1. Introduction
a. Motivation
The governing dynamics of supercells are a long-standing area of active research, often with a focus on processes relevant to the developing and mature stages of the storm (e.g., Klemp et al. 1981; Rotunno and Klemp 1982; Davies-Jones 1984, 2002; Droegemeier et al. 1993; Weisman and Rotunno 2000). Comparatively few studies, however, have examined the conditions and processes associated with supercell weakening and dissipation [two notable exceptions are Bluestein (2008) and Ziegler et al. (2010)]. An improved understanding of supercell demise would have implications for operational forecasting and may better inform our current understanding of supercell maintenance.
Observations of supercell dissipation are scarce (Bluestein 2008), but the recent second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2; Wurman et al. 2012) field campaign provided a unique dataset for 9 June 2009, with dense observations throughout the lifetime of an isolated supercell in south-central Kansas (including its demise). Exploring this dataset will help to bring into perspective the relationship between supercell storms and their environments, one of VORTEX2’s scientific areas of focus. An improved understanding of how supercells respond to and interact with their environments is key for anticipating storm evolution. Knowing whether a supercell will be maintained or dissipate is important for short-term forecasts, impacting warning issuance. More fundamentally, if we cannot predict supercell demise, then this implies that we do not have a complete understanding of the factors and processes that maintain supercells.
b. Background
Supercell thunderstorms are characterized by a single, dominant rotating updraft. The processes behind the generation of this rotating updraft have been well established in the literature (e.g., Rotunno 1981; Rotunno and Klemp 1982, 1985; Davies-Jones 1984; Brandes et al. 1988). Helical updraft rotation is partly a function of the extent to which the updraft is ingesting streamwise vorticity. From a vorticity perspective, this process occurs as a result of tilting, but it also benefits significantly from stretching (a process dependent on the strength of the updraft, which is a function of the instability of ingested parcels), since the vertical gradient in vertical velocity is positive up through the midlevels. This rotation of the supercell’s updraft has been thought to contribute to its longevity as a result of the suppression of turbulent dissipation (e.g., Lilly 1986; Brandes et al. 1988; Droegemeier et al. 1993), the separation of the updraft from the storm’s precipitation core (e.g., Browning 1964), as well as the favorable vertical accelerations that it promotes from perturbation pressure gradients (e.g., Rotunno and Klemp 1982). Accordingly, Bunkers et al. (2006b) found that long-lived supercells (at least 4 h in duration) tend to occur in environments with larger deep-layer1 shear.
Studies have also found that if a supercell is located near a preexisting boundary (and moving parallel to the boundary’s orientation), updraft rotation and storm longevity are enhanced as a consequence of mesoscale ascent, increased bulk shear, and the boundary’s solenoidal circulation acting to locally increase storm-relative helicity (SRH), a measure of the influx of horizontal streamwise vorticity (Lilly 1986; Atkins et al. 1999; Bunkers et al. 2006a,b). However, on the cool side of a preexisting boundary, the high-SRH air tends to be more stable and contains higher convective inhibition (CIN). Increases in low-level (i.e., 0–1 km) CIN are also usually present approaching nightfall as the boundary layer cools and stabilizes. If a surface-based storm moves into increasingly stable air, it may continue to ingest parcels from the stable layer and lift them to their levels of free convection (LFCs), it may become elevated and ingest parcels from above the stable layer, or it may dissipate. The processes that occur as the environment stabilizes and convection transitions from surface to elevated are a topic of ongoing research (e.g., Parker 2008; French and Parker 2010; Billings and Parker 2012). It is generally acknowledged that an increasingly stable environment (i.e., increasing CIN) is less favorable for supercell maintenance (e.g., Maddox et al. 1980; Doswell et al. 2002; Bunkers et al. 2006a,b), although our current understanding of the mechanisms that lift (or fail to lift) stable low-level parcels is fairly limited (e.g., Nowotarski et al. 2011).
Historically, supercells have been considered to contain two main downdraft regions: the forward-flank downdraft and the rear-flank downdraft (Lemon and Doswell 1979). The baroclinicity of the cold pools associated with these downdrafts provides a source of lifting of low-level parcels due to the presence of a density gradient along the leading edge of the cold pool. The strength of low-level lifting by cold pools has been hypothesized to be a function of the comparative balance of environmental vorticity generated from low-level shear perpendicular to the gust front and the baroclinically generated vorticity associated with the cold pool [Rotunno et al. (1988), also known as Rotunno–Klemp–Weisman (RKW) theory]. Within the context of squall-line maintenance and structure (Weisman et al. 1988), RKW theory states that if the circulation within the cold pool overwhelms the environmental shear, the updraft tilts rearward. With stronger line-perpendicular shear, the updraft tilts forward. If the strength of the cold pool roughly balances with the low-level line-perpendicular shear, then a more erect updraft is generated, allowing for a more direct path for parcels to reach their LFCs.
Although RKW theory has predominately been applied to understanding the maintenance and structure of squall lines, it could potentially be relevant to supercells. Specifically, if an optimal balance between the low-level shear and the cold pool was found beneath or near the main updraft, the supercell would likely benefit from parcels reaching their LFCs more directly, reducing entrainment, and thus generating a stronger updraft. Alternatively, an imbalance between the cold pool circulation and environmental vorticity near the main updraft would likely prevent some parcels from reaching their LFCs, thus weakening the updraft. However, the direct applicability of RKW theory is admittedly more complex for supercells, since the cold pool is not quasi-two-dimensional (as it is for squall lines), meaning that the balance of low-level shear and cold pool strength may vary significantly along the gust front. It is presumably the balance near the main updraft that is relevant to supercell maintenance.
Bluestein (2008) provides some observational evidence suggesting the pertinence of RKW theory in supercell maintenance. The study proposed a sequence of processes leading to the “downscale transition” of a supercell, based on photographic and mobile-radar observations of a limited number of cases of supercells with a cooling inflow environment. It was postulated that an increasingly stable environment reduces the baroclinic generation of horizontal vorticity along the rear-flank outflow of the supercell. Applying RKW theory within this context would be expected to promote a downshear-tilted updraft, which is indeed what was visually observed by Bluestein (2008).
A separate supercell dissipation study conducted by Ziegler et al. (2010), coming from a slightly different perspective, also concluded that the strength of the surface cold pool influenced storm maintenance. Quasi-idealized simulations of a supercell in an increasingly stable low-level environment demonstrated that weaker low-level lifting led to storm dissipation, though the simulations suggested that this was a result of a weaker forward-flank outflow temperature gradient that acted to suppress the main updraft. In these experiments, as the supercell encountered increasingly stable low levels, the forward-flank baroclinicity decreased, which consequently reduced lift along the boundary and in turn reduced updraft buoyancy as a result of fewer parcels sustaining the storm. A weaker updraft then resulted in less precipitation production, further diminishing cold pool baroclinicity, consequently resulting in stagnation or retrograde motion of the cold pool. Thus, a positive feedback loop was created whereby a weakening cold pool eventually led to the demise of the supercell.
The processes proposed by both Ziegler et al. (2010) and Bluestein (2008) are very similar in that a cooling environment leads to weaker low-level lifting as a result of reduced baroclinicity. Although the strength of low-level cold pool lifting appears to be important for supercell maintenance, supercells have traditionally been understood to be primarily driven by dynamic lifting produced by vertical perturbation pressure gradients (e.g., Rotunno and Klemp 1982, 1985), an aspect of supercell maintenance deemphasized by the aforementioned dissipation studies. Nowotarski et al. (2011) demonstrated that in a stable low-level environment, with a weak or negligible surface cold pool, supercell updraft parcels have origins from a variety of heights above the surface, including from within the near-surface stable layer. The authors argued that the dynamic vertical perturbation pressure gradient force, acting over a significant depth (at least the lowest 4 km) of the storm, primarily worked to lift parcels from the stable low levels, though any discernable cold pool at the surface could also play a role. Thus, the strength of deep dynamic lifting present in a supercell (i.e., a consequence of the presence of strong vertical wind shear) may be sufficient to maintain a supercell in an increasingly stable environment as cold pool lifting weakens, providing the vertical accelerations needed to lift higher-CIN parcels to their LFCs and/or to lift elevated parcels that have sufficient instability and small CIN.
The relative lack of studies on supercell maintenance and dissipation highlights a need to better understand the mechanisms and circumstances that promote or inhibit the lifting of parcels to their LFCs. What are the comparative contributions of low-level lifting by supercell outflow and lifting by the dynamic vertical perturbation pressure gradient force (a consequence of the storm developing in strong deep vertical wind shear) to the maintenance of a supercell? Under what circumstances will low-level lifting by the cold pool be more important than deep, storm-scale dynamic lifting and vice versa? How does an increasingly stable environment modify the contribution of the lifting mechanisms? These questions will be explored in this study through the examination of the evolution of an isolated supercell sampled by VORTEX2 on 9 June 2009. This case was chosen because of the breadth of data available throughout the storm’s lifetime, as well as the storm’s position on the cool side of a boundary, representing a scenario encompassing the largest fraction of shorter-lived supercells (e.g., Bunkers et al. 2006b). The present study serves as the foundation of a larger effort to determine the processes leading to dissipation on 9 June 2009 within the context of the storm interacting with the evolving inflow environment; a forthcoming study will use numerical experimentation to analyze the hypotheses described here. Thus, the purpose of this initial study is to document the evolution of the storm using the VORTEX2 observations and provide possible mechanisms leading to dissipation.
The next section details the instrumentation and analysis methods utilized in this study. Section 3 then focuses on the observations of the 9 June 2009 supercell, describing storm evolution within the context of the modifications in the local environment. Section 4 discusses the main findings and provides hypotheses explaining the relationship between the evolving environment and the dissipating supercell. These hypotheses provide the basis for modeling tests that will identify the relevant processes at work. Finally, section 5 provides a summary of the results and highlights areas of future work.
2. Analysis methods
Coordinated sampling of the 9 June 2009 supercell and its mesoscale environment was achieved with numerous VORTEX2 instrument platforms [for details, see Wurman et al. (2012)]. The environment surrounding the storm was sampled by radiosondes launched using the Mobile GPS Advanced Upper Air Sounding (MGAUS) system developed at the National Center for Atmospheric Research (NCAR). The MGAUS system utilized Vaisala RS92 radiosondes that measured pressure, temperature, and humidity, with winds computed from the GPS position of the sonde. Throughout the VORTEX2 campaign, four radiosondes were launched approximately once every hour (once a target storm was identified), distributed throughout the near- or far-inflow environment and the forward- and rear-flank regions (Wurman et al. 2012; Parker 2014). The present study of the 9 June 2009 supercell focused on the temporal variability of the inflow environment and how those changes impacted the evolution of the storm; thus, here we only examine the near-inflow soundings from this day. Soundings that did not reach the tropopause were completed using the nearest available sounding in time and space to achieve consistent calculations of integrated instability parameters, such as convective available potential energy (CAPE) and CIN. Quality control2 of the sounding data, including bias correction, was performed by staff at NCAR’s Earth Observing Laboratory (EOL).
In situ observations of the storm’s surface thermodynamic and kinematic fields were obtained via transects throughout the hook echo and inflow region by mobile mesonets (Straka et al. 1996) and StickNets (Weiss and Schroeder 2008). Each individual instrument underwent quality control, including bias correction based on data collected during a homogeneous period to ensure consistent measurements within each platform. Bias correction for the mobile mesonets was based on intercomparisons performed en route to a deployment in precipitation-free, well-aspirated conditions while the vehicles were traveling together at highway speeds. Bias correction for the StickNets was based on mass tests performed on days throughout the project when operations were not scheduled. For the StickNet bias correction applied to the 9 June 2009 data, the mass test performed on 15 June 2009 was utilized, since it was the mass test closest in time to the deployment of interest. Differences in the nature of the platforms (i.e., mobile versus stationary), as well as the instrument response times, provide a challenge in terms of direct comparability of the instruments, particularly near strong gradients. The time-to-space conversion technique is useful for visualizing mobile and stationary instrumentation (Skinner et al. 2010), though observations that overlap in space require further scrutiny of individual biases. The recommended method for achieving a useful analysis where the platforms overlap is to remove the platform that results in a noticeable discontinuity, indicating a larger spatial bias [based on subjective assessment; P. Skinner (2014, personal communication)], a technique applied in this study.
The radar data utilized for the purposes of examining the storm-scale evolution were collected by the two Shared Mobile Atmospheric Research and Teaching Radars (SMART-R1 and SMART-R2; Biggerstaff et al. 2005). These radars were operating utilizing a 5-cm wavelength (i.e., C band) and 1.5° beamwidth, and tended to be deployed 20–30 km south (direction dependent on storm motion) of the target storm to attain storm-scale data, resulting in a horizontal beamwidth between approximately 500 and 800 m. The dual-Doppler baseline for the deployment of interest was 28.4 km. Sampling occurred between 0.5° and 38° relative to the horizontal. Editing of the data to remove spurious echoes (e.g., ground clutter) as well as dealiasing of the velocity data was performed by staff at the National Severe Storms Laboratory (NSSL). Coordinated dual-Doppler scans were available from the SMART-Rs just prior to and during the dissipation of the 9 June 2009 supercell, from approximately 2345 to 0024 UTC. This period also represents the time frame within which the target storm experienced its most rapid weakening. Several other mobile radars sampled the 9 June 2009 supercell, including the Doppler on Wheels (DOW) radars (Wurman et al. 1997; Wurman 2001), a mobile phased-array weather radar [Meteorological Weather Radar X-band Phased Array (MWR-05XP); Bluestein et al. 2010], and the University of Massachusetts (UMass) dual-polarized X‐band mobile Doppler radar (X-Pol; Bluestein et al. 2007); these radars sampled the storm at a shorter range and smaller vertical depth than the SMART-Rs, often focusing on the low-level mesocyclone region. On 9 June 2009, sampling among these X-band radars was less well coordinated; as a result, there was insufficient dual-Doppler coverage (temporally and spatially), particularly when the most significant weakening occurred. This is the period of greatest interest to our study, so our analysis will focus on the storm-scale radar data collected by the SMART-Rs.
The SMART-R data were interpolated onto a 60 km × 60 km × 15 km Cartesian grid using a two-pass Barnes analysis scheme (Barnes 1964; Majcen et al. 2008), with a horizontal and vertical grid spacing of 0.5 km. The Barnes smoothing parameter κ was 1.9 km2, consistent with the recommendations of Pauley and Wu (1990) and Trapp and Doswell (2000). A between-pass convergence parameter of γ = 0.3 was also used [as recommended by Majcen et al. (2008)]. A correction for storm motion was also made to account for the time required to complete a volume scan (approximately 3 min), based on the observed motion of the mesocyclone. Once the storm significantly weakened such that it no longer contained a mesocyclone, the approximate center of the storm (centroid determined via radar reflectivity of the base scan) was tracked. A flat lower boundary was used for the dual-Doppler analysis domain, with the bottom of the grid defined to be the mean elevation of the two radars. The three-dimensional wind field was constructed using upward integration of mass continuity with a lower boundary condition of w = 0. Upward integration was utilized since we were most interested in velocities in the lower portion of the storm. Errors tend to accumulate with distance from the imposed boundary conditions; thus, integrating from the lower boundary upward minimizes error in the lower portion of the storm. No below-beam extrapolation of the radial velocity data was performed during the objective analysis stage to prevent synthesis of radial winds that would be extrapolated downward from different heights. During synthesis, downward extrapolation of velocity vectors from the lowest data-bearing level was allowed during the calculation of vertical velocity. These extrapolated winds were then discarded so that they were not included in the analysis (e.g., Markowski et al. 2011).
3. Observations
a. Preconvective environment and evolution
The supercell sampled by VORTEX2 on 9 June 2009 formed in south-central Kansas in an environment that underwent a somewhat complicated evolution prior to its development. Earlier in the day, convective storms in northern Kansas produced an outflow boundary that moved southward throughout the day, eventually merging with a northward-moving warm front, creating a quasi-stationary boundary by 1800 UTC (not shown). This boundary moved slowly southward in the late afternoon and then stalled (dashed line in Fig. 1). The superposition of the evolving boundary with an eastward-moving dryline (solid line in Fig. 1) served as a focal point for convective initiation, resulting in storm development around 2140 UTC (see box in Fig. 1c).
Regional observations near the time of storm initiation on 9 Jun 2009, including (a) subjectively analyzed temperature (°F) and (b) dewpoint contours (°F), as well as (c) visible satellite imagery. The solid and dashed black lines represent the dryline and outflow boundary, respectively, while the box in (c) indicates the storm of interest.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
The initial storm appeared to form just to the cool side of the east–west thermal boundary (Fig. 1c) and quickly split into left- and right-moving supercells (Fig. 2a). The right-moving supercell became the VORTEX2 target storm and rapidly developed strong low-level rotation (indicated in the reflectivity field as a hook echo) as it passed over the Dodge City, Kansas, WSR-88D around 2300 UTC (Figs. 2b and 3a). The rapid maturation of the right-moving supercell is consistent with previous studies demonstrating that supercells traveling along a preexisting boundary ingest larger amounts of SRH, which acts to enhance updraft rotation and more quickly organize the storm (e.g., Atkins et al. 1999; Bunkers et al. 2006b). However, the supercell was not observed to be long lived (Figs. 2 and 3), unlike many other supercells located near boundaries (e.g., Bunkers et al. 2006b; Ziegler et al. 2010). Examining the evolution of the combined outflow and east–west thermal boundary relative to the movement of the supercell illustrates that the storm did not remain anchored to the boundary (Fig. 4). The position of the boundary was determined via tracking its associated radar reflectivity fineline (e.g., as shown in Fig. 2) and radial velocity gradient, since the conventionally available surface observations are relatively sparse in south-central Kansas. Over a period of 4 h, the boundary moved approximately 30 km toward the south-southwest, while the supercell primarily tracked eastward, thus moving deeper into the cool air located north of the boundary (Figs. 1 and 4). Over time, as the storm progressed farther into the cool air, it developed a laminar cloud base with an increasingly narrow and downshear-tilted updraft [as in Bluestein (2008); Fig. 3b], completely dissipating within a few short hours of its birth (Figs. 2 and 3c,d).
Base reflectivity from the Dodge City WSR-88D [denoted by the black triangle in (a),(b)] depicting the evolution of the storm over time in relation to the launches of the near-inflow soundings [denoted by black squares in (b)–(d), starting at 2319 UTC]. The dashed lines represent the different boundaries discussed in the text, as detected by the radar finelines, with the supercell’s RFGF, the combined outflow boundary–synoptic front (“front”), and the dryline labeled appropriately.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
Photographs of the VORTEX2 target storm at (a) 2317, (b) 0009, (c) 0026, and (d) 0047 UTC 9 Jun 2009. [Photographs were taken by D. Dowell (NOAA/EOL) and were obtained from the VORTEX2 field catalog.]
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
Evolution of the radar fineline (from the Dodge City WSR-88D) associated with the combined outflow boundary–synoptic front on 9 Jun 2009. Colored squares represent the location of the low-level mesocyclone or (as the storm weakened) approximate storm centroid. The S1, S2, and S3 notations indicate the launch positions of the near-inflow soundings.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
b. Storm evolution
Figures 5 and 6 illustrate the evolution of the storm during the period of SMART-R1 (SR1)/SMART-R2 (SR2) dual-Doppler coverage, representing the first such analysis of a dissipating supercell. First, we review the storm’s fundamental structures, after which we discuss its relationship to the regional changes in the environment.
Horizontal cross sections of radar reflectivity (shaded), storm-relative wind vectors, and dual-Doppler-derived vertical velocity (contoured in white at 5 m s−1) at 1 km AGL from 2345 to 0021 UTC 9–10 Jun 2009. The x and y axes represent distances (km) from the center of the grid (the position of SMART-R1).
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
As in Fig. 5, but for horizontal cross sections at 5 km AGL. Vertical velocity at 10 m s−1 is contoured in white.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
At 1 km AGL, significant rotation was evident in the storm-relative wind field near the main updraft near the start of the dual-Doppler window, between 2345 and 2351 UTC, coinciding with a report of a brief tornado touchdown outside of Greensburg, Kansas (Fig. 5). Soon thereafter, starting around 2357 UTC, the storm-relative winds near the main updraft no longer indicated significant rotation and the overall low-level reflectivity began to decrease. By 0009 UTC, low-level rotation was virtually nonexistent, and the updraft, while still present (Fig. 5), had shifted to a more laminar appearance (Fig. 3b). This indicates that a significant shift in internal storm processes occurred within a short time frame. Interestingly, the size of the 1-km updraft (as indicated by the 5 m s−1 vertical velocity contour) did not change appreciably during the latter half of the dual-Doppler window as the overall storm weakened significantly. This suggests that the storm was able to continue lifting some low-level parcels (presumably more stable ones; Fig. 3b) into the updraft, even as it moved deeper into the cool air away from the boundary (Fig. 4), and as the rotation and reflectivity decreased (Fig. 5).
At 5 km AGL, rotation is indicated by the cyclonic shearing and slight curvature of the storm-relative wind field near the main updraft (the strength of this rotation will be confirmed shortly with dual-Doppler-derived vorticity; Fig. 6). By 0009 UTC, the storm-relative wind field near the updraft is nearly uniform in its direction, qualitatively indicating that updraft rotation weakened very quickly (less than 30 min from the start of the dual-Doppler period). After 2351 UTC, and especially after 0003 UTC, a notable decline in reflectivity is evident, which also corresponds to a rapid shrinking of the main updraft (outlined by the 10 m s−1 contour). In fact, by 0021 UTC, there are very few areas where the vertical velocity exceeds 10 m s−1 (Fig. 6). The rapidly shrinking midlevel updraft contrasts with the evolution at 1 km, where the updraft area was relatively unchanged. This suggests that the vertical profile of vertical accelerations was changing, which must be fundamentally tied to either parcel buoyancy or to evolving dynamical accelerations. Additional analysis is needed to identify the specific shifts in those mechanisms.
Figure 7 provides corresponding information for Figs. 5 and 6, quantifying the evolution in the updraft area and vertical vorticity area exceeding mesocyclone strength (i.e., 10−2 s−1) at 1 and 5 km AGL. Updraft helicity (Kain et al. 2008) is also shown, which represents the vertical integral of the product of vertical velocity and vertical vorticity.3 At 1 km, the updraft retained a similar vertical velocity after 0000 UTC, while vertical vorticity and updraft helicity continued their rapid decline (consistent with the qualitative evaluation of the storm-relative wind vectors; Fig. 7). In other words, the storm appears to have been lifting parcels to their LFCs and sustaining the low-level updraft, but these parcels did not continue to produce appreciable vertical vorticity. At 5 km, the parallel decreases in vertical vorticity and updraft helicity corresponded to the rapid shrinking of the updraft, suggesting that the magnitude of vertical acceleration was weakening over time (Fig. 7). The decline in updraft helicity would by definition be linked to the observed decreases in both the updraft vertical vorticity and vertical velocity.
Time series of the areal extent (km2) of the updraft (defined as dual-Doppler-derived vertical velocities at 5 km >10 m s−1), updraft vertical vorticity (ζ > 10−2 s−1 at 5 km), and updraft helicity [values > 250 m2 s−2, calculated in the manner of Kain et al. (2008) in the 2–5-km layer].
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
In terms of the individual contributions to the production of vertical vorticity, slightly different trends were observed. There was an overall decreasing trend in dual-Doppler estimates of the stretching term (Fig. 8a), corresponding to decreases in vertical velocity, which were larger aloft than in the low levels (Figs. 7 and 8c). In contrast, the tilting term (which is linked to the horizontal, rather than the vertical, gradient in vertical velocity) experienced slight overall decreases, but remained comparatively steady over time (Fig. 8b). Admittedly, the vorticity budget terms are noisy because of the uncertainty present in the dual-Doppler analysis, which is compounded by taking additional derivatives to compute the vorticity budget. Even so, the comparison of overall trends suggests that the decline in stretching was critical to the observed weakening of the midlevel mesocyclone. The picture that emerges is one in which the evolving environment leads to a comparatively weaker updraft in the midlevels, which in turn produces a much weaker midlevel mesocyclone. The weakened midlevel rotation could then impact the updraft strength by decreasing the upward dynamical accelerations. Modifications in the storms’ local environment and their contribution to the observed evolution are described in the next section.
Time series of the area-averaged (a) stretching term, (b) tilting term (10−4 s−2) within the updraft (w > 10 m s−1), and the (c) max updraft value at several heights.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
c. Environmental evolution
The 9 June 2009 supercell was observed to rapidly weaken over time, which was accompanied by significant decreases in dual-Doppler-derived vertical velocities, vertical vorticity, and updraft helicity. These changes occurred as the storm moved away from the surface boundary, farther into a cool air mass (e.g., Figs. 1 and 4). To determine how the environmental stability and vertical shear in the cold air mass may have contributed to the observed storm evolution, we next examine the inflow soundings sampled over the lifetime of the storm.
The evolution in the storm’s near-inflow environment was captured with three soundings launched throughout the lifetime of the supercell at 2319, 2354, and 0056 UTC (Figs. 2 and 9). Each sounding indicates a well-mixed elevated residual layer (ERL) of dry air from ~850 to 500 hPa, as well as a midlevel moist layer (near 450 hPa at 2319 and 2354 UTC, and 350 hPa at 0056 UTC). The 2354 UTC sounding notably contained a sharp low-level inversion below the ERL, which subsequently descended and became more intense by 0056 UTC as low-level temperatures continued to cool (Figs. 9b,c). A midlevel temperature inversion was also present, which appeared to descend over time from ~430 hPa at 2319 UTC to ~550 hPa at 0056 UTC. Such a feature, while not necessarily preventing convection, would reduce its intensity as a result of a reduction of CAPE associated with the elevated warm layer and broad descent.
Skew T–log p (i.e., skewed temperature, logarithmic pressure) diagrams of the soundings sampling the near-inflow environment during 9–10 Jun 2009. The time of the launch and the storm’s degree of maturity are indicated. The purple horizontal line in (a) denotes the starting level where the sounding was modified using data from the 2354 UTC sounding to have data up through the tropopause. The turquoise and orange dashed lines indicate the surface and most unstable parcel paths, respectively.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
Given that the low-level inflow environment was presumably undergoing rapid changes as the supercell moved deeper into the cool air (Figs. 2 and 4), we evaluated the shift in stability parameters throughout the lower troposphere to determine whether a shift from surface-based to elevated convection was possible. Using parcels originating from each individual level in the observed soundings, we constructed vertical profiles of CAPE, CIN, and Δz (defined as the vertical distance between a parcel’s starting height and its level of free convection, which was used as a proxy for the amount of lifting required for convection). In other words, each individual value of CAPE, CIN, and Δz in the vertical profile represents the integrated value for a parcel originating at that altitude. Thus, the profiles serve to illustrate modifications to the thermodynamic environment throughout the depth of the layer of potentially buoyant inflow parcels.
CAPE remained fairly consistent throughout the lifetime of the supercell, with values supportive of deep, moist convection up through nearly 2 km AGL (Fig. 10). In contrast, more meaningful modifications were present in the CIN profiles. As expected, the effects of diurnal cooling and the increasingly intense capping inversion (Fig. 9) resulted in inhibition notably increasing over time in the lowest 0.75 km AGL layer. This evolution is consistent with the observed progression of the storm (e.g., Fig. 5) and its movement deeper into the cool air away from the surface boundary (Fig. 4). CIN below 0.75 km AGL significantly increased after 2354 UTC, near the time of most significant storm weakening (e.g., Fig. 7) as well as a meaningful shift in the visual characteristics of the updraft (Fig. 3b). While an increase in low-level CIN was expected, the 0.75–1.5 km AGL layer contained CIN that decreased over time (tied to the moistening near the base of the ERL; Fig. 10), an observation that was somewhat unexpected, given the storm’s evolution. The combination of decreasing CIN between 0.75 and 1.5 km AGL and sufficient CAPE for convection present above 0.75 km AGL suggests an elevated environment favorable for convective maintenance; this occurred as the near-surface layer simultaneously became increasingly stable and less favorable. Additionally, the amount of lifting required for the 0.75–1.5 km AGL parcels to reach their levels of free convection (i.e., Δz in Fig. 10) was virtually unchanged during the storm’s lifetime. A logical question to ask is why the supercell was not sustained by the elevated parcels in the 0.75–1.5 km AGL layer, even as the low levels cooled and stability increased.
Vertical profiles of CAPE (J kg−1), CIN (J kg−1), and Δz (km AGL) over time from the near-inflow soundings on 9 Jun 2009.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
The idealized simulations of Nowotarski et al. (2011), performed in a more moist background environment, have shown that supercells’ dynamic lifting is quite capable of sustaining the main updraft with elevated (and even stable near surface) parcels. In this case, the presence of the dry ERL air within the elevated environment above the strengthening inversion (Fig. 9) likely played a role in suppressing elevated convection through entrainment, which has been shown to reduce buoyancy and weaken convection, even in mature supercells (e.g., James and Markowski 2010). While the ERL was not exceptionally dry in this case, entrainment of the dry air would nevertheless have a deleterious impact on the strength of the storm. Ongoing analysis of idealized simulations of this case study suggests that the dry elevated air was indeed detrimental to the prolonged maintenance of the supercell.
The observed increases in low-level CIN may very well explain much of the observed evolution of the target storm, but the movement of the storm deeper into the cool air and farther away from the surface boundary (Fig. 4) was also associated with changes in the near-storm wind profile. Figure 11 demonstrates low-level (0–0.5 km) winds backing over time, or becoming more northerly, resulting in the hairpin shape of the hodograph at 2354 and 0056 UTC, as well as a straightening of the hodograph in the midlevels. Shear and helicity parameters derived from the inflow soundings and the observed storm motion (estimated from the Dodge City WSR-88D) summarize the overall evolution of the wind profile. Slight increases in 0–0.5-, 0–1-, and 0–3-km bulk shear vector magnitudes were observed, as were decreases in 0–6-km bulk shear and 0–1- and 0–3-km storm-relative helicity (Table 1). For comparison, helicity parameters were also computed based on the Bunkers et al. (2000) storm motion estimate; the broad evolution was similar to the observed values, but more notable differences were present at 2354 UTC, where the Bunkers storm motion estimate was faster and farther to the right, resulting in slight increases in helicity (not shown).
Hodographs from the 9 Jun 2009 near-inflow soundings, with markers placed every 0.5 km, up to 10 km. Select height levels (km AGL) are labeled for reference. Storm motions are indicated by the crosses and enumerated on the bottom left.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
Table of shear and helicity parameters for the observed near-inflow soundings on 9 Jun 2009, where shear represents the bulk shear vector magnitude. The effective layer depth and SRH were defined as in Thompson et al. (2007), while effective shear is the bulk shear vector magnitude difference from the effective inflow base up to halfway to the equilibrium level. Storm-relative parameters were calculated based on storm motions estimated by the Dodge City WSR-88D tracking algorithm. Gust-front-perpendicular shear was calculated based on a mean 30° orientation of the RFGF.
In addition to shear and helicity parameters calculated over standard layers, we also examined effective-layer parameters, which are based on the (somewhat arbitrary) assumption that only parcels with CAPE ≥ 100 J kg−1 and CIN ≥ −250 J kg−1 actively sustain convective updrafts (Thompson et al. 2007). The depth of this layer remained consistent during the mature and weakening stages (2319 and 2354 UTC) of the storm, but decreased slightly at the dissipated stage (0056 UTC). Similarly, effective shear and helicity showed strong decreases in the last inflow observation (Table 1).
The overall weakening trends in SRH and deep-layer shear would indicate a reduction in horizontal vorticity (including streamwise vorticity) available for tilting. Based upon our dual-Doppler vorticity budget computations (Fig. 8), the changes in tilting appeared to be comparatively small. However, as indicated by the simulations of Nowotarski et al. (2011), it is also clear that the dynamic lifting associated with a supercell’s mesocyclone and updraft–shear interactions can be significant to the maintenance of a storm in an inhospitable environment. We therefore acknowledge the possibility that the weakening deep-layer shear and SRH in some way hastened the storm’s demise. Given that the transition from surface-based to elevated convection is not well understood (e.g., Parker 2008), this evolution in shear and helicity provides a potential explanation for why the storm was unable to persist by ingesting the parcels with sufficient CAPE and small CIN from the 0.75 to 1.5 km AGL layer. We will discuss the range of possible mechanisms within the context of previous work, as well as an ongoing companion modeling study, in the next section.
In addition to the potential contributions of weakening deep-layer shear and SRH to storm demise, we also postulate that changes in cold pool–shear interactions (via the application of RKW theory) were pertinent to the demise. As the storm weakened, it is evident that the 0–3-km shear vector lengthened and became increasingly orthogonal to the rear-flank gust front (RFGF; Fig. 12). Quantitatively, the 0–3-km shear perpendicular to the gust front jumped upward from 8.0 m s−1 at 2354 UTC to 13.1 m s−1 at 0056 UTC (Table 1). Within the context of RKW theory, increases in low-level shear orthogonal to the cold pool would promote a downshear-tilted updraft, consistent with visual observations of the 9 June 2009 case [as well as the cases discussed in Bluestein (2008); Fig. 3].
Evolution of the position of the RFGF over time during 9–10 Jun 2009. The x and y axes represent distance (km) from the center of the grid (which is the position of SMART-R1). The purple (green) vector approximates the 0–3-km shear vector at 2354 (0056) UTC, with the black vector in the bottom-right corner shown for scale.
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
As 0–3-km shear increased, the rear-flank outflow temperature gradient weakened over time. Virtual potential temperatures within the cold pool remained near 302–303 K, but the near-inflow sector 
Plan view of SMART-R reflectivity at (a) 2354, (b) 0003, and (c) 0012 UTC 9–10 Jun 2009. Time-to-space-converted mobile mesonet and StickNet tracks are overlaid. Each track is 3 min in length. Virtual potential temperature (K) is subjectively analyzed and contoured. The x and y axes represent distance (km) from the center of the grid (the position of SMART-R1).
Citation: Weather and Forecasting 30, 2; 10.1175/WAF-D-14-00087.1
4. Discussion
To our knowledge, this examination of the 9 June 2009 VORTEX2 target storm represents the first such analysis of a dissipating supercell. Although the increase in CIN in the lowest 0.75 km AGL provides an obvious catalyst for the observed weakening, an intriguing hypothesis is that the supercell would not have dissipated solely in response to the near-surface stabilization that it experienced. In other words, the changes in SRH and shear were also necessary contributions to the demise. While the thermodynamic changes were undoubtedly important to the storm’s weakening, the fact that the storm dissipated in a favorable elevated environment (between 0.75 and 1.5 km AGL; Fig. 10) suggests that other mechanisms may have contributed to storm weakening. It is well understood that the environmental wind profile influences storm organization and the internal dynamics; both low-level lifting by the cold pool (e.g., Rotunno et al. 1988; Moncrieff and Liu 1999) and deep, storm-scale dynamic lifting within a supercell (e.g., Weisman and Klemp 1982, 1984; Rotunno and Klemp 1982; Klemp 1987) are known to be strongly sensitive to the environmental vertical wind shear.
It is well known that spatial variations in environmental shear are commonly found in convective storm environments (e.g., Brooks et al. 1996; Markowski and Richardson 2007; Parker 2014) and subsequently impact storm development and evolution (e.g., Richardson et al. 2007). We speculate that in a case of a supercell with an increasingly inhospitable thermodynamic environment (e.g., the documented increases in low-level CIN and the presence of dry air aloft on 9 June 2009), the storm may become increasingly reliant upon upward dynamical accelerations. This is consistent with the findings from the low-CAPE simulations of McCaul and Weisman (1996) and the stable-layer simulations of Nowotarski et al. (2011).
Given this context, we hypothesize that the changing near-storm wind profile may have been harmful to the 9 June 2009 supercell for the following reasons: 1) decreases in the influx of horizontal streamwise vorticity (i.e., SRH) lead to weaker updraft rotation and thus weaker nonlinear dynamic updraft forcing (e.g., Rotunno and Klemp 1982), 2) decreases in the bulk deep-layer vertical wind shear also lead to decreases in dynamical lifting associated with linear updraft forcing [a consequence of the updraft’s existence in a sheared environment; Rotunno and Klemp (1982)], and 3) increases in low-level shear perpendicular to the rear-flank cold pool, coincident with a decrease in the strength of the cold pool, weakened low-level lifting as a consequence of modified cold pool–shear interactions (Rotunno et al. 1988; Bluestein 2008). The available observations do not fully permit quantitative assessment of these hypotheses. In a forthcoming companion article, we will present idealized numerical simulations supporting the important contributions of hypothesized process 1 (in addition to the primary role of increasing CIN).
The dissipation of the storm coincided neatly with the observed increase in low-level CIN (Figs. 2 and 10), suggesting that these modifications played a significant role in the storm’s subsequent evolution. Although this finding is in line with what might be expected of dissipating convection, the favorable elevated environment that apparently did not sustain the storm is intriguing, and prompts additional questions. For example, how strong would dynamic lifting need to be in a supercell to overcome the observed increases in CIN [e.g., as observed in the simulations of Nowotarski et al. (2011)]? And to what extent would the 9 June 2009 storm have been able to sustain itself on the more favorable elevated parcels containing less CIN (i.e., the 0.75–1.5 km AGL layer in Fig. 10)? Again, these questions will be addressed in a follow-up study through the use of controlled hypothesis tests in a numerical model. The fact that the 9 June 2009 case is one of the best-observed examples of supercell demise provides a strong constraint on these experiments.
5. Summary and future work
The dense observations of the 9 June 2009 supercell provided a unique opportunity to explore the physical processes associated with storm dissipation. The storm was observed to rapidly weaken and dissipate as its low-level inflow environment cooled and stabilized (Fig. 10), while bulk shear and helicity simultaneously weakened (Table 1). The storm was initiated along a preexisting thermal boundary (Fig. 1), with the subsequent motion of both the boundary and the storm placing the supercell deeper into the stable air mass (Fig. 4), which further cooled over time because of diurnal effects and weak cold-air advection. Large-scale sinking motion was also suggested by the presence of a descending midlevel inversion (Fig. 9). As a consequence of this evolving environment, inflow CIN within the lowest 0.75 km AGL strongly increased (Fig. 10). Additionally, the cooling low levels resulted in a weaker temperature gradient across the rear-flank outflow (Fig. 13), which would make it more difficult to lift low-level parcels to their levels of free convection and overcome their increasing convective inhibition [e.g., the mechanism proposed by Bluestein (2008) and Ziegler et al. (2010)]. Notably, the inflow environment contained elevated parcels in the 0.75–1.5-km layer with less CIN and sufficient CAPE for convection; however, they were also drier than the low-level parcels (Figs. 9 and 10), likely preventing the storm from becoming elevated because of the negative impacts of entrainment (e.g., James and Markowski 2010).
It is also hypothesized that a diminished ability to lift the increasingly stable low-level parcels may be partly attributed to the evolution in the wind profile. Decreases in 0–6-km bulk shear and storm-relative helicity might have impacted updraft forcing as a result of changes in linear and nonlinear dynamical perturbation pressure effects (Rotunno and Klemp 1982), while modifications to low-level shear could have modified gust front lifting due to cold pool–shear interactions (Rotunno et al. 1988; Bluestein 2008; Table 1; Fig. 12). The available evidence from dual-Doppler syntheses was not conclusive, but suggested that vertical vorticity weakened primarily as a result of decreases in the stretching term in the vorticity equation, though slight decreases in the tilting term were also observed (Fig. 8). Simple weakening of the updraft from the aforementioned thermodynamic changes would produce this decrease in stretching, after which there may be nonlinear feedbacks associated with the diminished midlevel updraft rotation.
Many of our findings are rather straightforward, but the simultaneous evolution of the near-storm wind profile also provides for some intriguing complementary hypotheses. The comparative roles of the wind profile versus the increasingly stable inflow environment in storm dissipation are difficult to determine using only the observations. These questions will be addressed in a forthcoming companion paper using a suite of idealized sensitivity experiments incorporating the varying inflow environment observed on 9 June 2009. The long-range goal of this work is to advance the understanding of how supercells are (or are not) maintained in increasingly inhospitable environments.
Acknowledgments
Funding for this research was provided by NSF Grants AGS-0758509 and AGS-1156123. The authors thank Dr. Ananatha Aiyyer, Dr. Gary Lackmann, Dr. Sandra Yuter, and Dr. Conrad Ziegler for their valuable input throughout this study. NSSL graciously provided the edited radar data. Dual-Doppler syntheses were constructed using code supplied by David Dowell. The authors also acknowledge constructive feedback provided by three anonymous reviewers that greatly improved this manuscript.
REFERENCES
Atkins, N. T., Weisman M. L. , and Wicker L. J. , 1999: The influence of preexisting boundaries on supercell evolution. Mon. Wea. Rev., 127, 2910–2927, doi:10.1175/1520-0493(1999)127<2910:TIOPBO>2.0.CO;2.
Barnes, S. L., 1964: A technique for maximizing details in numerical weather map analysis. J. Appl. Meteor., 3, 396–409, doi:10.1175/1520-0450(1964)003<0396:ATFMDI>2.0.CO;2.
Biggerstaff, M. I., and Coauthors, 2005: The Shared Mobile Atmospheric Research and Teaching Radar: A collaboration to enhance research and teaching. Bull. Amer. Meteor. Soc., 86, 1263–1274, doi:10.1175/BAMS-86-9-1263.
Billings, J. M., and Parker M. D. , 2012: Evolution and maintenance of the 22–23 June 2003 nocturnal convection during BAMEX. Wea. Forecasting, 27, 279–300, doi:10.1175/WAF-D-11-00056.1.
Bluestein, H. B., 2008: On the decay of supercells through a “downscale transition”: Visual documentation. Mon. Wea. Rev., 136, 4013–4028, doi:10.1175/2008MWR2358.1.
Bluestein, H. B., French M. M. , Tanamachi R. L. , Frasier S. , Hardwick K. , Junyent F. , and Pazmany A. L. , 2007: Close-range observations of tornadoes in supercells made with a dual-polarization, X-band, mobile Doppler radar. Mon. Wea. Rev., 135, 1522–1543, doi:10.1175/MWR3349.1.
Bluestein, H. B., French M. M. , PopStefanija I. , Bluth R. T. , and Knorr J. B. , 2010: A mobile, phased-array Doppler radar for the study of severe convective storms. Bull. Amer. Meteor. Soc., 91, 579–600, doi:10.1175/2009BAMS2914.1.
Brandes, E. A., Davies-Jones R. P. , and Johnson B. C. , 1988: Streamwise vorticity effects on supercell morphology and persistence. J. Atmos. Sci., 45, 947–963, doi:10.1175/1520-0469(1988)045<0947:SVEOSM>2.0.CO;2.
Brooks, H. E., Doswell C. A. III, Carr M. T. , and Ruthford J. E. , 1996: Preliminary analysis of VORTEX-95 soundings. Preprints, 18th Conf. on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 133–136.
Browning, K. A., 1964: Airflow and precipitation trajectories within severe local storms which travel to the right of the mean wind. J. Atmos. Sci., 21, 634–639, doi:10.1175/1520-0469(1964)021<0634:AAPTWS>2.0.CO;2.
Bunkers, M. J., Klimowski B. A. , Zeitler J. W. , Thompson R. L. , and Weisman M. L. , 2000: Predicting supercell motion using a new hodograph technique. Wea. Forecasting, 15, 61–79, doi:10.1175/1520-0434(2000)015<0061:PSMUAN>2.0.CO;2.
Bunkers, M. J., Hjelmfelt M. R. , and Smith P. L. , 2006a: An observational examination of long-lived supercells. Part I: Characteristics, evolution, and demise. Wea. Forecasting, 21, 673–688, doi:10.1175/WAF949.1.
Bunkers, M. J., Hjelmfelt M. R. , and Smith P. L. , 2006b: An observational examination of long-lived supercells. Part II: Environmental conditions and forecasting. Wea. Forecasting, 21, 689–714, doi:10.1175/WAF952.1.
Davies-Jones, R., 1984: Streamwise vorticity: The origin of updraft rotation in supercell storms. J. Atmos. Sci., 41, 2991–3006, doi:10.1175/1520-0469(1984)041<2991:SVTOOU>2.0.CO;2.
Davies-Jones, R., 2002: Linear and nonlinear propagation of supercell storms. J. Atmos. Sci., 59, 3178–3205, doi:10.1175/1520-0469(2003)059<3178:LANPOS>2.0.CO;2.
Doswell, C. A., III, Baker D. V. , and Liles C. A. , 2002: Recognition of negative mesoscale factors for severe-weather potential: A case study. Wea. Forecasting, 17, 937–954, doi:10.1175/1520-0434(2002)017<0937:RONMFF>2.0.CO;2.
Droegemeier, K. K., Lazarus S. M. , and Davies-Jones R. , 1993: The influence of helicity on numerically simulated convective storms. Mon. Wea. Rev., 121, 2005–2029, doi:10.1175/1520-0493(1993)121<2005:TIOHON>2.0.CO;2.
French, A. J., and Parker M. D. , 2010: The response of simulated nocturnal convective systems to a developing low-level jet. J. Atmos. Sci., 67, 3384–3408, doi:10.1175/2010JAS3329.1.
James, R. P., and Markowski P. M. , 2010: A numerical investigation of the effects of dry air aloft on deep convection. Mon. Wea. Rev., 138, 140–161, doi:10.1175/2009MWR3018.1.
Kain, J. S., and Coauthors, 2008: Some practical considerations regarding horizontal resolution in the first generation of operational convection-allowing NWP. Wea. Forecasting, 23, 931–952, doi:10.1175/WAF2007106.1.
Klemp, J. B., 1987: Dynamics of tornadic thunderstorms. Annu. Rev. Fluid Mech., 19, 369–402, doi:10.1146/annurev.fl.19.010187.002101.
Klemp, J. B., Wilhelmson R. B. , and Ray P. S. , 1981: Observed and numerically simulated structure of a mature supercell thunderstorm. J. Atmos. Sci., 38, 1558–1580, doi:10.1175/1520-0469(1981)038<1558:OANSSO>2.0.CO;2.
Lemon, L. R., and Doswell C. A. III, 1979: Severe thunderstorm evolution and mesocyclone structure as related to tornadogenesis. Mon. Wea. Rev., 107, 1184–1197, doi:10.1175/1520-0493(1979)107<1184:STEAMS>2.0.CO;2.
Lilly, D. K., 1986: The structure, energetics and propagation of rotating convective storms. Part II: Helicity and storm stabilization. J. Atmos. Sci., 43, 126–140, doi:10.1175/1520-0469(1986)043<0126:TSEAPO>2.0.CO;2.
Maddox, R. A., Hoxit L. R. , and Chappell C. F. , 1980: A study of tornadic thunderstorm interactions with thermal boundaries. Mon. Wea. Rev., 108, 322–336, doi:10.1175/1520-0493(1980)108<0322:ASOTTI>2.0.CO;2.
Majcen, M., Markowski P. , Richardson Y. , Dowell D. , and Wurman J. , 2008: Multipass objective analyses of Doppler radar data. J. Atmos. Oceanic Technol., 25, 1845–1858, doi:10.1175/2008JTECHA1089.1.
Markowski, P., and Richardson Y. , 2007: Observations of vertical wind shear heterogeneity in convective boundary layers. Mon. Wea. Rev., 135, 843–861, doi:10.1175/MWR3334.1.
Markowski, P., and Richardson Y. , 2010: Mesoscale Meteorology in Midlatitudes.Wiley, 430 pp.
Markowski, P., Majcen M. , Richardson Y. , Marquis J. , and Wurman J. , 2011: Characteristics of the wind field in three nontornadic low-level mesocyclones observed by the Doppler on Wheels radars. Electron. J. Severe Storms Meteor., 6 (3). [Available online at http://www.ejssm.org/ojs/index.php/ejssm/article/viewArticle/75.]
McCaul, E. W., and Weisman M. L. , 1996: Simulations of shallow supercell storms in landfalling hurricane environments. Mon. Wea. Rev., 124, 408–429, doi:10.1175/1520-0493(1996)124<0408:SOSSSI>2.0.CO;2.
Moncrieff, M. W., and Liu C. , 1999: Convection initiation by density currents: Role of convergence, shear, and dynamical organization. Mon. Wea. Rev., 127, 2455–2464, doi:10.1175/1520-0493(1999)127<2455:CIBDCR>2.0.CO;2.
Nowotarski, C. J., Markowski P. M. , and Richardson Y. P. , 2011: The characteristics of numerically simulated supercell storms situated over statically stable boundary layers. Mon. Wea. Rev., 139, 3139–3162, doi:10.1175/MWR-D-10-05087.1.
Parker, M. D., 2008: Response of simulated convective squall lines to low-level cooling. J. Atmos. Sci., 65, 1323–1341, doi:10.1175/2007JAS2507.1.
Parker, M. D., 2014: Composite VORTEX2 supercell environments from near-storm soundings. Mon. Wea. Rev., 142, 508–529, doi:10.1175/MWR-D-13-00167.1.
Pauley, P. M., and Wu X. , 1990: The theoretical, discrete, and actual response of the Barnes objective analysis scheme for one- and two-dimensional fields. Mon. Wea. Rev., 118, 1145–1163, doi:10.1175/1520-0493(1990)118<1145:TTDAAR>2.0.CO;2.
Richardson, Y. P., Droegemeier K. K. , and Davies-Jones R. P. , 2007: The influence of horizontal environmental variability on numerically simulated convective storms. Part I: Variations in vertical shear. Mon. Wea. Rev., 135, 3429–3455, doi:10.1175/MWR3463.1.
Rotunno, R., 1981: On the evolution of thunderstorm rotation. Mon. Wea. Rev., 109, 577–586, doi:10.1175/1520-0493(1981)109<0577:OTEOTR>2.0.CO;2.
Rotunno, R., and Klemp J. B. , 1982: The influence of the shear-induced pressure gradient on thunderstorm motion. Mon. Wea. Rev., 110, 136–151, doi:10.1175/1520-0493(1982)110<0136:TIOTSI>2.0.CO;2.
Rotunno, R., and Klemp J. B. , 1985: On the rotation and propagation of simulated supercell storms. J. Atmos. Sci., 42, 271–292, doi:10.1175/1520-0469(1985)042<0271:OTRAPO>2.0.CO;2.
Rotunno, R., Klemp J. B. , and Weisman M. L. , 1988: A theory for strong, long-lived squall lines. J. Atmos. Sci., 45, 463–485, doi:10.1175/1520-0469(1988)045<0463:ATFSLL>2.0.CO;2.
Skinner, P. S., Weiss C. C. , Richardson Y. P. , and Markowski P. M. , 2010: Intercomparison between mobile and stationary surface observing platforms in VORTEX2. Preprints, 25th Conf. of Severe Local Storms, Denver, CO, Amer. Meteor. Soc., P5.1. [Available online at https://ams.confex.com/ams/pdfpapers/176245.pdf.]
Straka, J. M., Rasmussen E. N. , and Fredrickson S. E. , 1996: A mobile mesonet for finescale meteorological observations. J. Atmos. Oceanic Technol., 13, 921–936, doi:10.1175/1520-0426(1996)013<0921:AMMFFM>2.0.CO;2.
Thompson, R. L., Mead C. M. , and Edwards R. , 2007: Effective storm-relative helicity and bulk shear in supercell thunderstorm environments. Wea. Forecasting, 22, 102–115, doi:10.1175/WAF969.1.
Trapp, R. J., and Doswell C. A. III, 2000: Radar data objective analysis. J. Atmos. Oceanic Technol., 17, 105–120, doi:10.1175/1520-0426(2000)017<0105:RDOA>2.0.CO;2.
Weisman, M. L., and Klemp J. B. , 1982: The dependence of numerically simulated convective storms on vertical wind shear and buoyancy. Mon. Wea. Rev., 110, 504–520, doi:10.1175/1520-0493(1982)110<0504:TDONSC>2.0.CO;2.
Weisman, M. L., and Klemp J. B. , 1984: The structure and classification of numerically simulated convective storms in directionally varying wind shears. Mon. Wea. Rev., 112, 2479–2498, doi:10.1175/1520-0493(1984)112<2479:TSACON>2.0.CO;2.
Weisman, M. L., and Rotunno R. , 2000: The use of vertical wind shear versus helicity in interpreting supercell dynamics. J. Atmos. Sci., 57, 1452–1472, doi:10.1175/1520-0469(2000)057<1452:TUOVWS>2.0.CO;2.
Weisman, M. L., Klemp J. B. , and Rotunno R. , 1988: Structure and evolution of numerically simulated squall lines. J. Atmos. Sci., 45, 1990–2013, doi:10.1175/1520-0469(1988)045<1990:SAEONS>2.0.CO;2.
Weiss, C. C., and Schroeder J. L. , 2008: StickNet—A new portable, rapidly-deployable, surface observing system. New Orleans, LA, Amer. Meteor. Soc., 4A.1. [Available online at https://ams.confex.com/ams/pdfpapers/134047.pdf.]
Wurman, J., 2001: The DOW mobile multiple-Doppler network. Preprints, 30th Int. Conf. on Radar Meteorology, Munich, Germany, Amer. Meteor. Soc., P3.3. [Available online at https://ams.confex.com/ams/pdfpapers/21572.pdf.]
Wurman, J., Straka J. , Rasmussen E. , Randall M. , and Zahrai A. , 1997: Design and deployment of a portable, pencil-beam, pulsed, 3-cm Doppler radar. J. Atmos. Oceanic Technol., 14, 1502–1512, doi:10.1175/1520-0426(1997)014<1502:DADOAP>2.0.CO;2.
Wurman, J., Dowell D. , Richardson Y. , Markowski P. , Rasmussen E. , Burgess D. , Wicker L. , and Bluestein H. B. , 2012: The second Verification of the Origins of Rotation in Tornadoes Experiment: VORTEX2. Bull. Amer. Meteor. Soc., 93, 1147–1170, doi:10.1175/BAMS-D-11-00010.1.
Ziegler, C. L., Mansell E. R. , Straka J. M. , MacGorman D. R. , and Burgess D. W. , 2010: The impact of spatial variations of low-level stability on the life cycle of a simulated supercell storm. Mon. Wea. Rev., 138, 1738–1766, doi:10.1175/2009MWR3010.1.
The Bunkers et al. (2006b) study defined deep-layer shear as being over the 0–6- and 0–8-km layers. In this study, we define the deep-layer shear as being the bulk wind difference from 0 to 6 km.
Readers may view the details of the quality control procedures in the EOL VORTEX2 data archive (http://data.eol.ucar.edu/master_list/?project=VORTEX2).
Updraft helicity, a surrogate for indicating the presence of a deep, persistent mesocyclone, was calculated in the manner of Kain et al. (2008) over a vertical distance dz between 2 and 5 km AGL, based on dual-Doppler estimates of vertical velocity w and vertical vorticity ζ: 