1. Introduction
Bow echoes are a well-known mode of severe convection that have long been recognized for producing damaging downburst winds and tornadoes. Early bow echo damage surveys by Fujita (1978, 1981), Fujita and Wakimoto (1981), and Forbes and Wakimoto (1983) showed that bow echo damage swaths can be a couple hundred kilometers in length and often contain F0–F2 damage.
Based on radar and damage survey data, Fujita (1978, 1979) put forth a conceptual model illustrating the morphology of a bow echo and attendant surface wind damage. Reproduced in Fig. 1, Fujita’s model shows how the convective system begins as a tall echo (stage A) where downburst winds then descend to the surface pushing the echo outward, distorting it into a bow-shaped bulge (stage B). Damaging winds may be experienced at the apex of the developing bow echo. The close association of surface wind damage at the bow apex has also been discussed by a number of investigators (e.g., Fujita 1981; Forbes and Wakimoto 1983; Wakimoto 1983; Przybylinski 1995; Funk et al. 1999; Atkins et al. 2004). Fujita hypothesized that a strong descending rear-inflow jet (RIJ) was the source of the damaging winds at the bow apex (Wakimoto 2001). Doppler radar studies have provided confirmatory data showing descending rear-inflow jets positioned at the apex of bow echoes (Burgess and Smull 1990; Jorgensen and Smull 1993; Schmidt and Cotton 1989; Atkins et al. 2004). Numerical simulations have since shown that the RIJ is a ubiquitous system-scale feature in mesoscale convective systems (MCSs) and is generated by a horizontal perturbation pressure gradient at midlevels behind the leading edge of the convective system (Lafore and Moncrieff 1989; Weisman 1993). Genesis of the RIJ has also been described as the result of horizontal buoyancy gradients associated with the warm air aloft in the ascending front to rear flow and the cold pool. If the buoyancy gradient associated with the cold pool is larger than with the warm air aloft, the RIJ will descend to the ground rearward of the leading edge of the convective system (Weisman 1993).
As the bow echo continues to intensify (stage C), it may take the shape of a spearhead, accompanied by a channel of weak echo, or rear-inflow notch, located along the central axis of the downburst flow. The operational significance of the rear-inflow notch on the back side of a bow echo signifying the presence of downburst winds has been noted by Przybylinski and Gery (1983), Przybylinski and DeCaire (1985), and Przybylinski (1995). Smull and Houze (1985, 1987) have shown that the rear-inflow notch is collocated with the RIJ that is transporting drier air into the rear of an MCS, thus locally enhancing evaporation of hydrometeors. Downburst-induced tornadoes are also likely to form during stages B and C of bow echo development at or north of the bow apex, confirmed by more recent radar studies (e.g., Forbes and Wakimoto 1983; Funk et al. 1999; Atkins et al. 2004). Atkins et al. (2004) have recently shown, however, that tornadoes can also form south of the bow apex.
As the bow echo enters the comma echo stage (stages D and E in Fig. 1), the downburst winds weaken and a prominent cyclonic mesoscale circulation can be found on the northern end collocated with a radar-detected comma-shaped echo. Numerical modeling studies of quasi-linear convective systems (QLCSs) such as bow echoes and squall lines have shown that midlevel system-scale vortices often form on the ends and behind the leading edge of the convective system (e.g., Weisman 1993; Weisman and Davis 1998). Their genesis has been attributed to the vertical tilting of baroclinically generated horizontal vorticity by the system-scale updraft along the gust front (Weisman and Davis 1998). Stretching of planetary vorticity enhances (weakens) the northern cyclonic (southern anticyclonic) vortex (Skamarock et al. 1994) producing an asymmetric system-scale structure with time (Houze et al. 1989). Note that damaging downburst winds and tornadoes are not observed during the weakening comma echo stage. Recent observations by Pfost and Gerard (1997) and Trapp et al. (1999), however, have documented straight-line and tornadic damage during the comma-echo stage of bow echo evolution.
While Fujita’s conceptual model suggests that the strongest damaging winds occur during the bow echo stage and are located at the apex, recent observational and numerical studies hypothesize that damaging surface winds may be created by system- and subsystem-scale vortices formed within the bow echo. In addition to the system-scale “book-end vortices” (e.g., Weisman 1993), subsystem-scale (1–10 km) “mesovortices” have also been observed to form along the leading edge of QLCSs at low levels (generally 0–3 km AGL) primarily at and north of the bow apex (e.g., Przybylinski 1995; Funk et al. 1999; Przybylinski et al. 2000; Atkins et al. 2004) although they have been observed south of the apex (Atkins et al. 2004). Indeed, observational analysis of a bow echo by Schmocker et al. (2000) showed that the strongest wind damage was observed along the southern periphery of a mesovortex that was located north of the apex. Miller and Johns (2000) discussed how extreme, damaging winds may be produced by embedded supercell storms and their attendant low-level mesocyclone within a derecho-producing MCS. Trapp and Weisman (2003) analyzed idealized bow echo simulations and showed that the strongest ground-relative winds were produced well north of the bow apex in association with a low-level mesovortex. The large horizontal pressure gradients associated with the mesovortex help to generate strong “straight line” winds at the surface. Moreover, Atkins et al. (2004) showed that enhanced damage may be generated by the midlevel cyclonic book-end vortex descending to the ground. Resolution of the damaging straight-line wind mechanism(s) within bow echoes may be accomplished by superimposing single- or dual-Doppler radar data on detailed damage survey analyses. The authors are unaware of any such analyses published in the refereed literature.
In addition to their possible role in generating damaging straight-line winds, it is well known that mesovortices also produce tornadoes within bow echoes (e.g., Forbes and Wakimoto 1983; Wakimoto 1983; Przybylinski 1995; Funk et al. 1999; Przybylinski et al. 2000; Atkins et al. 2004). Most bow echo tornadoes reported in the literature have produced F0–F2 damage; however, they are capable of producing F3–F4 damage (Trapp et al. 2005), while Wakimoto (1983) documented an F4 anticyclonic tornado located north of a bow echo apex.
Not all mesovortices formed within a bow echo become tornadic. While the bow echo tornado genesis mechanism(s) is (are) still not understood, recent observations by Atkins et al. (2004) suggest that it may be possible to differentiate between tornadic and nontornadic mesovortices. They showed that the tornadic vortices tended to be stronger, longer-lived, and deeper, as viewed by Doppler radar, than their nontornadic counterparts. While their analyses may have important forecasting implications, the generality of the results is not known.
During the afternoon hours on 10 June 2003, a severe QLCS formed over central Missouri and moved eastward over the greater Saint Louis area during the Bow Echo and Mesoscale Convective Vortex (MCV) Experiment (BAMEX; Davis et al. 2004). The QLCS produced two well-defined bow echoes. The first was observed approximately 100 km southwest of Saint Louis, while the second produced a damage swath approximately 50 km in length and a number of tornadoes east of Saint Louis in southwestern Illinois. BAMEX was not operational on the afternoon of 10 June because a mission had been executed during the late evening and early morning hours on 9 and 10 June, respectively, on a severe MCS that also passed over the Saint Louis area. Fortuitously, the Weather Surveillance Radar-1988 Doppler (WSR-88D) at Saint Charles, Missouri (KLSX), collected data on the second bow echo. Furthermore, detailed ground-based and aerial damage surveys were completed for both bow echoes. By combining the KLSX radar and detailed damage survey data, the objectives of this study are to 1) determine the mechanism(s) that produced the long swath of straight-line wind damage east of Saint Louis, and 2) examine the structural evolution of the tornadic and nontornadic mesovortices and compare the results to those discussed by Atkins et al. (2004).
The paper is organized as follows. A brief description of BAMEX and the damage survey methodology is discussed in section 2. Section 3 presents the synoptic and mesoscale environment observed on 10 June 2003, while section 4 provides a radar overview of the event and the damage that was created. Detailed radar and damage survey data are superimposed in section 5 to better understand the damaging straight-line wind mechanism, while section 6 discusses the differences between the tornadic and nontornadic mesovortices. Discussion and conclusions are finally presented in section 7.
2. Damage surveys during BAMEX
During the period from 20 May to 6 July 2003, the field phase of BAMEX was executed over a large portion of the central United States. BAMEX was a highly mobile airborne and ground-based experiment designed to collect observations throughout the life cycle of bow echoes and MCVs. For a detailed description of BAMEX, the reader is referred to Davis et al. (2004).
Within the experimental domain, detailed damage surveys were performed for most damaging bow echo events during BAMEX to collect the requisite ground truth information of surface wind damage. Both ground and aerial surveys were executed following techniques established by Fujita et al. (1976) and Fujita (1981, 1992). Radar data, local storm reports (LSRs), and storm information collected by local National Weather Service offices were used to guide the survey teams to the affected area. Thereafter, the survey teams carefully traversed the damage path mapping the damaging surface wind field based on observed tree fall along with damage to structures. A total of nine events were surveyed during BAMEX, six of which have been analyzed in a companion study by Wheatley et al. (2005, manuscript submitted to Mon. Wea. Rev.). The interested reader can find preliminary survey results and damage photographs on the BAMEX damage survey Web site (http://apollo.lsc.vsc.edu/bamex).
3. Synoptic and mesoscale environment on 10 June 2003
The synoptic-scale environment on 10 June 2003 was similar to the warm-season derecho environment described by Johns and Hirt (1987) and Johns (1993). Within this environment, the derecho moves along a stationary surface thermal boundary. Prominent thermal advection is present at low levels while moderate northwesterly flow is observed at midlevels, often in advance of a short-wave trough. Large values of convective available potential energy (CAPE) are observed in the derecho genesis region. At 850 hPa in Fig. 2a, broad southwesterly flow was observed advecting warm, high equivalent potential temperature (θe) air into southern and central Missouri ahead of a trough at 1200 UTC. At 500 hPa, west-southwesterly flow was observed over Missouri with speeds of 17–20 m s−1.
A more detailed depiction of the mesoscale environment is shown in Fig. 3 where surface analysis and data along with radar data are superimposed on visible Geostationary Operational Environmental Satellite-E (GOES-E) satellite imagery. At midday (1600 UTC), a cold front was moving southeastward and had entered northwestern Missouri. South of the front, warm, moist air was being transported northward by 5–8 m s−1 of south-southwesterly flow. This air was characterized by θe values in excess of 355 K. A decaying MCS and associated cold pool that moved through the Saint Louis area earlier in the day was observed primarily over southeastern Missouri and southern Illinois. By 1800 UTC (Fig. 3b), new convection that eventually evolved into the severe QLCS of interest, formed over west-central Missouri just east and west of the Kansas City, Missouri, WSR-88D. The new convection appeared to form where the axis of high θe air was impinging upon the cold front. Figure 4 shows a sounding launched at Springfield, Missouri, at 1800 UTC (location of star in Fig. 3b) and clearly indicates that the high θe air mass was unstable, with a surface-based CAPE of 2558 J kg−1. The hodograph reveals a moderate wind vector difference of 14 m s−1 from the surface to 2 km, with the shear vector oriented to the east. The wind vector difference through a deeper layer of 5 km was only 18 m s−1; thus, most of the shear was found at low levels. Previous numerical and observational studies have shown that bow echo environments are often characterized by large CAPE and moderate to strong low-level shear (e.g., Johns and Hirt 1987; Weisman 1993; Evans and Doswell 2001). A bulk Richardson number of 52 was calculated and would suggest organized convection to be multicellular (Weisman and Klemp 1982). The decaying MCS southeast of Saint Louis continued to weaken and move southeastward. During the next 2 h, the convection that formed near Kansas City, Missouri, intensified and began to move eastward (Fig. 3c). The developing QLCS propagated eastward along the θe gradient oriented west–east across central Missouri and appeared to organize roughly orthogonal to the low-level shear vector. By 2200 UTC (Fig. 3d), a well-defined bow echo was observed approximately 100 km to the southwest of Saint Louis. As will be shown in the next section, this was the first of two damaging bow echoes produced by the QLCS.
4. Radar and damage overview
a. Radar overview
As the QLCS approached Saint Louis, two active areas of convection were observed at 2059 UTC by the KLSX radar shown in Fig. 5. The southern group of convective cells appeared to be larger and stronger and eventually evolved into the first bow echo by 2159 UTC. This is the same bow echo that was shown in the satellite analysis at 2200 UTC in Fig. 3d. From 2059 to 2159 UTC, the northern area of convection appeared to grow modestly in size. Inbound velocities exceeding 25 m s−1 at the 0.5° elevation scan were observed within this area of convection, implying an increased potential for damage produced by strong surface flow. Also during this time period, the appearance of mesovortices was detected in the KLSX radial velocity data. Interestingly, many of the initial mesovortices that formed within the QLCS were nontornadic (thick-dashed lines) and generally produced little or no surface wind damage. By 2300 UTC, the first bow echo had become less defined in the reflectivity field while the second bow echo was beginning to develop just west of the MidAmerica Airport (MAA) in Mascoutah, Illinois (location of star in Fig. 5). Curiously, many of the mesovortices that formed between 2200 and 2300 UTC were tornadic (solid lines), in contrast to the nontornadic vortices that formed during the previous hour. As the second bow echo continued to mature, a well-defined comma head was observed at 2356 UTC on its northern end, suggesting a transition to the comma echo stage (Fig. 1).
b. Damage overview
On 11 June 2003, ground and aerial survey teams began the process of surveying damage produced by both bow echoes. The results are shown in Fig. 6. The resultant damage analysis for the first bow echo is likely incomplete and, therefore, should be interpreted with some caution for the following reasons. First, ground teams could not access much of the affected area due to the lack of good roads. Second, it took nearly 1.5 weeks for the survey teams to reach the area southwest of Saint Louis, as they first surveyed the large area affected by the second bow echo. Furthermore, damaging wind events occurred before and after the 10 June QLCS1; thus aerial survey efforts were hampered since it was often difficult to unambiguously determine whether observed damage was created by the 10 June event. Nonetheless, the limited damage survey data shows that the first bow echo appeared to create F0 or greater straight-line wind damage in the vicinity of the bow apex.
Notable in Fig. 6 is that the second bow echo produced a continuous swath of F0 or greater damage approximately 50 km in length in St. Clair and Clinton counties east of Saint Louis, reminiscent of bow echo damage swaths surveyed by Fujita (1978). Interestingly, the BAMEX Operations Center, located at the MidAmerica Airport, was directly in the primary damage swath. Indeed, wind gusts of 46 m s−1 were measured by a weather station at the nearby Scott Air Force Base. All BAMEX personnel were evacuated from the Operations Center just a few minutes prior to bow echo passage. The high winds knocked down nearby power lines and power to the airport. As a result, the BAMEX Operations Center was temporarily relocated to a nearby hotel while power was restored.
The damage survey efforts also uncovered six tornado tracks, five of which appeared to be associated with the second bow echo. The sixth tornado track was associated with a long-lived mesovortex that formed within the first bow echo. Tornado damage was unambiguously discernible from straight-line wind damage in the following ways. First, sharp gradients of damage were observed in the direction orthogonal to the tornado path orientation. The area of greatest damage within the track was often less than 100 m in width. Second, the damage vectors consistently showed a convergent signature in the wind field along the damage track, consistent with a translating rotational wind field embedded within updraft (Hall and Brewer 1959). Forbes and Wakimoto (1983) also showed similar damage patterns associated with bow echo tornadoes.
A more detailed analysis of the damage observed in southwestern Illinois largely associated with the second bow echo is shown in Fig. 7, akin to the bow echo damage analysis presented by Forbes and Wakimoto (1983). Clearly evident is the primary damage swath beginning just east of East Saint Louis. The damage swath was approximately 8 km in width. Damage was continuously produced over a length of 50 km; however, the damage path discontinuously extends 70 km from just east of Cahokia eastward to just north of Hoyleton. Divergent microburst damage was uncovered within and surrounding the primary damage swath. The six F0–F1 tornado tracks2 are also shown in more detail. As shown in Table 1, the tornadoes tended to be narrow with short track lengths and durations. Indeed, most of the tornadoes had lifetimes comparable to or less than the time required for the WSR-88D to complete a volume scan in severe weather mode (approximately 4.5 min in VCP-12).
As the second bow echo moved eastward over the Saint Louis area, the KLSX radar collected a comprehensive dataset owing to the short range from the radar to the developing bow echo, and storm motion was largely in the radial direction. Thus, superposition of the damage survey shown in Fig. 7 with KLSX radar data will help to determine where, within the bow echo, the primary damage swath was created, and therefore its generation mechanism. The structural evolution of the tornadic and nontornadic mesovortices can also be examined. These analyses are presented in the next two sections.
5. Primary damage swath generation mechanism
The precise location of the damage with respect to the second bow echo is shown in Fig. 8, in which the damage survey data in Fig. 7 is superimposed on radar reflectivity and ground relative radial velocities every 10 min from 2300 to 2340 UTC. At 2300 UTC, the second bow echo was bulging outward slightly, evident in the reflectivity field between azimuths of approximately 95° and 135°. The attendant RIJ was visible as a core of strong radial velocities in excess of 25 m s−1 in the 2.4° elevation scan. Radial velocities at lower levels along the leading edge of the bow echo were in excess of 30 m s−1, a likely result of the RIJ descending to the ground. Interestingly, damage was not observed over the entire area of enhanced surface winds. Rather, F0 straight-line wind damage was being generated immediately north of the axis of strong winds associated with the RIJ, collocated with mesovortex 6. The outbound radial velocity maximum at 2300 UTC associated with mesovortex 6 was, in fact, collocated with the western edge of the F0 primary damage swath. This is more clearly observed in the storm-relative radial velocities in the inset diagram (Fig. 8a) where the vortex couplet is more apparent. A smaller area of F0 wind damage was evident at 60 km, 125° from the radar, and was associated with mesovortex 8. The close association between the mesovortices and tornadic damage paths was also evident.
Ten minutes later at 2310 UTC (Fig. 8b), the primary damage swath was again collocated with mesovortex 6 well north of the bow apex and RIJ. Similar to 2300 UTC, no appreciable damage was observed in the vicinity of the RIJ at this time; however, mesovortices 8 and 9 had produced F0 and F1 tornadoes, respectively, sometime between 2300 and 2310 UTC. Between 2310 and 2320 UTC, the bow echo, and more specifically, mesovortex 6 passed over MidAmerica Airport. Wind gusts of 46 m s−1 were measured at the nearby Scott Air Force Base. Mesovortices 6 and 9 began to merge by 2320 UTC in Fig. 8c. Their combined outbound velocity maximum was positioned directly over the primary damage swath and contained ground-relative radial velocities in excess of 40 m s−1. At 2325 UTC, the vortices had merged and subsequently traveled along the path of mesovortex 9, dissipating by 2330 UTC. At 2330 UTC, the damage swath was still located along the southern periphery of the merged, dissipating mesovortex. The southwesterly divergent flow observed within the primary damage swath beyond 90 km suggests that embedded microbursts may have also contributed to the damaging winds at and after 2330 UTC. A comma-head structure and weak echo channel were becoming evident in Fig. 8d and were apparent by 2340 UTC (Fig. 8e). Note the northward displacement of the damage swath relative to the weak echo channel and radial velocity maximum at 0.5°. Even though the reflectivity data at 2340 UTC showed a well-defined bow echo structure, the primary damage swath was already created.3 In other words, much of the straight-line wind damage produced by the second bow echo occurred before a well-defined bow echo reflectivity structure was observed by the KLSX radar. This is consistent with Fujita’s conceptual model (Fig. 1) in which damaging surface winds are observed within the bow echo stage of development. However, the reflectivity structure of the second bow echo was not well defined until it had reached the comma-head stage of development, even though the RIJ was well resolved in the radial velocity data. This result highlights the importance of monitoring other radar-detected features often observed within bow echoes such as the midaltitude radial convergence (MARC) signature that often precedes the onset of damaging winds at the surface (Schmocker et al. 1996).
6. Tornadic and nontornadic mesovortices
Damage survey and radar data presented in Figs. 5 and 8 showed that five of the radar-detected mesovortices produced F0–F1 tornadoes. An example of the tornado damage uncovered by the damage surveys is shown in Fig. 9a for the tornado associated with mesovortex 9. A common characteristic of these weak, quickly translating tornadoes was the convergent nature of the damage pattern observed on the ground (Hall and Brewer 1959; Forbes and Wakimoto 1983). Figure 9b shows an aerial photograph over a field located along the northwestern part of the damage path clearly indicating convergence into a “litter line” as described by Fujita et al. (1976). Convergent tree damage was also observed farther southeast within the damage path in Figs. 9c and 9e. Examples of structural damage are shown in Figs. 9d and 9e. Located within the initial part of the damage path, the home in Fig. 9d sustained major damage, consistent with F2 damage described by Fujita (1971); however, the damage was rated F1 because of the compromised structural integrity of the home. Similarly, Fig. 9e shows considerable damage to the roof and walls of a metal shed, whereas the nearby garage and house received less severe damage having lost shingles from their respective roofs. Numerous trees surrounding the home were uprooted.
The damage survey results in Figs. 6 and 7 show that the most intense damage (F1) produced by the 10 June QLCS was associated with the tornadic mesovortices. Thus, an important detection and warning problem is to determine which mesovortices within QLCSs will become tornadic. Atkins et al. (2004) have addressed this question and found that tornadic mesovortices tended to be stronger, deeper, and longer-lived than their nontornadic counterparts. The radar and damage survey data for the 10 June event are now discussed to address the applicability of these results.
a. Mesovortex lifetimes
A histogram of mesovortex lifetimes is shown in Fig. 10c. The tornadic vortices tended to be longerlived (56 min) relative to the nontornadic vortices (19 min). This result is consistent with Atkins et al. (2004) and with simulated QLCSs in weak sheared environments (Weisman and Trapp 2003). Simulated QLCSs in stronger (e.g., 30 m s−1 over 5 km) sheared environments, however, had mesovortex lifetimes up to several hours long. Interestingly, the mean tornado lifetime of 5.5 min was quite short relative to the parent mesovortex. On average, only 12 min elapse from the genesis of the parent mesovortex until it produces a tornado, suggesting that forecasters have little time to determine which mesovortices will become tornadic. More cases, however, need to be analyzed to address the generality of this result.
b. Mesovortex strength
Figure 10 also offers an assessment of mesovortex strength using mean vertical profiles of Vr for all nontornadic and tornadic vortices. Here Vr is defined as (Vmax − Vmin)/2 where Vmax and Vmin are the outbound and inbound mesovortex velocity maximum, respectively. In Fig. 10a, only data prior to tornadogenesis were used for the tornadic vortices. Clearly, the tornadic vortices were stronger than the nontornadic vortices from the surface up to 2.5 km AGL. When data at all times were included for the tornadic vortices (Fig. 10b), the strength differences were slightly larger below 3 km AGL, suggesting that the tornadic vortices noticeably strengthen only in the lowest 3 km during and after tornadogenesis. Further details illustrating mesovortex evolution are presented next.
c. Mesovortex evolution
To better understand how the mesovortices evolved with time, values of Vr, ΔVr/Δt, mesovortex couplet diameter, and azimuthal shear were calculated for all times that a mesovortex was manually detected in the single-Doppler data. Analysis of these variables for all mesovortices shown in Fig. 5 suggests that the tornadic vortices evolve differently than the nontornadic ones. Time–height diagrams of the aforementioned variables for two tornadic vortices are shown in Figs. 11 and 12. The mesovortex shown in Fig. 11 produced two tornadoes along with straight-line wind damage. The first tornado formed approximately 15 min after vortexgenesis. Examination of the Vr and ΔVr/Δt fields prior to tornadogenesis revealed a mesovortex that was deepening and intensifying rapidly. Consistent with the results in Fig. 10, the mesovortex was strongest near the ground, with Vr exceeding 16 m s−1 just prior to tornadogenesis. The evolution of Vr was very consistent with a circulation that appeared to be building upward as larger values of Vr were found at progressively higher altitudes at later times (Wakimoto and Wilson 1989; Trapp et al. 1999; Przybylinski et al. 2000). The ΔVr/Δt field showed that most of the intensification was occurring just prior to tornadogenesis and within the lowest two km with positive values exceeding 10 × 10−3 m s−2. Interestingly, after the first tornado dissipated, the mesovortex appeared to weaken slightly and then reintensified and deepened just prior to and during the genesis of the second tornado approximately 35 min after vortexgenesis. The ΔVr/Δt field was again positive from the surface to about 1.5 km AGL just prior to the genesis of the second tornado. The couplet diameter (distance between maximum inbound and outbound Doppler velocities) for this mesovortex was relatively small, approximately 2 km at the time of vortexgenesis and grew in size to about 4 km by the time it dissipated. The azimuthal shear values strengthened dramatically prior to the genesis of the first tornado largely due to the increase in Vr at this time. The azimuthal shear increased modestly prior to the second tornado.
Similar evolution can be seen in Fig. 12 for mesovortex 9. Prior to tornadogenesis, the mesovortex was observed to deepen rapidly and strengthen with time; ΔVr/Δt again showed that most of the intensification was occurring prior to tornadogenesis below 2 km. The couplet diameter decreased somewhat prior to tornadogenesis, suggesting that the circulation was contracting as it intensified. Thereafter, the couplet diameter increased in size. Again, the azimuthal shear increased prior to and during tornadogenesis due to a combination of the increasing mesovortex strength and decreasing diameter.
The results in Figs. 11 and 12 show a consistent evolution of these two tornadic mesovortices prior to tornadogenesis. The mesovortex deepened rapidly and strengthened, with most of the strengthening occurring in the lowest 2 km. The mesovortex size decreased slightly as the circulation intensified and then grew modestly until it dissipated. The azimuthal shear increased in response to the strengthening mesovortex and, to a lesser extent, the decreasing mesovortex size.
In stark contrast to the tornadic vortices, the nontornadic vortices evolved in a much different manner, as illustrated by two nontornadic vortices in Figs. 13 and 14. Relative to the tornadic vortices, the nontornadic circulations were much shallower and weaker as revealed by Vr and azimuthal shear. The Vr magnitudes rarely exceeded 12 m s−1 for the nontornadic vortices. Interestingly, the couplet diameter showed no consistent evolution from case to case.
Consistent with the conclusions of Atkins et al. (2004), the results presented in Figs. 10 –14 suggest that it may be possible to discriminate between the tornadic and nontornadic mesovortices within QLCSs. A clearer conceptual model is beginning to emerge in which tornadic mesovortices within QLCSs tend to be longer-lived, deepen quickly, and strengthen dramatically, especially in the lowest 2 km prior to tornadogenesis. Funk et al. (1999), Schmocker et al. (2000), and Atkins and Przybylinski (2000) have also observed tornadic mesovortices to deepen and strengthen prior to tornadogenesis. Nontornadic mesovortices tended to be weaker and shallower. This distinction is important since the greatest surface wind damage observed with the 10 June QLCS event was created by the tornadic mesovortices. Generally, the nontornadic vortices produced weak F0 or no damage.
d. Tornadic vortex association with the RIJ
An interesting observation in Fig. 5 is that many of the initial mesovortices to form within the QLCS were nontornadic and not associated with an apparent bow echo structure. All but one of the tornadic vortices (6, 8, 9, and 11), however, formed later and in the vicinity of the second bow echo. Tornadic mesovortex 2 was an unusually long-lived mesovortex and appeared to form in the vicinity of the first bow echo.
The relationship between the tornadic vortices and the second developing bow echo is shown in Fig. 15, wherein radar data are shown in both plan position indicator (PPI) and reconstructed range–height indicator (RHI) displays from 2230 to 2300 UTC, every 5 min. Also shown in Fig. 15 are the mesovortex tracks and damage survey data. Note that the first analysis time (2230 UTC) is half an hour earlier than the first time presented in Fig. 8.
At 2230 UTC, the convective line is visible in the northwestern part of the PPI display with convergence along the gust front apparent. A multicellular structure was observed in the RHI display. The three tornadic mesovortices (6, 8, and 9) had not yet formed. An important development within the QLCS happened in the next 5 min. At 2235 UTC, the PPI display showed the QLCS advancing eastward with little change in the overall structure. However, a developing, elevated RIJ was now apparent in the RHI display. Radial velocities increased by at least 22 m s−1 since 2230 UTC, exceeding 40 m s−1 in the RIJ. Thus, the radar data suggested that 2235 UTC represented the approximate genesis time for the RIJ associated with the second bow echo. The multicellular structure of the QLCS was again evident at 2235 UTC. It was also at this time that mesovortex 6 formed along the gust front north of, but in the general vicinity of, the developing RIJ. The structure of the QLCS began to change rapidly in the next 5 min. At 2240 UTC low-level flow behind the leading edge of the gust front increased, suggested by the increased areal coverage of the 20 m s−1 radial velocity isopleth from 2235 to 2240 UTC. This may have been a response to the RIJ that continued to expand in size and strengthen. The 29 m s−1 radial velocity isopleth was observed at low levels in the reconstructed RHI, suggesting that the RIJ had started to descend to the ground at this time, consistent with the strengthening flow behind the gust front. By 2245 UTC, the initiation time for tornadic mesovortex 8, the RIJ was more clearly descending to the ground. The low-level flow behind the gust front appeared to respond by strengthening and expanding in areal coverage. Indeed, mesovortex 8 formed in an area of strong convergence in the general vicinity of the RIJ. The expanding and strengthening trend for the low-level flow continued at 2250 and 2255 UTC. The RIJ also expanded horizontally in size and was observed to descend to the ground farther rearward of the convective system’s leading edge at 2255 UTC. Note that the third tornadic mesovortex (9) formed at 2255 UTC north of the RIJ core, however, along the gust front where the convergence had increased in response to the descending RIJ. The enhanced convergence near mesovortex 9 was more clearly seen at 2300 UTC. Twenty-five minutes later, the fourth tornadic mesovortex (11) formed at the bow apex (see Fig. 8d) along the gust front.
While interpretation of single-Doppler data can be difficult at times, the results presented in Fig. 15 suggest that the tornadic mesovortices associated with the second bow echo formed in close association with a developing RIJ. None of the tornadic vortices formed prior to the RIJ genesis time. Rather, they formed concurrently or within one hour after the RIJ formed and at a location along the gust front where convergence was locally enhanced by descending rear-to-front flow associated with the RIJ.
7. Discussion and conclusions
Detailed radar and damage survey data for the 10 June 2003 Saint Louis QLCS event during BAMEX has been presented in an effort to understand how bow echoes create surface wind damage. A discussion of the principal conclusions drawn from the data follows.
While previous studies have discussed how long swaths of straight-line wind damage may be generated at the bow apex, presumably by a descending RIJ, the results presented herein suggest another mechanism that is illustrated in the schematic diagram shown in Fig. 16. Damage survey results showed that the second bow echo, which formed east of Saint Louis on 10 June 2003, produced a continuous swath of damage approximately 50 km in length and 8 km in width. When superimposed on the KLSX radar data, this primary damage swath was not located at the bow apex and descending RIJ, rather it was located north of the bow apex and RIJ core, collocated with a low-level mesovortex formed on the QLCS gust front. In fact, the majority of the 50-km-long damage swath was located north of the bow apex, collocated with at least one mesovortex. Amazingly, very little damage was found at the bow apex and core of the relatively strong RIJ where radial velocities of 40 m s−1 were observed. Damage at the bow apex was observed only in close proximity to a mesovortex. It is possible, therefore, that if mesovortices had not formed within the second bow echo that the 50-km-long swath of straight-line wind damage may not have been created. These results are consistent with previous observational (e.g., Schmocker et al. 2000) and numerical modeling studies (Trapp and Weisman 2003), which discussed the potential role that mesovortices may play in generating damaging winds within bow echoes. An important question that cannot be addressed in this study concerns how often bow echo straight-line wind damage is produced by a descending RIJ at the bow apex as in Fujita’s conceptual model (Fig. 1) or by mesovortices. Partial resolution of this question will be addressed in a companion paper by Wheatley et al. (2005, manuscript submitted to Mon. Wea. Rev.) in which damage survey analyses for other BAMEX events are presented. Full resolution of this question will require detailed damage surveys to be performed for many more bow echo events.
Another important finding concerns the timing of damage within the second bow echo. The combined radar and damage survey analysis showed that most of the damage within the primary damage swath was created before obvious bow echo features were observed within the radar reflectivity field such as a pronounced bowing segment of cells, weak echo channel, and comma-head echo structure. These features were observed in the KLSX radar data, but after the primary damage swath had been created. Thus, forecasters must be aware that QLCS mesovortex damage may be occurring before the bow echo reflectivity structure is evident.
In addition to producing straight-line wind damage, 5 of the 11 observed mesovortices also produced tornadoes. Close examination of the data suggests that it may be possible to distinguish between tornadic and nontornadic mesovortices within QLCSs. This distinction is very important since the most intense damage observed with the 10 June QLCS event was produced by the tornadic mesovortices. Tornadic mesovortices tended to be longer-lived and deepened and intensified rapidly prior to tornadogenesis. The majority of mesovortex intensification occurred in the lowest 2 km. The mean time from vortexgenesis to tornadogenesis was about 12 min, though analysis of more cases is necessary to ascertain the generality of this result. The nontornadic mesovortices tended to be weaker and shallower. These results are consistent with those of Atkins et al. (2004) and suggest that a consistent conceptual model for QLCS tornadic mesovortex evolution is emerging from Doppler radar observations. The results also highlight the need for high temporal and spatial resolution data at low levels to properly monitor mesovortex development within QLCSs, thereby allowing a forecaster to discriminate between tornadic and nontornadic vortices. Given their small size (typically 1–10 km in diameter) and rapid evolution at low levels, current detection and warning efforts will be difficult when relying on WSR-88D data unless the circulation is close to the radar.
Within the QLCS, the tornadic vortices formed concurrently with or shortly after the genesis of the RIJ along a portion of the gust front where convergence had locally been enhanced (Figs. 15 and 16). The increased convergence appeared to be a result of enhanced low-level flow behind the gust front associated with the descending RIJ. Based on these observations, it is hypothesized that mesovortices within QLCSs are more likely to become tornadic if they form along a portion of the gust front that has been strengthened by the RIJ. Presumably, stronger stretching of vertical vorticity would be present along the enhanced portion of the gust front. Future modeling studies will help to test this hypothesis.
It is also interesting to speculate whether the mesoscale downdraft associated with the RIJ has any role in mesovortex genesis as has been suggested by Trapp and Weisman (2003). While the genesis mechanism(s) is (are) not well understood within observed bow echoes, previous investigators have suggested that they are formed through a release of a horizontal shearing instability (e.g., Forbes and Wakimoto 1983; Przybylinski 1995), tilting of horizontal baroclinic vorticity at the intersection point of a preexisting boundary (e.g., Przybylinski et al. 2000), or tilting of crosswise vorticity by convective-scale downdrafts (Trapp and Weisman 2003). Resolution of the mesovortex genesis mechanism(s) will require further analysis of high-resolution simulations and observed bow echo events and detailed dual-Doppler radar and damage survey observations.
Acknowledgments
Research results presented in this paper were supported by the National Science Foundation under Grants ATM-0100016, ATM-0233178 (NTA), and ATM-0233344 (RJT). Dusty Wheatley and Jon Chamberlain ably assisted with the ground survey effort. The PPI and SOLO software packages developed within the Mesoscale and Microscale Meteorology (MMM) and Atmospheric Technology Divisions (ATD), respectively, at the National Center for Atmospheric Research (NCAR) were used to visualize much of the radar data presented in this paper. The WATADS radar software developed at the National Severe Storms Laboratory was used for the construction of the reflectivity of Doppler velocity cross sections. Comments by an anonymous reviewer greatly improved an earlier version of the manuscript.
REFERENCES
Atkins, N. T., and R. W. Przybylinski, 2000: Radar and damage analysis of the 27 May 2000 tornadic derecho event. Preprints, 21st Conf. on Severe Local Storms, San Antonio, TX, Amer. Meteor. Soc., 567–570.
Atkins, N. T., J. M. Arnott, R. W. Przybylinski, R. A. Wolf, and B. D. Ketcham, 2004: Vortex structure and evolution within bow echoes. Part I: Single-Doppler and damage analysis of the 29 June 1998 derecho. Mon. Wea. Rev., 132 , 2224–2242.
Burgess, D. W., and B. F. Smull, 1990: Doppler radar observations of a bow echo associated with a long-track severe windstorm. Preprints, 16th Conf. on Severe Local Storms, Kananaskis Park, AB, Canada, Amer. Meteor. Soc., 203–208.
Davis, C., and Coauthors, 2004: The bow echo and MCV experiment (BAMEX): Observations and opportunities. Bull. Amer. Meteor. Soc., 85 , 1075–1093.
Evans, J. S., and C. A. Doswell, 2001: Examination of derecho environments using proximity soundings. Wea. Forecasting, 16 , 329–342.
Forbes, G. S., and R. M. Wakimoto, 1983: A concentrated outbreak of tornadoes, downbursts and microbursts, and implications regarding vortex classification. Mon. Wea. Rev., 111 , 220–236.
Fujita, T. T., 1971: Proposed mechanism of suction spots accompanied by tornadoes. Preprints, Seventh Conf. on Severe Local Storms, Kansas City, MO, Amer. Meteor. Soc., 208–213.
Fujita, T. T., 1978: Manual of downburst identification for project Nimrod. Satellite and Mesometeorology Research Paper 156, Dept. of Geophysical Sciences, University of Chicago, 104 pp.
Fujita, T. T., 1979: Objectives, operation, and results of project NIMROD. Preprints, 11th Conf. on Severe Local Storms, Kansas City, MO, Amer. Meteor. Soc., 259–266.
Fujita, T. T., 1981: Tornadoes and downbursts in the context of generalized planetary scales. J. Atmos. Sci., 38 , 1511–1534.
Fujita, T. T., 1992: Mystery of severe storms. SMRP Research Paper 239, University of Chicago, 298 pp.
Fujita, T. T., and R. M. Wakimoto, 1981: Five scales of airflow associated with a series of downbursts of 16 July 1980. Mon. Wea. Rev., 109 , 1438–1456.
Fujita, T. T., G. S. Forbes, and T. A. Umenhofer, 1976: Close-up view of 20 March 1976 tornadoes: Sinking cloud tops to suction vortices. Weatherwise, 29 , 117–131.
Funk, T. W., K. E. Darmofal, J. D. Kirkpatrick, V. L. DeWald, R. W. Przybylinski, G. K. Schmocker, and Y-J. Lin, 1999: Storm reflectivity and mesocyclone evolution associated with the 15 April 1994 squall line over Kentucky and southern Indiana. Wea. Forecasting, 14 , 976–993.
Hall, F., and R. D. Brewer, 1959: A sequence of tornado damage patterns. Mon. Wea. Rev., 87 , 207–216.
Houze, R. A., S. A. Rutledge, M. I. Biggerstaff, and B. F. Smull, 1989: Interpretation of Doppler weather radar displays of midlatitude mesoscale convective systems. Bull. Amer. Meteor. Soc., 70 , 608–619.
Johns, R. H., 1993: Meteorological conditions associated with bow echo development in convective storms. Wea. Forecasting, 8 , 294–299.
Johns, R. H., and W. D. Hirt, 1987: Derechos: Widespread convectively induced wind-storms. Wea. Forecasting, 2 , 32–49.
Jorgensen, D. P., 1993: Meteorological conditions associated with bow echo development in convective storms. Wea. Forecasting, 8 , 294–299.
Jorgensen, D. P., and B. F. Smull, 1993: Mesovortex circulations seen by airborne Doppler radar within a bow-echo mesoscale convective system. Bull. Amer. Meteor. Soc., 74 , 2146–2157.
Lafore, J-P., and M. W. Moncrieff, 1989: A numerical investigation of the organization and interaction of the convective and stratiform regions of tropical squall lines. J. Atmos. Sci., 46 , 521–544.
Miller, D. J., and R. H. Johns, 2000: A detailed look at extreme wind damage in derecho events. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., 52–55.
Pfost, R. L., and A. E. Gerard, 1997: “Bookend vortex”—Induced tornadoes along the Natchez Trace. Wea. Forecasting, 12 , 572–580.
Przybylinski, R. W., 1995: The bow echo: Observations, numerical simulations, and severe weather detection methods. Wea. Forecasting, 10 , 203–218.
Przybylinski, R. W., and W. J. Gery, 1983: The reliability of the bow echo as an important severe weather signature. Preprints, 13th Conf. on Severe Local Storms, Tulsa, OK, Amer. Meteor. Soc., 270–273.
Przybylinski, R. W., and D. M. DeCaire, 1985: Radar signatures associated with the derecho, a type of mesoscale convective system. Preprints, 14th Conf. on Severe Local Storms, Indianapolis, IN, Amer. Meteor. Soc., 228–231.
Przybylinski, R. W., G. K. Schmocker, and Y-J. Lin, 2000: A study of storm and vortex morphology during the intensifying stage of severe wind mesoscale convective systems. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., 173–176.
Schmidt, J. M., and W. R. Cotton, 1989: A high plains squall line associated with severe surface winds. J. Atmos. Sci., 46 , 281–302.
Schmocker, G. K., R. W. Przybylinski, and Y. J. Lin, 1996: Forecasting the initial onset of damaging downburst winds associated with a mesoscale convective system (MCS) using the midlatitude radial convergence (MARC) signature. Preprints, 15th Conf. on Weather Analysis and Forecasting, Norfolk, VA, Amer. Meteor. Soc., 306–311.
Schmocker, G. K., R. W. Przybylinski, and E. N. Rasmussen, 2000: The severe bow echo event of 14 June 1998 over the mid-Mississippi valley region: A case of vortex development near the intersection of a preexisting boundary and a convective line. Preprints, 20th Conf. on Severe Local Storms, Orlando, FL, Amer. Meteor. Soc., 169–172.
Skamarock, W. C., M. L. Weisman, and J. B. Klemp, 1994: Three-dimensional evolution of simulated long-lived squall lines. J. Atmos. Sci., 51 , 2563–2584.
Smull, B. F., and R. A. Houze Jr., 1985: A midlatitude squall line with a trailing region of stratiform rain: Radar and satellite observations. Mon. Wea. Rev., 113 , 117–133.
Smull, B. F., and R. A. Houze Jr., 1987: Rear-inflow in squall lines with trailing stratiform precipiation. Mon. Wea. Rev., 115 , 2869–2889.
Trapp, R. J., and M. L. Weisman, 2003: Low-level mesovortices within squall lines and bow echoes. Part II: Their genesis and implications. Mon. Wea. Rev., 131 , 2804–2823.
Trapp, R. J., E. D. Mitchell, G. A. Tipton, D. W. Effertz, A. I. Watson, D. L. Andra Jr., and M. A. Magsig, 1999: Descending and non-descending tornadic vortex signatures detected by WSR-88Ds. Wea. Forecasting, 14 , 625–639.
Trapp, R. J., S. A. Tessendorf, E. S. Godfrey, and H. E. Brooks, 2005: Tornadoes from squall lines and bow echoes. Part I: Climatological distribution. Wea. Forecasting, 20 , 23–34.
Wakimoto, R. M., 1983: The West Bend, Wisconsin Storm of 4 April 1981: A problem in operational meteorology. J. Appl. Meteor., 22 , 181–189.
Wakimoto, R. M., 2001: Severe Convective Storms. Meteor. Monogr., No. 50, Amer. Meteor. Soc., 561 pp.
Wakimoto, R. M., and J. W. Wilson, 1989: Non-supercell tornadoes. Mon. Wea. Rev., 117 , 1113–1140.
Weisman, M. L., 1993: The genesis of severe, long-lived bow echoes. J. Atmos. Sci., 50 , 645–670.
Weisman, M. L., and J. B. Klemp, 1982: The dependence of numerically simulated convective storms on vertical wind shear and buoyancy. Mon. Wea. Rev., 110 , 504–520.
Weisman, M. L., and C. A. Davis, 1998: Mechanisms for the generation of mesoscale vortices within quasi-linear convective systems. J. Atmos. Sci., 55 , 2603–2622.
Weisman, M. L., and R. J. Trapp, 2003: Low-level mesovortices within squall lines and bow echoes: Part I. Overview and dependence on environmental shear. Mon. Wea. Rev., 131 , 2779–2803.
Schematic diagram illustrating bow echo evolution. Figure adapted from Fujita (1979).
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Schematic diagram illustrating bow echo evolution. Figure adapted from Fujita (1979).
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Schematic diagram illustrating bow echo evolution. Figure adapted from Fujita (1979).
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
The 1200 UTC upper-air analysis at 850 and 500 hPa. Solid and long-dashed lines show the geopotential height (gpm) and temperature fields (°C), respectively. Short-dashed lines are equivalent potential temperature, with values greater than 330 K shaded gray. Winds are also shown (half barb = 2.5 m s−1; full barb = 5 m s−1; flag = 25 m s−1).
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
The 1200 UTC upper-air analysis at 850 and 500 hPa. Solid and long-dashed lines show the geopotential height (gpm) and temperature fields (°C), respectively. Short-dashed lines are equivalent potential temperature, with values greater than 330 K shaded gray. Winds are also shown (half barb = 2.5 m s−1; full barb = 5 m s−1; flag = 25 m s−1).
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
The 1200 UTC upper-air analysis at 850 and 500 hPa. Solid and long-dashed lines show the geopotential height (gpm) and temperature fields (°C), respectively. Short-dashed lines are equivalent potential temperature, with values greater than 330 K shaded gray. Winds are also shown (half barb = 2.5 m s−1; full barb = 5 m s−1; flag = 25 m s−1).
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
GOES-E visible satellite imagery at (a) 1600, (b) 1800, (c) 2000, and (d) 2200 UTC. Frontal and cold pool locations are superimposed along with surface and radar data. Surface data include temperature and dewpoint (both in °C) along with winds following the convention in Fig. 2. Pressure (hPa) and equivalent potential temperature (K) are also contoured as solid and long-dashed black lines, respectively. Radar data from the KEAX and KLSX radars are also shown with values of 35 and 50 dBZ contoured and filled black, respectively.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
GOES-E visible satellite imagery at (a) 1600, (b) 1800, (c) 2000, and (d) 2200 UTC. Frontal and cold pool locations are superimposed along with surface and radar data. Surface data include temperature and dewpoint (both in °C) along with winds following the convention in Fig. 2. Pressure (hPa) and equivalent potential temperature (K) are also contoured as solid and long-dashed black lines, respectively. Radar data from the KEAX and KLSX radars are also shown with values of 35 and 50 dBZ contoured and filled black, respectively.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
GOES-E visible satellite imagery at (a) 1600, (b) 1800, (c) 2000, and (d) 2200 UTC. Frontal and cold pool locations are superimposed along with surface and radar data. Surface data include temperature and dewpoint (both in °C) along with winds following the convention in Fig. 2. Pressure (hPa) and equivalent potential temperature (K) are also contoured as solid and long-dashed black lines, respectively. Radar data from the KEAX and KLSX radars are also shown with values of 35 and 50 dBZ contoured and filled black, respectively.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Sounding and hodograph at 1800 UTC from Springfield, MO. Springfield is located at the star in Fig. 3b. Temperature and dewpoint are plotted on the sounding as solid black lines, while a surface-based parcel path is shown as the short-dashed line. The gray area represents the CAPE for the lifted parcel. Winds (half barb = 5 m s−1; full barb = 10 m s−1) are also shown.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Sounding and hodograph at 1800 UTC from Springfield, MO. Springfield is located at the star in Fig. 3b. Temperature and dewpoint are plotted on the sounding as solid black lines, while a surface-based parcel path is shown as the short-dashed line. The gray area represents the CAPE for the lifted parcel. Winds (half barb = 5 m s−1; full barb = 10 m s−1) are also shown.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Sounding and hodograph at 1800 UTC from Springfield, MO. Springfield is located at the star in Fig. 3b. Temperature and dewpoint are plotted on the sounding as solid black lines, while a surface-based parcel path is shown as the short-dashed line. The gray area represents the CAPE for the lifted parcel. Winds (half barb = 5 m s−1; full barb = 10 m s−1) are also shown.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Radar reflectivity data from the KLSX radar at 2059, 2159, 2300, and 2356 UTC. Solid and thick-dashed lines represent the locations of radar-detected tornadic and nontornadic mesovortices, respectively. Start and end times (UTC) for all mesovortices are also shown. Thick solid lines along the tornadic mesovortex paths represent the location of observed tornado damage. Thin-dashed lines are county boundaries. The long-dashed line separates the 2059 and 2159 UTC radar data.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Radar reflectivity data from the KLSX radar at 2059, 2159, 2300, and 2356 UTC. Solid and thick-dashed lines represent the locations of radar-detected tornadic and nontornadic mesovortices, respectively. Start and end times (UTC) for all mesovortices are also shown. Thick solid lines along the tornadic mesovortex paths represent the location of observed tornado damage. Thin-dashed lines are county boundaries. The long-dashed line separates the 2059 and 2159 UTC radar data.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Radar reflectivity data from the KLSX radar at 2059, 2159, 2300, and 2356 UTC. Solid and thick-dashed lines represent the locations of radar-detected tornadic and nontornadic mesovortices, respectively. Start and end times (UTC) for all mesovortices are also shown. Thick solid lines along the tornadic mesovortex paths represent the location of observed tornado damage. Thin-dashed lines are county boundaries. The long-dashed line separates the 2059 and 2159 UTC radar data.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Damage survey analysis for both bow echoes. Straight-line wind damage areas of F0 or greater intensity are shaded gray. Black lines delineate tornado tracks. Black dashed lines delineate the 35-dBZ contour illustrating the QLCS position west and east of Saint Louis at times when the bow echoes were producing surface wind damage. Gray dashed lines are county boundaries. The BAMEX Operations Center is located at the star. The Saint Louis city boundary is also shown.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Damage survey analysis for both bow echoes. Straight-line wind damage areas of F0 or greater intensity are shaded gray. Black lines delineate tornado tracks. Black dashed lines delineate the 35-dBZ contour illustrating the QLCS position west and east of Saint Louis at times when the bow echoes were producing surface wind damage. Gray dashed lines are county boundaries. The BAMEX Operations Center is located at the star. The Saint Louis city boundary is also shown.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Damage survey analysis for both bow echoes. Straight-line wind damage areas of F0 or greater intensity are shaded gray. Black lines delineate tornado tracks. Black dashed lines delineate the 35-dBZ contour illustrating the QLCS position west and east of Saint Louis at times when the bow echoes were producing surface wind damage. Gray dashed lines are county boundaries. The BAMEX Operations Center is located at the star. The Saint Louis city boundary is also shown.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Detailed damage survey analysis over southwestern Illinois. Straight-line wind damage areas F0 or greater are shaded in light gray while tornado tracks are delineated in black. Thin arrows are damaging wind streamlines. County lines and town boundaries are shown as long-dashed and thin solid black lines, respectively. Areas of water are shaded dark gray. The M’s denote the location of microburst damage.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Detailed damage survey analysis over southwestern Illinois. Straight-line wind damage areas F0 or greater are shaded in light gray while tornado tracks are delineated in black. Thin arrows are damaging wind streamlines. County lines and town boundaries are shown as long-dashed and thin solid black lines, respectively. Areas of water are shaded dark gray. The M’s denote the location of microburst damage.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Detailed damage survey analysis over southwestern Illinois. Straight-line wind damage areas F0 or greater are shaded in light gray while tornado tracks are delineated in black. Thin arrows are damaging wind streamlines. County lines and town boundaries are shown as long-dashed and thin solid black lines, respectively. Areas of water are shaded dark gray. The M’s denote the location of microburst damage.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Damage survey data from Fig. 7 superimposed on KLSX 0.5° radar reflectivity data (dBZ) and ground-relative radial velocities [m s−1; cool (warm) colors are flow toward (away) from the radar] at (a) 2300, (b) 2310, (c) 2320, (d) 2330, and (e) 2340 UTC. In all panels, the positions of tornadic and nontornadic mesovortices are shown as solid and dashed black lines, respectively. Couplet locations and time stamps are indicated along the mesovortex tracks. Tornado paths are colored purple. The blue contour represents the 25 m s−1 radial velocity isopleth observed at 2.4° with values greater 30 m s−1 filled. The inset radial velocity data in (a) are relative to mesovortex 6. The thick-dashed line represents the approximate location of the RIJ. Thin-dashed lines are the range–azimuth grid lines for the KLSX radar. Surface data are also plotted in (a), showing the winds, temperature, and dewpoint following the convention used in Fig. 3. The starred location is MAA. The inset diagram in (a) displays storm-relative radial velocities in the area of the dashed box.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Damage survey data from Fig. 7 superimposed on KLSX 0.5° radar reflectivity data (dBZ) and ground-relative radial velocities [m s−1; cool (warm) colors are flow toward (away) from the radar] at (a) 2300, (b) 2310, (c) 2320, (d) 2330, and (e) 2340 UTC. In all panels, the positions of tornadic and nontornadic mesovortices are shown as solid and dashed black lines, respectively. Couplet locations and time stamps are indicated along the mesovortex tracks. Tornado paths are colored purple. The blue contour represents the 25 m s−1 radial velocity isopleth observed at 2.4° with values greater 30 m s−1 filled. The inset radial velocity data in (a) are relative to mesovortex 6. The thick-dashed line represents the approximate location of the RIJ. Thin-dashed lines are the range–azimuth grid lines for the KLSX radar. Surface data are also plotted in (a), showing the winds, temperature, and dewpoint following the convention used in Fig. 3. The starred location is MAA. The inset diagram in (a) displays storm-relative radial velocities in the area of the dashed box.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Damage survey data from Fig. 7 superimposed on KLSX 0.5° radar reflectivity data (dBZ) and ground-relative radial velocities [m s−1; cool (warm) colors are flow toward (away) from the radar] at (a) 2300, (b) 2310, (c) 2320, (d) 2330, and (e) 2340 UTC. In all panels, the positions of tornadic and nontornadic mesovortices are shown as solid and dashed black lines, respectively. Couplet locations and time stamps are indicated along the mesovortex tracks. Tornado paths are colored purple. The blue contour represents the 25 m s−1 radial velocity isopleth observed at 2.4° with values greater 30 m s−1 filled. The inset radial velocity data in (a) are relative to mesovortex 6. The thick-dashed line represents the approximate location of the RIJ. Thin-dashed lines are the range–azimuth grid lines for the KLSX radar. Surface data are also plotted in (a), showing the winds, temperature, and dewpoint following the convention used in Fig. 3. The starred location is MAA. The inset diagram in (a) displays storm-relative radial velocities in the area of the dashed box.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
(a) Tornado damage survey analysis of the damage path near Caseyville, IL, collocated with mesovortex 9 in Fig. 8. (b)–(e) Photographs illustrating damage along the tornado path. Within the survey analysis frame, F scale is shaded gray, roads are shown as black lines, and the outline of Caseyville is dashed. Approximate start and end times of tornado damage are also indicated. Short black arrows in (c) and (e) indicate the direction of tree fall.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
(a) Tornado damage survey analysis of the damage path near Caseyville, IL, collocated with mesovortex 9 in Fig. 8. (b)–(e) Photographs illustrating damage along the tornado path. Within the survey analysis frame, F scale is shaded gray, roads are shown as black lines, and the outline of Caseyville is dashed. Approximate start and end times of tornado damage are also indicated. Short black arrows in (c) and (e) indicate the direction of tree fall.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
(a) Tornado damage survey analysis of the damage path near Caseyville, IL, collocated with mesovortex 9 in Fig. 8. (b)–(e) Photographs illustrating damage along the tornado path. Within the survey analysis frame, F scale is shaded gray, roads are shown as black lines, and the outline of Caseyville is dashed. Approximate start and end times of tornado damage are also indicated. Short black arrows in (c) and (e) indicate the direction of tree fall.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
(a) Vertical profiles of Vr [defined as (Vmax − Vmin)/2, where Vmax and Vmin are the outbound and inbound mesovortex velocity maximum, respectively] for all tornadic and nontornadic vortices. Only data prior to tornadogenesis were considered for the tornadic vortices. Error bars represent one standard deviation for all points at the respective level. (b) Same as in (a) except all data were considered for the tornadic vortices. (c) Histogram of mesovortex lifetimes for all vortices shown in Fig. 5.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
(a) Vertical profiles of Vr [defined as (Vmax − Vmin)/2, where Vmax and Vmin are the outbound and inbound mesovortex velocity maximum, respectively] for all tornadic and nontornadic vortices. Only data prior to tornadogenesis were considered for the tornadic vortices. Error bars represent one standard deviation for all points at the respective level. (b) Same as in (a) except all data were considered for the tornadic vortices. (c) Histogram of mesovortex lifetimes for all vortices shown in Fig. 5.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
(a) Vertical profiles of Vr [defined as (Vmax − Vmin)/2, where Vmax and Vmin are the outbound and inbound mesovortex velocity maximum, respectively] for all tornadic and nontornadic vortices. Only data prior to tornadogenesis were considered for the tornadic vortices. Error bars represent one standard deviation for all points at the respective level. (b) Same as in (a) except all data were considered for the tornadic vortices. (c) Histogram of mesovortex lifetimes for all vortices shown in Fig. 5.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Time–height profiles of Vr, ΔVr/Δt, mesovortex couplet diameter, and azimuthal shear for tornadic mesovortex 6. Here Vr is contoured every 4 m s−1 with values greater than 16 m s−1 shaded gray, and ΔVr/Δt (×10−3 m s−2) is contoured every 5 × 10−3 m s−2 with values greater than 5 × 10−3 m s−2 shaded gray. Couplet diameter is contoured every 2 km while azimuthal shear (×10−3 s−1) is contoured every 2 × 10−3 s−1 with values greater than 12 × 10−3 s−1 shaded gray. A distance scale relative to KLSX along with a time scale relative to vortexgenesis are indicated along the horizontal axis. The intensity and times of tornado and straight-line wind damage produced by the mesovortex are shown as black and gray bars, respectively, along the horizontal axis.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Time–height profiles of Vr, ΔVr/Δt, mesovortex couplet diameter, and azimuthal shear for tornadic mesovortex 6. Here Vr is contoured every 4 m s−1 with values greater than 16 m s−1 shaded gray, and ΔVr/Δt (×10−3 m s−2) is contoured every 5 × 10−3 m s−2 with values greater than 5 × 10−3 m s−2 shaded gray. Couplet diameter is contoured every 2 km while azimuthal shear (×10−3 s−1) is contoured every 2 × 10−3 s−1 with values greater than 12 × 10−3 s−1 shaded gray. A distance scale relative to KLSX along with a time scale relative to vortexgenesis are indicated along the horizontal axis. The intensity and times of tornado and straight-line wind damage produced by the mesovortex are shown as black and gray bars, respectively, along the horizontal axis.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Time–height profiles of Vr, ΔVr/Δt, mesovortex couplet diameter, and azimuthal shear for tornadic mesovortex 6. Here Vr is contoured every 4 m s−1 with values greater than 16 m s−1 shaded gray, and ΔVr/Δt (×10−3 m s−2) is contoured every 5 × 10−3 m s−2 with values greater than 5 × 10−3 m s−2 shaded gray. Couplet diameter is contoured every 2 km while azimuthal shear (×10−3 s−1) is contoured every 2 × 10−3 s−1 with values greater than 12 × 10−3 s−1 shaded gray. A distance scale relative to KLSX along with a time scale relative to vortexgenesis are indicated along the horizontal axis. The intensity and times of tornado and straight-line wind damage produced by the mesovortex are shown as black and gray bars, respectively, along the horizontal axis.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as in Fig. 11 but for tornadic mesovortex 9.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as in Fig. 11 but for tornadic mesovortex 9.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as in Fig. 11 but for tornadic mesovortex 9.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as Fig. 11 except for nontornadic mesovortex 4. The ΔVr/Δt field is not shown due to a lack of data.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as Fig. 11 except for nontornadic mesovortex 4. The ΔVr/Δt field is not shown due to a lack of data.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as Fig. 11 except for nontornadic mesovortex 4. The ΔVr/Δt field is not shown due to a lack of data.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as Fig. 13 except for nontornadic mesovortex 7.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as Fig. 13 except for nontornadic mesovortex 7.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Same as Fig. 13 except for nontornadic mesovortex 7.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
KLSX radar reflectivity (dBZ) and ground-relative radial velocity (m s−1) data in plan view (0.5°), and vertical cross sections from 2230 to 2300 UTC, every 5 min. Reflectivity data are in color. Radial velocities are contoured in light blue with solid (dashed) contours representing flow away from (toward) the radar. Radial velocities are contoured every 10 m s−1 (values greater than 30 m s−1 are shaded) in plan view and every 11 m s−1 (values greater than 40 m s−1 are shaded) in the vertical cross sections. Within the plan view plots, the thick black line represents the location of the vertical cross section at the respective times. Thin solid and dashed lines with time markers represent the locations of the tornadic and nontornadic mesovortices, respectively. Damage survey data from Fig. 7 are superimposed with tornado tracks colored purple and straight-line wind damage in gray.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
KLSX radar reflectivity (dBZ) and ground-relative radial velocity (m s−1) data in plan view (0.5°), and vertical cross sections from 2230 to 2300 UTC, every 5 min. Reflectivity data are in color. Radial velocities are contoured in light blue with solid (dashed) contours representing flow away from (toward) the radar. Radial velocities are contoured every 10 m s−1 (values greater than 30 m s−1 are shaded) in plan view and every 11 m s−1 (values greater than 40 m s−1 are shaded) in the vertical cross sections. Within the plan view plots, the thick black line represents the location of the vertical cross section at the respective times. Thin solid and dashed lines with time markers represent the locations of the tornadic and nontornadic mesovortices, respectively. Damage survey data from Fig. 7 are superimposed with tornado tracks colored purple and straight-line wind damage in gray.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
KLSX radar reflectivity (dBZ) and ground-relative radial velocity (m s−1) data in plan view (0.5°), and vertical cross sections from 2230 to 2300 UTC, every 5 min. Reflectivity data are in color. Radial velocities are contoured in light blue with solid (dashed) contours representing flow away from (toward) the radar. Radial velocities are contoured every 10 m s−1 (values greater than 30 m s−1 are shaded) in plan view and every 11 m s−1 (values greater than 40 m s−1 are shaded) in the vertical cross sections. Within the plan view plots, the thick black line represents the location of the vertical cross section at the respective times. Thin solid and dashed lines with time markers represent the locations of the tornadic and nontornadic mesovortices, respectively. Damage survey data from Fig. 7 are superimposed with tornado tracks colored purple and straight-line wind damage in gray.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Schematic model of damage produced by the bow echo observed on 10 Jun 2003 east of Saint Louis.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Schematic model of damage produced by the bow echo observed on 10 Jun 2003 east of Saint Louis.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
Schematic model of damage produced by the bow echo observed on 10 Jun 2003 east of Saint Louis.
Citation: Monthly Weather Review 133, 8; 10.1175/MWR2973.1
