on the southern side there being a large saturation deﬁcit, as measured by of the domain is relatively high and, judged directly θes − θe, exceeding 30 K. (Here θes is the saturation from these ﬁgures, these soundings have consider- value of θe.) These values indicate the presence of ably lower CAPE than the two moist soundings6. relatively dry air at these levels. Indeed, air parcels lifted from, or near, the surface • The LFC of the two moist soundings is much lower than the two dry ones and the CIN is correspond- 6Actual values of CAPE, calculated by the method given in an appendix, ingly lower. Moreover, the buoyancy of lifted air are shown in Figure 5. parcels is much larger, whereupon the values of

Figure 1. (continued) Objective analyses of surface pressure (left panels) and total precipitable water (TPW) (right panels) for ex-Tropical Storm Gaston on 06 and 07 September 2010. Contour interval 0.5 mb for pressure, 10 kg m −2 for TPW.

Figure 2. Frequency diagrams of TPW for ex-Gaston on 02, 03, 05, 06 and 07 September 2010. The approximate track of the pouch centre during the ﬂight with starting and ending times indicated is marked by the thick green line.

(a) (b) (c) (d)

Figure 3. Four soundings for ex-Gaston on 05 September: (a) 1448 UTC; (b) 1513 UTC; (c) 1809 UTC; (d) 1820 UTC. See Fig. 1 for positions in relation to TPW.

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 6 R. K. SMITH AND M. T. MONTGOMERY

Figure 4. All soundings of θv (red curves) and θe (blue curves) for Gaston on 02, 03 and 05-07 September 2010 as indicated. The pouch- mean sounding is indicated by the black curve. The saturation θe of the pouch-mean sounding is indicated by the right black curve. The left and right lower panels show soundings from the NASA DC8 ﬂights on 6 and 7 September.

CAPE are much larger also. For these soundings, of the θe curves lie to the left of the θes-curve for the mean the saturation deficit is relatively small throughout sounding. much of the troposphere. Figure 5 shows the spatial distribution of CAPE and CIN for the five missions into ex-Gaston and Figure 6 shows a scatter plot of CAPE and CIN for all8 soundings, Figure 4 shows similar profiles of all soundings with −2 identified by day. It shows also the track segment of a TPW equal to or higher than 50 kg m together with the sweet spot during the flight mission. The following their mean on each day of measurement7. The foregoing features are evident: criterion was found to correspond roughly with the region of closed streamline contours at 700 mb in the European • On all days there are some relatively high values Centre for Medium Range Weather Forecasts (ECMWF) of CAPE (> 2 kJ kg−1), with the maximum values analysis near the time of the mission and eliminated only a increasing from day to day. On last three missions, small fraction of soundings on the periphery of the system. a few soundings had values exceeding 4000 J kg−1. Therefore, this value of TPW is used throughout this study • On most days, the high CAPE and low CIN values as a rough proxy for the pouch. are located in the vicinity of the minimum surface Consistent with the four soundings shown in Fig- pressures, i.e. near the sweet spot. ure 3, the vertical structure of θv shows little variability • Soundings with high values of CAPE tend to have between soundings. However, there is considerable vari- relatively low values of CIN (Figure 6). ability in the vertical structure of θe on a particular day. The variability between the daily-mean soundings is dis- The statistics of the CAPE and CIN values can cussed later (see Figure 7). On all days, the vast majority be summarized by constructing frequency diagrams (not shown). The main findings are as follows: 7The number and percentage (in brackets) of soundings with TPW > 50 kg m−2 was 14 (70%), 20 (90%), 14 (67%), 12 (55%), 19 (86%) on 2, 8Out of the 106 soundings, only two have values of CIN > 160 J kg−1 3, 5, 6, 7 September, respectively putting them outside the plot.

Figure 5. Objective analyses of CAPE (left panels) and CIN (right panels) for ex-TS Gaston on 02, 03 and 05 September 2010. Contour interval for CAPE: 500 J kg−1, values larger than 2 kJ kg−1 in bold. Contour interval for CIN: thin lines 5 J kg−1, thick lines 25 J kg−1. The aircraft track is marked with the location of soundings along the track (small circles). The time of the sounding in UTC is given above the circle and the data below the circle. The approximate track of the pouch centre during the ﬂight with starting and ending times indicated is marked by the thick green line. Figure continued on the next page.

• On 02 September, only 10% of soundings had • The largest percentage of soundings with CIN val- −1 values of CAPE exceeding 2.5 kJ kg−1 and none ues below 40 J kg was on 03 September (71%) had values exceeding 3 kJ kg−1. However, the and the largest percentage with values below 20 J kg−1 were on the same day. Thereafter, CIN values CAPE increased during the ﬁve days of monitoring increased so that on the last day of monitoring, 07 and on 07 September, 73% of soundings had values September, CIN values had generally increased with of CAPE exceeding 2.5 kJ kg−1 and 50% had values only 32% of values below 40 J kg−1 and only 9% of −1 exceeding 3 kJ kg−1. values below 20 J kg . • Between 32% and 71% of soundings had values of These results show that as the system evolves there −1 CIN below 40 J kg and between 5% and 33% had is a trend towards the development of higher values values below 20 J kg−1. of CAPE, and also to higher values of CIN. However,

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 8 R. K. SMITH AND M. T. MONTGOMERY

Figure 5. (Continued) Objective analyses of CAPE (left panels) and CIN (right panels) for ex-TS Gaston on 06 and 07 September 2010. Contour interval for CAPE: 500 J kg−1, values larger than 2 kJ kg−1 in bold. Contour interval for CIN: thin lines 5 J kg−1, thick lines 25 Jkg−1. as shown in Figure 5 above, the region of increasing CAPE is roughly coincident with the region of minimum CIN. Thus, despite the increasing CAPE near the sweet spot, the system did not redevelop over the five days of measurements. Evidently, these metrics by themselves are insufficient indicators of the failure of the system to redevelop. Figure 7 shows the day-to-day evolution of the “pouch-mean” soundings obtained from the dropsondes during missions into ex-Gaston. These pouch-mean soundings are characterized by the virtual potential temperature (θv) and pseudo-equivalent potential temperature (θe) as a function of height. The two lower panels of Fig- ure 7 show the deviations of θv and θe from the five day mean. Features that stand out are as follows: • There is little difference in the day-to-day variation in the system-average θv. • Figure 6. Scatter plot of CAPE and CIN values for the 106 sound- There is a general day-to-day increase in the lower ings in ex-Gaston. Note the tendency for soundings with high CAPE tropospheric θe, but at mid-upper levels (z>3 km), to have low CIN. the mean θe decreases with time. • There is a monotonic warming of the lowest 2 km, exceeding 2 K near the surface. Above 3 km It is reasonable to speculate that the warming and at altitude, there is cooling on the order of 1 to 1.5 K least part of the increase in θe at low levels is due to the at all heights up to at least 10 km. increase in sea-surface temperature (on the order of 1.5 C) • There is a relatively large difference (typically 25 as ex-Gaston moved across the Atlantic Ocean and then K) between near-surface θe value and those of the over the warmer Caribbean Sea. mid-troposphere minimum. Traditional reasoning would suggest that ensuing

Figure 7. Vertical profiles of the day-to-day differences in system-mean virtual potential temperature and pseudo-equivalent potential temperature from the mean profile of all days for ex-Tropical Storm Gaston on 02, 03, 05, 06, 07 September, 2010. The thick curves mark the first and last days of the GV flights. Numbers on curves refer to the date. Curves for the NASA DC8 flights on 06 and 07 September are denoted D6 and D7 respectively. convection within a relatively dry mid-upper air environ- of a quasi-persistent area of convection along the ITCZ ment would lead to comparatively-strong downdraughts near the northern coast of the Continent. An easterly (e.g., Emanuel 1994). From the perspective of convective wave formed subsequently near this locale and propagated dynamics, stronger downdraughts (implied by the lower westwards towards the eastern Caribbean. Convection mid-tropospheric relative humidity in Gaston) would tend within the early pre-Karl disturbance was elongated east- to import low θe into the boundary layer and frustrate the west on 10 September. On this day, two missions were enhancement of boundary layer θe by sea-to-air moisture conducted, one centred on roughly 13 UTC, the other fluxes. This enhancement is necessary to fuel subsequent about 6 h later. Both flights revealed a broad cyclonic wind deep convective activity. However, recent numerical mod- gyre of diameter roughly 400 km that featured similar elling studies by James and Markowski (2009) and Kilroy tangential winds between 950 mb and 700 mb (Boothe, (2011, personal communication) suggest that the principal personnal communication). Pre-Karl was rather slow to effects of the dry air are to reduce the updraught strength develop over the next several days. Over the course of 11- and water loading, while the downdraught strengths are 14 September, convective activity gradually consolidated not changed appreciably. The dilution of the updraught from an initial southwest-northeast elongation to a more with dry air would make the CAPE, which is calculated on compact pattern encircling the sweet spot within the the assumption of undilute ascent, less effective in ampli- pouch, the latter being close to the location where Karl fying vertical vorticity (see section 5). formed later on 14 September. As to the source of this drier air in ex-Gaston, there Figure 8 shows objective analyses of the surface pres- is evidence of a strong, dry Saharan air mass surrounding sure for the first two flights, on 10 September, and the last the northern hemisphere of ex-Gaston in multiple satel- lite derived products. A Lagrangian analysis carried out two on 14 September. As in Figure 1, the flight tracks, by Rutherford and Montgomery (2011) indicates strongly drop points and drop times are shown also, together with that the dry air seen in the observations is the result of the track of the sweet spot during the flight. On 12 Septem- a complex process of intrusion of dry air from outside ber, the flight tracks as well as the location and timing of of the pouch region. Dry air continued to plague ex- dropsondes were unsuitable for objective analysis. As in Gaston during the subsequent days. The cyclonic circu- the case of ex-Gaston, the analyses show coherent pat- lation became shallower over time also (Davis, personnal terns on a scale larger than the spacing between sondes. communication, Boothe, personnal communication). By However, unlike ex-Gaston, the pressure patterns from the 5 September, the circulation was not evident at 500 mb analyses show multiple centres with closed isobars as late and by 7 September it was barely evident at 700 mb. The as the first GV mission on 14 September, approximately storm continued to produce intense convection intermit- 6 hours prior to the disturbance being declared a tropi- tently throughout the period, but the vortex systematically cal storm. Nevertheless, the later flight by the NASA DC8 weakened. found a single pressure minimum at the surface (cf. Figure 8d). Only three soundings (all on 13 September) had 4.2 Pre-Hurricane Karl: 10-14 September values of TPW less than 50 kg m−2, while 50% of −2 The disturbance that ultimately became Hurricane Karl soundings had values exceeding 60 kg m , compared originated within the Inter-Tropical Convergence Zone with only 26% for Gaston. For this reason, plots of TPW (ITCZ) region near the northern coast of South America. are not shown for Karl. However, we note that, as in A surge of southwesterlies over northern South America Gaston, the highest TPW values lay broadly within the around 8 September was accompanied by the formation region of lowest surface pressures on all days.

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 10 R. K. SMITH AND M. T. MONTGOMERY

Figure 8. Objective analyses of surface pressure for the two GV flights into pre-Hurricane Karl on the first day of observations (10 September) and the two flights on the last day (14 September), the last one being the DC8 flight. Contour interval 0.5 mb for pressure. The aircraft track is marked with the location of soundings along the track (small circles) and the data are given below the circle. The approximate track of the pouch centre during the flight with starting and ending times indicated is marked by the thick green line.

Figure 9 shows the spatial distribution of CAPE and • The day with the largest percentage of soundings CIN for the two flights on 10 September and the flights with CIN values below 40 J kg−1 was 14 September on 11, 13 and 14 September. The corresponding scatter (43%); the same day had the largest percentage with plot for these data analogous to that shown for ex-Gaston CIN values below 20 J kg−1 (41%). (Figure 6) are shown in Figure 10. Like ex-Gaston, on all days there are some relatively high values of CAPE. Figure 11 shows profiles of all soundings with a TPW ≥ 50 kg m−2 together with their mean on each day of The highest values of CAPE have relatively low values measurement in pre-Karl. As in the case of ex-Gaston, of CIN, but not the lowest values. However, unlike ex- the soundings are characterized by the vertical profiles of Gaston, there is not a systematic increase of CAPE from θv and θe. The main points to note are that there is little day to day. Furthermore, the highest values of CAPE are variation between soundings for θv and the scatter in θe is not near the sweet spot position. significantly less than that in Figure 4. In addition, the vast Some statistics can be derived also from these data majority of the θe curves lie to the left of the θes curves for using frequency diagrams (not shown): the mean sounding. The left panel of Figure 12 shows the day-to-day • Between 23% and 43% of soundings over the six evolution of the pouch-mean soundings obtained from GV missions had values of CAPE exceeding 2.5 kJ −1 the dropsondes. The middle and right panels show the kg and between 5% and 18% of soundings after θ θ −1 deviations of v and e from the five day mean. Features 10 September had values exceeding 3 kJ kg . that stand out are: • The days with the largest percentages of soundings −1 with CAPE ≥ 2.5 kJ kg were 12 September • As for ex-Gaston, there is little difference in the day- (41%) and 14 September (43%). These days had to-day variation in the system-average θv. also the largest percentages of CAPE 3 kJ ≥ kg−1 • There is a general day-to-day increase in the lower (18% on 12 September and 14% on 14 September). tropospheric θe. However, unlike the case of ex- • Between 19% and 48% of soundings had values of Gaston, the mean θe at mid-upper levels (z>3km) CIN below 40 J kg−1 and between 5% and 10% had increases over the period of observation, but this values below 20 J kg−1. increase is not monotonic.

Figure 9. Objective analyses of CAPE (left columns) and CIN (right columns) for pre-Hurricane Karl on 10 and 11 September 2010. There were two ﬂights on 10 September. Contour interval for CAPE: 500 J kg−1, values larger than 2 kJ kg−1 in bold. Contour interval for CIN: thin lines 5 J kg−1, thick lines 25 J kg−1. The aircraft track is marked with the location of soundings along the track (small circles). The time of the sounding in UTC is given above the circle and the data below the circle. The approximate track of the pouch centre during the ﬂight with starting and ending times indicated is marked by the thick green line. Figure continued on the next page.

• There is a systematic warming of the entire tropo- 4.3 Pre-Matthew: 20-24 September sphere between 10 and 14 September. Between the surface and 5 km altitude the warming is on the In contrast with Karl, Matthew was generally unfore- order of 0.7 K. Above 5 km, the warming increases seen as few as 5 days prior to its formation. While the with height and exceeds 3 K above about 9 km alti- PREDICT team was tracking weak vorticity features con- tude. nected with the ITCZ on 17 and 18 September, it was not • There is only a modest difference (∼ 15 K) between until deep convection erupted on 19 September that there was a feature sufficiently defined in the atmosphere and in the near-surface θe and the mid-tropospheric mini- global models to plan a mission. The first GV mission was ∼ mum. This is in contrast to the 25 K difference in centred roughly on 16 UTC 20 September. At this time, the non-developing ex-Gaston. the pre-Matthew disturbance appeared less organized than

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 12 R. K. SMITH AND M. T. MONTGOMERY

Figure 9. (Continued) Objective analyses of CAPE (left columns) and CIN (right columns) for pre-Hurricane Karl on 13 and 14 September 2010.

on 23 September, but despite forecasts of potential rapid intensification, it did not intensify further and made land- fall in Belize as a moderate tropical storm late on 24 September. Although it had similarities to Karl in its mid- ocean origin and track through the Caribbean, Matthew developed rapidly, but failed to become a hurricane. Figure 13 shows a scatter plot of CAPE against CIN for the GV soundings obtained in Matthew. Unlike the case of pre-Karl and ex-Gaston, CAPE values were generally largest on the first day of observation (20 September) and subsequently declined. Specifically, 38% of soundings on 20 September had CAPE values exceeding 2.5 kJ kg−1 and 14% had values exceeding 3 kJ kg−1.On subsequent days, only one sounding had a CAPE value exceeding 2.5 kJ kg−1, but this value was less than 3 kJ −1 Figure 10. Scatter plot of CAPE and CIN values for the 106 kg . This sounding was on 21 September. There were soundings in pre-Karl. Note the tendency for soundings with high one or two soundings per day with CIN values less than −1 values of CAPE to have low values of CIN. 20Jkg , except on 22 September, when there were none. In contrast, between 24% and 29% of soundings had CIN values less than 40 J kg−1, except again on 22 September, Karl when it was first flown. However, Matthew quickly when there were only 13%. organized on 21 and 22 September. The pouch was well defined on 22 September with The left panel of Figure 14 shows the day-to-day strong convection occurring close to the predicted max- evolution of the pouch-mean soundings obtained from imum in Okubo-Weiss parameter in the analysis from the dropsondes. The middle and right panels show the the ECMWF model (see Montgomery et al. for details). deviations of θv and θe from the four day mean. Features Matthew went on to develop into a strong tropical storm that stand out are:

Figure 11. All pouch soundings of θv (red curves) and θe (blue curves) for pre-Karl on 10-14 September 2010 as indicated. The pouch- mean sounding is indicated by the black curve. The saturation θe of the pouch-mean sounding is indicated by the right black curve. The upper left panel (a) contains data for both ﬂights on 10 September, while the upper right and two lower panels include both the GV and NASA DC8 data.

Figure 12. Vertical profiles of the day-to-day differences in system-mean virtual potential temperature and pseudo-equivalent potential temperature from the mean profile of all days for pre-Hurricane Karl on 10-14 September, 2010. The thick curves mark the first and last days of the sequence. Numbers on curves refer to the date. The profiles on 10 September include both GV flights and those for 12-14 September include both GV and DC8 data.

• As for ex-Gaston and pre-Karl, there is little dif- and then increases. There is little difference in the ference in the day-to-day variation in the system- mean θe over the ﬁrst three days at mid-upper levels average θv. (z>3 km), but over the next two days the θe • Unlike ex-Gaston and pre-Karl, there is an overall increases by approximately 2 K. Initially, there is increase in the lower tropospheric θe over the four only a modest difference (∼ 15 K) between the near- days of observation, but the θe initially decreases surface θe and the mid-tropospheric minimum. This

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 14 R. K. SMITH AND M. T. MONTGOMERY

On 27 September, there were no soundings with CIN < 20 Jkg−1, but there were 7 on 28 September. There was only one value of CIN < 40 Jkg−1 on 27 September, but 15 on 28 September. In essence, the CAPE decreased marginally with time in the pouch region, but there was a signiﬁcant reduction in CIN over the two days of observation. Figure 16 shows the day-to-day evolution of the pouch-mean soundings obtained from the dropsondes. The middle and right panels show the deviations of θv and θe from the two day mean. Features that stand out over the two days of observations are the following:

• There is a very slight cooling in θv below 2 km altitude, with a slight warming between 2 and 4 km altitude. Above 4 km altitude, there is a pronounced warming of approximately 2 K. • The mean θe increases at all heights, with the largest Figure 13. Scatter plot of CAPE and CIN values for the 79 sound- increase of 4 K occurring at the surface, and an ings in pre-Matthew. increase of approximately 2 K above 5 km.

difference is similar to that found in Karl. 4.5 Hurricane Earl: 1-2 September • Between 20 and 21 September there is a slight Figure 17 shows the profiles of all soundings together cooling in the virtual potential temperature below with their mean on two days of measurement during an altitude of 6 km of approximately 1 K. On NASA DC8 missions into Hurricane Earl. There were subsequent days there is a systematic warming on 25 soundings on 1 September and 29 soundings on 2 the order of 1.5 K. September. Although the mature hurricane was not a • Like the virtual potential temperature, the deviation theme of the PREDICT experiment, it is of interest to of θe from the five day mean falls about 4 K contrast the soundings in a fully-fledged hurricane with below 2 km altitude. Above this altitude, there is no those observed in the pre-storm stage of evolution. Both appreciable change up to 22 September, after which the θv and θe curves show much more variability than there is a systematic increase of approximately 2 K. found in ex-Gaston and pre-Karl (compare Figure 17 with Figures 4 and 11). A distinct feature of these profiles 4.4 Pre-Nichol: 27-28 September is the set of ”eye” and ”eyewall” wall soundings, which θ In the wake of Matthew a large gyre formed over Cen- have significantly higher values of e than were recorded tral America in which strong flow from the Eastern in any of the pre-storms. These soundings are distinctly Pacific extended across Panama and Costa Rica into the warmer in terms of virtual potential temperature than Caribbean. The NHC initiated a tropical depression advi- those outside of the eye-eyewall region. Note that the eye- eyewall profiles of θe are almost vertical and lie well to the sory to the east of Belize on the eastern side of this θ gyre at 15 UTC 28 September. The tropical depression right (by approximately 10 K) of the mean saturation e of tracked northeastwards and became a weak tropical storm the averaged temperature of all soundings. This feature is (Nicole) at 1500 UTC 29 September over central Cuba. indicative of moist adiabatic ascent up to flight level. Tropical storm Nicole tracked north-northeastwards and decayed prior to 2100 UTC 29 September in the Florida 5 Discussion and Interpretation Strait. The pre-genesis environment of Nicole was interesting because the storm appeared to originate out of a Over the years, the life cycle of individual deep convective broad area of multiple low-level vorticity maxima that clouds has been well documented and there have been stretched from the Gulf of Tehuantepec to the Bay of text books written on the subject (e.g. Emanuel 1994, Campeche and then eastward across northern Guatemala Cotton and Anthes 1994, Houze 1994). It is known that and Belize to the southwestern Caribbean Sea. The pres- such clouds have a lifetime of an hour or less. During ence of multiple low-level vorticity maxima on the east their developing phase they remove moist air from the side of the larger gyre created great uncertainty in prop- lowest 1-2 km and carry it into the upper troposphere. erly locating the pouch of pre-Nichol. As their updraughts become loaded with liquid water Figure 15 shows a scatter plot of CAPE against CIN and/or ice, the lower part of the updraught decays and for the GV soundings obtained in pre-Nichol. Again, like becomes a downdraught. The downdraught is driven by pre-Matthew, but unlike the ex-Gaston and pre-Karl cases, the collective drag of precipitation and, below cloud base, few soundings had CAPE values exceeding 2.5 kJ kg−1 also to the cooling of air resulting from evaporation of (only 2 out of 21 on 27 September and only 1 out of 21 on the precipitation as it falls through unsaturated air. The 28 September) and none had values exceeding 3 kJ kg−1. collective effect of downdraughts is to spread a thin carpet

Figure 14. Vertical profiles of the day-to-day differences in system-mean virtual potential temperature and pseudo-equivalent potential temperature from the mean profile of all days for pre-Matthew on 20-22 and 24 September. The thick curves mark the first and last days of the sequence. Numbers on curves refer to the date.

Figure 16. Vertical profiles of the pouch mean soundings of virtual potential temperature and pseudo-equivalent potential temperature Figure 15. Scatter plot of CAPE and CIN values for the 42 sound- for pre-Nichol on 27 and 28 September. Numbers on curves refer to ings in pre-Nichol. the date. of cool air near the surface that tends to stabilize the air to subsequent convection. Mixing and detrainment from a spectrum of convective cells with air in the cloud environment leads to a moistening of the environment in middle and upper levels. The traditional argument has been that downdraughts tend to be stronger when the middle-layer air in the cloud environment is dry, but this view has been challenged recently. In section 4.1, it was argued that the failure of Gaston to re-develop was traceable to the injection of dry air into Gaston’s pouch near the middle levels. If the foregoing mechanisms are operating, the implications of this dry air would be a cooling and drying of the near-surface air. However, precisely the opposite is observed! In par- Figure 17. All soundings of θv (red curves) and θe (blue curves) for ticular, the boundary layer θe increased progressively over Hurricane Earl on 01-02 September 2010 as indicated. The black the entire observation period and yet the system did not line is the saturation θe of the mean sounding. develop. While there may be several potential explana- tions for this behaviour, these findings raise a basic ques- When convection occurs in an environment of non- tion. What are the consequences of the dry air on the zero vertical vorticity, basic fluid dynamical considera- ability of the clouds to amplify the background rotation? tions suggest that updraughts would amplify the vorticity

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 16 R. K. SMITH AND M. T. MONTGOMERY by the process of vortex-tube stretching. Even for back- suggested that sustained precipitation in the stratiform ground rotation rates as low as that of an undisturbed cloud deck together with the evaporation of rain drops tropical atmosphere away from the equator, cloud model below would gradually cool and saturate the layer below simulations confirm this tendency to amplify planetary cloud base while advecting cyclonic vorticity downwards vorticity on time scales of an hour (Saunders and Mont- to the surface. Some questions about the dynamics of gomery, 2004; Wissmeier and Smith 2011). A striking this process have been raised by Tory and Montgomery result of such simulations is the exceedingly large ampli- (2006), but the thermodynamical aspects are interesting fication of the vertical component of relative vorticity, also and worthy of investigation using the data analysed which is a maximum at low levels and which persists long in section 4. after the initial updraught has decayed. These cloud model In the pouch-mean data presented for the three devel- simulations indicate also that the induced tangential wind oping storms (Figures 12, 14, and 16), we do not find speeds by a single updraught is typically no more than a cooling of air below 10 km in altitude. Rather, in all cases few meters per second with a horizontal scale of around 10 there is slight warming. km, and would be barely detectable by normal measure- Figure 18 shows the day-to-day evolution of the ment methods in the presence of an ambient wind field. system-mean relative humidity for pre-Karl and pre- Together these results suggest that all non-shallow tropi- Matthew. In both disturbances there was no significant cal convection away from the equator is vortical to some change in this quantity below about 3 km, but there was a degree and can amplify the vertical vorticity locally by clear moistening above this level between the first and last between one and two orders of magnitude. It is not hard to flight into the storm. In pre-Nichol, there was an increase imagine, then, that the stretching of vertical vortex tubes in relative humidity at all altitudes below 12 km between by a developing cumulus cloud is a fundamental process the two flights on 27 and 28 September (figure not shown). and that it may be an important process in tropical cyclo- These findings do not provide unequivocal support for genesis. the Bister and Emanuel picture described above. For one These vortical convective clouds have been identified thing we do not see a progressive increase in low- to as fundamental building blocks during both the tropical mid-troposphere relative humidity in all cases. Further- cyclone genesis and intensification process (Hendricks et more, the moistening that we do see cannot be explained al. 2004, Montgomery et al. 2006, Braun et al. 2010, Fang by evaporative cooling as it occurs in the presence of and Zhang. 2010, Nguyen et al. 2008, Montgomery and increasing temperature: it occurs also through a signifi- Smith 2011). The vertical vorticity that is generated by cant depth of the atmosphere, not just below cloud base. the clouds outlives the convection that produced them in Of course, there is the issue of the horizontal scale dif- the first place. The vortical remnants tend to aggregate ferences between the data gathered during TEXMEX and in a quasi two-dimensional manner with a corresponding those collected during PREDICT. The sampling strategy upscale energy cascade and some of these remnants will for TEXMEX was focused on the scale of a single meso- be intensified further by subsequent convective episodes. scale convective system and its associated mesoscale con- The amplification and aggregation of vorticity represents vective vortex, while that for the PREDICT experiment an increase in the relative circulation within a fixed circuit focused on the broader-scale aspects of genesis. encompassing the convective area. As the circulation Influenced by the TEXMEX study, Raymond and progressively increases in strength, there is some increase Sessions (2007) proposed an alternative model for trop- in the surface moisture fluxes. However, it is not necessary ical cyclogenesis that is linked with the gross moist sta- that the moisture fluxes continue to increase with surface bility of the tropical atmosphere. They investigated the wind speed (Montgomery et al. 2009). This research hypothesis that an increase in the gross moist stability forms the basis of a unified view of tropical cyclogenesis would lead to a fundamental change in the vertical pro- and intensification (Montgomery and Smith 2010). In this file of convergence for a developing tropical disturbance. view, the separate stages proposed in previous significant To this end, they carried out a series of two-dimensional studies and reviews (e.g. Frank, 1987; Emanuel 1989; (zonal-height) cumulus ensemble simulations in the con- McBride 1995; Karyampudi and Pierce 2002; Tory and text of the weak temperature gradient approximation with- Frank, 2010) are unnecessary. out background rotation. The experiments imposed small Another line of research on tropical cyclogenesis 9 thermal and moisture perturbations to a state of radiative- emerged from the TEXMEX experiment . This research convective equilibrium as a possible way of investigating emphasized the importance of thermodynamical pro- the evolution of convection in thermodynamic environ- cesses within the so-called “mesoscale convective vortex ments conducive for tropical cyclogenesis. Environments embryo.” Bister and Emanuel (1997) proposed that the that were cooler at low levels and warmer at upper levels development of a cool, moist environment resulting from on the order 1 K were found to lower the level of max- stratiform rain serves as the incubation region for the for- imum vertical mass flux from 10 km to approximately 5 mation of a low-level, warm-core cyclonic vortex. They km; such environments were found to increase the precipitation rate as well. The net result of these effects is to 9 TEXMEX is the acronym for the Tropical Experiment in Mexico, increase the inflow into the convection. The suggestion which was conducted in the eastern Pacific basin out of Acapulco, Mexico. This experiment is described more fully in Bister and Emanuel is that if realistic values of ambient rotation associated (1997) and Raymond et al. (1998). with a tropical wave or monsoon trough were included,

pertinent to enquire whether there is observational support in the PREDICT data for the conﬁguration envisaged by Raymond and Sessions. As noted above, from the data shown in (Figures 12, 14, and 16), we do not ﬁnd cooling of air below 10 km in altitude. Rather, in all developing cases there is slight warming.

6 Conclusions This work was carried out to determine the salient thermodynamic characteristics of the “pouch region” of developing and nondeveloping tropical disturbances. To this end we have presented an analysis of the dropwindsonde data obtained for the most extensively documented pre- depression disturbances during the PREDICT and GRIP experiments, which were conducted during the peak of the 2010 Atlantic hurricane season. In particular, we have chosen three cases that were extremely well sampled both spatially and temporally. These data are compared with those of a fully-fledged hurricane obtained during the GRIP portion of the experiment. The analyses provide unprecedented quantitative information about the vertical structure, magnitude, and range of variability of important thermodynamic quantities such as virtual potential temperature and equivalent potential temperature. We calculated the mean sounding of virtual potential temperature and pseudo-equivalent potential temperature on each day for the pouch region of the disturbances. The departures from the system mean averaged over the four or five (mostly consecutive) days of observations was presented also. For the three developing disturbances (Karl, Figure 18. Vertical profiles of the day-to-day differences in system- Matthew and Nichol) in the Caribbean region and for the mean relative humidity from the mean profile of all days for (a) system that failed to redevelop (ex-Gaston), there was a pre-Karl during the period 10-14 September, and (b) pre-Matthew during the period 20-24 September. The thick curves mark the first slight to moderate warming in virtual potential tempera- and last days of the sequence. Numbers on curves refer to the date. ture through much of the troposphere during the four or A letter ‘D’ refers to a DC8 flight. five days of observations, but no obvious change in dry static stability. For the case of pre-Nichol, observations were collected only for two consecutive days. this inflow would cause a stronger vorticity convergence Two notable paradigms that have been offered pre- at lower levels and thus contribute to the spin up of the viously to explain tropical cyclogenesis have assumed a system. This effect is argued to be an explanation for modest cooling of the lower troposphere. This presump- why tropical-wave-scale mid-level vortices foster tropical tion is not supported by the PREDICT-GRIP observations. storm formation. In contrast to the developing cases, the equivalent While we have several concerns about Raymond and potential temperature showed a higher degree of variabil- Sessions’ formulation of the problem10, it is nonetheless ity in the non-developing ex-Gaston, and even more variability in the case of mature Hurricane Earl. In ex-Gaston, this variability was found to be largely associated with that 10 First, the temperature and moisture perturbations imposed in the in the water vapour mixing ratio. model are based on observational data taken within a mesoscale convective vortex embryo and proximity soundings of the same mesoscale Perhaps the most potentially significant difference convective system. It would be fortuitous if these proximity soundings between the developing and non-developing systems was were representative of the radiative-convective equilibrium state of the the smaller difference in pseudo-equivalent potential tem- entire tropical disturbance. Indeed, the entire system may not be in a radiative-convective equilibrium state (Brown and Bretherton 1997). perature at the surface and the mid-tropospheric minimum These perturbations are treated as perturbations to the radiative convec- of this quantity. This difference was about 15-17 K in tive equilibrium state of the tropical disturbance. Second, the research the developing systems compared with about 25 K in the flights were typically at 700 mb and thus the data obtained were unable to document the convective environment through the troposphere. Third, non-developing system, ex-Gaston. All systems had sig- without ambient rotation and without a truly three-dimensional config- nificant amounts of CAPE, the largest values (> 4000 J uration of the model, these experiments cannot offer insight into the kg−1) being found in ex-Gaston and soundings with high rotational dynamics of vorticity aggregation as discussed above. It is yet to be demonstrated whether the thermodynamic control as envisaged by values of CAPE tended to have relatively low values of Raymond and Sessions is essential in a rotational environment. CIN. However, the evolution and distribution of CAPE

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj 18 R. K. SMITH AND M. T. MONTGOMERY and CIN by themselves did not reveal an obvious distinc- Emanuel KA. 1994 Atmospheric convection. Oxford Uni- tion between developing and nondeveloping systems. versity Press, 580pp. These observations provide a context for future study Frank WM. 1987: Tropical cyclone formation. In: A global view of tropical cyclones. Ed. R. L. Elsberry, Office of Naval of the interaction between the pouch and its environment Research, 53-90. and of the development of rotating deep convection within Guimond S. R., G. M. Heymsfield, and F. J. Turk, 2010: the pouch. Both of these topics are important elements Multiscale observations of Hurricane Dennis (2005): The effects of the marsupial paradigm and research efforts in both of hot towers on rapid intensification. J. Atmos. Sci., 67, 633- directions are currently underway. 654. Holton JR. 2004 An introduction to dynamic meteorology Academic Press, London, pp535. Houze RA. 2004: Mesoscale Convective Systems. Reviews 7 Acknowledgements of Geophysics, 42, RG4003, 1-43. Houze, RA Lee W-C Bell MM. 2009 Convective contri- We would like to thank the United States National Science bution to the genesis of Hurricane Ophelia (2005). Mon. Wea. Foundation for their support of the experiment called Pre- Rev., 137, 2778-2800. Depression Investigation of Cloud Systems in the Trop- James RP Markowski PM. 2009: A numerical investigation ics (PREDICT). MTM acknowledges the support of NSF of the effects of dry air aloft on deep convection. Mon. Wea. ATM-0733380, NOAAs Hurricane Research Division Rev., 137, 140-161. Karyampudi, V. M., and H. F. Pierce, 2002: Synoptic-scale and NASA grants NNH09AK561 and NNG09HG031. influence of the Saharan air layer on tropical cyclogenesis over RKS acknowledges financial support for tropical-cyclone the eastern Atlantic. Mon. Wea. Rev., 130, 3100-3128. research from the German Research Council (Deutsche McBride JL. 1995: Tropical cyclone formation. In Global Forschungsgemeinschaft). Perspectives on Tropical Cyclones. pp21-62. WMO/TD-No 693 (Ed. R. L. Elsberry), World Meteorological Organization, Appendix: Calculation of CAPE and CIN Geneva, 289pp. Montgomery MT Smith RK. 2010: Tropical-Cyclone For- The CAPE and CIN values given in this paper are cal- mation: Theory and Idealized modelling. Report for the Seventh culated by lifting hypothetical air parcels from the surface International Workshop on Tropical Cyclones, La Réunion, Nov. 2010. World Meteorological Organization, Geneva, Switzerland. and from a height of 100 m above the surface, assum- Montgomery M. T., S. V. Nguyen, and R. K. Smith, 2009: ing pseudo-adiabatic ascent until the LNB. Integrals of the Do tropical cyclones intensify by WISHE? Q. J. R. Meteorol. buoyancy force, proportional to the difference between the Soc., 135, 1697-1714. virtual temperature of a lifted parcel and that of its envi- Ooyama KV. 1969 Numerical simulation of the life cycle ronment at a given height, are performed over the height of tropical cyclones. J. Atmos. Sci., 26, 3-40. Raymond D. J., and S. L. Sessions, 2007: Evolution of ranges of positive buoyancy and negative buoyancy using convection during tropical cyclogenesis, Geophys. Res. Lett., 34, a trapezoidal method with a height interval of 100 m. The L06811, doi:10.1029/2006GL028607 positive integrals for the two lifted parcels are then aver- Reasor, P. D., M. T. Montgomery and L. F. Bosart, 2005: aged to give the CAPE and the two negative integrals are Mesoscale observations of the genesis of Hurricane Dolly averaged to give the CIN. The reader is reminded that this (1996). J. Atmos. Sci., 62, 31513171. Ritchie, E. A. and G. J. Holland, 1999: Large-scale patterns method, like any other, is arbitrary and other methods will associated with tropical cyclogenesis in the western Pacific. give different values. For example, if the averages are per- Mon. Wea. Rev., 127, 2027-2043. formed for air parcels lifted from the surface and heights Rutherford B Montgomery MT. 2011: A Lagrangian char- at intervals of 100 m to an altitude of 500 m, much lower acterization of a developing and non-developing disturbance values of CAPE and higher values of CIN may arise. observed during the PREDICT Experiment. Submitted to Atmos. Chem. Phys. The calculations use Bolton’s formula (Bolton 1980) Saunders and Montgomery MT. 2004: A closer look at to evaluate the pseudo-equivalent potential temperature. vortical hot towers within a tropical cyclogenesis environment. Colorado State University, Atmospheric Science Bluebook No. 752. 8 References Shay, LK. 2010: Air-sea interactions. Chapter 42 of Global perspectives on Tropical cyclones: From science to mitigation. Bister M Emanuel KA. 1997 The genesis of Hurricane (Ed. Kepert JD. Chan JCL.) World Scientific Series on Asia- Guillermo: TEXMEX analyses and a modeling study. Mon. Pacific Weather and Climate, Vol. 4, 448pp. Wea. Rev., 125, 2662-2682. Tory KJ Montgomery MT. 2006: Internal influences on Bolton D. 1980 The computation of equivalent potential tropical cyclone formation. The Sixth WMO International Work- temperature. Mon. Wea. Rev., 108, 10461053. shop on Tropical Cyclones (IWTC-VI), San Jos, Costa Rica, Brown RG Bretherton CS. 1997: A test of the strict quasi- 2006. World Meteorological Organization, Geneva, Switzerland. equilibrium theory on long time and space scales. J. Atmos. Sci., Tory K Frank WM. 2010: Tropical Cyclone Formation. 54, 624-638. Chapter 2 of Global perspectives on Tropical cyclones: From Dunkerton TJ Montgomery MT Wang Z. 2009 Tropical science to mitigation. (Ed. Kepert JD. Chan JCL.) World Scien- cyclogenesis in a tropical wave critical layer: easterly waves. tific Series on Asia-Pacific Weather and Climate, Vol. 4, 448pp. Atmos. Chem. Phys., 9, 55875646. Elsberry R Harr P. 2008 Tropical cyclone structure (TCS08) Field experiment scientific basis, observational plat- forms, and strategy. Asia-Pacific Journal of Atmospheric Sci- ences, 44, 1-23. Emanuel KA. 1989 The finite amplitude nature of tropical cyclogenesis. J. Atmos. Sci., 46, 3431-3456.

Copyright c 2011 Royal Meteorological Society Q. J. R. Meteorol. Soc. 137: 1–18 (2011) Prepared using qjrms3.cls DOI: 10.1002/qj