Simulation of Dryline Misovortex Dynamics and Cumulus Formation

Michael S. Buban Cooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma, and NOAA/OAR/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Michael S. Buban in
Current site
Google Scholar
PubMed
Close
,
Conrad L. Ziegler NOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Conrad L. Ziegler in
Current site
Google Scholar
PubMed
Close
,
Edward R. Mansell NOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by Edward R. Mansell in
Current site
Google Scholar
PubMed
Close
, and
Yvette P. Richardson Department of Meteorology, The Pennsylvania State University, State College, Pennsylvania

Search for other papers by Yvette P. Richardson in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

A dryline and misocyclones have been simulated in a cloud-resolving model by applying specified initial and time-dependent lateral boundary conditions obtained from analyses of the 22 May 2002 International H2O Project (IHOP_2002) dataset. The initial and lateral boundary conditions were obtained from a combination of the time–spaced Lagrangian analyses for temperature and moisture with horizontal velocities from multiple-Doppler wind syntheses. The simulated dryline, horizontal dry-convective rolls (HCRs) and open cells (OCCs), misocyclones, and cumulus clouds are similar to the corresponding observed features. The misocyclones move northward at nearly the mean boundary layer (BL) wind speed, rotate dryline gradients owing to their circulations, and move the local dryline eastward via their passage. Cumuli develop along a secondary dryline, along HCR and OCC segments between the primary and secondary drylines, along HCR and OCC segments that have moved over the dryline, and within the dryline updraft. After the initial shearing instability develops, misocyclogenesis proceeds from tilting and stretching of vorticity by the persistent secondary dryline circulation. The resulting misocyclone evolution is discussed.

Corresponding author address: Michael S. Buban, Forecast Research and Development Division, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072. E-mail: michael.buban@noaa.gov

Abstract

A dryline and misocyclones have been simulated in a cloud-resolving model by applying specified initial and time-dependent lateral boundary conditions obtained from analyses of the 22 May 2002 International H2O Project (IHOP_2002) dataset. The initial and lateral boundary conditions were obtained from a combination of the time–spaced Lagrangian analyses for temperature and moisture with horizontal velocities from multiple-Doppler wind syntheses. The simulated dryline, horizontal dry-convective rolls (HCRs) and open cells (OCCs), misocyclones, and cumulus clouds are similar to the corresponding observed features. The misocyclones move northward at nearly the mean boundary layer (BL) wind speed, rotate dryline gradients owing to their circulations, and move the local dryline eastward via their passage. Cumuli develop along a secondary dryline, along HCR and OCC segments between the primary and secondary drylines, along HCR and OCC segments that have moved over the dryline, and within the dryline updraft. After the initial shearing instability develops, misocyclogenesis proceeds from tilting and stretching of vorticity by the persistent secondary dryline circulation. The resulting misocyclone evolution is discussed.

Corresponding author address: Michael S. Buban, Forecast Research and Development Division, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072. E-mail: michael.buban@noaa.gov

1. Introduction

High-resolution observations of the planetary boundary layer (BL) have revealed the presence of small-scale (i.e., 1–6 km in width) misovortices along various near-surface boundaries (Crook et al. 1991; Atkins et al. 1995; Pietrycha and Rasmussen 2004; Kawashima and Fujiyoshi 2005; Arnott et al. 2006; Murphey et al. 2006; Marquis et al. 2007; Buban et al. 2007). Vortices having a core circulation diameter between 40 m and 4 km are termed “misocyclones” (Fujita 1981; Wakimoto and Wilson 1989). Recent field observations have demonstrated that misocyclones on this scale are characteristically coherent structures (e.g., Marquis et al. 2007; Buban et al. 2007), lasting upward of 30 min or longer and remaining attached to the near-surface boundaries that they move along. Several studies have hypothesized that these misocyclones may assist convection initiation (CI) by rearranging the moisture and convergence fields (Pietrycha and Rasmussen 2004; Buban et al. 2007; Lee et al. 2000; Kanak 2008). Other studies have suggested that misocyclones may be associated with nonsupercell tornadogenesis by providing a pre-existing source of vertical vorticity that convective updrafts can subsequently intensify via stretching (Carbone 1982, 1983; Wakimoto and Wilson 1989; Wakimoto and Atkins 1996; Lee and Wilhelmson 1997a,b; Wakimoto et al. 1998; Ziegler et al. 2001).

Although misocyclones have been documented, and aspects of their development along outflow boundaries (e.g., Lee and Wilhelmson 1997a) and in a pure convective boundary layer (Kanak et al. 2000) have been explored in idealized numerical simulations, the dynamics that govern their formation and evolution along a dryline are poorly understood. Specifically, it is unclear how these concentrated vortices arise and are maintained, whether by tilting and/or stretching of environmental vorticity or generation by other means. A reason for the lack of understanding is the small scales at which misocyclones occur. Detailed observations of these features have been made, but the data resolution is typically not fine enough in space and time to capture the underlying dynamic processes. The numerical simulation of convective-scale motions at high resolution is necessary to capture the scales important for dryline and BL dynamics, misocyclones, and CI. A primary goal of this study is to use the comprehensive, internally consistent, high-resolution dataset provided by a simulation of the dryline and its BL environment to further our understanding of misocyclones.

Many studies have been conducted over mesoscale- to regional-scale domains by forcing simulations via boundary conditions obtained from larger-scale analyses or forecasts. In the present study, misoscale BL analyses are employed to force the solution of a cloud-resolving model. In cases where larger-scale simulations force smaller, higher-resolution domains, both one- and two-way interactive nesting may be used. When two-way nesting is used, information at the lateral boundaries of the smaller domain is interpolated back to the larger domain, where it subsequently influences the larger-scale results. Thus, small-scale information (having been interpolated to the larger domain and integrated forward in time) may feed back into the smaller domain via application of boundary conditions at subsequent time steps. In the case where analyses are used but influences of the smaller scale on the larger scale are unwanted, one-way nesting via the application of large-scale information at lateral inflow boundary points is employed. Outflow boundary points are typically treated as wave radiating to allow features to pass through the boundaries with minimal influence on the solution. Considerable caution must be exercised to reduce potential sources of simulation error caused by the boundary conditions. For example, analysis errors at the boundaries can propagate into the interior of the model domain, and input fields are not in balance when they enter the domain, necessitating a short period of adjustment. At outflow boundaries, various modes can be reflected back into the interior depending upon the numerical technique used. A review of the limitations of lateral boundary conditions can be found in Warner et al. (1997).

Time-varying radar and Lagrangian analyses (Buban et al. 2007) produced from data obtained during the International H2O Project (IHOP_2002) on 22 May 2002 have been used in the present study as initial and time-dependent lateral inflow boundary conditions to conduct a high-resolution simulation of the dryline and its surrounding BL. The dryline simulation employs a three-dimensional, nonhydrostatic cloud-mesoscale model that includes shortwave and longwave radiation and force–restore surface physics parameterization options. The effectiveness of the newly developed time-dependent lateral boundary conditions at communicating information about the observed finescale structure can be deduced by comparing simulated results with the Lagrangian and radar analyses. The simulation reproduces a nearly north–south-oriented dryline with horizontal moisture and temperature gradients similar to observed values, as well as misocyclones, horizontal convective rolls (HCRs), transverse rolls, open convective cells (OCCs), and cumulus clouds. These features of the simulated BL are similar to analogous structures manifested in the observations, although the modeled features are typically of higher spatial and temporal resolutions and may have larger amplitudes than the equivalent observed features. The simulation output provides a high-resolution internally consistent dataset from which analysis can be conducted. These analyses facilitate our understanding of the relevant dynamics of the dryline, BL convection, and cumulus formation. For example, vorticity and vorticity tendency terms are computed to understand the physical processes involved in the misocyclone life cycle. The solenoidal term for horizontal vorticity tendency and the two-dimensional accumulation of scalar gradients are computed to analyze the structure of the dryline. Backward trajectories emanating from cumuli are calculated to characterize the cumulus formation process.

2. Data analysis and model description

a. Lagrangian and radar analyses

Radial velocities from multiple Doppler radars were used to synthesize the evolving 3D boundary layer airflow near surface-based mesoscale boundaries observed during IHOP_2002 (e.g., Buban et al. 2007; Ziegler et al. 2007). Using these wind fields, a Lagrangian analysis was performed to retrieve the evolving 3D thermodynamic structure of the BL (Buban et al. 2007). The Lagrangian analysis advects in situ observations from mobile mesonets, aircraft, and mobile soundings, as well as pseudosounding gridpoint data both upstream and downstream along trajectories computed from the radar-synthesized wind fields (Ziegler et al. 2007). The Lagrangian data are then temporally and spatially weighted and objectively analyzed to the grid using a two-pass Barnes objective analysis scheme (Barnes 1973). Both the radar and Lagrangian analyses have a grid spacing of 500 m in the horizontal and 250 m in the vertical, with a total horizontal extent of 30 km and a vertical extent of 2.5 km (above which there is little or no Doppler velocity data). Examples of Lagrangian and multiple-Doppler wind analyses are shown in Figs. 1c,d.

Fig. 1.
Fig. 1.

Water vapor mixing ratio (color filled), horizontal wind vectors (1 km = 20 m s−1), and vertical vorticity every (a),(b) 20 × 10−3 s−1 starting at ±20 × 10−3 s−1 and (c),(d) 3 × 10−3 s−1 starting at ±3 × 10−3 s−1, with positive values solid and negative values dashed. (a) Simulation at 2327 UTC, (b) simulation at 0012 UTC, (c) Lagrangian analysis at 2327 UTC, and (d) Lagrangian analysis at 0012 UTC. The longer dashed line in (a) locates cross sections shown in Fig. 3, while the shorter dashed lines in (a) locate cross sections shown in Fig. 17. The vertical levels are at (left) 20 m AGL and (right) ground level. The solid and dashed black curves denote the horizontal component of the moist (MT) and dry (DT) trajectories shown in Fig. 17. The short-dashed and dotted white curves denote the dryline and eastern dryline, respectively.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

b. Numerical model description

The dryline-BL simulations in the present study are conducted using the Collaborative Model for Multiscale Atmospheric Simulation (COMMAS; Wicker and Wilhelmson 1995; Coniglio et al. 2006; Mansell et al. 2010). COMMAS is a cloud-resolving and nonhydrostatic model, and it includes fifth-order advection (Wicker and Skamarock 2002) and a 1.5-order subgrid-scale turbulence parameterization. The lower boundary was treated as free slip and the upper boundary as rigid. Since only cumulus clouds developed on 22 May, a simple Kessler warm-rain scheme was used to model the cloud microphysics.

Several new features were added to COMMAS to study BL circulations and misocyclones. Time-varying specified inflow boundary conditions were incorporated to allow nudging from the Lagrangian analyses. Since surface fluxes are important to boundary layer structures in the dryline environment (Sun and Wu 1992; Ziegler et al. 1995, 1997), the model was also modified to include surface fluxes as calculated with a modified version of the Deardorff (1978) force–restore land surface–atmosphere exchange model (Peckham et al. 2004). Parameters of the surface physics scheme are listed in Table 1, while a comparison of modeled and observed surface fluxes for this case study are discussed in the appendix. A surface shortwave and longwave radiation parameterization (Benjamin and Carlson 1986; Peckham et al. 2004) that includes the cloud-shadowing effect was also added.

Table 1.

Parameters for the surface physics scheme. The letter (i) corresponds to an initial condition (i.e., these quantities are subsequently predicted) and (c) represents a constant.

Table 1.

The type of lateral boundary condition used is critically important to the model solution. Whether in smaller domains, or where the speed of meteorological features is such that the advective time scale across the domain is small, the impact of the boundary conditions on the solution may be comparable to that of the initial conditions. As the initial conditions are advected out of the domain, features introduced at the lateral boundaries and acted on by the model physics replace them. It is necessary to apply the lateral inflow boundary conditions with a fine enough temporal and spatial resolution to introduce misoscale features of interest into the domain, where they may subsequently evolve according to the fully nonlinear physics. For example, preliminary tests using 9-min-interval multiple-Doppler wind analyses applied at the boundary proved too coarse, and no misocyclones developed in the model interior. Only when the 3-min-interval analyses were used was the information communicated to the model interior fine enough for misocyclones to form. It is also important that, in conjunction with the specified inflow boundary conditions, lateral outflow boundaries (where the normal component of the wind is exiting the domain) are wave radiating to allow for features to exit the domain with minimal feedback.

c. Model configuration and initialization

To resolve small-scale features, the model’s horizontal grid spacing was set to 150 m. This resolution was chosen to adequately resolve small-scale features, yet coarse enough to be compatible with the Lagrangian and multiple-Doppler radar analyses. In the vertical, the grid had a lowest-layer thickness of 40 m at 20 m AGL. The model grid spacing is smaller than the Lagrangian analysis grid spacing of 500 m. Since BL structures were emphasized and since only rather shallow cumuli were observed on 22 May, the upper boundary had a layer thickness of 200 m at 6 km AGL. The 30 km × 30 km × 6 km simulation model grid thus contains 201 × 201 × 61 grid points. The model was integrated with a time step of 2 s. The model’s base-state profiles of pressure, potential temperature, vapor mixing ratio, and u- and υ-wind components were prescribed from a mobile ground-based sounding (i.e., as depicted in Fig. 3c of Buban et al. 2007).

The initial conditions for the dryline simulation were provided by spatially interpolating the radar wind synthesis and Lagrangian analysis fields corresponding to the initial model time onto the model grid from the surface through 2.5 km AGL. The horizontally homogeneous initial model fields above 2.5 km were prescribed from the base-state sounding. The initial pressure field was calculated by applying a hydrostatic balance constraint within each grid column.

The time-dependent lateral inflow boundary conditions for the dryline simulation were obtained by spatially and temporally interpolating the 3-min interval multiple-Doppler wind fields and 9-min interval Lagrangian analyses via nudging to the (fixed, ground relative) model grid at each time step. Above the 2.5-km level within the lateral inflow boundary surfaces, time-invariant and horizontally homogeneous fields were specified from the base state. Thus, observed finescale BL structures obtained from the radar and Lagrangian analyses were communicated into the model domain where they were subsequently forced by the nonlinear physics.

3. Model results

a. Mesoscale dryline environment

The dryline and surrounding BL were simulated over an intensive observing region (IOR) where data were collected on 22 May 2002 during IHOP_2002. The fixed model domain had the same dimensions and location as the IOR, facilitating both data assimilation and model validation. The initial conditions were specified at 2242 UTC from the Lagrangian analysis for temperature and moisture and the multiple-Doppler radar analysis of horizontal velocity. The simulations were run for 90 min (2242–0012 UTC) corresponding to the time of the first and last Lagrangian analyses.

Both the analysis and the simulation contain a nearly north–south-oriented dryline that vacillates initially before retrograding to the west later in the period (Fig. 1). The modeled and observed drylines are characterized by horizontal confluence and a strong moisture gradient (2–3 g kg−1 km−1), as also evidenced by the individual mobile mesonet traverses (Buban et al. 2007). The modeled dryline has a tendency to form along vortex sheet segments indicative of concentrated across-dryline shear of the dryline-parallel wind component. Variability in the moisture fields is manifest as small undulations or ripples that move northward along the dryline at nearly the speed of the mean boundary layer flow. Because the thermally forced BL circulations take on order 10 min to develop within the roughly 20 m s−1 southerly flow downstream from the inflow boundaries, analysis of the fully formed misoscale BL structure is valid only within about the northern two-thirds of the domain. Hence, subsequent discussion of the local airflow perturbations (e.g., updrafts cells, misocyclones, etc.) will emphasize the portion of the domain from y = 10 to 30 km.

The 2D horizontal accumulation of a scalar gradient (analogous to the frontogenesis function for the rate of change of the potential temperature gradient), neglecting source, sink, and mixing terms, can be expressed in the following form (Sanders 1955; Bluestein 1993; Ziegler et al. 1995; Grasso 2000; Buban et al. 2007):
e1
where H = (∂/∂x)i + (∂/∂y)j and the scalar s is either water vapor mixing ratio or virtual potential temperature in the present study. The dryline is dominated by persistent accumulation of water vapor mixing ratio gradients and virtual potential temperature gradients at low levels (Fig. 2), as also shown from the observations and Lagrangian analysis of Buban et al. (2007, see their Fig. 17). The quantity defined as accumulation (Saucier 1955) in the present study has been called frontogenesis in many previous dryline studies. A degree of localized along-dryline variability, however, exists in the structure of the accumulation field, especially during the early parts of the simulation. In the vertical, the dryline tends to have strong accumulation at low levels and negative accumulation at upper levels, consistent with convergence near the surface and divergence aloft as parcels exit the dryline updraft. This vertical structure of persisting airflow circulation and horizontal accumulation was also analyzed by Buban et al. (2007) and simulated in an earlier dryline case by Ziegler et al. (1995).
Fig. 2.
Fig. 2.

Simulated horizontal accumulation (color filled) of (a)–(d) water vapor mixing ratio and (e)–(h) virtual potential temperature at 250 m AGL and 2309–0003 UTC. Also shown are misocyclone-relative wind vectors (1 km = 5 m s−1) calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind, and vertical vorticity every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values black and negative values white. The black dashed curves denote dryline locations.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

The vacillating dryline location is marked by a persistent, strong localized maximum of solenoidal forcing (via the horizontal vorticity equation) that collocates with the maximum updraft (Fig. 3). The persistent updraft core located at the dryline manifests the upward branch of a persistent secondary circulation that is maintained by persistent solenoidal forcing that is evident in the mean flow (e.g., as also shown in Fig. 18 of Buban et al. 2007). The solenoidal secondary circulation assists in forming and maintaining the dryline by generating convergent, accumulative flow and vertical motion along the dryline. The solenoidal generation is a maximum at low levels and tilts downshear with height, as also shown in the Lagrangian analysis of Buban et al. (2007) and the modeling study of Ziegler et al. (1995). Wakimoto and Murphey (2009) analyzed the 22 May dryline on the mesoscale via dropsondes, and they documented the existence of a mesoscale virtual potential temperature gradient in the BL, which contributed via solenoidal forcing to a mesoscale secondary circulation with a maximum at low levels. Miao and Geerts (2007) list several other observed drylines that are all collocated with density gradients and secondary circulations. Schultz et al. (2007) discuss the relationship between the strength of the dryline gradient and the synoptic-scale confluence. From 2242 through 0012 UTC, both the analysis and simulation maintain strong horizontal convergence and updraft speeds along the dryline, as also shown by Weiss et al. (2006).

Fig. 3.
Fig. 3.

Vertical cross sections of the simulated solenoidal generation (color filled) of the along-dryline component of vorticity (×10−6 s−2). Also plotted are ground-relative wind vectors in the plane (1 km = 5 m s−1) and vertical vorticity every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values black and negative values white. Also shown are the locations of the dryline (DL) and a horizontal convective roll (HCR). The cross-sectional locations are shown in Fig. 1.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Although the strongest vertical motion tends to develop along the dryline, comparably intense localized cells or bands of vertical motion are also seen to the east and west of the dryline (Figs. 34). The presence of HCRs, transverse rolls, and OCCs are noted to the west of the primary dryline (Fig. 4) resulting from unstable stratification via surface heating (thus a net upward sensible heat flux). The HCRs, transverse rolls, and OCCs that develop in the simulation have a similar structure to those in the Lagrangian analyses (Buban et al. 2007, see their Fig. 5). As localized updraft cells evolve, cumuli develop at times within deeper updrafts in the northern part of the domain (Fig. 5). Higher-based cumuli develop both along a secondary dryline to the west of the primary dryline and along stronger plumes associated with HCRs and OCC segments. Lower-based cumuli develop to the east of the primary dryline where BL circulations have interacted with and crossed over the surface dryline location. Cumuli also develop along and east of the dryline where updrafts associated with solenoidal forcing locally lift a mixture of moist and dry air within a mesoscale updraft. As the simulation progresses toward early evening, the dryline retrogrades westward and BL convection west of the dryline weakens owing to the loss of surface heating as outgoing longwave radiation exceeds insolation.

Fig. 4.
Fig. 4.

Simulated vertical velocity (color filled) and misocyclone-relative wind at 500 m AGL (vectors) calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind, with a vector length of 1 km = 5 m s−1. Also shown are contours of vertical vorticity at 5 × 10−3 s−1 interval starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values solid and negative values dashed. The white dashed lines indicate OCCs. The dashed black curves denote the dryline locations.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Fig. 5.
Fig. 5.

Simulated water vapor mixing ratio (color filled) at the lowest model level, cloud water mixing ratio >0.05 g kg−1 (gray shaded) at ~3.5 km AGL, and misocyclone-relative wind vectors at the lowest model level calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind, with 1-km length equal to 5 m s−1. Vertical velocity is contoured at 2 m s−1 intervals starting at 1 (−1) m s−1 with positive (negative) values black (white) and is shown at ~3.5 km AGL. The black dashed curves denote the dryline location. The black dashed line indicates the cross section shown in Fig. 15.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

b. Misocyclones

The horizontal scale of the misocyclones on 22 May is generally about 1–3 km in the simulations. This is likely a result of the grid spacing differences between the radar analyses (500 m) and the model (150 m). The vortical circulations in the simulations are able to contract in scale via persistent convergent forcing owing to the finer grid spacing, allowing them to match the scales observed in the raw single-Doppler data. Conversely, the radar analysis of Buban et al. (2007) employed a one-pass Barnes radar data interpolation, which in combination with the relatively coarse grid resulted in spatial smoothing of the objectively analyzed misocyclones (e.g., Majcen et al. 2008). The maximum vertical vorticity within the simulated misocyclones (~30 × 10−3 s−1) is also stronger than the radar-synthesized misocyclones (~10 × 10−3 s−1), again likely owing to the difference in grid resolution and radar analysis smoothing.

Observed airflow undulations and bands of concentrated υ-component wind shear in the x direction that coincide with the dryline are introduced at the southern model domain boundary via the time-dependent lateral inflow boundary conditions, as prescribed by the radar analyses (e.g., Fig. 6). In contrast, the u-component wind shear in the y direction that is introduced at the southern boundary is about half as strong as the υ-component shear. These wavelike perturbations, which have a horizontal length scale of about 10 km, subsequently collapse in scale and intensify to form misocyclones as they move downstream inside the model domain. The resulting misocyclones move north-northeastward along the dryline at 18 m s−1 (nearly the speed of the mean BL flow of 19.2 m s−1). This speed estimate of a misocyclone was based on the movement of its vertical vorticity core at approximately 3-min intervals. The standard deviation of the speeds of all simulated misocyclones was 1.4 m s−1, while the standard deviation of the wind speed from the mean BL state was 3.3 m s−1. Although small differences exist in the structure of the various misocyclones, the main features common to the misocyclones can be presented by focusing on one particularly intense misocyclone (hereafter labeled “M1”).

Fig. 6.
Fig. 6.

Vertical cross section along the southern boundary with simulated vertical vorticity (color filled × 10−3 s−1), υ component of the wind (contoured), and plane-parallel ground-relative wind vectors (every other vector, with 1-km length = 15 m s−1). The black dashed curve denotes the dryline location. This vorticity perturbation amplifies into misocyclone M1 in Fig. 7.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Growth of misocyclone M1 is illustrated in a Lagrangian, vortex-following subdomain (Fig. 7). The misocyclone-relative horizontal wind in the figures was obtained by subtracting the misocyclone mean speed from the total wind. An undulation introduced at the southern lateral boundary (Fig. 6) is initially manifest as a concentrated but elongated1 region of enhanced positive vertical vorticity near the surface centered on the dryline (Fig. 7a). The maximum vertical vorticity associated with the initial shear zone introduced at the southern inflow boundary that subsequently amplifies into M1 has a magnitude of about 10 × 10−3 s−1 (Fig. 6). Additionally, an elongated enhanced region of negative vertical vorticity is present along a much weaker moisture gradient in the dry air west of the dryline. Convergence along and north of the developing misocyclone causes the region between the dryline and western weak moisture gradient to contract, creating a stronger dryline segment (Figs. 7a–d). As this occurs, the region of negative vertical vorticity along the western moisture gradient weakens because of a decrease in convergence, and the maximum vorticity within the misocyclone increases to around its peak value exceeding 30 × 10−3 s−1. The vorticity within the misocyclone at 2321 UTC (Fig. 7d) is strongest near the surface and weakens with height (Figs. 8g–i). South and west of the misocyclone, a secondary moisture gradient associated with an HCR extends from the misocyclone center southwest into the dry air (Fig. 7d). The developing misocyclone takes on an elliptical shape with the major axis initially oriented north–south, with vortex tails on the northern and southern ends. The misocyclonic axis precesses counterclockwise as the dryline moisture gradient is rotated.

Fig. 7.
Fig. 7.

Simulated water vapor mixing ratio (color filled), horizontal misocyclone-relative wind vectors (1 km = 5 m s−1) calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind, and vertical vorticity every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values solid and negative values dashed. (a)–(h) All fields are at the lowest model level (20 m AGL). The dashed black lines in (f) denote cross-sectional locations in Figs. 9 and 10.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Fig. 8.
Fig. 8.

Simulated vertical velocity (color filled), horizontal misocyclone-relative wind vectors (1 km = 5 m s−1) calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind, and vertical vorticity every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values black and negative values white. Vortex motion is from south to north (i.e., bottom to top). Heights are at (a),(d),(g) 234 m AGL; (b),(e),(h) 480 m AGL; and (c),(f),(i) 763 m AGL. Each row is at the same horizontal location. The black dashed line indicates an HCR extending to the southwest of the misocyclone.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

The genesis and roll-up of an elliptical vortex with vortex tails that subsequently precesses has been described by many studies (e.g., Goldstein 1931; Rosenhead 1931; Drazin and Reid 1981; Corcos and Sherman 1984; Pozrikidis and Higdon 1985). The most common explanation of the roll-up process of a shear band is due to nonlinear Kelvin–Helmholtz instability. In the special case where density gradients are absent and the shear is in the horizontal plane, the term “horizontal shearing instability” is used (Rayleigh 1880; Miles and Howard 1964; Lee and Wilhelmson 1997b). Periodic, small-amplitude disturbances along either a vortex sheet or a shear layer within some optimal range of width, shear magnitude, and vorticity are predicted to be unstable and subject to exponential initial growth according to linear theory. As nonlinear processes emerge, vorticity accumulates into localized elliptical cores, with the shear layer or vortex sheet stretched between the cores. These cores are connected to each other by vorticity “tails” or “wings,” and they rotate or precess with nearly constant angular velocity. The misocyclone adjusts the moisture field along the dryline as it moves northward, bringing higher moisture westward preceding and drier air eastward after vortex passage. The wrap-up of a gradient in a scalar field by a vortex has been demonstrated and theoretically explained in several studies (Doswell 1984, 1985; Davies-Jones 1985).

As the misocyclone approaches its mature phase or maximum vorticity, its major axis rotates more than 90° as moisture advects southward (i.e., relative to the vortex motion) to the west of the vortex (Fig. 7e). To the east of the misocyclone, dry air is simultaneously advected northward relative to the vortex motion. This process continues as the misocyclone migrates northward into the wrapping moist air (Fig. 7f). Eventually, moisture is wrapped completely around a sequestered pocket of dry air to the east of the misocyclone core. As this occurs, the misocyclone develops an “ inverted U shaped” asymmetry with the highest vorticity values residing on its southwest flank (Figs. 7g,h). As moisture wraps completely around the misocyclone center, a small region of negative vorticity moves northeast and strengthens just south of the strongest positive vorticity values on the south side of the misocyclone in association with a strengthening HCR (Fig. 7g). As the misocyclone interacts with the HCR, the southern part of the misocyclone reintensifies with vertical vorticity subsequently approaching its earlier peak value (Fig. 7h).

The strongest updrafts are initially collocated with the misocyclone and the dryline moisture gradient at 2321 UTC (i.e., when the vortex is farthest south), with enhanced updrafts along a developing HCR extending southwest into the dry air (Figs. 8g–i). The misocyclonic vorticity is strongest near the surface and weakens as the center tilts downshear with height. As the misocyclone reaches a mature phase (2327 UTC), the updrafts on both the northern dryline segment and the HCR strengthen as a downdraft develops within the misocyclone center (Figs. 8d–f). A separate updraft core is located along the eastern edge of the misocyclone at 2327 UTC. As the misocyclone decays by 2333 UTC, the updraft on the eastern edge of the vortex is rotated and relocated along the northeast portion of the misocyclone (Figs. 8a–c). The strongest updraft at 2333 UTC extends to the southwest of the misocyclone along the merged dryline and HCR, with only a weak updraft along the northern dryline segment. The central downdraft now resides in an area of weaker vertical vorticity at the center of the inverted U-shaped vortex, and the downdraft is flanked by regions of stronger vorticity on the northeast and southwest sides (Figs. 8a–c).

During the mature phase of the misocyclone, deep-layer convergence deepens the moist layer to the north of the vortex center (Fig. 9a). The dryline tilts eastward with height and remains associated with enhanced vertical vorticity due to across-dryline horizontal shears (Fig. 9a). An eastward surge of dry air to the south of the vortex effectively shifts the dryline eastward (Fig. 9c). The wrapping eastern and western drylines can be seen through the center of the misocyclone as dry air enters the misocyclone core during the wrap-up phase (Fig. 9b). Downdrafts associated with the dryline’s secondary circulation depress the top of the moist BL to the northeast and southeast of the misocyclone center (Figs. 9a,c).

Fig. 9.
Fig. 9.

Vertical cross sections of (a)–(c) simulated water vapor mixing ratio and (d)–(f) virtual potential temperature north, through the center, and to the south of a misocyclone at 2327 UTC. Also plotted are vortex-relative wind vectors (1 km = 15 m s−1) and vertical vorticity every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values black and negative values white. Cross-sectional locations are shown in Fig. 7f.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

For the most part, the virtual potential temperature fields tend to be negatively correlated with the water vapor mixing ratio fields. Relatively low θυ values are collocated with higher qυ values, while relatively high θυ values are collocated with lower qυ values (Figs. 9d–f vs Figs. 9a–c). However, isolated locations near the surface dryline location may have positively correlated temperature and moisture fields. As previously shown by Ziegler et al. (1997), the mesoscale BL circulations and updrafts associated with the dryline and HCR tend to transport the very warm, unstable lower-BL air into buoyant dry-convective plumes that may subsequently strengthen the mesoscale updrafts (Figs. 9d–f).

A pattern of 2D horizontal qυ accumulation at lower levels and negative qυ accumulation at upper levels is seen to the north of the misocyclone along the dryline (Figs. 10a–c). A similar pattern is found in terms of 2D horizontal θυ accumulation, although the negative accumulation at upper levels is much weaker (Figs. 10d–f). The dryline and HCRs have intense accumulation in the lowest ~200–500 m AGL. Accumulation occurs along both dryline gradients through the center of a mature misocyclone. The dryline gradients accumulate to the south, however, the accumulation zone is not as deep as to the north or along the eastern moisture gradient through the center of the misocyclone. It has been shown that accumulation (frontogenesis) is favored to the northwest and southeast of a cyclonically rotating vortex as it deforms a scalar field characterized by initially north–south-oriented isopleths associated with an east–west horizontal gradient (Doswell 1984, 1985; Davies-Jones 1985; Schultz et al. 1998). The updraft associated with the HCR contains pronounced θυ accumulation due to locally strong thermal gradients in the dry BL.

Fig. 10.
Fig. 10.

Vertical cross sections of simulated (a)–(c) accumulation of water vapor mixing ratio and (d)–(f) virtual potential temperature north, through the center, and south of a misocyclone at 2327 UTC. Also plotted are vortex-relative wind vectors (1 km = 15 m s−1) and vertical vorticity every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 (−5 × 10−3 s−1), with positive values black and negative values white. Cross-section locations are shown in Fig. 7f.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

4. Discussion

a. Comparison of simulated and observed misocyclones

Several of the simulated misocyclones in this study have a similar structure and evolution to M1 (discussed in section 3). The misocyclones intensify within a zone of convergence, concentrated shear, updrafts, and vertical vorticity, along or just east of the dryline moisture gradient (e.g., Fig. 11). As misocyclones M1, M2, and M4 continue to intensify, their major axes precess counterclockwise to form a characteristic “S shaped” gradient (Figs. 12a,b,d). In contrast, misocyclone M3 develops a more circular structure and does not precess significantly as it moves downstream (Fig. 12c). Misocyclones M1, M2, and M4 reach a maximum intensity after precessing about 90°. As each misocyclone reaches the mature phase, an axial downdraft develops as updrafts persist in convergent regions to the north and south of the circulation centers. Continued advection around the misocyclone eventually leads to a seclusion, wherein moisture wraps completely around the vortex and closes off a pocket of dry air.

Fig. 11.
Fig. 11.

(a)–(d) Simulated water vapor mixing ratio (color filled) and vertical vorticity (contoured every 5 × 10−3 s−1 starting at 5 × 10−3 s−1) at the lowest model level, with positive values black and negative values white and cloud water mixing ratio >0.05 g kg−1 (gray shaded) at ~3.5 km AGL. Also shown are the misocyclone-relative wind vectors at the lowest model level (1 km = 5 m s−1) calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind, with 1-km length equal to 5 m s−1. The letter c in (b) and (c) indicate the same cumulus cloud. The black dashed lines indicate cross-sectional locations shown in Fig. 17.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Fig. 12.
Fig. 12.

(a)–(p) Evolution of four misocyclones (M1–M4) at a 6-min interval from the simulation (i.e., with time and panel label increasing from bottom to top). Shown are the lowest model level positive vertical vorticity values (contoured) every 5 × 10−3 s−1 starting at 5 × 10−3 s−1 and horizontal vortex-relative wind vectors. Also shown is the 7.5 g kg−1 mixing ratio isopleth along the dryline (gray curve) at the lowest model level and estimated motion (black dashed line).

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Several notable features of the simulated misocyclones (Figs. 7 and 12) are consistent with observations for this case. These simulated features are also consistent with observations of other observed misocyclones (e.g., Pietrycha and Rasmussen 2004; Arnott et al. 2006; Marquis et al. 2007). The characteristic misocyclonic structure, which features an absolute maximum in vertical vorticity near the surface and a relative maximum that decreases and slopes downshear with height, is similar in both the simulation and the observations (Fig. 7; Buban et al. 2007, their Fig. 19). Both the simulated and observed misocyclones are centered and remain along the dryline gradient through the majority of their life cycles. The simulated and observed misocyclones also tend to be elliptically shaped. The major axes of elliptical misocyclones are oriented along the dryline gradient and precess counterclockwise with time. However, the precession is much slower in the observations than in the simulation. The misocyclones are initially collocated with an updraft that extends both north and south along the dryline, with the strongest updrafts located north of the misocyclones. The central updraft subsequently weakens in both the simulated and observed misocyclones, and a downdraft eventually replaces the weakening updraft in the simulation of M1. The limited period of radar measurements on 22 May precluded radar observation of a central downdraft. However, central downdrafts have been observed in other dryline misocyclones (Marquis et al. 2007). At later times, the updraft north of the misocyclone weakens whereas the updraft south of the misocyclone strengthens.

In both the simulations and the Lagrangian analyses, the misocyclones adjust the moisture fields along the dryline by advecting moisture to the west north of the vortex and advecting dry air to the east south of the vortex. A wrapping pattern of moist and dry air around the simulated misocyclones (e.g., as described in section 3, Fig. 9) is similar in the Lagrangian analyses (Buban et al. 2007, their Fig. 21) and trajectories of Marquis et al. (2007). Although the general effects of the misocyclones on the moisture fields are similar in both the Lagrangian analyses and the simulations, much more detail is seen in the simulations owing to both the finer grid resolutions and filtering in the radar and Lagrangian analyses (Ziegler et al. 2007). Additionally, the misocyclones were not observed long enough to capture the decay process, and therefore cannot be compared to the simulations in this regard.

b. Misocyclone dynamics

Given the similar structure and evolution of several simulated and observed misocyclones, it is likely that the simulation is resolving a consistent underlying dynamical evolution process. The formation of observed and simulated misocyclones occurs along dryline segments distinguished by convergence, shear, updrafts, and bands of vertical vorticity, suggesting that an essentially barotropic horizontal shearing instability may be playing an important role in misocyclogenesis (see previous discussion in section 3b). However, determination of the exact instability from which the misocyclones eventually grow is beyond the scope of this study. The simulation provides a high-resolution dataset from which subsequent misocyclone evolution and important dynamical forcing processes may be characterized.

Using the anelastic approximation of the momentum equations, the vertical vorticity equation can be written in the following form:
e2
where ζ is the vertical vorticity and the effects of friction and the earth’s rotation have been neglected (Shapiro et al. 2009). The baroclinic term has also been omitted from Eq. (2), as it is several orders of magnitude smaller than the remaining calculated terms (not shown). Thus, the local vertical vorticity tendency is due to the four right-hand-side terms, namely the tendencies due to the horizontal advection of ζ, the vertical advection of ζ, the tilting of the horizontal vorticity into the vertical axis, and the stretching of ζ, respectively. Since we are interested in the change in intensity following the vortex, the horizontal wind components in Eq. (2) are misocyclone relative. The dynamics of the misocyclone can be described in terms of vorticity tendencies as it evolves through three distinct phases.

During the initial growth phase (~2306–2312 UTC), the development of the misocyclone is primarily due to the stretching of vertical vorticity. During this phase the stretching mainly occurs along and east of the updraft core (Figs. 13a and 14a–d, bullet 1). West of the updraft core the vorticity tendency is primarily negative (Figs. 13a and 14a–d, bullets 2 and 3) and due to the tilting of westward-directed horizontal vorticity vectors downward (i.e., , , ). Weak negative advection of vertical vorticity into the misocyclone is present at this stage (i.e., the vertical vorticity is a local maximum in the misocyclone and the flow is convergent), while positive vertical advection of vertical vorticity is also acting through the top of the misocyclone (not shown).

Fig. 13.
Fig. 13.

Simulated vertical vorticity at the lowest model level (color filled) with other overlay fields at 2312–2330 UTC for misocyclone M1. (a),(c),(e),(g) Tilting production of vertical vorticity (contoured in black every 50 × 10−6 s−2 starting at 50 × 10−6 s−2), with positive values solid and negative values dashed, and stretching of vertical vorticity (contoured in white every 50 × 10−6 s−2 starting at 50 × 10−6 s−2), with positive values solid and negative values dashed. Also shown are the misocyclone-relative wind vectors at the lowest model level (1 km = 5 m s−1) calculated by subtracting a mean wind of 18 m s−1 at 190° from the total wind. (b),(d),(f),(h) Vertical velocity (contoured in black every 0.5 m s−1 starting at 0.5 m s−1) with positive values solid and negative values dashed, and horizontal vorticity vectors with 1 km length equal to 15 × 10−3 s−1. Black dashed lines indicate cross-sectional locations in Fig. 14.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Fig. 14.
Fig. 14.

Vorticity vectors in the plane of the cross-section (150-m length equal to 15 × 10−3 s−1) with kinematic fields at 2306–2330 UTC. (a),(c),(e),(g),(i) Vertical velocity (color filled), vertical tilting of horizontal vorticity (contoured in black every 200 × 10−6 s−2 starting at 50 × 10−6 s−2), with positive values solid and negative values dashed, and stretching of vertical vorticity (contoured in white every 100 × 10−6 s−2 starting at 50 × 10−6 s−2), with positive values solid and negative values dashed. (b),(d),(f),(h),(j) Vertical vorticity (color filled), the sum of tilting plus stretching of vertical vorticity tendency (contoured in black every 200 × 10−6 s−2 starting at 50 × 10−6 s−2), with positive values solid and negative values dashed. The black dashed line indicates the axis of the updraft core. Numbered white dots locate vorticity forcing features described and referred to as “bullets” in the text. Cross-sectional locations for (c)–(j) are located in Fig. 13.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

During the ensuing rapid growth phase (~2312–2324 UTC), the stretching of vertical vorticity intensifies as convergence and vertical motion increase within the misocyclone (refer to figures at 2318 UTC). Thus, during the rapid growth phase w increases with height more rapidly since w vanishes at the lower boundary. As increasing becomes located over increasing ζ, stretching intensifies along the updraft core (Figs. 13c,d and 14e,f, bullet 4). The vertical vorticity tendency west of the updraft core is negative and remains dominated by the tilting of westward-directed horizontal vorticity vectors downward (Figs. 13c,d and 14e,f, bullet 5). Although there is negative horizontal vertical vorticity advection, the strong stretching allows the misocyclone to grow in both intensity and diameter (Figs. 13c–f and 14e,f). Strong vertical advection of vertical vorticity also contributes to the deepening of the misocyclone through 2324 UTC. The misocyclone reaches its maximum intensity during this phase.

The onset of the decay phase is marked by simultaneous weakening of the tilting and stretching tendency terms. At the onset of misovortex decay, the horizontal and vertical advection terms are balanced with the tilting and stretching terms. From the initiation through the growth phase, the pressure within the misocyclone center falls as a cyclostrophic balance between the pressure and angular momentum is rapidly achieved and subsequently sustained. That is, the simulation rapidly adjusts the pressure in response to the circulation to remain dynamically balanced. In fact, calculations confirm that the pressure perturbation near the surface within the misocyclone is almost exactly what would be expected for the misocyclone to be in cyclostrophic balance given the misocyclone-relative horizontal velocity field (not shown). As the misocyclone continues to strengthen, the pressure continues to fall in the center. Since the misocyclonic vorticity decreases with height, there is a net downward-directed perturbation pressure gradient force. As the perturbation pressure gradient strengthens, the updraft weakens and subsequently transitions to a central downdraft (Figs. 14g,h, bullet 6). The transition process from axial updraft to downdraft in a misocyclone is similar to the development of an occlusion downdraft in the intensifying low-level mesocyclone of a supercell storm (e.g., Brandes 1984). This process limits the extent of misocyclone growth and initiates the onset of the decay phase. With the downdraft continuing to strengthen, the misocyclonic vertical vorticity weakens from the center outward. During this phase, compression of vortex tubes (Figs. 13g,h and 14i,j, bullet 7) along with negative tilting of eastward-directed vorticity vectors (Fig. 14i,j, bullet 8) lead to negative vertical vorticity tendency in the center of the misocyclone. Along the periphery of the central downdraft, areas of updraft and positive vertical vorticity stretching persist (Figs. 13g,h and 14i,j, bullet 9). Thus, the decaying misocyclone takes the form of an (inverted U shaped) arc of enhanced vorticity along the western edge of the broader circulation (Figs. 13g,h). As the misocyclone continues to decay, the remaining vertical vorticity weakens by 2330 UTC except for a residual vortex at the southern edge of the broader circulation (Figs. 13g,h). The weakening broader misocyclone then moves out of the model domain.

c. Cumulus formation

Shallow cumuli were observed both visually and by satellite within the IOR on 22 May (Buban et al. 2007). The Lagrangian analyses were unable to represent cumuli since the cloud base was higher than the top of the analysis domain. Cumuli developed in the simulation, thus providing a high-resolution dataset with which to demonstrate the transportive nature of the cumulus formation process. Ziegler and Rasmussen (1998) proposed a “parcel continuity principle,” which states that cumulus will form only if air parcels reach their LCL prior to exiting the mesoscale updraft that provides the lift needed to achieve water saturation. That is, the time required for an air parcel to cross and eventually exit the updraft (defined as w > 0) horizontally must equal or exceed the time required for the parcel to rise from its entrainment level to its LCL (Ziegler et al. 2007; Ziegler and Rasmussen 1998). Depending on the saturation point level of the source region of air and the time duration and intensity of lifting, only a rather small subset of all BL updraft air parcels would normally achieve their LCLs (Ziegler et al. 2007). The BL air that eventually formed simulated cumuli was hot and dry (θυ ~ 315 K and qυ ~ 5 g kg−1), thus requiring strong, sustained lifting through a rather deep layer that contained potentially inhibitive vertical wind shear to reach the LCL height of ~3.0–3.5 km. The latter conditions are achieved in several areas within the northeastern quadrant of the simulation domain (Fig. 5).

West of the dryline, high-based cumuli develop within stronger updrafts associated with a weak secondary dryline, HCRs, and OCCs. Along the dryline, lower-based cumuli develop within strong updraft bands (Fig. 5). Cumuli also develop where HCR or OCC segments have moved from southwest to northeast across the dryline (Fig. 5). Measurement of 30 separate updraft cores yielded a mean horizontal speed of 19.0 m s−1 with a standard deviation of 1.1 m s−1. Updraft cores thus tend to move near the mean BL velocity (19.2 m s−1), allowing air parcels to spend a sufficient amount of time within the updraft to reach their LCLs.

Cloud-base heights tend to decrease from west to east across the dryline as a result of the differing origins and saturation points of rising air (Figs. 15 and 16 ) as previously discussed by Ziegler and Rasmussen (1998). The range of simulated cloud-base heights are ~2.9–3.2 km AGL along and east of the dryline, ~3.30–3.35 km AGL just west of the dryline, and ~3.35–3.5 km AGL along the western secondary dryline. These LCL heights are broadly consistent with an LCL height of ~3.4 km estimated from mobile soundings launched west of the dryline (Buban et al. 2007). The detailed differences in cloud-base height are a manifestation of the differing thermodynamic characteristics of originating BL air masses of the air parcels that subsequently reach the cloud base.

Fig. 15.
Fig. 15.

Simulated relative humidity (color filled), water vapor mixing ratio (black contours at 0.5 g kg−1 interval), cloud water mixing ratio greater than 0.01 g kg−1 (gray shading), and plane-parallel wind vectors with 1 km = 5 m s−1 at 2305 UTC. Also shown are dryline (DL) and secondary dryline (SDL). The numbers within each cumulus cloud indicate regions in Fig. 16. The cross-sectional location is shown in Fig. 5.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

Fig. 16.
Fig. 16.

Cloud-base height vs east–west distance from the dryline. 1) Cumuli formed from air lifted along secondary dryline, 2) cumuli formed from dry air lifted west of the dryline, 3) cumuli formed from dry air lifted over the dryline, 4) cumuli formed from a mixture of dry and moist air within the dryline, and 5) cumuli formed from a mixture of dry and more moist air lifted just east of the dryline.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

The simulated cumuli can be organized into five groups according to individual cloud-base heights and locations (Fig. 16). Backward air trajectories calculated from cumuli in the first group (not shown) with the highest cloud base (~3.5 km) reveal that parcels eventually reaching their LCL originate in the middle to upper BL, and are lifted gradually while traversing the updraft along the secondary dryline. All air trajectories described in the present study are calculated using three iterations of a quadratic Runge–Kutta predictor-corrector scheme (McCalla 1967). In the second group, somewhat lower-based cumuli with cloud bases around 3.3 km form within locally stronger updrafts along HCR and OCC segments between the secondary and primary drylines. Air that reaches the LCL in this latter area has a source region in the lower to middle BL. Both higher-based and lower-based cumuli form along and just east of the dryline. The higher-based cumuli compose the third group and form along HCR and OCC segments that have crossed over the dryline. These latter cumuli have similar cloud-base heights and parcel source regions to the cumuli formed along HCR and OCC segments that have not crossed over the dryline. Air within these higher-based cumuli originates at various levels within the lower to middle BL west of the dryline. In the fourth group, lower-based cumuli form within the dryline updraft and consist of a mixture of parcels originating within the dryline moisture gradient at very low levels and drier air in the middle BL west of the dryline. These cumuli have bases ~3.2 km. Those cumuli farthest east with the lowest cloud base compose the fifth group, consisting of a mixture of moist air at low levels within the dryline gradient and dry air west of the dryline in the middle BL; however, those low-level parcels arise from the moist side of the dryline gradient.

Backward air trajectories illustrate parcel source regions (e.g., Figs. 1a and 17b) for a cumulus cloud east of the dryline. Air parcels that subsequently enter the cumulus cloud originate at the southern boundary from both the moist side of the dryline gradient and about 4 km west of the dryline (Fig. 1a). As these air parcels of differing saturation points enter the dryline updraft and are brought into closer proximity by convergence (Fig. 17b), they mix as they are accelerated upward (not shown). In fact, because of the helical nature of the flow, these two air parcels actually twist around one another as they approach the LCL (Figs. 1a and 17b), with the eastern and lower air parcel eventually becoming farther west and higher than the other parcel. It is found that the simulated cumuli all form from a mixture of air from differing source regions, that these source regions are spaced much farther apart than the scale of the individual clouds, and that air travels a considerable horizontal distance (~20 km) within the updraft prior to reaching the LCL.

Fig. 17.
Fig. 17.

(a),(c),(d) Relative humidity (color filled), water vapor mixing ratio (black contours at 0.5 g kg−1 interval), virtual potential temperature (white contours at 0.25-K interval), cloud water mixing ratio greater than 0.01 g kg−1 (gray shading), and plane-parallel wind vectors with 1 km = 15 m s−1. Cross-sectional locations are shown in Fig. 11. (b, left panel) Water vapor mixing ratio (color filled) and vertical velocity (every 1 m s−1 starting at ±1 m s−1) with positive (negative) values indicated by the solid (dashed) black contours. Also shown are plane-parallel wind vectors with 1 km = 20 m s−1. (b, right panel) As in (a), but omitting temperature. Also shown are plane-parallel wind vectors with 1 km = 20 m s−1. In (b), the thick black solid and dashed curves are the plane projections of the trajectories shown in Fig. 1. Cross-sectional locations for (b) are shown in Fig. 1.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

The simulated cumuli both east and west of the dryline are small in horizontal extent (Fig. 11) and rather shallow (Figs. 15 and 17). These characteristics are consistent with photographic images of cumuli on 22 May (Buban et al. 2007, their Fig. 14c). Cumuli formed by a strengthening HCR west of the dryline (Figs. 11c and 17c) and cumuli formed from HCR segments that have moved over the dryline (Figs. 11a,d and 17a,d) have nearly the same cloud-base height, reside within a larger area of higher relative humidity, and are forced by a rather erect updraft plume associated with the parent HCR. A significant distinction between these cumuli is that the plumes that develop cumuli east of the dryline are no longer connected to their near-surface updrafts due to the stratification introduced by the (virtually cooler) moist BL. Cumuli formed by a mixture of dry air from the west of the dryline and moist air within the dryline zone have a lower cloud base (Figs. 11b and 17b) and are forced by the along-dryline updraft whose enhanced tilt is a consequence of the solenoidally forced secondary circulation that also enhances westerly shear.

Several cumuli develop near misocyclones in the simulations (Fig. 11). Although some appear to be a result of previously hypothesized processes (e.g., deepening of the moist layer and enhanced convergence north of the misocyclone due to the misocyclone’s interaction with the dryline), the number of cumuli associated with misocyclones is small compared to the total number of cumuli. Furthermore, cumuli do not develop within the vicinity of several misocyclones. Therefore, no inferences can be drawn from the present simulation regarding the influence of misocyclones on cumulus formation. This conclusion is not surprising, since misocyclones are relatively shallow features (i.e., ~1 km deep) in comparison to the deep, dry-convective updrafts and since the LCLs on this day are much higher (~3–3.5 km).

5. Conclusions

This paper presents results of a COMMAS model simulation of the dryline and associated misocyclones and cumulus clouds using initial and lateral boundary conditions obtained from Lagrangian and multiple-Doppler wind analyses of the 22 May 2002 IHOP_2002 case. The time-dependent, prescribed lateral inflow boundary conditions are spatially and temporally interpolated to the model grid from the Lagrangian and multiple-Doppler wind analyses, where they are applied at every time step. The outflow boundaries are treated as wave radiating. The model has been further modified to include a force–restore surface flux scheme following Deardorff (1978) and includes both shortwave and longwave radiation. Cumulus clouds are explicitly resolved in the simulation (including cloud-shadow effects), and a simple Kessler microphysics scheme has been implemented here since the observed cumulus clouds on 22 May did not reach the precipitation stage.

The simulation reproduces the dryline, misocyclones, and convective BL features such as HCRs and OCCs, with similar structures to observed features of equivalent scale. However, some differences in the details exist. The dryline is simulated as a nearly north–south-oriented zone marked by qυ and θυ gradients of about 2–3 g kg−1 km−1 and 1 K km−1, respectively. The dryline is persistently convergent with accumulation in lower levels, and maintains a persistent solenoidally driven secondary circulation. These characteristics of the dryline were also observed, albeit weaker in the observational analyses than the simulations mainly due to coarser grid spacing and necessary analysis filtering. These characteristics are also consistent with both observed and simulated drylines in other past studies. Features on the dry and moist sides of the dryline (e.g., moisture plumes, HCRs, and OCCs) also had higher amplitudes and more detailed structure in the simulations than in the analyses as a result of the finer model grid resolution.

The misocyclones that develop along the dryline in the simulation are of particular interest. Observed-prescribed shear perturbations along the southern inflow boundary enter the model domain as elongated enhanced regions of vertical vorticity along the dryline. As the misocyclones develop, they take on an elliptical shape with northern and southern vorticity tails, vertical vorticity within their interior increases, and they begin to precess counterclockwise. The observed fully developed misocyclones were also elliptical and precessed counterclockwise with time. These characteristics are consistent with previous numerical studies of vortices formed from the wrap-up of vortex sheets or shear layers. The vertical vorticity within the misocyclones has a maximum near the ground and decreases with height in the downshear-sloping core, as also seen in the observations. The misocyclones rotate the moisture gradient along the dryline, advecting moisture north of the vortex to the west and dry air south of the vortex to the east. Just prior to misocyclone decay, the process of moisture wrapping completely around the vortex ultimately closes off a pocket of dry air near the center.

The general misocyclone evolution (as characterized by the evolution of misocyclone M1) can be described in three phases. In the initial growth phase, the elongated area of enhanced vertical vorticity grows primarily in response to positive stretching of vertical vorticity along and east of the dryline updraft core. During the rapid growth phase, the stretching of vertical vorticity by the dryline updraft intensifies as convergence and vertical motion locally increase within the misocyclone. Strong negative tilting of horizontal vorticity into the vertical persists along the western edge of the misocyclone. In spite of the negative tendency due to tilting combined with negative horizontal advection of vertical vorticity into the misocyclone, the strong stretching combined with strong vertical advection of vertical vorticity allows the misocyclone to continue growing in intensity and in diameter. As the misocyclone achieves its maximum intensity, an axial downdraft develops due to a downward-directed vertical perturbation pressure force in response to the near-surface maximum vertical vorticity strongly decreasing with height. This central downdraft limits the extent of misocyclone growth and subsequently initiates the decay phase via the downward tilting of horizontal vorticity vectors and negative stretching of vertical vorticity as vortex tubes are compressed. Persistent negative tilting and stretching during the decay phase eventually results in the erosion of the misocyclone from the center outward.

Simulated cumuli have similar characteristics to observed cumuli. High-based simulated cumuli develop along a weak secondary dryline west of the primary dryline and contain a mixture of air originating from the middle and upper BL. Somewhat lower-based cumuli develop along stronger plumes associated with HCRs and OCCs between the primary and secondary drylines, and consist of a mixture of air with source regions in the lower and middle BL. Cumuli also develop east of the dryline and have similar cloud-base heights and air source regions as those between the primary and secondary drylines. These cumuli arise within updrafts associated with HCR and OCC segments from the west that move over the dryline. Lower-based cumuli also develop within the dryline updraft east of the surface dryline position. These cumuli contain a mixture of dry air from the middle BL west of the dryline and varying degrees of moist air from very low levels within the dryline gradient. All simulated cumuli contain at least some dry air from west of the dryline. Furthermore, all cumuli form from a mixture of air from more than one source region and thus air with a range of water vapor mixing ratios and saturation points. The cumulus source regions are spaced much farther apart than the scale of the individual clouds while air travels a considerable horizontal distance prior to reaching its LCL. Orientation of HCR and dryline updrafts along the mean wind vector allows lifted parcels to spend a sufficient amount of time within the updraft to reach their LCL.

Acknowledgments

The authors gratefully acknowledge contributions by David Schultz and three anonymous reviewers whose comments improved the manuscript. Louis Wicker offered constructive comments about model numerics, while David Stensrud helpfully critiqued the comparison between modeled and observed surface fluxes. Funding for this research was provided by National Science Foundation Grants ATM-0130316, ATM-0638572, and AGS-0638512, and the National Severe Storms Laboratory.

APPENDIX

Surface Fluxes

Having realistic surface fluxes is important owing to their influence on the BL structure. For example, inclusion of surface heating and specification of unstable stratification at the lateral boundaries from the Lagrangian analyses forces the development of dry-convective plumelike structures in the BL (Figs. 9d–f). Observed surface fluxes obtained from the National Center for Atmospheric Research (NCAR) Integrated Surface Flux Facility (ISFF) located at Elmwood, Oklahoma, during IHOP_2002 (e.g., Conzemius and Fedorovich 2008) are compared to simulated fluxes calculated at the closest grid point to the ISFF site (Table A1 and Fig. A1). The observed and simulated sensible and latent heat fluxes are similar in both magnitude and evolution throughout the 90-min period. The RMS error between the observed and simulated sensible heat fluxes (30.52 W m−2) is small relative to the flux values (~125 W m−2). The RMS error for the latent heat flux (17.31 W m−2) is smaller as are the actual values (~50 W m−2); hence, the normalized RMS error is larger in comparison. However, the RMSE of latent heat flux seems reasonable given the larger range of possible values. Similarly, the RMSEs are also small for the observed and modeled skin temperatures (0.96 K) compared to the magnitude of temperatures and their ranges. Likewise, the observed and modeled surface temperatures also match closely with a RMSE of 0.76 K. However, there is more variance in the observations than in the simulations. The larger observed variance could be a consequence of the sensors measuring transient features unresolved by the model, the passage of observed (but not modeled) cumuli over the ISFF sensors, or the passage of modeled (but not observed) cumuli over the site location in the model domain. In addition, the formulation of the surface flux parameterization is such that the surface fluxes are rather smooth in the model.

Table A1.

RMSE and bias between the simulated and observed sensible heat flux, latent heat flux, skin temperature, and surface temperature. Observed values are from the Elmwood, OK, flux station and simulated values are from the grid point nearest to this station in the model.

Table A1.
Fig. A1.
Fig. A1.

Time series of (right) simulated and observed sensible and latent heat fluxes and (left) surface and skin temperatures. Also shown are the (solid) linear trend lines.

Citation: Monthly Weather Review 140, 11; 10.1175/MWR-D-11-00189.1

REFERENCES

  • Arnott, N. R., Y. P. Richardson, J. M. Wurman, and E. M. Rasmussen, 2006: Relationship between a weakening cold front, misocyclones, and cloud development on 10 June 2002 during IHOP. Mon. Wea. Rev., 134, 311335.

    • Search Google Scholar
    • Export Citation
  • Atkins, N. T., R. M. Wakimoto, and T. M. Weckwerth, 1995: Observations of the sea-breeze front during CaPE. Part II: Dual-Doppler and aircraft analysis. Mon. Wea. Rev., 123, 944969.

    • Search Google Scholar
    • Export Citation
  • Barnes, S. L., 1973: Mesoscale objective analysis using weighted time-series observations. NOAA Tech. Memo. ERL NSSL-62, National Severe Storms Laboratory, 60 pp.

  • Benjamin, S. G., and T. N. Carlson, 1986: Some effects of surface heating and topography on the regional severe storm environment. Part I: Three-dimensional simulations. Mon. Wea. Rev., 114, 307329.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Vol. 2, Observations and Theory of Weather Systems, Oxford University Press, 594 pp.

  • Brandes, E. A., 1984: Relationships between radar-derived thermodynamic variables and tornadogenesis. Mon. Wea. Rev., 112, 10331052.

  • Buban, M. S., C. L. Ziegler, E. N. Rasmussen, and Y. P. Richardson, 2007: The dryline on 22 May 2002 during IHOP: Ground-radar and in situ data analyses of the dryline and boundary layer evolution. Mon. Wea. Rev., 135, 24732505.

    • Search Google Scholar
    • Export Citation
  • Carbone, R. E., 1982: A severe frontal rainband. Part I: Stormwide hydrodynamic structure. J. Atmos. Sci., 39, 258279.

  • Carbone, R. E., 1983: A severe frontal rainband. Part II: Tornado parent vortex circulation. J. Atmos. Sci., 40, 26392654.

  • Coniglio, M. C., D. J. Stensrud, and L. J. Wicker, 2006: Effects of upper-level shear on the structure and maintenance of strong quasi-linear mesoscale convective systems. J. Atmos. Sci., 63, 12311252.

    • Search Google Scholar
    • Export Citation
  • Conzemius, R. J., and E. Fedorovich, 2008: A case study of convective boundary layer development during IHOP_2002: Numerical simulations compared to observations. Mon. Wea. Rev., 136, 23052320.

    • Search Google Scholar
    • Export Citation
  • Corcos, G. M., and F. S. Sherman, 1984: The mixing layer: Deterministic models of a turbulent flow. Part 1. Introduction and the two-dimensional flow. J. Fluid Mech., 139, 2965.

    • Search Google Scholar
    • Export Citation
  • Crook, N. A., T. L. Clark, and M. W. Moncrieff, 1991: The Denver cyclone. Part II: Interaction with the convective boundary layer. J. Atmos. Sci., 48, 21092126.

    • Search Google Scholar
    • Export Citation
  • Davies-Jones, R., 1985: Comments on “A kinematic analysis of frontogenesis associated with a nondivergent vortex.” J. Atmos. Sci., 42, 20732075.

    • Search Google Scholar
    • Export Citation
  • Deardorff, J. W., 1978: Efficient prediction of ground surface temperature and moisture, with inclusion of a layer of vegetation. J. Geophys. Res., 83 (C4), 18891903.

    • Search Google Scholar
    • Export Citation
  • Doswell, C. A., 1984: A kinematic analysis of frontogenesis associated with a nondivergent vortex. J. Atmos. Sci., 41, 12421248.

  • Doswell, C. A., 1985: Reply. J. Atmos. Sci., 42, 20762079.

  • Drazin, P. G., and W. H. Reid, 1981: Hydrodynamic Stability. Cambridge University Press, 525 pp.

  • Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38, 15111534.

  • Goldstein, S., 1931: On the stability of superposed streams of fluids of different densities. Proc. Roy. Soc. London, 132, 524548.

  • Grasso, L. D., 2000: A numerical simulation of dryline sensitivity to soil moisture. Mon. Wea. Rev., 128, 28162834.

  • Kanak, K., 2008: Vortical structures in convective boundary layers and implications for the initiation of deep convection. Preprints, 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., 18.3. [Available online at https://ams.confex.com/ams/24SLS/techprogram/paper_142080.htm.]

  • Kanak, K., D. K. Lilly, and J. T. Snow, 2000: The formation of vertical vortices in the convective boundary layer. Quart. J. Roy. Meteor. Soc., 126A, 27892810.

    • Search Google Scholar
    • Export Citation
  • Kawashima, M., and Y. Fujiyoshi, 2005: Shear instability wave along a snowband: Instability structure, evolution, and energetics derived from dual-Doppler radar data. J. Atmos. Sci., 62, 351370.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., and R. B. Wilhelmson, 1997a: The numerical simulation of non-supercell tornadogenesis. Part I: Initiation and evolution of pretornadic misocyclone circulations along a dry outflow boundary. J. Atmos. Sci., 54, 3260.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., and R. B. Wilhelmson, 1997b: The numerical simulation of non-supercell tornadogenesis. Part II: Evolution of a family of tornadoes along a weak outflow boundary. J. Atmos. Sci., 54, 23872415.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., C. A. Finley, and R. B. Wilhelmson, 2000: Simulating deep convection initiation by misocyclones. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., P2.3. [Available online at https://ams.confex.com/ams/Sept2000/techprogram/paper_16291.htm.]

  • Majcen, M., P. Markowski, Y. Richardson, D. Dowell, and J. Wurman, 2008: Multipass objective analyses of Doppler radar data. J. Atmos. Oceanic Technol., 25, 18451858.

    • Search Google Scholar
    • Export Citation
  • Mansell, E. R., C. L. Ziegler, and E. C. Bruning, 2010: Simulated electrification of a small thunderstorm with two-moment bulk microphysics. J. Atmos. Sci., 67, 171194.

    • Search Google Scholar
    • Export Citation
  • Marquis, J., Y. P. Richardson, and J. M. Wurman, 2007: Kinematic observations of misocyclones along boundaries during IHOP. Mon. Wea. Rev., 135, 17491768.

    • Search Google Scholar
    • Export Citation
  • McCalla, T. R., 1967: Introduction to Numerical Methods and FORTRAN Programming. Wiley, 359 pp.

  • Miao, Q., and B. Geerts, 2007: Finescale vertical structure and dynamics of some dryline boundaries observed in IHOP. Mon. Wea. Rev., 135, 41614184.

    • Search Google Scholar
    • Export Citation
  • Miles, J. W., and L. N. Howard, 1964: Note on a heterogeneous shear flow. J. Fluid Mech., 20, 331336.

  • Murphey, H. V., R. M. Wakimoto, C. Flamant, and D. E. Kingsmill, 2006: Dryline on 19 June 2002 during IHOP. Part I: Airborne Doppler and LEANDRE II analyses of the thin line structure and convection initiation. Mon. Wea. Rev., 134, 406430.

    • Search Google Scholar
    • Export Citation
  • Peckham, S. E., R. B. Wilhelmson, L. J. Wicker, and C. L. Ziegler, 2004: Numerical simulation of the interaction between the dryline and horizontal convective rolls. Mon. Wea. Rev., 132, 17921812.

    • Search Google Scholar
    • Export Citation
  • Pietrycha, A. E., and E. N. Rasmussen, 2004: Finescale surface observations of the dryline: A mobile mesonet perspective. Wea. Forecasting, 19, 10751088.

    • Search Google Scholar
    • Export Citation
  • Pozrikidis, C., and J. J. L. Higdon, 1985: Nonlinear Kelvin-Helmholtz instability of a finite vortex layer. J. Fluid Mech., 157, 225263.

    • Search Google Scholar
    • Export Citation
  • Rayleigh, L., 1880: On the stability, or instability, of certain fluid motions. Proc. London Math. Soc., XI, 5770.

  • Rosenhead, L., 1931: The formation of vortices from a surface of discontinuity. Proc. Roy. Soc. London, 134, 170192.

  • Sanders, F., 1955: An investigation of the structure and dynamics of an intense frontal zone. J. Meteor., 12, 542552.

  • Saucier, W. J., 1955: Principles of Meteorological Analysis. Dover Publications, Inc., 438 pp.

  • Schultz, D. M., D. Keyser, and L. F. Bosart, 1998: The effect of large-scale flow on low-level frontal structure and evolution in midlatitude cyclones. Mon. Wea. Rev., 126, 17671791.

    • Search Google Scholar
    • Export Citation
  • Schultz, D. M., C. C. Weiss, and P. M. Hoffman, 2007: The synoptic regulation of dryline intensity. Mon. Wea. Rev., 135, 16991709.

  • Shapiro, A., C. K. Potvin, and G. Jidong, 2009: Use of a vertical vorticity equation in variational dual-Doppler wind analysis. J. Atmos. Oceanic Technol., 26, 20892106.

    • Search Google Scholar
    • Export Citation
  • Sun, W., and C. Wu, 1992: Formation and diurnal variation of the dryline. J. Atmos. Sci., 49, 16061619.

  • Wakimoto, R. M., and J. W. Wilson, 1989: Non-supercell tornadoes. Mon. Wea. Rev., 117, 11131140.

  • Wakimoto, R. M., and N. T. Atkins, 1996: Observations on the origins of rotation: The Newcastle tornado during VORTEX 94. Mon. Wea. Rev., 124, 384407.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., and H. V. Murphey, 2009: Analysis of a dryline during IHOP: Implications for convection initiation. Mon. Wea. Rev., 137, 912936.

    • Search Google Scholar
    • Export Citation
  • Wakimoto, R. M., C. Liu, and H. Cai, 1998: The Garden City, Kansas, storm during VORTEX 95. Part I: Overview of the storm’s life cycle and mesocyclogenesis. Mon. Wea. Rev., 126, 372392.

    • Search Google Scholar
    • Export Citation
  • Warner, T. T., R. A. Peterson, and R. E. Treadon, 1997: A tutorial on lateral boundary conditions as a basic and potentially serious limitation to regional numerical weather prediction. Bull. Amer. Meteor. Soc., 78, 25992617.

    • Search Google Scholar
    • Export Citation
  • Weiss, C. C., H. B. Bluestein, and A. L. Pazmany, 2006: Finescale radar observations of the 22 May 2002 dryline during the International H2O project (IHOP). Mon. Wea. Rev., 134, 273293.

    • Search Google Scholar
    • Export Citation
  • Wicker, L. J., and R. B. Wilhelmson, 1995: Simulation and analysis of tornado development and decay within a three-dimensional supercell thunderstorm. J. Atmos. Sci., 52, 26752703.

    • Search Google Scholar
    • Export Citation
  • Wicker, L. J., and W. Skamarock, 2002: Time-splitting methods for elastic models using forward time schemes. Mon. Wea. Rev., 130, 20882097.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., and E. N. Rasmussen, 1998: The initiation of moist convection at the dryline: Forecasting issues from a case study perspective. Wea. Forecasting, 13, 11061131.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., W. J. Martin, R. A. Pielke, and R. L. Walko, 1995: A modeling study of the dryline. J. Atmos. Sci., 52, 263285.

  • Ziegler, C. L., T. J. Lee, and R. A. Pielke Sr., 1997: Convective initiation at the dryline: A modeling study. Mon. Wea. Rev., 125, 10011026.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., E. N. Rasmussen, T. R. Shepherd, A. I. Watson, and J. M. Straka, 2001: The evolution of low-level rotation in the 29 May 1994 Newcastle–Graham, Texas, storm complex during VORTEX. Mon. Wea. Rev., 129, 13391368.

    • Search Google Scholar
    • Export Citation
  • Ziegler, C. L., M. S. Buban, and E. N. Rasmussen, 2007: A Lagrangian objective analysis technique for assimilating in situ observations with multiple-radar-derived airflow. Mon. Wea. Rev., 135, 24172442.

    • Search Google Scholar
    • Export Citation
1

The introduction of persistent, slowly evolving radar-observed shear at the inflow boundary coupled with strong horizontal advection produces the elongated shear bands within the model domain.

Save
  • Arnott, N. R., Y. P. Richardson, J. M. Wurman, and E. M. Rasmussen, 2006: Relationship between a weakening cold front, misocyclones, and cloud development on 10 June 2002 during IHOP. Mon. Wea. Rev., 134, 311335.

    • Search Google Scholar
    • Export Citation
  • Atkins, N. T., R. M. Wakimoto, and T. M. Weckwerth, 1995: Observations of the sea-breeze front during CaPE. Part II: Dual-Doppler and aircraft analysis. Mon. Wea. Rev., 123, 944969.

    • Search Google Scholar
    • Export Citation
  • Barnes, S. L., 1973: Mesoscale objective analysis using weighted time-series observations. NOAA Tech. Memo. ERL NSSL-62, National Severe Storms Laboratory, 60 pp.

  • Benjamin, S. G., and T. N. Carlson, 1986: Some effects of surface heating and topography on the regional severe storm environment. Part I: Three-dimensional simulations. Mon. Wea. Rev., 114, 307329.

    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., 1993: Synoptic-Dynamic Meteorology in Midlatitudes. Vol. 2, Observations and Theory of Weather Systems, Oxford University Press, 594 pp.

  • Brandes, E. A., 1984: Relationships between radar-derived thermodynamic variables and tornadogenesis. Mon. Wea. Rev., 112, 10331052.

  • Buban, M. S., C. L. Ziegler, E. N. Rasmussen, and Y. P. Richardson, 2007: The dryline on 22 May 2002 during IHOP: Ground-radar and in situ data analyses of the dryline and boundary layer evolution. Mon. Wea. Rev., 135, 24732505.

    • Search Google Scholar
    • Export Citation
  • Carbone, R. E., 1982: A severe frontal rainband. Part I: Stormwide hydrodynamic structure. J. Atmos. Sci., 39, 258279.

  • Carbone, R. E., 1983: A severe frontal rainband. Part II: Tornado parent vortex circulation. J. Atmos. Sci., 40, 26392654.

  • Coniglio, M. C., D. J. Stensrud, and L. J. Wicker, 2006: Effects of upper-level shear on the structure and maintenance of strong quasi-linear mesoscale convective systems. J. Atmos. Sci., 63, 12311252.

    • Search Google Scholar
    • Export Citation
  • Conzemius, R. J., and E. Fedorovich, 2008: A case study of convective boundary layer development during IHOP_2002: Numerical simulations compared to observations. Mon. Wea. Rev., 136, 23052320.

    • Search Google Scholar
    • Export Citation
  • Corcos, G. M., and F. S. Sherman, 1984: The mixing layer: Deterministic models of a turbulent flow. Part 1. Introduction and the two-dimensional flow. J. Fluid Mech., 139, 2965.

    • Search Google Scholar
    • Export Citation
  • Crook, N. A., T. L. Clark, and M. W. Moncrieff, 1991: The Denver cyclone. Part II: Interaction with the convective boundary layer. J. Atmos. Sci., 48, 21092126.

    • Search Google Scholar
    • Export Citation
  • Davies-Jones, R., 1985: Comments on “A kinematic analysis of frontogenesis associated with a nondivergent vortex.” J. Atmos. Sci., 42, 20732075.

    • Search Google Scholar
    • Export Citation
  • Deardorff, J. W., 1978: Efficient prediction of ground surface temperature and moisture, with inclusion of a layer of vegetation. J. Geophys. Res., 83 (C4), 18891903.

    • Search Google Scholar
    • Export Citation
  • Doswell, C. A., 1984: A kinematic analysis of frontogenesis associated with a nondivergent vortex. J. Atmos. Sci., 41, 12421248.

  • Doswell, C. A., 1985: Reply. J. Atmos. Sci., 42, 20762079.

  • Drazin, P. G., and W. H. Reid, 1981: Hydrodynamic Stability. Cambridge University Press, 525 pp.

  • Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38, 15111534.

  • Goldstein, S., 1931: On the stability of superposed streams of fluids of different densities. Proc. Roy. Soc. London, 132, 524548.

  • Grasso, L. D., 2000: A numerical simulation of dryline sensitivity to soil moisture. Mon. Wea. Rev., 128, 28162834.

  • Kanak, K., 2008: Vortical structures in convective boundary layers and implications for the initiation of deep convection. Preprints, 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., 18.3. [Available online at https://ams.confex.com/ams/24SLS/techprogram/paper_142080.htm.]

  • Kanak, K., D. K. Lilly, and J. T. Snow, 2000: The formation of vertical vortices in the convective boundary layer. Quart. J. Roy. Meteor. Soc., 126A, 27892810.

    • Search Google Scholar
    • Export Citation
  • Kawashima, M., and Y. Fujiyoshi, 2005: Shear instability wave along a snowband: Instability structure, evolution, and energetics derived from dual-Doppler radar data. J. Atmos. Sci., 62, 351370.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., and R. B. Wilhelmson, 1997a: The numerical simulation of non-supercell tornadogenesis. Part I: Initiation and evolution of pretornadic misocyclone circulations along a dry outflow boundary. J. Atmos. Sci., 54, 3260.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., and R. B. Wilhelmson, 1997b: The numerical simulation of non-supercell tornadogenesis. Part II: Evolution of a family of tornadoes along a weak outflow boundary. J. Atmos. Sci., 54, 23872415.

    • Search Google Scholar
    • Export Citation
  • Lee, B. D., C. A. Finley, and R. B. Wilhelmson, 2000: Simulating deep convection initiation by misocyclones. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., P2.3. [Available online at https://ams.confex.com/ams/Sept2000/techprogram/paper_16291.htm.]

  • Majcen, M., P. Markowski, Y. Richardson, D. Dowell, and J. Wurman, 2008: Multipass objective analyses of Doppler radar data. J. Atmos. Oceanic Technol., 25, 18451858.

    • Search Google Scholar
    • Export Citation
  • Mansell, E. R., C. L. Ziegler, and E. C. Bruning, 2010: Simulated electrification of a small thunderstorm with two-moment bulk microphysics. J. Atmos. Sci., 67, 171194.

    • Search Google Scholar
    • Export Citation
  • Marquis, J., Y. P. Richardson, and J. M. Wurman, 2007: Kinematic observations of misocyclones along boundaries during IHOP. Mon. Wea. Rev., 135, 17491768.

    • Search Google Scholar
    • Export Citation
  • McCalla, T. R., 1967: Introduction to Numerical Methods and FORTRAN Programming. Wiley, 359 pp.

  • Miao, Q., and B. Geerts, 2007: Finescale vertical structure and dynamics of some dryline boundaries observed in IHOP. Mon. Wea. Rev., 135, 41614184.

    • Search Google Scholar
    • Export Citation
  • Miles, J. W., and L. N. Howard, 1964: Note on a heterogeneous shear flow. J. Fluid Mech., 20, 331336.

  • Murphey, H. V., R. M. Wakimoto, C. Flamant, and D. E. Kingsmill, 2006: Dryline on 19 June 2002 during IHOP. Part I: Airborne Doppler and LEANDRE II analyses of the thin line structure and convection initiation. Mon. Wea. Rev., 134, 406430.

    • Search Google Scholar
    • Export Citation
  • Peckham, S. E., R. B. Wilhelmson, L. J. Wicker, and C. L. Ziegler, 2004: Numerical simulation of the interaction between the dryline and horizontal convective rolls. Mon. Wea. Rev., 132, 17921812.

    • Search Google Scholar
    • Export Citation
  • Pietrycha, A. E., and E. N. Rasmussen, 2004: Finescale surface observations of the dryline: A mobile mesonet perspective. Wea. Forecasting, 19, 10751088.

    • Search Google Scholar
    • Export Citation
  • Pozrikidis, C., and J. J. L. Higdon, 1985: Nonlinear Kelvin-Helmholtz instability of a finite vortex layer. J. Fluid Mech., 157, 225263.

    • Search Google Scholar
    • Export Citation
  • Rayleigh, L., 1880: On the stability, or instability, of certain fluid motions. Proc. London Math. Soc., XI, 5770.

  • Rosenhead, L., 1931: The formation of vortices from a surface of discontinuity. Proc. Roy. Soc. London, 134, 170192.

  • Sanders, F., 1955: An investigation of the structure and dynamics of an intense frontal zone. J. Meteor., 12, 542552.

  • Saucier, W. J., 1955: Principles of Meteorological Analysis. Dover Publications, Inc., 438 pp.

  • Schultz, D. M., D. Keyser, and L. F. Bosart, 1998: The effect of large-scale flow on low-level frontal structure and evolution in midlatitude cyclones. Mon. Wea. Rev., 126, 17671791.

    • Search Google Scholar
    • Export Citation
  • Schultz, D. M., C. C. Weiss, and P. M. Hoffman, 2007: The synoptic regulation of dryline intensity. Mon. Wea. Rev., 135, 16991709.

  • Shapiro, A., C. K. Potvin, and G. Jidong, 2009: Use of a vertical vorticity equation in variational dual-Doppler wind analysis. J. Atmos. Oceanic Technol., 26, 20892106.

    • Search Google Scholar
    • Export Citation
  • Sun, W., and C. Wu, 1992: Formation and diurnal variation of the dryline. J. Atmos. Sci., 49, 16061619.

  • Wakimoto, R. M., and J. W. Wilson, 1989: Non-supercell tornadoes. Mon. Wea. Rev., 117, 11131140.

  • Wakimoto, R. M., and N. T. Atkins, 1996: Observations on the origins of rotation: The Newcastle tornado during VORTEX 94. Mon. Wea. Rev., 124, 384407.