1. Introduction
Tornadoes are dangerous, often highly destructive phenomena that can threaten life and/or property. The tornado outbreak of 25–28 April 2011 alone killed 321 people and caused billions of dollars in damage (NOAA 2011). Currently, forecasters can sometimes predict with accuracy hours to days in advance if such tornado outbreaks will occur, as was the case in April 2011. However, determining which particular supercells will generate tornadoes is a serious challenge; nontornadic and tornadic supercells alike can have significant low-level rotation on the mesocyclone scale (Markowski et al. 2011), and the majority of supercells never produce tornadoes (Trapp et al. 2005). Increasing our understanding of the differences in the environments and internal processes and characteristics of nontornadic and tornadic supercells is, therefore, crucial to better forecasting.
The second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2) was designed to study such processes and environmental characteristics by collecting wind (radar) and thermodynamic (sounding, mobile mesonet, and deployable mesonet) observations within tornadic and nontornadic supercells (Wurman et al. 2012) and their environments. Given that synoptic-scale forcing can vary substantially from case to case, it can be difficult to compare tornadic and nontornadic supercells across different days. The optimal observational approach to do so, therefore, may be to examine supercells occurring in the same region on the same day. The VORTEX2 armada collected a rare dataset like this on 10 June 2010, when it deployed on one nontornadic and one tornadic supercell evolving in close proximity to each other in northeastern Colorado (Fig. 1). The VORTEX2 armada intercepted the northern of the two supercells, which was never tornadic, from 2345 to 0040 UTC, before redeploying to intercept the more impressive supercell to the south from 0110 to 0230 UTC (Fig. 2). This southern storm generated two tornadoes from 0109 to 0115 UTC and 0122 to 0126 UTC, respectively; the two tornadoes were weak [enhanced Fujita (EF0) rating according to Storm Data (NCDC 2010)], and no damage was found in a survey performed by Lyndon State University (N. Atkins 2015, personal communication). To examine why one storm produced no tornadoes and the other produced two, we investigate the storms’ interactions with other cells, as well as the storms’ environments and kinematic fields. We study the tornadic supercell’s tornado production, as the tornadoes are the crucial distinguishing factor between the two storms. We also examine the tornadic supercell’s subsequent mesocyclone behavior, as it may have inhibited further tornado production in this storm.
Progression of interactions between original cells and cells A and B (indicated in red). Contours are of KFTG WSR-88D logarithmic equivalent reflectivity factor 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
(a) Deployment map for 10 Jun 2010. The 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
To understand how environmental characteristics influence tornado potential, we first briefly review the current understanding of the genesis and maintenance of tornadoes associated with a mesocylone (i.e., ignoring nonmesocyclonic tornadoes such as landspouts). To produce a tornado, a supercell must 1) generate rotation (a mesocyclone) at midlevels, 2) generate rotation near the surface, and 3) significantly strengthen rotation near the surface (Davies-Jones 2015). The first step is well understood and occurs when an updraft [whose strength, at least the part owing to buoyancy, depends on the convective available potential energy (CAPE)] tilts and stretches horizontal vorticity associated with the mean vertical wind shear in the environment (Rotunno 1981; Lilly 1982; Davies-Jones 1984). If the horizontal vorticity is mostly streamwise (i.e., if the environmental vorticity vector is nearly parallel to the horizontal storm-relative winds) and the storm-relative winds are sufficiently strong (
However, the tilting of horizontal vorticity by the updraft will not generate vertical vorticity very close to the surface, as the updraft advects the vorticity away from the ground as it is produced. Thus, a downdraft is crucial for developing near-surface rotation if none preexists the storm. Current theories suggest the downdraft tilts horizontal vorticity that is baroclinically (or possibly frictionally) generated in the outflow into the vertical and advects the resulting vertical vorticity toward the surface (Klemp and Rotunno 1983; Rotunno and Klemp 1985; Davies-Jones and Brooks 1993; Davies-Jones 2015; Markowski and Richardson 2014; Dahl et al. 2014; Schenkman et al. 2014). An alternative to the baroclinic or frictional mechanisms was offered by Davies-Jones (2008) who showed that an imposed rain curtain could instigate tornadogenesis, given an initial midlevel mesocyclone, through an enhanced downdraft that transports moderately high-angular-momentum air toward the surface.
In the final step, the near-surface vertical vorticity must then become collocated with updraft forcing, so it can strengthen through vortex stretching. Vorticity stretching sufficient for tornadogenesis is contingent upon the ability of surface outflow parcels to be lifted, which is most easily accomplished if a strong vertical dynamic perturbation pressure gradient force is present and if the parcels are not too negatively buoyant (Markowski and Richardson 2014).
The strong upward dynamic perturbation pressure gradient force is often associated with rotation aloft, whose strength and distance above the ground determine the magnitude of the pressure gradient (Markowski and Richardson 2014). The strength of this dynamic lifting has been tied to the 0–1-km storm-relative helicity (SRH1) (Markowski and Richardson 2014), a measure of the low-level environmental streamwise vorticity (Davies-Jones et al. 1990), which is typically higher in tornadic supercell environments than nontornadic supercell environments (Rasmussen 2003; Thompson et al. 2012). Indeed, composite environments created by Parker (2014) for VORTEX2 tornadic and nontornadic supercells, based on targeted VORTEX2 soundings, show increased alignment between the 0–1-km environmental horizontal vorticity and the storm-relative winds in the tornadic cases, although it is not clear that this relationship holds in shallower layers near the surface.
Given these favorable conditions for lifting beneath the mesocyclone, it is reasonable to expect the most likely scenario for tornado formation to involve vertical alignment between the low- and midlevel circulations, with the midlevel circulation also coincident with the buoyant updraft to provide further support for lifting. Skinner et al. (2014) found that the low-level mesocyclone in their observed case intensified and grew upward when aligned with the midlevel mesocyclone, whereas Burgess et al. (1982) describe storm-relative rearward motion of vortices that become separated from the updraft as outflow surges around them to the east, resulting in cyclic behavior as new mesocyclones form at the subsequent occlusion point. Dowell and Bluestein (2002b) link the dissipation of a tornado to its displacement from the midlevel updraft owing to changes in the low-level outflow winds in which it was embedded. Similarly, French et al. (2008) noted significant rearward (with respect to storm motion) advection of pretornadic (i.e., those forming prior to the vortex that became tornadic) vortices, presumably limiting their ability to experience the forcing necessary to become tornadic. The tornadic vortex in their case, in contrast, exhibited much weaker rearward motion. Tanamachi et al. (2012) suggest more similar updraft motion and tornado motion for a long-lived tornado than a short-lived tornado. Marquis et al. (2012) also found that the longest-lived tornado in their study was maintained under the midlevel updraft, but that the shortest-lived tornado was similarly located below the midlevel updraft, suggesting that this vertical alignment, while important, may not always be sufficient for successful tornado maintenance.
Also crucial to lifting are the thermodynamic characteristics of the outflow parcels, with less-negatively buoyant parcels more easily lifted. Measurements made within the outflow of supercells suggest that tornadic supercells tend to have warmer outflows than nontornadic supercells (Markowski et al. 2002; Shabbott and Markowski 2006; Hirth et al. 2008; Weiss et al. 2015). Markowski et al. (2002) found, using mobile mesonet observations, that RFDs in “weakly” tornadic (i.e., producing EF0 or EF1 tornadoes lasting less than 5 min) and nontornadic supercells had maximum virtual potential temperature deficits of 4°–7°C, compared to less than 2°C in significantly tornadic supercells (producing EF2 or greater tornadoes, or tornadoes of any strength lasting greater than 5 min). Markowski et al. (2002) also found equivalent potential temperature deficits in nontornadic supercells of 10°–12°C, compared to less than 4°C in significantly tornadic supercells. The advantage of having outflow that is not too cold is consistent with climatological observations showing that tornadic supercells tend to have a lower environmental lifting condensation level (LCL), indicating higher relative humidity and presumably less evaporation of precipitation (all else being equal), than nontornadic supercells (Rasmussen and Blanchard 1998; Thompson et al. 2012).
One significant complication to the forecasting problem is the potential for spatial heterogeneity of each of these environmental characteristics and the temporal evolution of these quantities over the lifetime of a storm system. Markowski et al. (1998), using soundings from the 1994–95 VORTEX field campaign, showed large spatial variations in storm-relative helicity, particularly for storms occurring in the vicinity of an airmass boundary. Richardson et al. (2007) showed that idealized spatial variations in shear could strongly impact ongoing storm systems that travel through these variations. We might expect mesoscale subregions, likely at scales not well resolved by the operational observing network, to be more favorable for tornadoes. These subregions could develop through many different mesoscale processes including differential heating, mesoscale moisture advection, terrain influences, etc. Similarly, environments can evolve on temporal scales well below those of the observing network. Richardson and Droegemeier (1996), Kost (2004), Letkewicz et al. (2013), and Davenport and Parker (2015) all showed that temporal variations in a storm environment can lead to changes in storm outflow and rotational characteristics compared to a storm staying in the original environment. Coffer and Parker (2015) document large changes in hodograph shape during the early evening transition using soundings from VORTEX2, and use an idealized modeling framework to show these hodograph changes are linked to increases in dynamic lifting that aid the intensification of near-surface vorticity in the presence of increasing low-level stability.
Thus far, we have considered only influences on isolated cells, but often cells do not remain isolated over their entire lifetime. When two storms no longer have separate reflectivity maxima [in this study, based on a threshold of 35 dBZ, per Rogers (2012)], they have officially begun merging. Cell mergers can lead to changes in supercell behavior, including increased tornado production or the demise of the original supercell.
A merger may amplify low-level vertical vorticity, in agreement with some observations of tornadogenesis near the time of cell mergers (Lee et al. 2006; Wurman et al. 2007; Rogers and Weiss 2008). Enhancement of precipitation during a merger alters baroclinic zones and leads to a surge of outflow that can help stretch vorticity to tornado strength (e.g., Finley et al. 2001). In a study over five years, 27% of significant tornadoes occurred within 15 min of a merger event (Rogers 2012), as did 54% of nearly 100 tornadoes over 10 days (Rogers and Weiss 2008). Highlighting the complexity of merger outcomes, the observational study of Tanamachi et al. (2015) found that the 2011 El Reno supercell’s updraft weakened leading up to a merger, but the merger then led to the development of updraft pulses, which ultimately increased the vertical vorticity in the supercell. Alternatively, a merger can have adverse impacts, overall, on the supercell. Numerical simulations show that a merging (younger) cell may change the source of the supercell’s inflow, leading to the supercell’s demise as it ingests cold air from the young storm’s outflow (Hastings and Richardson 2016). We will see this scenario play out for the nontornadic supercell on 10 June.
In this study, we will examine rare data collected on one tornadic and one nontornadic supercell evolving in close proximity to each other to address the overarching question of why one supercell produced no tornadoes on this day while the other produced two. A description of the data and analysis methods is given in section 2. We first examine interactions between the supercells and nearby cells (section 3). We then compare environmental characteristics near the two storms and document the spatial and temporal variability (section 4). We next examine the storms’ outflow characteristics and kinematic fields to look for any obvious differences (section 5). Then, we analyze the evolution of the two tornadoes in the tornadic supercell (section 6). Finally, we study the motion of vortices in the tornadic supercell during its ensuing long nontornadic period (section 7). Concluding thoughts are offered in section 8.
2. Data and methods
The VORTEX2 armada collected an extensive dataset on 10 June 2010. For this case study, the focus will be on single-Doppler radar observations from the WSR-88D (KFTG) over the whole time period; from NOAA X-band dual-Polarization radar (NOXP; e.g., Burgess et al. 2010) and the Shared Mobile Atmospheric Research and Teaching (SMART) radars (SR1 and SR2; Biggerstaff et al. 2005) during the late stages of the tornadic storm; two dual-Doppler (using SR1 and SR2) deployments (one on each storm); single-Doppler radar observations from a Doppler on Wheels (DOW7) radar (Wurman et al. 1997) for the tornadic supercell during the two tornadoes; mobile mesonet and sounding data during both storms; and surface meteorological data from StickNet (Weiss and Schroeder 2008; Weiss et al. 2015) during the tornadic supercell (Fig. 2).
a. Radar data
Table 1 lists the mobile radar locations and general elevation angles of their data, whereas the locations are shown graphically in Fig. 2. Radar data were edited using the National Center for Atmospheric Research (NCAR) Solo II radar data editing and visualization software (Oye et al. 1995). Four independent estimates of velocity are available for DOW7, which used two different frequencies (9.35 and 9.50 GHz), each having two different pulse repetition times (the long and short pulse of the stagger). For each frequency, a radial velocity estimate is derived using standard staggered-pulse pulse-pair velocity retrieval, and then the resulting two estimates are averaged together to give the final estimated value for that radar gate. Velocities are filtered based on the normalized coherent power (NCP), eliminating velocities for which NCP is below 0.2–0.3 (with some subjectivity). The beam along the center of the tornado often spans both sides of an underresolved tornado, resulting in radial velocity estimates that are inconsistent with one another and a noisy averaged field. When that occurred, those center radial pixels were deleted. For the Smart Radar data, velocities are filtered based on returned power (DM) and spectrum width (SW), eliminating velocities for which DM is below around −88 dBZ and/or SW exceeds 8–10 m s−1 (with some subjectivity). For all radars, data believed to result from sidelobe contamination, ground clutter, etc., were removed separately. Of particular note, the SR1 elevation angles were corrected by approximately −1.5° using the azimuth-dependent correction equation in Rilling and Schumacher (2013), who originally identified this angle offset in SR1 data collected during the Dynamics of the Madden–Julian Oscillation (DYNAMO) experiment in 2011–12. Our independent analysis of reflectivity features from different radars suggest this offset must be applied to our VORTEX2 data as well, although we do not know if this applies to data collected earlier in the VORTEX2 project.
Radar deployment details.
Edited radar data were objectively analyzed to a Cartesian grid using the two-pass Barnes successive corrections method (Barnes 1964; Koch et al. 1983; Majcen et al. 2008). An isotropic Barnes weighting function was used within this method to retain scales appropriate for the data spacing δ, approximated as δ = rθ, where r is the radial distance from the radar and θ is the beamwidth. For the first pass, the smoothing parameter κ was set to 
The spacing of the Cartesian grid was set to 
Radial velocities were synthesized in regions where the between-beam angle was between 30° and 150°. The three-dimensional wind field was obtained by applying an upward integration (assuming 
For both storms, dual-Doppler syntheses were completed using 3-min volumes from SR1 and SR2, for 0006–0033 UTC for the nontornadic supercell and 0200–0230 UTC for the tornadic supercell. Details of the dual-Doppler analyses are summarized in Table 2. In addition to standard kinematic fields such as divergence and vorticity, other quantities derived from the dual-Doppler syntheses include updraft mass flux and circulation. Circulation (
Parameters used in dual-Doppler syntheses.
b. Sounding data
On 10 June 2010, four mobile GPS Advanced Upper-Air Sounding (MGAUS) systems from the National Severe Storms Laboratory (NSSL) and NCAR were operated by teams from North Carolina State University and NCAR to launch 21 radiosondes in northeastern Colorado, measuring pressure, temperature, relative humidity, and wind velocity every second. Data were quality controlled according to the methods of Loehrer et al. (1996, 1998).
Six radiosondes were launched prior to the storms (we use those launched after 
Hodographs constructed from the raw sounding data were extremely noisy; thus, we applied smoothing in a manner similar to Parker (2014). In particular, we used a one-pass Barnes (1964) filter with κ = 6.25 × 10−4 km2 within the lowest 1500 m and κ = 2.5 × 10−3 km2 over the rest of the sounding. The smaller κ at lower levels was used to retain as much detail as practical in this layer.
In addition to examining the individual sounding characteristics, the radiosonde data were used to help characterize the environments in which the supercells evolved in a bulk sense. For each sounding, numerous thermodynamic and kinematic parameters, as well as composite indices, were calculated and analyzed, such as MLCAPE,1 6BWD, SRH1, SRH3, MLLCL, and fixed-layer STP.
c. Mobile mesonet data
On 10 June 2010, six Pennsylvania State University–NSSL mobile mesonet vehicles with mounted instruments (Straka et al. 1996; Waugh and Fredrickson 2010) were deployed. The first supercell was sampled from 2345 to 0040 UTC, with spatial data coverage in both the forward and rear flanks of the storm. The second supercell was sampled from 0110 to 0230 UTC. Extensive data collection during the two tornadoes was hindered by a sparse road network and the inability to redeploy crews in time, especially during the first tornado. Data sampling improved following the second tornado as teams made their way deeper into the storm.
The mobile mesonet instrumentation collected GPS position, temperature, relative humidity, pressure, and wind velocity data every second. The “U tube” temperature shield design was used, allowing us to retain data from stationary mobile mesonet vehicles (Waugh and Fredrickson 2010) (unlike past studies using the “J tube” design), owing to the superior aspiration of the U-tube shield. Data were checked to remove obvious outliers and unreasonable values. For specifications of the mobile mesonet instruments, potential errors, and quality control procedures, refer to Straka et al. (1996), Markowski et al. (2002), and Waugh and Fredrickson (2010).
Two passes of a triangular weighting filter (using data within 10 s on either side) were applied to the mobile mesonet data to remove high-frequency noise, in effect smoothing the data. In total, 6 min of mobile mesonet data were used in each analysis (i.e., the storm was assumed to maintain a steady state over a 6-min period), with a time–space conversion done to produce analyses valid at a given reference time as in Markowski et al. (2002), Shabbott and Markowski (2006), and Markowski et al. (2012).
For each analysis, virtual potential temperature and pseudoequivalent potential temperature (Bolton 1980) were calculated and used to evaluate the thermodynamic characteristics within and near the supercells relative to a base state. The base state was calculated using observations collected by the mobile mesonet as the fleet approached the storms, for 15 min prior to deployment.
d. StickNet data
A spatially extensive StickNet dataset was collected during the posttornadic phase of the second storm. From 0142 to 0154 UTC, 17 StickNets were deployed predominantly in a north–south line, with a few in an adjacent east–west line, and captured thermodynamic and wind data as the storm passed over. Only data from 15 StickNets were used in our analyses, as two platforms collected temperature and relative humidity data that appeared unrepresentative when compared against other nearby StickNets over the same time period. Processing/filtering was applied in a similar fashion as with the mobile mesonet data (e.g., using a 6-min steady-state assumption). Note that while StickNet and mobile mesonet data must be compared with caution owing to their different time constants (Skinner et al. 2010), in this study, mobile mesonet and StickNet measurements in close proximity were generally consistent (within 1°C of each other), except in regions having strong gradients, where they differed by up to 2°–3°C. For further details on StickNet data, especially quality control procedures, refer to Weiss et al. (2015) and Skinner et al. (2011).
3. Cell interactions
On 10 June 2010, the nontornadic and tornadic supercells occurred in a region of low CIN conducive to formation of additional cells. Here, we investigate the ensuing interactions between the supercells and some of these younger cells. Both supercells interacted with a new storm that developed between them (hereafter referred to as “cell A”) and a small fast-moving reflectivity feature (presumed to be a very young cell, “cell B”) (Fig. 1). The tornadic supercell also seemed to interact with a small flanking line cell (“cell C”) (Fig. 3a). The merger of the nontornadic supercell with cell A was associated with the nontornadic supercell’s demise.
KFTG WSR-88D (left) 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
By 0014 UTC, cell A had initiated between the nontornadic and tornadic supercells (Fig. 1), near the intersection of their gust fronts (not shown). Over time, cell A intensified, increased in size, and began merging with the nontornadic supercell (Figs. 1b–e). Dual-Doppler analyses indicate general weakening of the updraft and mesocyclone in the nontornadic supercell during the merger. Updraft mass flux decreased from 0027 UTC onward (Fig. 4). Similarly, analyzed mesocyclone strength, measured in terms of circulation (at a radius of 1 km), decreased significantly from 0027 to 0033 UTC, dropping by at least 20 000 m2 s−1 over this time period at various heights (Fig. 4).
Impacts of the merger with cell A on characteristics of the nontornadic supercell, in terms of the time evolution of updraft strength at 2.1 km (average updraft mass flux, dashed black curve; 10−2 kg s−1 m−4) and mesocyclone strength (maximum circulation at a radius of 1 km, various heights denoted by colored solid curves; 104 m2 s−1). The brown dotted line denotes the maximum circulation in the tornadic supercell (based on the times when dual-Doppler data were available) at 1.2 km.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
The weakening of the updraft and mesocyclone of the nontornadic storm during the merger was associated with 1) inflow cooling and 2) rain incursion into the updraft. First, the maximum 
Mobile mesonet–measured virtual potential temperature perturbations (color-coded circles, relative to base state of 315.9 K) in the inflow region of the nontornadic supercell at (a) 0012 UTC, overlaid on 750-m objectively analyzed SR2 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
During this time, outflow from cell A was also interacting with the northern flank of the tornadic supercell and continued to do so leading up to the time of tornadogenesis (Fig. 3). These interactions may have favorably influenced baroclinicity and/or convergence in the tornadic supercell. In addition, around 0051 UTC, approximately 18 min prior to tornado formation, cell C may have merged into the rear flank of the tornadic supercell (the uncertainty is due to insufficient temporal resolution of the WSR-88D data; none of the mobile radars captured cell C, as they were repositioning during this time) (Fig. 3a). If it did merge into the rear flank, it is not clear from the WSR-88D data if this facilitated tornadogenesis or was merely coincidental, but mergers into the rear flank have been found in simulations (Hastings and Richardson 2016) to amplify low-level rotation. Observational studies have suggested a link between tornado events and rear-flank mergers as well (Rogers and Weiss 2008; Rogers 2012). Finally, just prior to tornadogenesis, another small new cell (cell B), that had formed to the south, merged into the far forward flank of the tornadic supercell (Figs. 1d–g). Numerical simulations by Hastings and Richardson (2016) predict a multicore system for updrafts interacting in this manner. In the present case, however, the new cell merged with the mass of precipitation to the north from the previous mergers, rather than remaining a separate entity.
Several interactions between the supercells and smaller cells occurred in this case. In particular, cell interactions limited the ability of the nontornadic cell to remain supercellular for as long as the tornadic supercell. The influence of cell interactions on the tornadic supercell is less clear, given available observational data. We now compare their environments for further clues about why one supercell produced tornadoes and the other did not.
4. Storm environments
Given the relationship between environmental characteristics and the likelihood for tornado formation, as discussed in the introduction, we next examine the environment. A total of 13 soundings were launched in the environment ahead of the storms from 2235 to 0230 UTC (Fig. 6a). Of these, the four (two for each storm) soundings closest in space and time to the two storms are used to characterize the near-storm environmental characteristics of each (Fig. 6a).When two soundings were a similar distance from the storm, we chose the one more likely to represent properties of air entering the storm at low levels based on the storm-relative wind directions. For example, we use the (NSSL1) 2342 UTC sounding rather than the (NCAR2) 2346 UTC sounding for the nontornadic storm, and we use the (NSSL1) 0138 UTC sounding rather than the (NCAR1) 0137 UTC sounding for the tornadic storm. For the nontornadic supercell, we focus on the time period during which it was mature until just after its interaction with cell A (2342–0042 UTC). For the tornadic supercell, we cover the time period within 30 min on either side of the tornadic phase2 (0042–0138 UTC).
(a) Map showing the soundings used to define the environmental conditions of the nontornadic and tornadic storms. The colored dots show the midlevel mesocyclone locations for the nontornadic supercell (bright green) and tornadic supercell (purple), every 15 min. The bright green (purple) circuit denotes the soundings most representative of the storm environment of the nontornadic (tornadic) supercell, based on temporal and spatial proximity. (b)–(d) Adapted from Thompson et al. (2012). Box-and-whisker plots of climatology of nontornadic, EF0 (weakly tornadic), and EF2+ (significantly tornadic) storm environments for discrete right-moving supercells represented by (b) 0–1-km storm-relative helicity, (c) mixed-layer lifting condensation level, and (d) significant tornado parameter overlaid with the ranges from 10 Jun 2010. Green dots show the values of the given parameter in the environment of the nontornadic supercell, and purple dots show the values in the environment of the tornadic supercell.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
In general, the tornadic storm environment fits the climatology for a weakly tornadic storm, while the nontornadic storm environment fits better with the nontornadic storm climatology based on Thompson et al. (2012). To better understand the spatial and temporal variations leading to these differences, we first examine the individual soundings as well as surface and 700-hPa analyses.
On 10 June 2010, the northeastern corner of Colorado resided in an environment containing both thermal and moisture gradients (Fig. 7). A dryline was evident near the Colorado–Kansas border and was associated with a low pressure center in southeastern Colorado, with easterly low-level flow in northeastern Colorado that advected higher dewpoints into the region from Kansas and southwestern Nebraska (Figs. 7a–c). The result was a complex moisture distribution, with a fairly steady moist axis running from Imperial, Nebraska (IML), to Akron, Colorado (AKO). Dewpoints over 50°F were confined to north of the Palmer Divide, with a strong moisture gradient south of Limon, Colorado (LIC). This surface temperature and moisture pattern will be evident in the environmental soundings discussed below. The pattern was similar at 700 hPa (Fig. 7d), with mesoanalyses from the Storm Prediction Center (Bothwell et al. 2002) indicating a strong moisture gradient in a similar location south of LIC.
(a)–(c) Global Telecommunications System (GTS) surface station plots from 10 Jun at 2233, 2343, and 0043 UTC (11 Jun), courtesy of NOAA. Times were chosen to be as close as possible to those of the mobile soundings. Station models show the temperature, dewpoint temperature (both in °F), wind speed and direction, and cloud cover at particular locations. Green and red (dashed) lines are subjectively analyzed isodrosotherms and isotherms, respectively. Here “L” indicates the low pressure center in southeast Colorado. (d) Mesoanalysis from the Storm Prediction Center at 700 hPa. Station plot shows temperature, dewpoint temperature (both in °C), pressure, and wind speed and direction. Red, dashed lines are isotherms. Green lines are isodrosotherms for dewpoint temperatures greater than −4°C.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
The synoptic-scale features on 10 June, combined with the terrain, generated a favorable environment for the development of severe weather near Denver, Colorado. At 2232 UTC, the two eventual supercells of interest were visible in satellite imagery (Fig. 8a) and as small echoes on radar (Fig. 9a) approximately 50 km apart along a north–south line. The southern cell (which later became tornadic) formed near the intersection of the Palmer Divide and the Rockies. Another cell (hereafter, cell D) formed slightly earlier on the southeast side of the Palmer Divide (
Visible satellite images from GOES-13 at (a) 2232, (b) 2345, (c) 0045, and (d) 0145 UTC. Stars correspond to sounding locations shown in Figs. 9–13. Cells of interest are labeled in each image.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
(a) (left) 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
Two soundings were taken around 2230 UTC at locations shown in Fig. 9a. These soundings, which are north of the eventual storm tracks, confirm the basic pattern of boundary layer moisture increasing from west to east across that part of the domain, with a small temperature contrast at the surface, in agreement with the surface analysis. Deep-layer shear (6BWD) was sufficient for supercells (i.e., 
By 2347 UTC, the two storms of interest were well-developed supercells (Fig. 10a). Three soundings taken ahead of the storms in a northwest–southeast line at about this time indicate very small values of CIN (
As in Fig. 9, but at 2347 UTC, and reflectivities below 2 dBZ have been removed for clarity. The pink, bright blue, and dark blue colors correspond to the NCAR2, NSSL1, and NCAR1 soundings at 2346, 2342, and 2354 UTC, respectively. The two unlabeled circles denote heights of 250 and 500 m. SRH1_T and SRH3_T refer to SRH1 and SRH3 computed using the motion of the tornadic supercell.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
Skew T–logp diagrams and hodographs showing the temporal evolution from (a),(b) 2235–2346 UTC for the NCAR1 (dark blue) and NCAR2 (pink) soundings and (c),(d) 2354–0230 UTC for the NCAR1 (dark blue and gray) and NSSL1 (bright blue and pale blue) soundings. Soundings are taken at similar locations within each panel.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
The wind profiles at 
Both storms of interest progressed eastward over the next hour, and soundings taken around 0042 UTC (Fig. 12b) show a substantial change in the environment ahead of the storms, particularly in the 750–600-hPa (
As in Fig. 10, but at 0042 UTC. The bright blue and dark blue colors correspond to the NSSL1 and NCAR1 soundings, respectively, at 0042 UTC.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
Coincident with these large thermodynamic changes were significant changes in the hodographs below 3 km (Figs. 12c and 11d), likely related to a sharp decrease in mixing in this layer as the environment stabilized. Mixing may have been further affected by cooling under the anvil as shown by Frame and Markowski (2013). A shortwave trough at 700 hPa also was entering the region (Fig. 7d) and likely affecting the wind profiles. These wind profile changes are similar to those documented during the early evening transition by Coffer and Parker (2015). SRH3 increased by 42%, while SRH1 tripled near the tornadic storm between 2354 and 0042 UTC (Fig. 11c), suggesting an environment much more supportive of tornado formation.
Although the 2354 UTC sounding is slightly warmer near the surface than the later soundings (Fig. 11c), little else is different thermodynamically below 750 hPa or above about 600 hPa, despite the drastic changes between these levels (Fig. 11c). The thermodynamic environment at the locations in this temporal series changed little after 0042 UTC, while the hodographs continued to evolve, with winds at 1 km changing from southeasterly at 0042 UTC to easterly thereafter (Fig. 11d). Overall, SRH1 in this series is a maximum at 0042 UTC but still supportive of at least weak tornadoes (i.e., 
As in Fig. 10, but at 0142 UTC. The pink, bright blue, and dark blue colors correspond to the NCAR2, NSSL1, and NCAR1 soundings at 0140, 0139, and 0137 UTC, respectively.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
As in Fig. 10, but at 0229 UTC. The pink, bright blue, and dark blue colors correspond to the NCAR2, NSSL1, and NCAR1 soundings at 0225, 0230, and 0226 UTC, respectively.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
To summarize thus far, the environment in which the supercells developed and evolved varied both spatially and temporally. Even early in the storms’ evolution, there was slightly greater low-level storm-relative helicity in the environment of the tornadic supercell. Both 0–1- and 0–3-km storm-relative helicity increased significantly as the storms continued to evolve, along with a strengthening of the storm-relative winds and a reduction of the mixed-layer depth. Over time, the environment became more favorable for tornado production. Although the two supercells were in close proximity, there were differences in their environments over their lifetimes, partially owing to spatial heterogeneity and partially owing to temporal evolution.
All values of 6BWD (17 and 22 m s−1 for the NT storm; 21 m s−1 for the T storm) used for the storm environments are within the lower half of climatological values for weakly tornadic (i.e., those producing only EF0 tornadoes) discrete, right-moving supercells (hereafter referred to simply as supercells) and below the 25th percentile for supercells producing EF2+ tornadoes (Thompson et al. 2012). The small range of values is consistent with similar gross storm characteristics for the two supercells. Values of MLCAPE (3174 and 4656 J kg−1 for the NT storm; 3379 and 3627 J kg−1 for the T storm) fall well above the 75th percentile of climatological values for all tornadic supercells (Thompson et al. 2012). MLCAPE is spatially variable (Figs. 10, 12, and 13), with the largest value occurring at 2342 UTC within the moist axis east of the nontornadic supercell. This high CAPE was short lived, returning to values below 3500 J kg−1 within an hour. Neither 6BWD nor MLCAPE discriminate well between the tornadic and nontornadic environments, as expected given that these parameters are more related to isolated storm type than to tornadic potential within the supercell type.
The LCL, traditionally found to be lower in tornadic supercell environments (e.g., Thompson et al. 2012), is actually a bit lower in the environment of the nontornadic supercell (924 and 1235 m) than in that of the tornadic supercell (1173 and 1304 m) (Fig. 6c). We note that the 924-m value was observed in only one sounding location at one time; the rest of the LCL values in both environments fit best with the distribution for weakly tornadic or nontornadic storms (Fig. 6c) and are above (i.e., worse than) the 75th percentile for significantly tornadic storms. Values of SRH1 are higher in the environment of the tornadic supercell (150 and 241 m2 s−2) than in that of the nontornadic supercell (44 and 166 m2 s−2) (Fig. 6b), suggesting the environment near the tornadic supercell was more favorable for tornado generation, with both values above the median for weakly tornadic supercells. Only at the later time (0138 UTC; just after the final tornado) is SRH1 above the 25th percentile range for significantly tornadic supercells. The nontornadic supercell experienced SRH1 below the 25th percentile for weakly tornadic supercells for most of its time as an isolated cell, reaching just above the median value for weakly tornadic storms only at the final time used (0042 UTC), when it was being affected by cell A (see discussion in section 3).
Values of the STP are higher in the environment of the tornadic supercell (3.1 and 4.0) than that of the nontornadic supercell (1.6 and 2.3) (Fig. 6d), consistent with the higher values of SRH1 in the former environment. STP values near the tornadic supercell fall above the 75th percentile for weakly tornadic supercells, but below the median for significantly tornadic supercells (Fig. 6d). The STP in the nontornadic supercell environment is above the median but below the 75th percentile of climatological values for weakly tornadic supercells, and only exceeds the 25th percentile for significantly tornadic supercells near the end of its life (e.g., 0042 UTC, at which time this storm was no longer isolated) (Fig. 6d).
5. Comparison of kinematic features and outflow characteristics
We now examine the basic flow and cold pool characteristics of each supercell. Dual-Doppler analyses of the nontornadic supercell show a mesocyclone at altitudes above 2 km AGL at 0021 UTC (Fig. 15b) but a lack of strong, well-developed rotation close to the surface (Fig. 15a). The lack of near-surface rotation may have been caused by relatively weak (
Dual-Doppler-derived, horizontal, storm-relative winds (blue vectors) and vertical vorticity (contoured every 4 × 10−3 s−1, positive in black, negative in white, with the zero contour omitted) overlaid on objectively analyzed SR2 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
Regarding the cold pool, measurements collected by the mobile mesonet fleet suggest that neither storm on 10 June 2010 had particularly large virtual potential temperature (
Mobile mesonet–measured virtual potential temperature perturbations (color-coded circles) in the outflow of (a) the nontornadic supercell at 0018 UTC (base state: 315.9 K), (b) the tornadic supercell at 0116 UTC (base state: 314.7 K), (c) the tornadic supercell at 0148 UTC (base state: 314.7 K), and (d) the tornadic supercell at 0202 UTC (base state: 314.7 K). Temperature perturbations are overlaid on objectively analyzed 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
Additionally, mobile mesonet measurements suggest that maximum equivalent potential temperature deficits across the outflows rarely exceeded 8°C in either storm prior to 0200 UTC (11 June). For example, at 0018 UTC in the nontornadic supercell, the maximum 
As in Fig. 16, but for equivalent potential temperature, relative to a base state of (a) 356.8 and (b)–(d) 358.9 K.
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
At most times in both storms, outflow temperatures were consistent with storms that produce weak tornadoes but also with storms that are nontornadic. Both outflows were, however, cold relative to typical outflows in significantly tornadic supercells. This highlights one of the forecasting challenges in marginal environments; storm characteristics (e.g., outflow temperature) and associated environmental parameters (e.g., LCL heights) may not be significantly different for weakly tornadic and nontornadic storms, resulting in a limited ability to forecast tornadoes.
6. Tornado evolution
We next briefly document the evolution3 of the two (visible) tornadoes4 produced by the tornadic supercell, as tornado production is the defining difference between the two supercells. The first tornado developed at approximately 0109 UTC (Fig. 18) and had a visually well-defined funnel (Fig. 19b), as it was located to the east of the bulk of the precipitation. It maintained a radial velocity couplet at or above tornado strength [defined here as 40 m s−1 following Alexander and Wurman (2008), no more than two gates apart (
Evolution of both tornadoes (and transition period in-between), shown by (left) 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
(a) Radial velocity differential (m s−1; red if greater than tornado threshold of 40 m s−1, blue if less than 40 m s−1) calculated using gate-to-gate (or one gate separation due to noise in the data; such differentials are denoted by an asterisk) radial velocity data from the raw 0.5°, 1°, 2°, 3°, 4°, 5°, and 6° DOW7 sweeps during the first tornado (0109–0115 UTC), transition period in-between (0115–0122 UTC), and the second tornado (0122–0126 UTC). Gray shading denotes the duration of the first and second tornado, respectively. (b),(c) Photographs of the (b) first tornado at 0114 UTC and (c) second tornado at 0124 UTC. Both photographs were taken by the LSC/NCAR Photogrammetry team and are looking to the west-southwest (0114 UTC) or ~west (0124 UTC).
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
For approximately the next 7 min, there was no visible condensation funnel present. During this transition time, the circulation aloft (1–2.5 km) generally maintained at least tornado strength (Fig. 19a). However, closer to the surface, around 0.5 km, the circulation had weakened by 0116 UTC, and had diminished below tornado strength after 0118 UTC. Even when the strength of the velocity couplet exceeded tornado strength, there was no accompanying visible funnel.
The second visible tornado began at approximately 0122 UTC (Fig. 19c). The strength of the circulation weakened at all heights throughout the tornado’s short life, especially from 0124 to 0126 UTC. During these last two minutes of the visible tornado’s life, the radial velocity differential was typically below tornado strength, even dipping as low as 20–25 m s−1 by 0126 UTC (Fig. 19a). Throughout its lifetime, the second tornado was surrounded by rain, likely due to an amplification of a surge in the rear-flank downdraft region to its south that resulted in the tornado experiencing westward motion relative to the eastward extent of high reflectivity. This surge was likely a rear-flank downdraft internal surge (e.g., Marquis et al. 2012), manifest here as eastward movement of a region of enhanced inbound radial velocities (i.e., up to 25–29 m s−1) (Figs. 18h–j), and seemed to be related to the leading edge of high reflectivity (i.e., greater than 25 dBZ). The tornado’s location within the rain and relatively humid air likely aided its ability to produce a condensation funnel despite the relatively weak circulation.
The evolution of the 10 June 2010 tornadoes did not follow the classic model of cyclic mesocyclogenesis prior to the production of a new tornado (Burgess et al. 1982). Rather, the same mesocyclone that produced the first tornado generated the second tornado minutes later, similar to behavior observed by Alexander and Wurman (2005) for the Spencer, South Dakota, tornadic supercell.
7. Posttornadic kinematic evolution
As previously discussed, the tornadic supercell’s mesocyclone underwent a cyclic evolution following the two tornadoes. This evolution is relevant to examine, as it may help explain why the tornadic supercell underwent a long nontornadic phase. Thus, it provides further insight into what makes tornado production more or less favorable, beyond that gleaned from comparing the nontornadic and tornadic supercells. Aloft, at approximately 2–2.5 km AGL, at least two new cyclonic circulations were observed to develop in the rear-flank region of the storm and move rearward relative to the rest of the storm, consistent with the behavior of some circulations in previous studies (French et al. 2008; Dowell and Bluestein 2002a). In contrast, an anticyclonic circulation remained relatively stationary in the storm-relative framework during this time period.
Single-Doppler and limited dual-Doppler data were used to analyze the circulations, due to severe sidelobe issues encountered by the radars collecting dual-Doppler data in low-reflectivity regions of the storm. The circulations were defined as having a diameter of 1–10 km and a radial velocity differential of at least 20 m s−1, per French et al. (2008). The first cyclonic circulation identified, “C1,” consisted of the remaining circulation from the second tornado. By 0131 UTC, C1 had moved backward in the rear flank (Fig. 20a) and then continued its rearward motion (relative to the motion of the storm). By 0147 UTC, the first “new” circulation (“C2”) had formed in the reflectivity hook (Fig. 20b) and subsequently traveled rearward in the storm (Figs. 20c–f). During this time, an anticyclonic circulation, “A1,” began significantly strengthening (e.g., Fig. 20e). Circulation A1 had been present, albeit weaker, prior to 0131 UTC. It is noteworthy that for much of the subsequent time analyzed, A1’s strength was comparable to or greater than that of the cyclonic circulations present (based on the radial velocity differential).
(left) 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
By 0204 UTC, a new circulation, “C3,” was rapidly developing well to the east-northeast of C2 and northwest of the anticyclonic vortex, along the rear-flank gust front (Fig. 20f). Like C1 and C2, over time, C3 moved rearward relative to the rest of the storm (Figs. 20g–j; Fig. 15d). Throughout this cycling, A1 experienced very little storm-relative motion. Analyses of dual-Doppler storm-relative streamlines from 0209 to 0218 UTC show that A1 was in virtually zero storm-relative flow, whereas the cyclonic circulations were embedded in strong midlevel rearward storm-relative flow (Fig. 21). The interplay between these midlevel circulations and the low-level (
Dual-Doppler-derived fields in the tornadic supercell during its later, nontornadic phase. Vorticity contours are black (cyclonic vorticity) and white (anticyclonic vorticity), every 0.5 × 10−2 s−1 beginning at 
Citation: Monthly Weather Review 144, 9; 10.1175/MWR-D-15-0345.1
8. Summary and conclusions
In this study, we analyzed a nontornadic and a tornadic supercell intercepted by VORTEX2 on 10 June 2010, and compared the interactions with other convective features, the storm environments, kinematic evolution, and outflow strengths. The goal was to identify differences and use these to hypothesize why one supercell never produced a tornado and the other produced at least two. Additionally, for the tornadic supercell, the evolution of the two tornadoes and the mesocyclone were studied.
One important difference between the two supercells appeared to be how each interacted with cell A, the storm which initiated between them (Fig. 1). The nontornadic supercell merged with cell A, and this merger led to the demise of the nontornadic supercell by weakening the updraft and mesocyclone through a combination of raining into the updraft and cooling the inflow of the supercell (Fig. 4). Had this merger not happened, would this supercell have been able to produce a tornado? It took the tornadic supercell nearly three hours after initiation to begin producing tornadoes, while the nontornadic supercell began weakening about two hours into its life. It is likely the nontornadic supercell merely did not have the opportunity to experience the increasingly favorable environment that evolved in time.
Both storms formed in environments that were initially similar and, for the most part, somewhat marginal for the development of tornadoes (especially “significant” tornadoes). The main difference at this early stage was slightly greater low-level storm-relative helicity in the environment of the tornadic supercell (e.g., Fig. 9). As the storms progressed, the environment evolved rapidly, with large increases in both 0–3- and 0–1-km storm-relative helicity and strengthening storm-relative winds, with a reduction of the mixed-layer depth (Fig. 11). While the environment was somewhat heterogeneous from the start, temporal variations were perhaps even more substantial. Because the environment became increasingly favorable for tornado production in time, the ability of the southern storm to remain isolated (at least with respect to its inflow sector) for a much longer time likely played a critical role in its ability to remain supercellular (Bunkers et al. 2006) long enough to experience these conditions and produce tornadoes.
Comparing the storm attributes, the two storms had cold pools with similar thermodynamic characteristics (at locations/times at which there were data), with outflows that were cold relative to those typical of supercells producing significant tornadoes (e.g., Fig. 16). The nontornadic storm had a significant midlevel mesocyclone but much weaker rotation at low levels. This may be related to the relatively weak storm-relative winds at low levels over most of its lifetime. The tornadic storm, on the other hand, was able to maintain significant low-level rotation (Fig. 15).
The evolution of the two tornadoes and the mesocyclone in the tornadic supercell were also analyzed. During the time period between the two tornadoes, the circulation generally maintained or exceeded tornado strength at most heights, while closer to the surface, the circulation weakened and there was no visible condensation funnel (Fig. 19). The same mesocyclone produced both tornadoes, but after the second tornado, cycling of the mesocyclone occurred. Two new midlevel circulations developed in the rear-flank region and moved rearward relatively quickly aloft. An anticyclonic circulation, on the other hand, remained almost stationary in the hook (Fig. 20). The long nontornadic phase of this storm is puzzling given the increases in STP in its environment. Perhaps the difficulty in producing a deep coherent vortex in a favorable location (i.e., near the main updraft), given the rearward motion of the mesocyclones that did develop, was a critical factor.
Future work will focus on how interactions with cells A and B may have helped the tornadic supercell produce a tornado. To study this, as well as what could have happened in the nontornadic supercell had it not experienced the detrimental merger, model simulations of this case using data assimilation techniques will be performed.
Acknowledgments
We are grateful to the Penn State Convective Storms Research Group, particularly Jim Marquis, for helpful discussion and support throughout the project. Additionally, we thank Michael Biggerstaff, Gordon Carrie, and Don Burgess at the University of Oklahoma and the National Severe Storms Laboratory for the SR1, SR2, and NOXP data. We are also grateful to all VORTEX2 participants for their dedication in collecting these data. Hans Verlinde provided helpful input as a member of the first author’s M.S. thesis committee. Richardson, Markowski, and Klees were supported by NSF-AGS-1157646 and NSF-AGS-1536460. The Doppler on Wheels NSF Lower Atmospheric Observing Facility is supported by NSF-AGS-1361237. Kosiba and Wurman were supported by NSF-AGS-1211132. Sounding and surface data were provided by NCAR/EOL under sponsorship of the National Science Foundation. VORTEX2 datasets can be requested through the EOL data archive at http://data.eol.ucar.edu/. Data also are available through the Penn State data commons (datacommons.psu.edu).
REFERENCES
Alexander, C. R., and J. Wurman, 2005: The 30 May 1998 Spencer, South Dakota, storm. Part I: The structural evolution and environment of the tornadoes. Mon. Wea. Rev., 133, 72–96, doi:10.1175/MWR-2855.1.
Alexander, C. R., and J. M. Wurman, 2008: Updated mobile radar climatology of supercell tornado structures and dynamics. 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., 19.4. [Available online at https://ams.confex.com/ams/pdfpapers/141821.pdf.]
Barnes, S., 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.
Beck, J., and C. Weiss, 2013: An assessment of low-level baroclinity and vorticity within a simulated supercell. Mon. Wea. Rev., 141, 649–669, doi:10.1175/MWR-D-11-00115.1.
Biggerstaff, M., 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.
Bolton, D., 1980: The computation of equivalent potential temperature. Mon. Wea. Rev., 108, 1046–1053, doi:10.1175/1520-0493(1980)108<1046:TCOEPT>2.0.CO;2.
Bothwell, P. D., J. A. Hart, and R. L. Thompson, 2002: An integrated three-dimensional objective analysis scheme in use at the Storm Prediction Center. Preprints, 21st Conf. on Severe Local Storms, San Antonio, TX, Amer. Meteor. Soc., JP3.1. [Available online at https://ams.confex.com/ams/SLS_WAF_NWP/techprogram/paper_47482.htm.]
Brandes, E., 1977: Flow in severe thunderstorms observed by dual-Doppler radar. Mon. Wea. Rev., 105, 113–120, doi:10.1175/1520-0493(1977)105<0113:FISTOB>2.0.CO;2.
Bunkers, M. J., M. R. Hjelmfelt, and P. L. Smith, 2006: An observational examination of long-lived supercells. Part I: Characteristics, evolution, and demise. Wea. Forecasting, 21, 673–688, doi:10.1175/WAF949.1.
Burgess, D., V. Wood, and R. Brown, 1982: Mesocyclone evolution statistics. Preprints, 12th Conf. on Severe Local Storms, San Antonio, TX, Amer. Meteor. Soc., 422–424.
Burgess, D., E. R. Mansell, C. M. Schwarz, and B. J. Allen, 2010: Tornado and tornadogenesis events seen by the NOXP x-band, dual-polarization radar during VORTEX2 2010. 25th Conf. on Severe Local Storms, Denver, CO, Amer. Meteor. Soc., 5.2. [Available online at https://ams.confex.com/ams/25SLS/techprogram/paper_176164.htm.]
Coffer, B. E., and M. D. Parker, 2015: Impacts of increasing low-level shear on supercells during the early evening transition. Mon. Wea. Rev., 143, 1945–1969, doi:10.1175/MWR-D-14-00328.1.
Dahl, J. M. L., M. D. Parker, and L. J. Wicker, 2014: Imported and storm-generated near-ground vertical vorticity in a simulated supercell. J. Atmos. Sci., 71, 3027–3051, doi:10.1175/JAS-D-13-0123.1.
Davenport, C. E., and M. D. Parker, 2015: Impact of environmental heterogeneity on the dynamics of a dissipating supercell thunderstorm. Mon. Wea. Rev., 143, 4244–4277, doi:10.1175/MWR-D-15-0072.1.
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., 2008: Can a descending rain curtain in a supercell instigate tornadogenesis barotropically? J. Atmos. Sci., 65, 2469–2497, doi:10.1175/2007JAS2516.1.
Davies-Jones, R., 2015: A review of supercell and tornado dynamics. Atmos. Res., 158–159, 274–291, doi:10.1016/j.atmosres.2014.04.007.
Davies-Jones, R., and H. Brooks, 1993: Mesocyclogenesis from a theoretical perspective. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, Geophys. Monogr., Vol. 79, Amer. Geophys. Union, 105–114.
Davies-Jones, R., D. W. Burgess, and M. Foster, 1990: Test of helicity as a tornado forecast parameter. Preprints, 16th Conf. on Severe Local Storms, Kananaskis Park, AB, Canada, Amer. Meteor. Soc., 588–592.
Dowell, D., and H. Bluestein, 2002a: The 8 June 1995 McLean, Texas, storm. Part I: Observations of cyclic tornadogenesis. Mon. Wea. Rev., 130, 2626–2648, doi:10.1175/1520-0493(2002)130<2626:TJMTSP>2.0.CO;2.
Dowell, D., and H. Bluestein, 2002b: The 8 June 1995 McLean, Texas, storm. Part II: Cyclic tornado formation, maintenance, and dissipation. Mon. Wea. Rev., 130, 2649–2670, doi:10.1175/1520-0493(2002)130<2649:TJMTSP>2.0.CO;2.
Dowell, D., and A. Shapiro, 2003: Stability of an iterative dual-Doppler wind synthesis in Cartesian coordinates. J. Atmos. Oceanic Technol., 20, 1552–1559, doi:10.1175/1520-0426(2003)020<1552:SOAIDW>2.0.CO;2.
Finley, C., W. Cotton, and R. S. Pielke, 2001: Numerical simulation of tornadogenesis in a high-precipitation supercell. Part I: Storm evolution and transition into a bow echo. J. Atmos. Sci., 58, 1597–1629, doi:10.1175/1520-0469(2001)058<1597:NSOTIA>2.0.CO;2.
Frame, J., and P. Markowski, 2013: Dynamical influences of anvil shading on simulated supercell thunderstorms. Mon. Wea. Rev., 141, 2802–2820, doi:10.1175/MWR-D-12-00146.1.
French, M., H. Bluestein, D. Dowell, L. Wicker, M. Kramar, and A. Pazmany, 2008: High-resolution, mobile Doppler radar observations of cyclic mesocyclogenesis in a supercell. Mon. Wea. Rev., 136, 4997–5016, doi:10.1175/2008MWR2407.1.
Hastings, R., and Y. Richardson, 2016: Long-term morphological changes in simulated supercells following mergers with nascent supercells in directionally varying shear. Mon. Wea. Rev., 144, 471–499, doi:10.1175/MWR-D-15-0193.1.
Hirth, B. D., J. L. Schroeder, and C. C. Weiss, 2008: Surface analysis of the rear-flank downdraft outflow in two tornadic supercells. Mon. Wea. Rev., 136, 2344–2363, doi:10.1175/2007MWR2285.1.
Klemp, J., and R. Rotunno, 1983: A study of the tornadic region within a supercell thunderstorm. J. Atmos. Sci., 40, 359–377, doi:10.1175/1520-0469(1983)040<0359:ASOTTR>2.0.CO;2.
Koch, S., M. DesJardins, and P. Kocin, 1983: An interactive Barnes objective map analysis scheme for use with satellite and conventional data. J. Climate Appl. Meteor., 22, 1487–1503, doi:10.1175/1520-0450(1983)022<1487:AIBOMA>2.0.CO;2.
Kost, J., 2004: Impacts of Temporally Variable Environmental Vertical Wind Shear upon Numerically Simulated Convective Storms. The Pennsylvania State University, 106 pp.
Lebo, Z. J., and H. Morrison, 2014: Dynamical effects of aerosol perturbations on simulated idealized squall lines. Mon. Wea. Rev., 142, 991–1009, doi:10.1175/MWR-D-13-00156.1.
Lee, B., B. Jewett, and R. Wilhelmson, 2006: The 19 April 1996 Illinois tornado outbreak. Part II: Cell mergers and associated tornado incidence. Wea. Forecasting, 21, 449–464, doi:10.1175/WAF943.1.
Letkewicz, C. E., A. J. French, and M. D. Parker, 2013: Base-state substitution: An idealized modeling technique for approximating environmental variability. Mon. Wea. Rev., 141, 3062–3086, doi:10.1175/MWR-D-12-00200.1.
Lilly, D., 1982: The development and maintenance of rotation in convective storms. Intense Atmospheric Vortices, L. Bengtsson and J. Lighthill, Eds., Springer-Verlag, 149–160.
Loehrer, S., T. Edmands, and J. Moore, 1996: TOGA COARE upper-air sounding data archive: Development and quality control procedures. Bull. Amer. Meteor. Soc., 77, 2651–2671, doi:10.1175/1520-0477(1996)077<2651:TCUASD>2.0.CO;2.
Loehrer, S., S. Williams, and J. Moore, 1998: Results from UCAR/JOSS quality control of atmospheric soundings from field projects. Preprints, 10th Symp. on Meteorological Observations and Instrumentation, Phoenix, AZ, Amer. Meteor. Soc., 1–6.
Majcen, M., P. Markowski, Y. Richardson, D. Dowell, and J. Wurman, 2008: Multipass objective analyses of Doppler radar data. J. Atmos. Oceanic Technol., 25, 1845–1858, doi:10.1175/2008JTECHA1089.1.
Markowski, P., and Y. Richardson, 2014: The influence of environmental low-level shear and cold pools on tornadogenesis: Insights from idealized simulations. J. Atmos. Sci., 71, 243–275, doi:10.1175/JAS-D-13-0159.1.
Markowski, P., J. M. Straka, E. N. Rasmussen, and D. O. Blanchard, 1998: Variability of storm-relative helicity during VORTEX. Mon. Wea. Rev., 126, 2959–2971, doi:10.1175/1520-0493(1998)126<2959:VOSRHD>2.0.CO;2.
Markowski, P., J. M. Straka, and E. N. Rasmussen, 2002: Direct surface thermodynamic observations within the rear-flank downdrafts of nontornadic and tornadic supercells. Mon. Wea. Rev., 130, 1692–1721, doi:10.1175/1520-0493(2002)130<1692:DSTOWT>2.0.CO;2.
Markowski, P., Y. Richardson, M. Majcen, J. Marquis, and J. Wurman, 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, 1–48.
Markowski, P., and Coauthors, 2012: The pretornadic phase of the Goshen County, Wyoming, supercell of 5 June 2009 intercepted by VORTEX2. Part I: Evolution of kinematic and surface thermodynamic fields. Mon. Wea. Rev., 140, 2887–2915, doi:10.1175/MWR-D-11-00336.1.
Marquis, J., Y. Richardson, P. Markowski, D. Dowell, and J. Wurman, 2012: Tornado maintenance investigated with high-resolution dual-Doppler and EnKF Analysis. Mon. Wea. Rev., 140, 3–27, doi:10.1175/MWR-D-11-00025.1.
NCDC, 2010: Storm Data. Vol. 52, No. 6, 1022 pp. [Available from National Centers for Environmental Information, 151 Patton Ave., Asheville, NC 28801-5001.]
NOAA, 2011: 2011 tornado information. Accessed 7 December 2013. [Available online at http://www.noaanews.noaa.gov/2011_tornado_information.html.]
Oye, R., C. Mueller, and S. Smith, 1995: Software for radar translation, visualization, editing, and interpolation. Preprints, 27th Conf. on Radar Meteorology, Vail, CO, Amer. Meteor. Soc., 359–361.
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.
Parker, M. D., and J. M. L. Dahl, 2015: Production of near-surface vertical vorticity by idealized downdrafts. Mon. Wea. Rev., 143, 2795–2816, doi:10.1175/MWR-D-14-00310.1.
Pauley, P., and X. Wu, 1990: The theoretical, discrete, and actual response of the Barnes objective analysis scheme for one- and two-dimensional fields. Mon. Wea. Rev., 118, 1145–1164, doi:10.1175/1520-0493(1990)118<1145:TTDAAR>2.0.CO;2.
Rasmussen, E., 2003: Refined supercell and tornado forecast parameters. Wea. Forecasting, 18, 530–535, doi:10.1175/1520-0434(2003)18<530:RSATFP>2.0.CO;2.
Rasmussen, E., and D. Blanchard, 1998: A baseline climatology of sounding-derived supercell and tornado forecast parameters. Wea. Forecasting, 13, 1148–1164, doi:10.1175/1520-0434(1998)013<1148:ABCOSD>2.0.CO;2.
Richardson, Y. P., and K. K. Droegemeier, 1996: An investigation of the dynamics governing organized multicell rotation and transition. Preprints, 18th Conf. on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 195–199.
Richardson, Y. P., K. K. Droegemeier, and R. P. Davies-Jones, 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.
Rilling, R., and C. Schumacher, 2013: SMART-R during DYNAMO: A technique to diagnose elevation angle errors. 36th Conf. on Radar Meteorology, Breckenridge, CO, Amer. Meteor. Soc., P238.
Rogers, J., 2012: Significant tornado events associated with cell mergers. 26th Conf. on Severe Local Storms, Nashville, TN, Amer. Meteor. Soc., 9.4. [Available online at https://ams.confex.com/ams/26SLS/webprogram/Paper211575.html.]
Rogers, J., and C. Weiss, 2008: The association of cell mergers with tornado occurrence. 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., P3.23. [Available online at https://ams.confex.com/ams/pdfpapers/141784.pdf.]
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 J. Klemp, 1985: On the rotation and propagation of simulated supercell thunderstorms. J. Atmos. Sci., 42, 271–292, doi:10.1175/1520-0469(1985)042<0271:OTRAPO>2.0.CO;2.
Schenkman, A. D., M. Xue, and M. Hu, 2014: Tornadogenesis in a high-resolution simulation of the 8 May 2003 Oklahoma City supercell. J. Atmos. Sci., 71, 130–154, doi:10.1175/JAS-D-13-073.1.
Shabbott, C. J., and P. M. Markowski, 2006: Surface in situ observations within the outflow of forward-flank downdrafts of supercell thunderstorms. Mon. Wea. Rev., 134, 1422–1441, doi:10.1175/MWR3131.1.
Skinner, P. S., C. C. Weiss, P. M. Markowski, and Y. P. Richardson, 2010: Intercomparison between mobile and stationary surface observing platforms in VORTEX2. 25th Conf. on Severe Local Storms, Denver, CO, Amer. Meteor. Soc., P5.1. [Available online at https://ams.confex.com/ams/25SLS/techprogram/paper_176245.htm.]
Skinner, P. S., C. C. Weiss, J. L. Schroeder, L. J. Wicker, and M. I. Biggerstaff, 2011: Observations of the surface boundary structure within the 23 May 2007 Perryton, Texas, supercell. Mon. Wea. Rev., 139, 3730–3749, doi:10.1175/MWR-D-10-05078.1.
Skinner, P. S., C. C. Weiss, M. M. French, H. B. Bluestein, P. M. Markowski, and Y. P. Richardson, 2014: VORTEX2 observations of a low-level mesocyclone with multiple internal rear-flank downdraft momentum surges in the 18 May 2010 Dumas, Texas, supercell. Mon. Wea. Rev., 142, 2935–2960, doi:10.1175/MWR-D-13-00240.1.
Straka, J. M., E. N. Rasmussen, and S. E. Fredrickson, 1996: A mobile mesonet for fine-scale meteorological observations. J. Atmos. Oceanic Technol., 13, 921–936, doi:10.1175/1520-0426(1996)013<0921:AMMFFM>2.0.CO;2.
Tanamachi, R. L., H. B. Bluestein, J. B. Houser, S. J. Frasier, and K. M. Hardwick, 2012: Mobile, X-band, polarimetric Doppler radar observations of the 4 May 2007 Greensburg, Kansas, tornadic supercell. Mon. Wea. Rev., 140, 2103–2125, doi:10.1175/MWR-D-11-00142.1.
Tanamachi, R. L., P. L. Heinselman, and L. J. Wicker, 2015: Impacts of a storm merger on the 24 May 2011 El Reno, Oklahoma, tornadic supercell. Wea. Forecasting, 30, 501–524, doi:10.1175/WAF-D-14-00164.1.
Thompson, R., R. Edwards, and J. Hart, 2002: Evaluation and interpretation of the supercell composite and significant tornado parameters at the Storm Prediction Center. 21st Conf. on Severe Local Storms, San Antonio, TX, Amer. Meteor. Soc., J3.2. [Available online at https://ams.confex.com/ams/SLS_WAF_NWP/techprogram/paper_46942.htm.]
Thompson, R., R. Edwards, and J. Hart, 2003: Close proximity soundings within supercell environments obtained from the Rapid Update Cycle. Wea. Forecasting, 18, 1243–1261, doi:10.1175/1520-0434(2003)018<1243:CPSWSE>2.0.CO;2.
Thompson, R., B. Smith, J. Grams, A. Dean, and C. Broyles, 2012: Convective modes for significant severe thunderstorms in the contiguous United States. Wea. Forecasting, 27, 1136–1154, doi:10.1175/WAF-D-11-00116.1.
Trapp, R., and C. Doswell, 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.
Trapp, R., G. Stumpf, and K. Manross, 2005: A reassessment of the percentage of tornadic mesocyclones. Wea. Forecasting, 20, 680–687, doi:10.1175/WAF864.1.
Waugh, S., and S. E. Fredrickson, 2010: An improved aspirated temperature system for mobile meteorological observations, especially in severe weather. 25th Conf. on Severe Local Storms, Denver, CO, Amer. Meteor. Soc., P5.2. [Available online at https://ams.confex.com/ams/25SLS/techprogram/paper_176205.htm.]
Weiss, C., and J. Schroeder, 2008: StickNet: A new portable, rapidly deployable surface observation system. Bull. Amer. Meteor. Soc., 89, 1502–1503.
Weiss, C., D. C. Dowell, J. L. Schroeder, P. S. Skinner, A. E. Reinhart, P. M. Markowski, and Y. P. Richardson, 2015: A comparison of near-surface buoyancy and baroclinity across three VORTEX2 supercell intercepts. Mon. Wea. Rev., 143, 2736–2753, doi:10.1175/MWR-D-14-00307.1.
Wurman, J., J. Straka, E. Rasmussen, M. Randall, and A. Zahrai, 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., Y. Richardson, C. Alexander, S. Weygandt, and P. F. Zhang, 2007: Dual-Doppler and single-Doppler analysis of a tornadic storm undergoing mergers and repeated tornadogenesis. Mon. Wea. Rev., 133, 97–119, doi:10.1175/MWR3276.1.
Wurman, J., D. Dowell, Y. Richardson, P. Markowski, E. Rasmussen, D. Burgess, L. Wicker, and H. Bluestein, 2012: The second Verification of the Origins of Rotation in Tornadoes Experiment. Bull. Amer. Meteor. Soc., 93, 1147–1170, doi:10.1175/BAMS-D-11-00010.1.
As some soundings were prematurely cut off prior to reaching their equilibrium level, data at higher levels from the 2342 UTC sounding were used so that contributions to CAPE at upper levels could be estimated. This sounding was chosen based on a comparison of all full-depth soundings during the analysis period. The 2342 UTC sounding reasonably represented the general upper-level conditions.
The grouping we chose to characterize the nontornadic and tornadic supercells is different than that used by Parker (2014) for this case, owing to our stricter spatial and temporal criteria.
For consistency, all of the radial velocity differential estimates use the four-field-averaged velocity (see section 2a), even though this sometimes required ignoring data in the center radial of the tornado. Inspection of individual fields containing data along the center radial (not shown) confirms the general temporal patterns of velocity differential, although exact amplitudes are slightly different.
We define tornado occurrence by the presence of a visible condensation funnel associated with a strong vortex. Although a visible funnel is not necessary for a tornado to exist, on this day the visible funnels were our best indicator of the tornadic vortices. That being said, as the definition of a tornado is subjective (as is the radial velocity differential threshold we use below), we recognize the possibility that there was just one continuous tornado on 10 June.
