• Blake, E. S., 2010: Tropical cyclone report—Tropical Storm Gaston (AL092010), 1-2 September 2010. NHC Rep., 9 pp. [Available online at http://www.nhc.noaa.gov/pdf/TCR-AL092010_Gaston.pdf.]

  • Braun, S. A., 2006: High-resolution simulation of Hurricane Bonnie (1998). Part II: Water budget. J. Atmos. Sci., 63, 4364.

  • Braun, S. A., 2010: Reevaluating the role of the Saharan air layer in Atlantic tropical cyclogenesis and evolution. Mon. Wea. Rev., 138, 20072037.

    • Search Google Scholar
    • Export Citation
  • Braun, S. A., , J. A. Sippel, , and D. S. Nolan, 2012: The impact of dry midlevel air on hurricane intensity in idealized simulations with no mean flow. J. Atmos. Sci., 69, 236257.

    • Search Google Scholar
    • Export Citation
  • Brown, D. P., , J. L. Beven, , J. L. Franklin, , and E. S. Blake, 2010: Atlantic hurricane season of 2008. Mon. Wea. Rev., 138, 19752001.

  • Bryan, G. H., , and H. Morrison, 2012: Sensitivity of a simulated squall line to horizontal resolution and parameterization of microphysics. Mon. Wea. Rev., 140, 202225.

    • Search Google Scholar
    • Export Citation
  • Bryan, G. H., , J. C. Wyngaard, , and J. M. Fritsch, 2003: Resolution requirements for the simulation of deep moist convection. Mon. Wea. Rev., 131, 23942416.

    • Search Google Scholar
    • Export Citation
  • Carlson, T. N., , and J. M. Prospero, 1972: The large-scale movement of Saharan air outbreaks over the northern equatorial Atlantic. J. Appl. Meteor., 11, 283297.

    • Search Google Scholar
    • Export Citation
  • Davis, C. A., , and D. A. Ahijevych, 2012: Mesoscale structural evolution of three tropical weather systems observed during PREDICT. J. Atmos. Sci., 69, 12481305.

    • Search Google Scholar
    • Export Citation
  • Draxler, R. R., , and G. D. Hess, 1998: An overview of the HYSPLIT_4 modeling system of trajectories, dispersion, and deposition. Aust. Meteor. Mag., 47, 295308.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107.

    • Search Google Scholar
    • Export Citation
  • Dunion, J. P., , and C. S. Velden, 2004: The impact of the Saharan air layer on Atlantic tropical cyclone activity. Bull. Amer. Meteor. Soc., 85, 353365.

    • Search Google Scholar
    • Export Citation
  • Dunkerton, T. J., , M. T. Montgomery, , and Z. Wang, 2009: Tropical cyclogenesis in a tropical wave critical layer: Easterly waves. Atmos. Chem. Phys., 9, 55875646.

    • Search Google Scholar
    • Export Citation
  • Evan, A. T., , J. Dunion, , J. A. Foley, , A. K. Heidinger, , and C. S. Velden, 2006: New evidence for a relationship between Atlantic tropical cyclone activity and African dust outbreaks. Geophys. Res. Lett., 33, L19813, doi:10.1029/2006GL026408.

    • Search Google Scholar
    • Export Citation
  • Fang, J., , and F. Zhang, 2010: Initial development and genesis of Hurricane Dolly (2008). J. Atmos. Sci., 67, 655672.

  • Gamache, J. F., , R. A. Houze, Jr., , and F. D. Marks, Jr., 1993: Dual-aircraft investigation of the inner core of Hurricane Norbert. Part III: Water budget. J. Atmos. Sci., 50, 32213243.

    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., , and J.-O. J. Lim, 2006: The WRF single-moment microphysics scheme (WSM6). J. Korean Meteor. Soc., 42, 129151.

  • Hong, S.-Y., , Y. Noh, , and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341.

    • Search Google Scholar
    • Export Citation
  • Hopsch, S. B., , C. D. Thorncroft, , and K. R. Tyle, 2010: Analysis of African easterly wave structures and their role in influencing tropical cyclogenesis. Mon. Wea. Rev., 138, 10571077.

    • Search Google Scholar
    • Export Citation
  • James, R. P., , and P. M. Markowski, 2009: A numerical investigation of the effects of dry air aloft on deep convection. Mon. Wea. Rev., 137, 140161.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181.

  • Karyampudi, V. M., , and T. N. Carlson, 1988: Analysis and numerical simulations of the Saharan air layer and its effect on easterly wave disturbances. J. Atmos. Sci., 45, 31023136.

    • Search Google Scholar
    • Export Citation
  • Karyampudi, V. M., , and H. F. Pierce, 2002: Synoptic-scale influence of the Saharan air layer on tropical cyclogenesis over the eastern Atlantic. Mon. Wea. Rev., 130, 31003128.

    • Search Google Scholar
    • Export Citation
  • Landsea, C. W., 1993: A climatology of intense (or major) Atlantic hurricanes. Mon. Wea. Rev., 121, 17031713.

  • Lau, W. K. M., , and K.-M. Kim, 2007a: Cooling of the Atlantic by Saharan dust. Geophys. Res. Lett., 34, L23811, doi:10.1029/2007GL031538.

    • Search Google Scholar
    • Export Citation
  • Lau, W. K. M., , and K.-M. Kim, 2007b: How nature foiled the 2006 hurricane forecasts. Eos, Trans. Amer. Geophys. Union, 88, 105, doi:10.1029/2007EO090002.

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

    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., , S. J. Taubman, , P. D. Brown, , M. J. Iacono, , and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmosphere: RRTM, a validated correlated-k model for the long wave. J. Geophys. Res., 102 (D14), 16 66316 682.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , L. L. Lussier III, , R. W. Moore, , and Z. Wang, 2010a: The genesis of Typhoon Nuri as observed during the Tropical Cyclone Structure 2008 (TCS-08) field experiment – Part 1: The role of the easterly wave critical layer. Atmos. Chem. Phys., 10, 98799900.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , Z. Wang, , and T. J. Dunkerton, 2010b: Coarse, intermediate and high resolution numerical simulations of the transition of a tropical wave critical layer to a tropical storm. Atmos. Chem. Phys., 10, 10 80310 827.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., and Coauthors, 2012: The Pre-Depression Investigation of Cloud-Systems in the Tropics (PREDICT) experiment: Scientific basis, new analysis tools, and some first results. Bull. Amer. Meteor. Soc., 93, 153172.

    • Search Google Scholar
    • Export Citation
  • Nolan, D. S., 2007: What is the trigger for tropical cyclogenesis? Aust. Meteor. Mag., 56, 241266.

  • Raymond, D. J., , and C. Lopez Carrillo, 2011: The vorticity budget of developing Typhoon Nuri (2008). Atmos. Chem. Phys., 11, 147163.

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

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , Y. Chen, , W. Wang, , C. Davis, , J. Dudhia, , and G. J. Holland, 2009: Large-eddy simulation of an idealized tropical cyclone. Bull. Amer. Meteor. Soc., 90, 17831788.

    • Search Google Scholar
    • Export Citation
  • Rozoff, C. M., , J. P. Kossin, , W. H. Schubert, , and P. J. Mulero, 2009: Internal control of hurricane intensity variability: The dual nature of potential vorticity mixing. J. Atmos. Sci., 66, 133147.

    • Search Google Scholar
    • Export Citation
  • Rutherford, B., , and M. T. Montgomery, 2011: A Lagrangian analysis of a developing and non-developing disturbance observed during the PREDICT experiment. Atmos. Chem. Phys. Discuss., 11, 33 27333 323.

    • Search Google Scholar
    • Export Citation
  • Sippel, J. A., , S. A. Braun, , and C.-L. Shie, 2011: Environmental influences on the strength of Tropical Storm Debby (2006). J. Atmos. Sci., 68, 25572581.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., , J. B. Klemp, , J. Dudhia, , D. O. Gill, , D. M. Barker, , W. Wang, , and J. G. Powers, 2005: A description of the Advanced Research WRF version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp.

  • Smith, R. K., , and M. T. Montgomery, 2011: Observations of the convective environment in developing and non-developing tropical disturbances. Quart. J. Roy. Meteor. Soc., 137, 120.

    • Search Google Scholar
    • Export Citation
  • Stewart, S. R., , and J. L. Beven II, 2008: Tropical cyclone report—Tropical Storm Fay (AL062008), 15-26 August 2008. NHC Rep., 29 pp. [Available online at http://www.nhc.noaa.gov/pdf/TCR-AL062008_Fay.pdf.]

  • Sun, D., , K.-M. Lau, , and M. Kafatos, 2008: Contrasting the 2007 and 2005 hurricane seasons: Evidence of possible impacts of Saharan dry air and dust on tropical cyclone activity in the Atlantic basin. Geophys. Res. Lett., 35, L15405, doi:10.1029/2008GL034529.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., 2012a: Role of intense updrafts in tropical cyclone formation. Preprints, 30th Conf. on Hurricanes and Tropical Meteorology, Ponte Vedra Beach, FL, Amer. Meteor. Soc., P1.28. [Available online at https://ams.confex.com/ams/30Hurricane/webprogram/Paper205467.html.]

  • Wang, Z., 2012b: Thermodynamic aspects of tropical cyclone formation. J. Atmos. Sci., 69, 24332451.

  • Wang, Z., , M. T. Montgomery, , and T. J. Dunkerton, 2009: A dynamically-based method for forecasting tropical cyclogenesis location in the Atlantic sector using global model products. Geophys. Res. Lett., 36, L03801, doi:10.1029/2008GL035586.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , M. T. Montgomery, , and T. J. Dunkerton, 2010a: Genesis of pre–Hurricane Felix (2007). Part I: The role of the wave critical layer. J. Atmos. Sci., 67, 17111729.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , M. T. Montgomery, , and T. J. Dunkerton, 2010b: Genesis of pre–Hurricane Felix (2007). Part II: Warm core formation, precipitation evolution and predictability. J. Atmos. Sci., 67, 17301744.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , T. J. Dunkerton, , and M. T. Montgomery, 2012a: Application of the marsupial paradigm to tropical cyclone formation from northwestward-propagating disturbances. Mon. Wea. Rev., 140, 6676.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , M. T. Montgomery, , and C. Fritz, 2012b: A first look at the structure of the wave pouch during the 2009 PREDICT–GRIP dry runs over the Atlantic. Mon. Wea. Rev., 140, 11441163.

    • Search Google Scholar
    • Export Citation
  • Zhang, D.-L., , Y. Liu, , and M. K. Yau, 2002: A multiscale numerical study of Hurricane Andrew (1992). Part V: Inner-core thermodynamics. Mon. Wea. Rev., 130, 27452763.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (left) The CIMSS TPW (shading) and 850-hPa storm-relative streamlines. (right) The GOES IR imagery (shading) and the 700-hPa storm-relative streamlines from (top to bottom) 1200 UTC 11 Aug 2008 to 1200 UTC 15 Aug 2008. The pouch center of the pre-Fay disturbance is indicated by the intersection of the critical latitude (pink lines) and the easterly wave trough axis (black lines) in the panels.

  • View in gallery

    CIMSS mid- to upper-level wind analyses (wind barbs: blue—100–250, yellow—251–350, and green—351–500 hPa) and water vapor imagery for (a) pre-Fay at 1200 UTC 13 Aug 2008 and (b) ex-Gaston at 1800 UTC 4 Sep 2010. The red circles identify the tropical disturbances.

  • View in gallery

    A dropsonde sounding at 1824 UTC 5 Sep 2010 at 16.0521°N, 49.0578°W (less than 1° southwest of the pouch center). [Data from the Earth Observing Laboratory (EOL)’s PREDICT field catalog.]

  • View in gallery

    Time–height cross sections of (a) relative vorticity, (b) OW parameter (10−9 s−2), (c) RH, and (d) equivalent potential temperature averaged in a 2° × 2° box following the pouch center for Fay (2008) between (left to right) 0000 UTC 13 Aug and 0000 UTC 16 Aug 2008.

  • View in gallery

    Azimuthally averaged water vapor budget fields for Fay (2008) averaged between 36 and 60 h: (a) net water vapor tendency, (b) mean horizontal flux convergence, (c) mean vertical flux convergence, (d) net condensation, (e) eddy horizontal flux convergence, and (f) eddy vertical flux convergence. The units of the variables are g kg−1 day−1.

  • View in gallery

    Azimuthally averaged vertical motion (shading; cm s−1) and radial flow (contours; m s−1) for (a) pre-Fay during 36–60 h, (b) pre-Fay during 29–34 h, and (c) ex-Gaston during 32–37 h.

  • View in gallery

    As in Fig. 5, but all terms are averaged over the period 29–34 h.

  • View in gallery

    Trajectory analysis of pre-Fay: forward trajectory and streamlines in the comoving frame of reference superimposed on RH (shading; %) for (a) 25 and (c) 60 h. (b),(d) Time series of RH and pressure along the parcel trajectory. In (a),(c) the points along the trajectories represent the positions of the parcel every 6 h and the yellow arrows denote the beginning point of the trajectory.

  • View in gallery

    As in Fig. 4, but for Gaston (2010) from 0000 UTC 4 Sep 2010 to 0000 UTC 7 Sep 2010.

  • View in gallery

    (a) The 3-km relative humidity and storm-relative streamlines for Gaston (2010) at 0800 UTC 5 Sep 2010 with a group of ensemble forward parcel trajectories (gray). (b) Vertical cross section of RH along 17.5°N (contour intervals are set to 15%) and backward trajectories (gray) projected onto the longitude–height plane. The box in (a) highlights a pocket of dry air near the pouch center and the line indicates the cross section location shown in (b). Both the particle trajectories and streamlines are shown in a wave comoving framework. The yellow dots indicate the initialization location of the ensemble trajectories.

  • View in gallery

    As in Fig. 4, but for the simulation of ex-Gaston averaged over 32–37 h.

  • View in gallery

    As in Fig. 8, but for the 3-km resolution simulation of Gaston (2010).

  • View in gallery

    Time–height cross section of vertical velocity (w) for ex-Gaston (2010) from (a) the control run and (b) the high-resolution simulation, averaged in a 2° × 2° box following the pouch center. The contour levels are −0.04, −0.03, −0.02, −0.01, 0, 0.1, 0.15, and 0.2 m s−1, and negative values are shaded.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1 1 1
PDF Downloads 2 2 2

A Numerical Study of the Impacts of Dry Air on Tropical Cyclone Formation: A Development Case and a Nondevelopment Case

View More View Less
  • 1 Department of Atmospheric Sciences, University of Illinois at Urbana–Champaign, Urbana, Illinois
© Get Permissions
Full access

Abstract

The impacts of dry air on tropical cyclone formation are examined in the numerical model simulations of ex-Gaston (2010) and pre-Fay (2008). The former, a remnant low downgraded from a short-lived tropical cyclone, can be regarded as a nondeveloping system because it failed to redevelop, and the latter developed into a tropical cyclone despite lateral dry air entrainment and a transient upper-level dry air intrusion. Water vapor budget analysis suggests that the mean vertical moisture transport plays the dominant role in moistening the free atmosphere. Backward trajectory analysis and water budget analysis show that vertical transport of dry air from the middle and upper troposphere, where a well-defined wave pouch is absent, contributes to the midlevel drying near the pouch center in ex-Gaston. The midlevel drying suppresses deep convection, reduces moisture supply from the boundary layer, and contributes to the nondevelopment of ex-Gaston. Three-dimensional trajectory analysis based on the numerical model simulation of Fay suggests that dry air entrained at the pouch periphery tends to stay off the pouch center because of the weak midlevel inflow or gets moistened along its path even if it is being wrapped into the wave pouch. Lateral entrainment in the middle troposphere thus does not suppress convection near the pouch center or prevent the development of Tropical Storm Fay. This study suggests that the upper troposphere is a weak spot of the wave pouch at the early formation stage and that the vertical transport is likely a more direct pathway for dry air to influence moist convection near the pouch center.

Corresponding author address: Zhuo Wang, Department of Atmospheric Sciences, University of Illinois at Urbana–Champaign, 105 South Gregory St., Urbana, IL 61801. E-mail: zhuowang@illinois.edu

Abstract

The impacts of dry air on tropical cyclone formation are examined in the numerical model simulations of ex-Gaston (2010) and pre-Fay (2008). The former, a remnant low downgraded from a short-lived tropical cyclone, can be regarded as a nondeveloping system because it failed to redevelop, and the latter developed into a tropical cyclone despite lateral dry air entrainment and a transient upper-level dry air intrusion. Water vapor budget analysis suggests that the mean vertical moisture transport plays the dominant role in moistening the free atmosphere. Backward trajectory analysis and water budget analysis show that vertical transport of dry air from the middle and upper troposphere, where a well-defined wave pouch is absent, contributes to the midlevel drying near the pouch center in ex-Gaston. The midlevel drying suppresses deep convection, reduces moisture supply from the boundary layer, and contributes to the nondevelopment of ex-Gaston. Three-dimensional trajectory analysis based on the numerical model simulation of Fay suggests that dry air entrained at the pouch periphery tends to stay off the pouch center because of the weak midlevel inflow or gets moistened along its path even if it is being wrapped into the wave pouch. Lateral entrainment in the middle troposphere thus does not suppress convection near the pouch center or prevent the development of Tropical Storm Fay. This study suggests that the upper troposphere is a weak spot of the wave pouch at the early formation stage and that the vertical transport is likely a more direct pathway for dry air to influence moist convection near the pouch center.

Corresponding author address: Zhuo Wang, Department of Atmospheric Sciences, University of Illinois at Urbana–Champaign, 105 South Gregory St., Urbana, IL 61801. E-mail: zhuowang@illinois.edu

1. Introduction

Dry air intrusion is one of many factors that affect tropical cyclone formation, particularly over the North Atlantic, where dry air can be attributed to frequent Saharan dust outbreaks and large-scale subsidence (Carlson and Prospero 1972; Karyampudi and Carlson 1988; Braun 2010). Using Geostationary Operational Environmental Satellite (GOES) split-window imagery and GPS dropsonde data, Dunion and Velden (2004) suggested that the Saharan air layer (SAL) suppressed tropical cyclone activity over the Atlantic. They hypothesized that the SAL may inhibit tropical cyclone intensification in three main ways: (i) dry, stable air entrained into the storm may promote convectively driven downdrafts; (ii) the warm, dry air mass north of the midlevel easterly jet may enhance the meridional temperature gradient and increase the wind shear; and (iii) radiative heating of the dust in the warm SAL air mass would enhance the preexisting trade wind inversion and stabilize the environment. Based on a comparison to the 2005 hurricane season, Lau and Kim (2007b,a) and Sun et al. (2008) attributed the reduced tropical cyclone activity during the 2006 and 2007 hurricane seasons to the increased SAL activity in those 2 yr. Evan et al. (2006) identified an inverse relationship between the North Atlantic tropical cyclone days and the atmospheric dust cover for the period 1982–2005.

Some earlier studies proposed that the SAL might have positive influences on tropical cyclone development. Karyampudi and Carlson (1988) and Karyampudi and Pierce (2002), for example, suggested the SAL is an important, “if not necessary,” condition for easterly wave growth because the SAL leads to strong baroclinicity along its southern border and strengthens the midlevel easterly jet. The enhanced positive vorticity advection in the left exit quadrant of the jet promotes upward motion in the lower troposphere and enhances convective precipitation, which is favorable for the maintenance and intensification of tropical wave disturbances.

The negative influences of the SAL on tropical cyclone intensification were also disputed recently by Braun (2010). He showed that comparisons between strengthening and weakening storms provide little evidence for significant negative impacts of the SAL. Braun (2010) further suggested that the SAL might even aid tropical cyclone development by focusing convection on the cyclonic (southern) side of the easterly jet. Using idealized numerical model simulations without mean flow, Braun et al. (2012) showed that dry air only slows down storm intensification if it gets near the storm center at the early stage. Without mean flow convection eventually moistens the free atmosphere, and all the storms reach the same intensity despite the different initial dry air distributions.

A majority of the tropical cyclones over the Atlantic originate from tropical easterly waves (e.g., Landsea 1993). The impacts of dry air on tropical cyclone formation thus need to be studied in the context of these synoptic-scale easterly waves. A new framework for tropical cyclone formation within tropical waves was recently proposed by Dunkerton et al. (2009, hereafter DMW09). By examining tropical easterly waves that developed into named tropical storms, DMW09 demonstrated that the critical layer of a tropical easterly wave, which forms from the nonlinear interaction of the wave with the mean flow, is the preferred location for tropical cyclone formation. The cat’s eye in the wave critical layer is a region of weak strain–shear deformation and provides a favorable environment for deep convection and vorticity aggregation in the lower troposphere. As a region of approximately closed Lagrangian circulation, the cat’s eye also protects the tropical cyclone protovortex1 from the generally hostile environment (i.e., dry air intrusion and shear deformation) to some extent. This can be regarded as the “marsupial paradigm,” as the entire sequence is likened to the development of a marsupial infant in its mother’s pouch, wherein the juvenile protovortex is carried along by the mother wave until it is strengthened into a self-sustaining entity. The closed circulation within the wave critical layer is also called the “wave pouch.”

The marsupial paradigm is supported by both observational diagnoses (DMW09; Wang et al. 2009; Montgomery et al. 2010a; Wang et al. 2012a,b) and numerical model simulations (Wang et al. 2010a,b; Montgomery et al. 2010b; Wang 2012a; Fang and Zhang 2010). Based on a high-resolution numerical model simulation and the analysis of dropsonde data from a field experiment, Wang (2012b) showed that the meso-β area near the pouch center or the pouch core region is characterized by high saturation fraction, small differences in equivalent potential temperature between the surface and the midtroposphere, and a short incubation time scale. Although stratiform precipitation prevails within the wave pouch, convective precipitation tends to recur near the pouch center. The associated convective heating, due to its strong radial and vertical gradients, can effectively drive a transverse circulation and spin up a surface vortex near the pouch center. Therefore, the thermodynamic conditions and moist convection near the pouch center are critically important for tropical cyclone formation.

The marsupial paradigm provides a framework to systematically examine the dynamic and thermodynamic evolution of precursor disturbances. The objective of our study is to understand how dry air affects tropical cyclone formation through the analysis of numerical model simulations in the marsupial framework, particularly how dry air may get into a wave pouch and influence the moist convection near the pouch center. The pregenesis evolution of Tropical Storm Fay (2008) and the poststorm evolution of Tropical Storm Gaston (2010) are simulated using the Weather Research and Forecasting (WRF) model. The former was a developing easterly wave, while the later can be regarded as a nondeveloper (see section 2 for more details). Both disturbances encountered dry air entrainment–intrusion during their lifetime, but the impact of the dry air on the evolution of Gaston (2010) differs substantially from that of Fay (2008) as shown in the following sections. By comparing the evolutions of the two storms, we hope to have a better understanding of the impacts of dry air on tropical cyclone formation.

An outline of the remaining paper is as follows. The synoptic overview of both Gaston (2010) and Fay (2008) are presented in section 2. In section 3, descriptions of the numerical model simulations, three-dimensional trajectory analysis, and water vapor budget formulation are presented. Sections 4 and 5 present diagnoses of Fay and Gaston, respectively, followed by a summary and discussion in section 6.

2. Synoptic overview of Fay (2008) and Gaston (2010)

Tropical Storm Fay (2008) developed from a tropical easterly wave that departed the African coast on 6 August 2008 (Brown et al. 2010). Fay was declared a tropical depression at 1200 UTC 15 August 2008 and upgraded to a tropical storm 6 h later. According to the National Hurricane Center (NHC) tropical cyclone report (Stewart and Beven 2008), it remained as a tropical storm until it weakened to a tropical depression at 0000 UTC 24 August. Prior to genesis, the pre-Fay disturbance encountered dry air from the northwest. Figure 1 (left panels) shows the Cooperative Institute for Meteorological Satellite Studies’ (CIMSS) total precipitable water (TPW) superimposed onto the 850-hPa streamlines in the frame of reference moving at the same speed with the wave from 1200 UTC 11 August (day −4) through 1200 UTC 15 August (day 0; genesis day). The flow becomes quasi stationary in the frame of reference that moves zonally at the same speed with the wave, and the streamlines in this comoving frame of reference can therefore be used as a good approximation to parcel trajectories. The wave pouch is depicted by the closed streamlines in Fig. 1 and is relatively moist compared to the surrounding air. CIMSS TPW shows that dry air was entrained into the wave pouch at the northwestern and southern quadrants on 12–13 August, and that convective activity, as indicated by GOES IR (right panels in Fig. 1), diminished on 13 August at the pouch periphery. The mid- to upper-level wind analyses and water vapor imagery (Fig. 2a) illustrate a well-defined upper-level low centered near 30°N, 50°W, just to the northeast of pre-Fay at this time. The passage of the upper-level low likely induced increased flow deformation along the storm’s northwestern flank, where a shearing deformation axis is evident. The flow deformation would facilitate dry air entrainment into the wave pouch and lead to the disorganization of convection and the misalignment of the wave pouch. However, the dry air did not seem to penetrate to the center of the circulation, and convection near the pouch center remained vigorous and became better organized on 14–15 August (Fig. 1). The disturbance developed into a tropical storm on 1800 UTC 15 August according to the NHC best-track data. Our simulation focuses on the pregenesis evolution of Fay.

Fig. 1.
Fig. 1.

(left) The CIMSS TPW (shading) and 850-hPa storm-relative streamlines. (right) The GOES IR imagery (shading) and the 700-hPa storm-relative streamlines from (top to bottom) 1200 UTC 11 Aug 2008 to 1200 UTC 15 Aug 2008. The pouch center of the pre-Fay disturbance is indicated by the intersection of the critical latitude (pink lines) and the easterly wave trough axis (black lines) in the panels.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

Fig. 2.
Fig. 2.

CIMSS mid- to upper-level wind analyses (wind barbs: blue—100–250, yellow—251–350, and green—351–500 hPa) and water vapor imagery for (a) pre-Fay at 1200 UTC 13 Aug 2008 and (b) ex-Gaston at 1800 UTC 4 Sep 2010. The red circles identify the tropical disturbances.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

Another storm examined in this study is Gaston, which was a short-lived tropical storm (Blake 2010). According to the NHC tropical cyclone report (Blake 2010), Tropical Storm Gaston (2010) developed from an African easterly wave (AEW) near 13°N, 35°W on 1 September 2010 that tracked west-northwestward with maximum sustained surface wind speeds of 18 m s−1. Despite warm sea surface temperatures (SSTs) (28°–31°C) and moderate wind shear as indicated by the NHC observational analysis, Gaston (2010) quickly weakened to a tropical depression within 24 h and then was downgraded to a remnant low. The remnant low continued moving westward but did not reintensify. Our simulation focuses on the poststorm stage of Gaston, or the ex-Gaston disturbance. It thus can be regarded as a nondeveloping disturbance.

Ex-Gaston was closely monitored during the National Science Foundation (NSF)-sponsored PREDICT field experiment (Montgomery et al. 2012) from 2 September 2010 to 7 September 2010. The Pre-Depression Investigation of Cloud-Systems in the Tropics (PREDICT) dropsonde data indicated that the atmosphere was relatively dry in the middle and upper troposphere (Fig. 3). The dryness is likely associated with subsidence on the southwest flank of an upper-level anticyclone (Fig. 2b). The observational analysis during the third flight mission (5 September 2010) showed that relative humidity at 700 mb just exceeded 60% near the core and much less to the south and east farther away from the core (less than 20% at 700 mb) (Davis and Ahijevych 2012). Following this period, soundings from latter missions continued to show extreme dryness in the eastern and southern portions of the system, where the west remained relatively moist.

Fig. 3.
Fig. 3.

A dropsonde sounding at 1824 UTC 5 Sep 2010 at 16.0521°N, 49.0578°W (less than 1° southwest of the pouch center). [Data from the Earth Observing Laboratory (EOL)’s PREDICT field catalog.]

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

3. Model simulations and analysis methodology

a. Numerical model simulation description

The model used in this study is the Advanced Research WRF model (ARW-WRF), version 3.2.1 (Skamarock et al. 2005). The model is fully compressible and nonhydrostatic. A high-resolution numerical model simulation was conducted for Tropical Storm Fay (2008) by adopting a four-grid nested domain with a horizontal grid spacing of 27–9–3–1 km, respectively. The outer two grids are fixed, while the inner two grids move with the pouch center (Wang 2012a). Convection was resolved explicitly at the grid scale except on the outermost mesh (27-km resolution), where the Kain–Fritcsh cumulus scheme (Kain 2004) was used. Other physics options include the Yonsei University planetary boundary layer scheme (Hong et. al. 2006), the Dudhia (1989) shortwave radiation scheme, the Rapid Radiative Transfer Model (RRTM) longwave radiation scheme (Mlawer et al. 1997), and the WRF single-moment 6-class (WSM6) microphysics scheme. Prognostic water substance variables in the WSM6 scheme include mixing ratios of water vapor, cloud water, cloud ice, snow, rain, and graupel (Hong and Lim 2006). Initial and boundary conditions were obtained from the European Centre for Medium-Range Weather Forecasts (ECMWF) Interim Re-Analysis (ERA-Interim) 6-hourly data. The simulation was started at 0000 UTC 13 August 2008 and run 84 h until 1200 UTC 16 August 2008, which covers the tropical wave, the tropical depression, and the tropical storm stages.

The evolution of ex-Gaston from 0000 UTC 4 September to 0000 UTC 7 September 2010 was also simulated using the WRF model. It turned out to be very challenging to simulate the nondevelopment of ex-Gaston. Several simulations were carried out with different domain sizes or horizontal resolutions. The disturbance intensifies in most of the simulations. Our diagnoses of ex-Gaston are based on a two-grid nested simulation with a horizontal grid spacing of 27–9-km resolution, in which the disturbance weakens as observed. The same physics options were adopted as in the Fay (2008) simulation, and the Kain–Fritcsh cumulus scheme was only used on the outermost grid with a resolution of 27 km. Similar to Fay, initial and boundary conditions were obtained from the ERA-Interim 6-hourly data. The sensitivity to the model resolution is briefly discussed in section 5c.

For each disturbance, two propagation speeds were determined based on the Hovmöller diagrams of total precipitable water and 700-hPa meridional wind, respectively. The average of the two estimated speeds (Gaston: −6.3 m s−1; Fay: −7.0 m s−1) was used to determine the critical latitude,2 and the pouch center was determined as the intersection of the critical latitude and the wave trough axis at 3 km (near 700 hPa).

b. 3D trajectory analysis description

The Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model was used in this study (Draxler and Hess 1998). The model is used to assess the three-dimensional (3D) trajectories of dry air parcels. The HYSPLIT model has compared favorably with a variety of real-time data observations (measured balloon trajectories, air concentrations of inert tracers, etc.) with some limitations pertaining to boundary layer trajectory calculations due to sensitivity to the low-level wind profile (Draxler and Hess 1998). In addition to trajectories, relative humidity and pressure are calculated along the path of a parcel by interpolating the value of a variable to the parcel location along the trajectory at the corresponding time. To examine the source of dry air, 3D parcel trajectory analyses are conducted for both ex-Gaston (2010) and pre-Fay (2008) using the 10-min output from the 9-km resolution grid. Examination of the paths of dry air parcels aids understanding of the impacts of dry air on mesoscale convection and the subsequent development–nondevelopment of a tropical wave.

To test the robustness of the results, ensemble parcel trajectories are carried out using the trajectory ensemble option in HYSPLIT. Multiple trajectories are initialized around a selected starting location by introducing a small offset in their initial location (one grid point in the horizontal and 0.01 sigma units in the vertical); this method results in 27 members for all possible offsets in x, y, and z for one ensemble group. To compare with streamlines in the comoving frame of reference, trajectories are displayed in the wave-relative frame of reference.

c. Budget formulation

The water vapor budget of mature tropical storms has been examined in many previous studies based on numerical model simulations or radar observations (e.g., Marks and Houze 1987; Gamache et al. 1993; Zhang et al. 2002; Braun 2006), but it has not been well studied at the pregenesis stage. Since moist convection is the driving force for tropical cyclone formation, a better understanding of the water vapor budget will help us better understand the impacts of dry air on tropical cyclone formation.

The azimuthal-mean water vapor budget equation in cylindrical coordinates can be written as
e1
where qυ is the water vapor mixing ratio; r is the radius with respect to the pouch center; u and w are the radial and vertical velocities, respectively; and the overbar denotes azimuthal average. The term on the left-hand side (LHS) of the equation is the water vapor tendency in the wave comoving frame of reference; the first two terms on the right-hand side (RHS) are the horizontal moisture flux convergence and the vertical moisture flux convergence, respectively. The third term is the divergence term, which is generally negligible; represents the net condensation; represents the contribution from the planetary boundary layer parameterization to the vapor budget; and the residual term includes the effects of turbulent diffusion and artificial source terms by setting negative mixing ratios to 0 (Braun 2006). In the equation, and are output directly from the model, and the remaining terms are derived from the WRF model output.
The first two RHS terms can be further separated as the azimuthal mean and the eddy term as follows:
e2
where a prime denotes the asymmetric component with respect to the azimuthal average. The first two terms on the RHS are the mean horizontal flux convergence term and the mean vertical flux convergence term, respectively, which are associated with the azimuthal-mean transverse circulation; and the third and fourth terms are the eddy horizontal and vertical flux convergence terms, respectively, which are associated with the pouch-relative flow or asymmetric eddies. A maximum radius of 500 km with the bin size 10 km is used to evaluate each term in Eqs. (1) and (2). We will focus on the net tendency, the mean and eddy flux convergence terms, and the net condensation term.

4. Tropical Storm Fay

The pregenesis evolution of Tropical Storm Fay (2008) is examined in this section. The evolution of Fay in the model simulation can be divided into three stages of development: tropical wave stage, tropical depression stage, and tropical storm stage. In this study, we define the genesis of a tropical depression as the formation of a closed surface circulation in the earth-relative frame of reference with the maximum 10-m wind speed greater than 10 m s−1 but less than 17 m s−1, which occurs at 0600 UTC 15 August 2008 in the model simulation. A tropical storm forms at 1200 UTC 15 August 2008 in the model simulation when the maximum 10-m wind speed exceeds 17 m s−1. The tropical depression formation and the tropical storm formation in the model simulation both occur 6 h earlier than those according to the NHC best track.

The time–height cross sections of relative vorticity (zeta), Okubo–Weiss (OW) parameter (Rozoff et al. 2009; DMW09), relative humidity (RH), and equivalent potential temperature (θe) from the numerical model simulation of Fay (2008) (derived from 9-km grid output) are shown in Fig. 4. The 2° × 2° box averages are calculated following the pouch center at 3-km altitude and represent the dynamic and thermodynamic conditions of the air column close to the pouch center, which is the preferred location for tropical cyclogenesis (Wang 2012b). Maximum cyclonic vorticity is initially located around 4 km, associated with the precursor wave. With the increase in cyclonic vorticity near the surface, the maximum vorticity (~3 × 10−4 s−1) occurs below 2 km between 1000 UTC 14 August and 1800 UTC 15 August, which implies a warm-core structure. The storm is subject to the influence of land near the end of the calculation period.

Fig. 4.
Fig. 4.

Time–height cross sections of (a) relative vorticity, (b) OW parameter (10−9 s−2), (c) RH, and (d) equivalent potential temperature averaged in a 2° × 2° box following the pouch center for Fay (2008) between (left to right) 0000 UTC 13 Aug and 0000 UTC 16 Aug 2008.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

The OW parameter is a measure of the “shape preserving” component of a vortical flow (DMW09) and is defined as OW = , where is relative vorticity and represents the strain rate. Positive values of OW (Fig. 4b) indicate that the flow is rotation dominant, and negative values of OW suggest that the flow is susceptible to rapid filamentation. Positive OW is largely confined below 6 km prior to 1000 UTC 15 August. Just before the formation of the tropical depression, there is a significant increase of OW near the surface, and positive OW extends to the upper troposphere shortly after 1200 UTC 15 August, indicating the formation and intensification of a deep vortex.

In the time–height cross section of relative humidity (Fig. 4c), there are two prominent intrusions of dry air in the middle and upper troposphere prior to the formation of a tropical cyclone. The primary intrusion of dry air takes place between 0000 and 1400 UTC 13 August above 6 km, where RH decreases to as low as 30%. RH below 5 km, however, increases at the same time. A secondary intrusion of dry air starts around 0000 UTC 14 August 2008. The extreme dry air (RH < 55%) is confined above 8 km. Although RH values are reduced to less than 60% above 6 km, the air below 4 km remains quite moist (RH > 85%), and RH starts to gradually increase 6 h later, which suggests that the dry air does not suppress convection near the pouch center. It is worth noting that the lateral entrainment indicated by CIMSS TPW (left panels in Fig. 2) is not represented in this figure, as the 2° × 2° box only represents a small area near the pouch center.

The time–height cross section of θe is displayed in Fig. 4d. The vertical profile of θe is characterized by a midlevel minimum between 3 and 8 km. The two aforementioned dry air intrusion events are accompanied by a θe minimum less than 340 K around 6 km. The midlevel θe shows an overall increasing trend, and the difference between the midlevel θe minimum and the surface θe is reduced from 15.6 K at 0600 UTC 13 August to 4.9 K at 1200 UTC 15 August. The reduction of the θe difference prior to genesis is consistent with the observational diagnosis by Wang (2012b) and Smith and Montgomery (2011).

The water vapor budget analysis is used to further examine the impacts of dry air intrusion. To put things into perspective, we first evaluated the budget terms 1 day prior to the formation of the tropical storm, 36–60 h. The net tendency (Fig. 5a) is rather weak, with a magnitude less than 1 g kg−1 day−1, except for the moistening over the pouch center between 5.0- and 7.5-km altitude and the drying around 2 km at radii larger than 300 km.

Fig. 5.
Fig. 5.

Azimuthally averaged water vapor budget fields for Fay (2008) averaged between 36 and 60 h: (a) net water vapor tendency, (b) mean horizontal flux convergence, (c) mean vertical flux convergence, (d) net condensation, (e) eddy horizontal flux convergence, and (f) eddy vertical flux convergence. The units of the variables are g kg−1 day−1.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

The mean horizontal and vertical flux convergence terms are shown in Figs. 5b,c, respectively. To better understand these two terms, the transverse circulation averaged in the same period is shown in Fig. 6a. The mean radial flow is characterized by inflow of 1–1.5 m s−1 within the boundary layer extending to a large radius (>350 km). Above the boundary layer, there is a deep layer of weak inflow below 9.5 km. However, the inflow does not extend all the way to the pouch center in the middle troposphere (between 3.5 and 7 km). There is weak outflow within the 100-km radius and strong outflow occurs above 10 km. Figure 6a also shows that upward motion greater than 7 cm s−1 is confined within the 100-km radius, with weaker upward motion at larger radii.

Fig. 6.
Fig. 6.

Azimuthally averaged vertical motion (shading; cm s−1) and radial flow (contours; m s−1) for (a) pre-Fay during 36–60 h, (b) pre-Fay during 29–34 h, and (c) ex-Gaston during 32–37 h.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

The mean radial flow converges moisture in the boundary layer and contributes to a positive tendency (>20 g kg−1 day−1) within the 50-km radius and a negative tendency between 90- and 200-km radii below 2 km (Fig. 5b). Above the boundary layer, there is a negative tendency near the pouch center between 4 and 7 km due to the outflow, and a weak positive tendency is found below ~8 km for radii less than ~220 km. The mean vertical motion (Fig. 5c) transports moisture from the boundary layer to the free atmosphere. It contributes to a negative tendency in the boundary layer and the lower troposphere and a positive tendency above, which is similar to a mature storm (Braun 2006).

The net condensation term represents the major moisture sink above the boundary layer (Fig. 5d). The maximum condensation (>20 g kg−1 day−1) occurs near the pouch center around 4 km. Within the ~100-km radius, net condensation stronger than 5 g kg−1 day−1 extends from near the surface to above 10-km altitude. At radii larger than 100 km, net condensation exceeds 5 g kg−1 day−1 above the freezing level (~3 km) and is generally weak below the freezing level. This indicates that convective precipitation is dominant near the pouch center and that the contribution of stratiform precipitation increases at the large radii.

The eddy horizontal flux convergence term (Fig. 5e) contributes to drying greater than 1 g kg−1 day−1 between 5 and 10 km and around 1-km altitude within the 120-km radius as well as between 1 and 4 km at radii larger than 290 km. The eddy vertical flux convergence term, Fig. 5f, contributes to negative tendency below 3 km and between 5 and 7 km and moistening between 8 and 11 km. Overall, the eddy flux convergence terms are much weaker than the mean flux convergence terms, suggesting relatively weak pouch relative flow and weak asymmetric eddies, and the mean vertical motion plays the dominant role in moistening the free atmosphere.

Next we examine the water vapor budget from 29 to 34 h, a period covering the second dry air intrusion prior to the formation of a tropical depression. The azimuthally averaged radial flow and vertical velocity are shown in Fig. 6b. Compared to Fig. 6a, the upward motion within the 100-km radius is significantly weakened, and there is even downward motion near the pouch center. However, upward motion at some larger radii, such as ~200 km, is stronger. It is interesting to note that the boundary layer inflow is not weakened. The outflow above the boundary is enhanced and becomes more extensive, and the upper-level outflow is reduced.

The azimuthally averaged water vapor budget fields are displayed in Fig. 7. The water vapor tendency field (Fig. 7a) shows drying above 4 km, moistening below, and drying near the surface over the pouch center. The mean horizontal flux convergence term (Fig. 7b) averaged over 29–34 h has a pattern similar to that averaged over 36–60 h, except that the drying tendency above 2 km over the pouch center is more extensive because of the stronger outflow above the boundary layer. The mean vertical flux convergence term (Fig. 7c) still contributes to drying below ~2 km; however, the moistening tendency above the boundary layer is confined below 6–7 km, and there is a drying tendency above 7 km over the pouch center, which is consistent with the weaker upward motion as shown in Fig. 6b. Figure 7d also shows reduced net condensation above the freezing level over the pouch center. The eddy horizontal flux convergence term (Fig. 7e) shows a large area of drying in the middle to upper troposphere within the 300-km radius and drying below 5 km at radii larger than 200 km. This may be due to the pouch-relative flow in the middle and upper troposphere and the mesoscale eddies at the pouch periphery. The vertical eddy flux convergence term (Fig. 7f) contributes to drying below 4–6 km and moistening above, and is particularly strong between 100 and 200 km radii, probably due to the stronger updrafts in this region. Comparison between Fig. 5c and Fig. 7c suggests a reduced upward moisture transport to the middle and upper troposphere at the inner pouch region during the dry air intrusion episode, which further enhances the drying in the middle and upper troposphere.

Fig. 7.
Fig. 7.

As in Fig. 5, but all terms are averaged over the period 29–34 h.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

To examine the dry air pathway and evolution, three-dimensional trajectory analyses are conducted using the HYSPLIT model at the tropical wave stage and at the tropical cyclone stage.

At the tropical wave stage, 30 h prior to the formation of the tropical depression (0000 UTC 14 August), a dry air slot is wrapped around the wave pouch along its eastern periphery (Fig. 8a). The HYSPLIT trajectory model shows that an air parcel initialized in the dry air region at ~650 hPa stays at the wave pouch periphery and does not move to the pouch center region. The air parcel trajectory illustrates good agreement with the translated streamlines in the first 24 h. The time series of parcel pressure (solid curve in Fig. 8b) shows that the air parcel stays around its initial pressure level (650–700 hPa). The relative humidity of the parcel (dashed curve in Fig. 8b) increases from 30% to 45% in the first 10 h and then fluctuates between 40% and 60% in the following 40 h.

Fig. 8.
Fig. 8.

Trajectory analysis of pre-Fay: forward trajectory and streamlines in the comoving frame of reference superimposed on RH (shading; %) for (a) 25 and (c) 60 h. (b),(d) Time series of RH and pressure along the parcel trajectory. In (a),(c) the points along the trajectories represent the positions of the parcel every 6 h and the yellow arrows denote the beginning point of the trajectory.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

During the cyclone stage (Figs. 8c,d), the trajectory is initialized at 1000 UTC 15 August (2 h prior to the formation of the tropical storm), and the dry air entrainment along the northwestern periphery of Fay is examined. The air parcel moves counterclockwise and gets closer to the pouch center along its path. The pressure time series (Fig. 8d) shows that the air parcel ascends steadily from 640 to 490 hPa within 20 h (vertical velocity ~0.21 Pa s−1), and the relative humidity of the air parcel increases from 35% to more than 80%. Although the air parcel is less than 100 km from the pouch center around 18 h, it has become quite moist by then.

Integration errors of the HYSPLIT model were estimated by computing backward trajectories from the forward trajectory end positions (not shown), and it was found that the numerical errors were relatively small. A large number of trajectories were evaluated for both stages, and they all showed dry air entrained at the pouch periphery tends to stay off the pouch center or gets moistened along its path even if it is being wrapped into the wave pouch. The evolution of the model precipitation suggests that the lateral dry air entrainment suppresses convection at the pouch periphery but does not affect convection near the pouch center. This is consistent with Sippel et al. (2011). They examined sets of ensemble members in simulations of Tropical Storm Debby (2006) and found that impacts of dry air depended on the storm strength: dry air possibly hinders development during the pregenesis stage but the impacts become insignificant once a tropical storm forms.

5. Nondeveloper—Gaston (2010)

a. Thermodynamic evolution and 3D trajectory analysis

The nondevelopment of ex-Gaston (2010) is examined in this section. The simulation of ex-Gaston started 2 days after Gaston weakened to a remnant low. Similar to the observation, ex-Gaston does not intensify in the model simulation with a 27–9-km resolution nested configuration. The pouch path was examined at various levels (not shown). A wave pouch is present below 3 km throughout the simulation. At 5 km, however, the wave pouch is less well defined and is displaced from the low-level pouch by 200–300 km after 1200 UTC 5 September. At 6 km and above (i.e., above ~500 hPa) a wave pouch does not exist, which makes the storm susceptible to dry air intrusion at the middle to upper troposphere. The vertical misalignment of the wave pouch in ex-Gaston was examined by Davis and Ahijevych (2012). A misaligned pouch is generally regarded as an unfavorable condition for tropical cyclone development (DMW09; Wang et al. 2012b; Raymond and Lopez Carrillo 2011).

The time–height cross sections of relative vorticity, OW, RH, and θe derived from the 9-km grid output are displayed in Fig. 9. Figure 9a shows that the maximum relative vorticity of magnitude around 2 × 10−4 s−1 occurs near the surface before 1800 UTC 5 September and then weakens in time. The vertical profile of OW (Fig. 9b) is characterized by prevailing negative values, with weak, positive values (0.5 × 10−8 s−2) present only intermittently below 5 km. This suggests that the environment is not favorable for vorticity aggregation or tropical cyclone development.

Fig. 9.
Fig. 9.

As in Fig. 4, but for Gaston (2010) from 0000 UTC 4 Sep 2010 to 0000 UTC 7 Sep 2010.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

The time–height cross section of RH (Fig. 9c) shows the presence of dry air above 7 km throughout the simulation, with RH as low as 40%, which is in contrast to the transient dry air intrusion in Fay. RH decreases steadily above 2 km in the first 2 days of the model simulation. Although the middle troposphere is moistened slightly during 1200–1800 UTC 6 September, RH remains extremely dry above 7–8 km.

The time–height cross section of θe is displayed in Fig. 9d. The vertical profile of θe is characterized by a midlevel minimum at 5–6-km altitude, which decreases from 1800 UTC 4 September to 0600 UTC 6 September by about 4 K. The midlevel θe remains below 340 K throughout the simulation, and the difference in θe between the surface (~357 K) and the 5-km level remains well above 15 K, which is in sharp contrast to pre-Fay. The altitude of the θe minimum also slightly descends during the simulation, suggesting drying of the middle troposphere. We further examined the time–height cross section of the specific humidity and potential temperature (not shown). From 1200 UTC 4 September to 1200 UTC 6 September, the 2° box averaged specific humidity dropped about 1–1.5 g kg−1 around 5 km, while there is no systematic change in potential temperature in the middle troposphere. This suggests that the decrease of the midlevel θe is mainly due to the midlevel drying.

The large θe difference between the surface and the middle troposphere in the model simulation is consistent with the dropsonde analysis by Davis and Ahijevych (2012), Smith and Montgomery (2011), and Wang (2012b). Rotunno and Emanuel (1987) suggested that one of the inhibiting factors for tropical cyclone formation is convective downdrafts, which transport air of low θe from the middle troposphere to the boundary layer and suppresses convection. The recent studies suggest that the midlevel moistening does not eliminate downdrafts (Nolan 2007; Wang et al. 2010b). Instead, it promotes vigorous updrafts by reducing the lateral entrainment of dry air (James and Markowski 2009; Wang 2012b; Smith and Montgomery 2011). The decrease of the midlevel θe in Gaston implies a stronger dilution of convective plumes by lateral dry air entrainment as well as stronger inimical impacts of downdrafts due to the large vertical gradient of θe. This likely contributes to the nondevelopment of ex-Gaston.

The vertical wind shear for Gaston and Fay were also examined (not shown). Vertical shear averaged over a 5° × 5° box is less than 8 m s−1 for both storms over 850–200 and 850–500 hPa. The middle troposphere shear (850–500 hPa) for ex-Gaston remain less than ~6 m s −1 and is slightly weaker than that of Fay. This suggests that the nondevelopment of Gaston is unlikely due to the vertical wind shear. Davis and Ahijevych (2012) also found that the environmental vertical shear for Gaston was rather weak, but the vertical misalignment of the wave pouch increased the total shear.

As mentioned in section 4, the 2° × 2° box averages represent the dynamic and thermodynamic conditions near the pouch center, which Wang (2012b) suggests are critical for tropical cyclone formation. DMW09 hypothesized that the wave pouch, as a region of approximately closed Lagrangian circulation, protects its interior from dry air intrusion to some extent. This leads to the question: what causes the midlevel drying near the pouch center and how does dry air get there? A snapshot of relative humidity at 3 km is shown in Fig. 10a. The wave pouch is depicted by the closed streamlines in the wave’s comoving frame of reference. The most striking feature is that the wave pouch retains a higher moisture content (RH generally larger than 75%) than the surrounding air (RH as low as 35%). This suggests that the pouch does prevent dry air intrusion to some extent. To confirm this, we carried out 3D trajectory analyses using the HYSPLIT model. The ensemble forward trajectories are superimposed on the translated streamlines in Fig. 10a. The parcels are initialized at 0800 UTC 5 September 2010 at the southern boundary of the wave pouch, where RH is less than 35%. The air parcels take a cyclonic route and all stay off the pouch center during the 40-h trajectory evolution. Ensemble trajectories with different initial locations were also tested, and they all suggested that the air parcels outside or at the periphery of the wave pouch tend to stay off the pouch center. Moreover, most parcel trajectories follow largely the streamlines in the comoving frame of reference, suggesting a region of approximately closed Lagrangian circulation.

Fig. 10.
Fig. 10.

(a) The 3-km relative humidity and storm-relative streamlines for Gaston (2010) at 0800 UTC 5 Sep 2010 with a group of ensemble forward parcel trajectories (gray). (b) Vertical cross section of RH along 17.5°N (contour intervals are set to 15%) and backward trajectories (gray) projected onto the longitude–height plane. The box in (a) highlights a pocket of dry air near the pouch center and the line indicates the cross section location shown in (b). Both the particle trajectories and streamlines are shown in a wave comoving framework. The yellow dots indicate the initialization location of the ensemble trajectories.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

A closer look at Fig. 10a, however, reveals pockets of isolated dry air near the pouch center. The vertical cross section going through a dry air pocket along 17.5°N (Fig. 10b) reveals a deep structure of the dry air (indicated by a box in Fig. 10a) extending from 9 km all the way down to 3 km. This suggests that the dry air may come from the upper troposphere. To confirm this, we employed the backward trajectory analysis using the HYSPLIT model. Four groups of ensemble backward trajectories (i.e., there are total 4 groups × 27 ensemble members = 108 trajectories) were initialized randomly within the dry air pocket at 0800 UTC 5 September, running backward in time for 33 h (ending at 0000 UTC 4 September). It was found that most of the air parcels reside above their initial pressure level during the period of the trajectory analysis, and that the air parcels at a higher altitude tend to have lower relative humidity (not shown). A group of ensemble trajectories are projected on the longitude–height plane in Fig. 10b. It shows that air parcels originate from different levels below 8.5 km, with a large fraction of parcels clustering between 6.0 and 8.5 km. Further examination suggests that these parcels are outside of the upper-level wave pouch or that a wave pouch does not exist at that level at all, while the parcels originating between 2 and 6 km are within the wave pouch at 0000 UTC 4 September. The parcels originating below 2 km are subject to the boundary layer inflow but are still within the wave pouch at 0000 UTC 4 September. The backward trajectory analysis confirms that dry air is transported downward and contributes to the midlevel drying near the pouch center.

b. Water vapor budget analysis

To further examine the potential impacts of dry air on the evolution of Gaston, the water vapor budget was examined. The budget terms are derived from the 9-km grid output and averaged over 32–37 h, a period when drying takes place above the boundary layer and an equivalent potential temperature decreases in the middle troposphere. As shown in Fig. 11a, the mixing ratio decreases near the pouch center from 1 to 12 km, with the maximum drying of about 2–4 g kg−1 day−1 occurring from 2 to 8 km. Given the generally low moisture content in the middle and upper troposphere, this indicates a significant drop of relative humidity as well as a decrease of equivalent potential temperature, consistent with Fig. 9. Figure 11a also shows a drying tendency of 2–6 g kg−1 day−1 at radii larger than 200 km below 5 km.

Fig. 11.
Fig. 11.

As in Fig. 4, but for the simulation of ex-Gaston averaged over 32–37 h.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

The mean horizontal flux convergence term is shown in Fig. 11b. It contributes to moistening between 20- and 250-km radii in the boundary layer. Compared to pre-Fay, the overall pattern is much noisier, especially above the boundary layer. The mean vertical flux convergence term (Fig. 11c) contributes to drying in a thin layer near the surface. Areas of drying are also found above 1 km, in particular from 3 to 10 km over the pouch center. As discussed in the last section, the mean vertical flux convergence term provides the major moisture source in the free atmosphere for pre-Fay. The reduced moistening plus areas of drying over the pouch center in ex-Gaston is a substantial difference from pre-Fay. We also examined other periods. The drying signals are intermittent, but the reduced moistening is always evident. The transverse circulation (Fig. 6c) suggests that this is associated with the downward motion over the pouch center. Compared to pre-Fay (Figs. 6a,b), the low-level inflow in ex-Gaston is stronger but much shallower. It decelerates significantly between 100- and 200-km radii. The radial flow above the boundary layer is less well organized in ex-Gaston, which explains the noisy pattern of the mean horizontal flux convergence term.

Similar to pre-Fay, the net condensation term is the major sink term in ex-Gaston. The vertical profile of the net condensation term, however, is different from that of pre-Fay. Figure 11d shows that the net condensation at multiple radii peaks at two altitudes, one around 8 km and the other around 4 km. The former is associated with ice microphysics processes. Because of suppressed convection near the pouch center (Fig. 6c), net evaporation (i.e., positive tendency) occurs near the pouch center and at several other radii. The maximum net condensation (i.e., negative tendency) occurs between 170- and 200-km radius.

The eddy horizontal flux convergence term (Fig. 11e) contributes to drying above 5 km within the 200-km radius and drying below 5 km at larger radii. The former can be attributed to the pouch-relative flow in the mid-to-upper troposphere, where a closed circulation is absent, and the latter represents lateral dry air entrainment due to the pouch relative flow and mesoscale eddies. The eddy vertical flux convergence terms (Fig. 11f) contribute to drying in the lower troposphere and moistening in the middle to upper troposphere. The overall pattern is similar to that for Fay (Fig. 7f) but much noisier.

In summary, the budget analysis suggests that both lateral dry air entrainment and downward dry air transport contribute to the drying in the middle troposphere. The midlevel dry air entrainment was also studied by Rutherford and Montgomery (2011) using ECMWF analysis data. Our diagnosis here suggests an additional pathway for dry air to enter the wave pouch: the pouch-relative flow in the upper troposphere brings dry air above the low-level pouch center, which is then transported downward by vertical motion. The midlevel drying weakens convective updrafts and reduces the moisture supply from the boundary layer, which further enhances the midlevel drying. The downward transport may be a more direct pathway for dry air to influence convection near the pouch center, as dry air entrained at the pouch periphery has to undergo several revolutions before getting to the pouch center and gets moistened along its path.

c. Sensitivity test—Gaston (2010)

As mentioned in section 3, several simulations were conducted for ex-Gaston. The ex-Gaston disturbance intensifies and develops into a tropical storm in most of the simulations, one of which is shown in Fig. 12. This simulation is the same as the control simulation examined in sections 5a and 5b, except that an inner grid with the horizontal resolution of 3 km was added. The disturbance develops into a weak tropical storm with a maximum surface wind speed of about 18 m s−1, while in the control run the maximum surface wind speed is about 14 m s−1 at the end of the 3-day simulation. The storm in the higher-resolution simulation also has a well-defined wave pouch extending to 8 km, while the disturbance in the coarse-resolution simulation does not have a well-defined pouch above 5 km.

Fig. 12.
Fig. 12.

As in Fig. 8, but for the 3-km resolution simulation of Gaston (2010).

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

A snapshot of the moisture field in this simulation, at the same time and location as Fig. 11a, is shown in Fig. 12a. Although the synoptic-scale patterns of the wind field and moisture field are quite similar to those in the control run, the dry air pockets are less pronounced, and the dry air entrainment is also evidently weaker at the pouch periphery. A vertical cross section of the relative humidity field is shown in Fig. 12b. Despite the presence of ambient dry air above 6 km, the air over the low-level pouch center is moist (the 75% contour extends above 9 km). The moisture field was examined at other times in the higher-resolution simulation, and only very few pockets of dry air were discovered and none was as extensive or persistent as what was shown in the coarse-resolution simulation. It was also found that the downdrafts (Fig. 13) near the wave pouch center are much weaker and less frequent in the high-resolution model simulation.

Fig. 13.
Fig. 13.

Time–height cross section of vertical velocity (w) for ex-Gaston (2010) from (a) the control run and (b) the high-resolution simulation, averaged in a 2° × 2° box following the pouch center. The contour levels are −0.04, −0.03, −0.02, −0.01, 0, 0.1, 0.15, and 0.2 m s−1, and negative values are shaded.

Citation: Journal of the Atmospheric Sciences 70, 1; 10.1175/JAS-D-12-018.1

The comparison of the two numerical model simulations reveals the sensitivity of a simulated convective storm to the model resolution, which has been examined in some previous studies (e.g., Bryan et al. 2003; Rotunno et al. 2009; Bryan and Morrison 2012). With higher resolution a model tends to simulate a storm of stronger intensity. This may be partly because it is harder to initiate explicit convection at coarse resolution than at finer resolution. The comparison of the two simulations suggests that downward dry air transport associated with the stronger downdrafts in the coarse-resolution simulation may also hinder tropical cyclone development. It is possible that some errors due to model physics are compensated by the model errors because of the coarse resolution, and thus a higher resolution does not produce a “better” simulation. Further study is necessary to better understand the sensitivity issue, which is beyond the scope of this study.

6. Conclusions and discussion

In this study, ex-Gaston (2010) and pre-Fay (2008) were simulated using the WRF model to examine the impacts of dry air on the storm evolution. Pre-Fay was subject to transient dry air impacts that can be linked to the passage of an upper-level low, while ex-Gaston was under the influence of subsidence associated with a midlatitude anticyclone, and the upper-level dry air was much more persistent. The ex-Gaston disturbance failed to redevelop after being downgraded to a remnant low, and the pre-Fay disturbance developed into a tropical storm.

Three-dimensional trajectory analysis suggests that dry air entrained at the pouch periphery tends to stay off the pouch center due to weak midlevel inflow in the simulation of pre-Fay. It is also shown that dry air gets moistened even if it is being wrapped into the wave pouch. Lateral entrainment in the middle troposphere thus does not suppress convection near the pouch center or prevent the formation of tropical cyclone Fay. However, ex-Gaston is subject to the persistent impacts of mid- to upper-level dry air intrusion, and the model simulation shows a decrease in the midlevel equivalent potential temperature near the pouch center due to midlevel drying, which is consistent with the observation. Backward trajectory analysis based on the simulation of ex-Gaston indicates a downward transport of dry air from the middle and upper troposphere, where a well-defined wave pouch is absent.

The water vapor budget analysis of Fay shows that the low-level inflow converges moisture near the pouch center in the boundary layer and that the mean vertical moisture transport plays the dominant role in moistening the free atmosphere. During the episodes of dry air intrusion, the upward moisture transport is reduced above 6–7 km in ex-Gaston. The budget analysis of ex-Gaston confirms the contribution of the downward transport of dry air to midlevel drying. In the presence of the midlevel drying, convection is suppressed and the midlevel drying is further enhanced because of the lack of moisture supply from the boundary layer. The eddy horizontal flux convergence term also indicates dry air entrainment at the pouch periphery in both storms.

As a region of approximately closed Lagrangian circulation, a wave pouch provides some protection to the moist convection within. However, dry air entrainment can be induced by mesoscale eddies, the pouch-relative flow, and the divergent flow component [see Fig. 3 in Wang et al. (2010a)]. Mesoscale eddies can result in lateral mixing at the pouch periphery between the pouch interior flow and the surrounding environment (DMW09). For a well-defined wave pouch, the pouch relative flow is weak, and the time scale of midlevel entrainment is determined by the radial inflow speed. As shown in Fig. 6, the azimuthally mean radial inflow speed is about 1 m s−1 or weaker, and it thus takes about 2–3 days for an air parcel to travel from the pouch periphery to the pouch center as it undergoes cyclonic revolutions. For a pouch with vigorous convection, dry air parcels will be moistened before they get to the pouch center. In other words, the entrainment time scale is much longer than the moistening time scale. It suggests that the circulation center is generally well protected from lateral dry air intrusion at the middle troposphere if a well-defined wave pouch exists. On the other hand, a wave pouch is usually confined to the lower troposphere at the early stage of the tropical cyclone formation, and the flow remains open in the upper troposphere. Dry air can be advected directly above the midlevel or low-level pouch center and then transported to the middle or lower troposphere, which can reduce the intensity of updrafts near the pouch center and hinder tropical cyclone formation (Wang 2012b). The upper troposphere is thus a weak spot of the wave pouch at the early stage, and the vertical transport is likely a more direct pathway for dry air to influence moist convection near the pouch center. This is consistent with the observational study by Hopsch et al. (2010). Using the 40-yr ECMWF Re-Analysis (ERA-40) data and satellite brightness temperature between 1979 and 2001, they compared composite structures of African easterly waves (AEWs) that developed into tropical cyclones to waves that did not, and they noted that the nondeveloper composite has mid- to upper-level dry air just ahead of the wave trough. Wang et al. (2012b) also showed that the upper- and midlevel dry air intrusion likely prevented the development of a tropical wave (PGI26L).

The impacts of upper-level dry air intrusion have not attracted as much attention as midlevel dry air intrusion in the past, probably because of the low moisture content in the upper troposphere. Nolan (2007) and Braun et al. (2012) studied the impacts of dry air on tropical cyclone development using idealized numerical simulations. Nolan (2007) showed that a dry upper troposphere does not have significant impacts on tropical cyclone formation because the upper troposphere is moistened quickly by deep convection. In Braun et al.’s (2012) idealized simulations, the storms in all the simulations eventually intensify to the same strength after the free atmosphere is moistened by convection. Since an environment with zero mean flow was used in both studies, moisture lofted by deep convection can accumulate and moisten the upper troposphere quickly. This setup is similar to a transient dry air intrusion event in pre-Fay. By contrast, if large-scale subsidence induces persistent dry air intrusion or if strong pouch-relative flow advects away moisture in the upper levels, then the upper troposphere can stay dry. Deep convection can be suppressed and tropical cyclone formation may be hindered or prevented, in particular when the upper-level synoptic-scale forcing induces descent over the wave pouch. This may help to explain why strong vertical shear between the upper and lower troposphere is detrimental for tropical cyclone formation. Further study is required to better understand the impacts of the upper-level flow on tropical cyclone formation.

Based on the azimuthally averaged radial inflow speed, we estimated that the midlevel entrainment time scale is 2–3 days for a wave pouch of typical size. In pre-Fay, the eddy horizontal flux convergence term is much weaker than the mean horizontal flux convergence term, and the azimuthally averaged radial inflow thus provides a reasonable estimate. However, we should not rule out asymmetric inflow. Asymmetric inflow may be due to pouch-relative flow or mesovortices inside the pouch. As formulated by DMW09, a dominant master protovortex is favorable to inner-core protection, while a suite of lesser vortices would invite asymmetric penetration to the center. The existence of asymmetric inflow and its relationship with the wave pouch kinematics and the environmental flow warrant further study.

Acknowledgments

This research was supported by National Science Foundation Grants ATM-1016095 and ATM-1118429. We thank Dr. Chris Landsea and two anonymous reviewers for their constructive comments, and we thank NCAR CISL for providing the computational resources.

REFERENCES

  • Blake, E. S., 2010: Tropical cyclone report—Tropical Storm Gaston (AL092010), 1-2 September 2010. NHC Rep., 9 pp. [Available online at http://www.nhc.noaa.gov/pdf/TCR-AL092010_Gaston.pdf.]

  • Braun, S. A., 2006: High-resolution simulation of Hurricane Bonnie (1998). Part II: Water budget. J. Atmos. Sci., 63, 4364.

  • Braun, S. A., 2010: Reevaluating the role of the Saharan air layer in Atlantic tropical cyclogenesis and evolution. Mon. Wea. Rev., 138, 20072037.

    • Search Google Scholar
    • Export Citation
  • Braun, S. A., , J. A. Sippel, , and D. S. Nolan, 2012: The impact of dry midlevel air on hurricane intensity in idealized simulations with no mean flow. J. Atmos. Sci., 69, 236257.

    • Search Google Scholar
    • Export Citation
  • Brown, D. P., , J. L. Beven, , J. L. Franklin, , and E. S. Blake, 2010: Atlantic hurricane season of 2008. Mon. Wea. Rev., 138, 19752001.

  • Bryan, G. H., , and H. Morrison, 2012: Sensitivity of a simulated squall line to horizontal resolution and parameterization of microphysics. Mon. Wea. Rev., 140, 202225.

    • Search Google Scholar
    • Export Citation
  • Bryan, G. H., , J. C. Wyngaard, , and J. M. Fritsch, 2003: Resolution requirements for the simulation of deep moist convection. Mon. Wea. Rev., 131, 23942416.

    • Search Google Scholar
    • Export Citation
  • Carlson, T. N., , and J. M. Prospero, 1972: The large-scale movement of Saharan air outbreaks over the northern equatorial Atlantic. J. Appl. Meteor., 11, 283297.

    • Search Google Scholar
    • Export Citation
  • Davis, C. A., , and D. A. Ahijevych, 2012: Mesoscale structural evolution of three tropical weather systems observed during PREDICT. J. Atmos. Sci., 69, 12481305.

    • Search Google Scholar
    • Export Citation
  • Draxler, R. R., , and G. D. Hess, 1998: An overview of the HYSPLIT_4 modeling system of trajectories, dispersion, and deposition. Aust. Meteor. Mag., 47, 295308.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1989: Numerical study of convection observed during the Winter Monsoon Experiment using a mesoscale two-dimensional model. J. Atmos. Sci., 46, 30773107.

    • Search Google Scholar
    • Export Citation
  • Dunion, J. P., , and C. S. Velden, 2004: The impact of the Saharan air layer on Atlantic tropical cyclone activity. Bull. Amer. Meteor. Soc., 85, 353365.

    • Search Google Scholar
    • Export Citation
  • Dunkerton, T. J., , M. T. Montgomery, , and Z. Wang, 2009: Tropical cyclogenesis in a tropical wave critical layer: Easterly waves. Atmos. Chem. Phys., 9, 55875646.

    • Search Google Scholar
    • Export Citation
  • Evan, A. T., , J. Dunion, , J. A. Foley, , A. K. Heidinger, , and C. S. Velden, 2006: New evidence for a relationship between Atlantic tropical cyclone activity and African dust outbreaks. Geophys. Res. Lett., 33, L19813, doi:10.1029/2006GL026408.

    • Search Google Scholar
    • Export Citation
  • Fang, J., , and F. Zhang, 2010: Initial development and genesis of Hurricane Dolly (2008). J. Atmos. Sci., 67, 655672.

  • Gamache, J. F., , R. A. Houze, Jr., , and F. D. Marks, Jr., 1993: Dual-aircraft investigation of the inner core of Hurricane Norbert. Part III: Water budget. J. Atmos. Sci., 50, 32213243.

    • Search Google Scholar
    • Export Citation
  • Hong, S.-Y., , and J.-O. J. Lim, 2006: The WRF single-moment microphysics scheme (WSM6). J. Korean Meteor. Soc., 42, 129151.

  • Hong, S.-Y., , Y. Noh, , and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 23182341.

    • Search Google Scholar
    • Export Citation
  • Hopsch, S. B., , C. D. Thorncroft, , and K. R. Tyle, 2010: Analysis of African easterly wave structures and their role in influencing tropical cyclogenesis. Mon. Wea. Rev., 138, 10571077.

    • Search Google Scholar
    • Export Citation
  • James, R. P., , and P. M. Markowski, 2009: A numerical investigation of the effects of dry air aloft on deep convection. Mon. Wea. Rev., 137, 140161.

    • Search Google Scholar
    • Export Citation
  • Kain, J. S., 2004: The Kain–Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170181.

  • Karyampudi, V. M., , and T. N. Carlson, 1988: Analysis and numerical simulations of the Saharan air layer and its effect on easterly wave disturbances. J. Atmos. Sci., 45, 31023136.

    • Search Google Scholar
    • Export Citation
  • Karyampudi, V. M., , and H. F. Pierce, 2002: Synoptic-scale influence of the Saharan air layer on tropical cyclogenesis over the eastern Atlantic. Mon. Wea. Rev., 130, 31003128.

    • Search Google Scholar
    • Export Citation
  • Landsea, C. W., 1993: A climatology of intense (or major) Atlantic hurricanes. Mon. Wea. Rev., 121, 17031713.

  • Lau, W. K. M., , and K.-M. Kim, 2007a: Cooling of the Atlantic by Saharan dust. Geophys. Res. Lett., 34, L23811, doi:10.1029/2007GL031538.

    • Search Google Scholar
    • Export Citation
  • Lau, W. K. M., , and K.-M. Kim, 2007b: How nature foiled the 2006 hurricane forecasts. Eos, Trans. Amer. Geophys. Union, 88, 105, doi:10.1029/2007EO090002.

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

    • Search Google Scholar
    • Export Citation
  • Mlawer, E. J., , S. J. Taubman, , P. D. Brown, , M. J. Iacono, , and S. A. Clough, 1997: Radiative transfer for inhomogeneous atmosphere: RRTM, a validated correlated-k model for the long wave. J. Geophys. Res., 102 (D14), 16 66316 682.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , L. L. Lussier III, , R. W. Moore, , and Z. Wang, 2010a: The genesis of Typhoon Nuri as observed during the Tropical Cyclone Structure 2008 (TCS-08) field experiment – Part 1: The role of the easterly wave critical layer. Atmos. Chem. Phys., 10, 98799900.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., , Z. Wang, , and T. J. Dunkerton, 2010b: Coarse, intermediate and high resolution numerical simulations of the transition of a tropical wave critical layer to a tropical storm. Atmos. Chem. Phys., 10, 10 80310 827.

    • Search Google Scholar
    • Export Citation
  • Montgomery, M. T., and Coauthors, 2012: The Pre-Depression Investigation of Cloud-Systems in the Tropics (PREDICT) experiment: Scientific basis, new analysis tools, and some first results. Bull. Amer. Meteor. Soc., 93, 153172.

    • Search Google Scholar
    • Export Citation
  • Nolan, D. S., 2007: What is the trigger for tropical cyclogenesis? Aust. Meteor. Mag., 56, 241266.

  • Raymond, D. J., , and C. Lopez Carrillo, 2011: The vorticity budget of developing Typhoon Nuri (2008). Atmos. Chem. Phys., 11, 147163.

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

    • Search Google Scholar
    • Export Citation
  • Rotunno, R., , Y. Chen, , W. Wang, , C. Davis, , J. Dudhia, , and G. J. Holland, 2009: Large-eddy simulation of an idealized tropical cyclone. Bull. Amer. Meteor. Soc., 90, 17831788.

    • Search Google Scholar
    • Export Citation
  • Rozoff, C. M., , J. P. Kossin, , W. H. Schubert, , and P. J. Mulero, 2009: Internal control of hurricane intensity variability: The dual nature of potential vorticity mixing. J. Atmos. Sci., 66, 133147.

    • Search Google Scholar
    • Export Citation
  • Rutherford, B., , and M. T. Montgomery, 2011: A Lagrangian analysis of a developing and non-developing disturbance observed during the PREDICT experiment. Atmos. Chem. Phys. Discuss., 11, 33 27333 323.

    • Search Google Scholar
    • Export Citation
  • Sippel, J. A., , S. A. Braun, , and C.-L. Shie, 2011: Environmental influences on the strength of Tropical Storm Debby (2006). J. Atmos. Sci., 68, 25572581.

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., , J. B. Klemp, , J. Dudhia, , D. O. Gill, , D. M. Barker, , W. Wang, , and J. G. Powers, 2005: A description of the Advanced Research WRF version 2. NCAR Tech. Note NCAR/TN-468+STR, 88 pp.

  • Smith, R. K., , and M. T. Montgomery, 2011: Observations of the convective environment in developing and non-developing tropical disturbances. Quart. J. Roy. Meteor. Soc., 137, 120.

    • Search Google Scholar
    • Export Citation
  • Stewart, S. R., , and J. L. Beven II, 2008: Tropical cyclone report—Tropical Storm Fay (AL062008), 15-26 August 2008. NHC Rep., 29 pp. [Available online at http://www.nhc.noaa.gov/pdf/TCR-AL062008_Fay.pdf.]

  • Sun, D., , K.-M. Lau, , and M. Kafatos, 2008: Contrasting the 2007 and 2005 hurricane seasons: Evidence of possible impacts of Saharan dry air and dust on tropical cyclone activity in the Atlantic basin. Geophys. Res. Lett., 35, L15405, doi:10.1029/2008GL034529.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., 2012a: Role of intense updrafts in tropical cyclone formation. Preprints, 30th Conf. on Hurricanes and Tropical Meteorology, Ponte Vedra Beach, FL, Amer. Meteor. Soc., P1.28. [Available online at https://ams.confex.com/ams/30Hurricane/webprogram/Paper205467.html.]

  • Wang, Z., 2012b: Thermodynamic aspects of tropical cyclone formation. J. Atmos. Sci., 69, 24332451.

  • Wang, Z., , M. T. Montgomery, , and T. J. Dunkerton, 2009: A dynamically-based method for forecasting tropical cyclogenesis location in the Atlantic sector using global model products. Geophys. Res. Lett., 36, L03801, doi:10.1029/2008GL035586.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , M. T. Montgomery, , and T. J. Dunkerton, 2010a: Genesis of pre–Hurricane Felix (2007). Part I: The role of the wave critical layer. J. Atmos. Sci., 67, 17111729.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , M. T. Montgomery, , and T. J. Dunkerton, 2010b: Genesis of pre–Hurricane Felix (2007). Part II: Warm core formation, precipitation evolution and predictability. J. Atmos. Sci., 67, 17301744.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , T. J. Dunkerton, , and M. T. Montgomery, 2012a: Application of the marsupial paradigm to tropical cyclone formation from northwestward-propagating disturbances. Mon. Wea. Rev., 140, 6676.

    • Search Google Scholar
    • Export Citation
  • Wang, Z., , M. T. Montgomery, , and C. Fritz, 2012b: A first look at the structure of the wave pouch during the 2009 PREDICT–GRIP dry runs over the Atlantic. Mon. Wea. Rev., 140, 11441163.

    • Search Google Scholar
    • Export Citation
  • Zhang, D.-L., , Y. Liu, , and M. K. Yau, 2002: A multiscale numerical study of Hurricane Andrew (1992). Part V: Inner-core thermodynamics. Mon. Wea. Rev., 130, 27452763.

    • Search Google Scholar
    • Export Citation
1

The protovortex is the initial vortical structure within a hybrid diabatic Rossby wave–vortex that may subsequently grow to a tropical depression–strength vortex (DMW09).

2

The critical latitude is defined as the latitude at which the wave phase speed is equal to the speed of the mean zonal flow.

Save