Tornadoes Strike Near Fallston
and Lincolnton on April 19, 2008
Patrick D. Moore
A funnel cloud was observed near Lawndale, North Carolina, shortly before a tornado touched down near Fallston on April 19, 2008. The image was provided by WSOC-TV Charlotte and taken by an unknown viewer. Used by permission.
Author's Note: The following report has not been subjected to the scientific peer review process.
An unusually small but well-organized thunderstorm developed over Cleveland County, North Carolina, during the evening of Saturday, April 19, 2008. The storm produced two tornadoes as it moved across the eastern part of Cleveland County and the western part of Lincoln County. The first tornado touched down at 805 pm EDT 19 April (0005 UTC 20 April) near the Double Shoals community and lifted at 811 pm (0011 UTC) near Fallston. Houses and trees were damaged along Double Shoals Road and Fallston-Waco Road. The second tornado touched down near the intersection of State Highways 182 and 274 in southwest Lincoln County at 820 pm (0020 UTC) and produced an intermittent damage path that extended to the northwest side of Lincolnton. The tornado damaged a chicken farm on Guy Heavener Road, and trees and homes on Howard's Creek School Road and Betterbrook Lane. The second tornado dissipated at 837 pm (0037 UTC). Both tornadoes produced damage that was rated EF-1 on the Enhanced Fujita Scale. A few reports of large hail were received as the storms moved across the northwest Piedmont of North Carolina, but overall the severe weather was relatively localized (Fig. 1). [Note: all times hereafter will be referenced to Universal Time Coordinated (UTC), which is Eastern Daylight Time (EDT) plus four hours.]
Figure 1. Severe thunderstorm and tornado reports received by the Storm Prediction Center for the 24 hours ending 1200 UTC 20 April.
The events of 19 April 2008 were interesting in that the storms developed in an environment that was initially thought to be not conducive to thunderstorms with vigorous updrafts, because of a lack of instability and buoyancy. In fact, several small thunderstorms developed and they exhibited many of the same characteristics of storms that are classified as supercells (Doswell and Burgess 1993). Even more interesting was the absence of cloud-to-ground lightning in the thunderstorm for 20 minutes prior to tornadogenesis and while the tornadoes were on the ground. Similar characteristics were noted by Markowski and Straka (2000) in a study of a miniature tornadic supercell thunderstorm event in central Oklahoma.
2. Synoptic Features and Pre-Storm Environment
A vertically stacked low pressure system was centered over Illinois on the morning of 19 April. The 500 mb analysis at 1200 UTC showed a mid-level jet of 50 to 80 kt winds bringing colder temperatures around the bottom of the upper low and toward the Carolinas (Fig. 2). Meanwhile, a short wave trough manifested as a swirl in the water vapor satellite imagery over Missouri was expected to rotate across the Tennessee Valley and into the western Carolinas late in the day (Fig. 3). Of note on the 850 mb analysis was the tongue of moisture moving up from the northeastern Gulf of Mexico and stretching across Georgia (Fig. 4). The moisture was moving northward ahead of an approaching cold front that stretched from a surface low located over northern Illinois, down across the Ohio and Tennessee valleys to the Florida Panhandle (Fig. 5).
Figure 2. SPC objective analysis of 500 mb geopotential height, temperature, and wind barbs at 1200 UTC 19 April. Click to enlarge.
Figure 3. GOES-12 water vapor imagery at 1145 UTC on 19 April. The warmer colors represent relatively dry air at mid-levels while the cooler or white shades represent moisture at mid levels. Click to enlarge.
Figure 4. SPC objective analysis of 850 mb geopotential height, temperature, dewpoint, and wind barbs at 1200 UTC 19 April. Click to enlarge.
Figure 5. Surface fronts and pressure analysis from the Hydrometeorological Prediction Center at 1200 UTC 19 April. Click to enlarge.
The combination of strong winds at mid levels, dynamic forcing from the passage of the upper level short wave trough, and a conditionally unstable atmosphere ahead of the surface front was expected to support the formation of a few severe thunderstorms in the afternoon and evening. However, the threat was not enough to warrant a Slight Risk on the updated Day 1 Convective Outlook issued at 1555 UTC by the Storm Prediction Center (SPC). A relative lack of instability was the limiting factor, mainly due to extensive morning cloudiness ahead of the front and dewpoints only in the lower to middle 50s at around 1600 UTC.
The perceived threat of severe weather had not changed by the late afternoon. The Day 1 Convective Outlook updated at 1934 UTC continued to indicate the possibility of isolated strong to severe thunderstorms. Dewpoints climbed into the upper 50s and lower 60s in some locations immediately ahead of the front which was crossing the Appalachians at 2100 UTC (Fig. 6). Cloud cover thinned across the North Carolina Foothills which allowed temperatures to reach the middle 60s (Fig. 7). However, low levels of the atmosphere were only weakly buoyant with Convective Available Potential Energy (CAPE) under 500 J/kg.
Figure 6. Regional surface fronts and pressure analysis at 2100 UTC 19 April. Observations are shown using the conventional station model. Click to enlarge.
Figure 7. GOES-12 visible satellite image at 2045 UTC 19 April. Click to enlarge.
By the early part of the evening, the front triggered a line of showers and thunderstorms across the central Appalachians and northwest North Carolina (Fig. 8). Meanwhile, an isolated thunderstorm had formed over the southern Foothills of North Carolina. By all measures, buoyancy and instability ahead of the front were weak at 2300 UTC (Fig. 9). However, east of the mountains the effective bulk shear surpassed 40 kt (Fig. 10) and Storm Relative Helicity (SRH) was nearly 150 m2/s2 (Fig. 11). Both values were in the range that supported the development of supercell thunderstorms. An upper air sounding released into the pre-convective environment near Greensboro, North Carolina (GSO), around 2300 UTC showed similar low values of CAPE and high values of SRH and shear (Fig. 12).
Figure 8. Regional radar reflectivity mosaic at 2300 UTC 19 April 2008. Click to enlarge.
Figure 9. SPC objective analysis of most unstable CAPE (contours) and lifted parcel level (shaded) at 2300 UTC on 19 April 2008. Click to enlarge.
Figure 10. SPC objective analysis of effective bulk shear (contours and barbs) at 2300 UTC 19 April 2008. Click to enlarge.
Figure 11. SPC objective analysis of SRH in the 0-1 km layer (contours) and storm motion (barbs) at 2300 UTC on 19 April 2008. Click to enlarge.
Figure 12. Skew T - log P diagram (upper left) and hodograph (upper right) of upper air sounding taken at GSO at 0000 UTC on 20 April 2008. A table of convective parameters and indices is shown at the bottom. Click to enlarge.
3. Radar observations
Convective initiation occurred around 2130 UTC just to the northwest of the National Weather Service office located at the Greenville - Spartanburg Airport (GSP). The showers moved northeast over the next hour and gradually organized into a thunderstorm over southern Rutherford County, North Carolina, by 2300 UTC. The thunderstorm crossed into west central Cleveland County around 2338 UTC, as seen by the Weather Surveillance Radar at KGSP (referred to as the KGSP radar) (Fig. 13). At this time the storm produced a cloud-to-ground lightning strike, which would be its last for the next hour.
Figure 13. KGSP composite radar reflectivity at 2338 UTC on 19 April 2008. Intensity of precipitation is given by the color table at the lower right. Click to enlarge.
The thunderstorm organized rapidly over the next ten minutes. Cyclonic rotation was evident in the storm relative velocity observed on the southern flank of the storm on the lowest elevation scan from the KGSP radar at 2348 UTC (Fig. 14). The corresponding reflectivity image from that time showed an appendage on the southwest flank of the storm (Fig. 15), but interpretation of the data was difficult because of the small size of the cell (about 8 miles in diameter) and the resolution of the radar beam. The reflectivity appendage was more apparent when the data were viewed using the GR2Analyst software package, which smoothed the data (Fig. 16a). Consideration of the lowest four elevation scans from KGSP at that time revealed a weak echo region underneath a suspended high reflectivity core, indicative of an inflow of unstable air feeding an updraft (Fig. 16). A cross-section through the high reflectivity core shows the shallow nature of the cell (Fig. 17). The sounding in figure 12 showed the freezing level just above 9,000 feet. The high reflectivity (i.e. greater than 40 dbZ) in the Cleveland County storm extended only to about 15,000 feet. This is well below the rule of thumb forecasters often use to identify cells that could produce cloud-to-ground lightning, which is 40 dbZ radar echoes 10,000 feet above the freezing level.
Figure 14. KGSP storm relative motion on the 0.5 degree scan at 2348 UTC 19 April 2008. The warmer colors represent motion away from the radar and the cooler colors represent motion toward the radar, which is located off the lower left corner of the image. Click to enlarge.
Figure 15. KGSP base reflectivity on the 0.5 degree scan at 2348 UTC 19 April 2008. The color table depicts the intensity of the precipitation. Click to enlarge.
Figure 16. KGSP base reflectivity at 0.5 degrees (a), 1.5 degrees (b), 2.4 degrees (c), and 3.4 degrees (d) at 2348 UTC on 19 April 2008. The color bar depicts the precipitation intensity. The white arrow in (a) shows the location of an inflow notch to the right of the reflectivity appendage. The line in (d) denotes the location of the vertical cross section in Figure 17. Images created using GR2Analyst. Click to enlarge.
Figure 17. KGSP base reflectivity vertical cross section at 2348 UTC on 19 April 2008. The cross section was cut along the line shown in Figure 16 (d). Image created with GR2Analyst. Click to enlarge.
The rotational velocity, defined as the average of the absolute value of the maximum inbound and maximum outbound velocity, remained around 20 kts for the next 20 minutes (Fig. 18), which is considered weak to minimal strength at a distance of 40 to 50 nautical miles. In the mean time, the first tornado touched down in Cleveland County near the Double Shoals community at 0005 UTC on 20 April. A significant increase in rotation, in this case taken as a 50 percent increase in rotational velocity, did not occur until the 0012 UTC scan (Fig. 19), by which time the first tornado had already lifted near Fallston.
Figure 18. KGSP rotational velocity from 2313 UTC 19 April to 0047 UTC 20 April. The distance to the radar was approximately 30 to 60 nautical miles during the time period. The times that tornadoes were on the ground are indicated by the pink bars.
Figure 19. As in Figure 14, except for 0012 UTC on 20 April 2008. Click to enlarge.
The location of the Double Shoals - Fallston tornado was nearly equidistant from the KGSP radar and the Terminal Doppler Weather Radar north of the Charlotte - Douglas International Airport (referred to as the TCLT radar). However, the TCLT radar did not offer any additional meaningful clues in the minutes leading up to the Double Shoals - Fallston tornado, in spite of the better resolution of the data. Rotational velocities were also in the weak to minimal mesocyclone range through approximately 0006 UTC (Fig. 20). A significant increase in strength did not occur until the tornado was already on the ground.
Figure 20. As in Figure 19, except for the TCLT radar. The distance from the storm to the radar was approximately 20 to 45 nautical miles during the time period. Range folded scans were indicated by a zero velocity.
A Tornado Warning was issued at 0015 UTC for northeastern Cleveland County and western Lincoln County based on the upward trend of the rotation observed on both the KGSP and TCLT radars. Although the rotation weakened over the next few minutes, a second tornado touched down in the southwestern corner of Lincoln County at 0020 UTC, near the intersection of State Highways 182 and 274. This tornado had an intermittent path along the ground that lasted for 17 minutes before it finally lifted on the west side of Lincolnton. The radar presentation at low levels was not as impressive as earlier, but the storm gradually gained strenth during the lifetime of the Western Lincoln tornado, as evidenced by the higher reflectivity in the core of the storm and its greater vertical extent at 0033 UTC (Figs. 21 and 22).
Figure 21. As in Figure 16, except at 0033 UTC on 20 April 2008. The line in (d) denotes the location of the vertical cross section in Figure 22. Click to enlarge.
Figure 22. KGSP base reflectivity vertical cross section at 0033 UTC on 20 April 2008. The cross section was cut along the line shown in Figure 21 (d). Click to enlarge.
The thunderstorm that produced the Double Shoals - Fallston and Western Lincoln tornadoes continued on a northeast path across the northwest Piedmont of North Carolina. No additional reports of tornadoes or wind damage were received, but penny to nickel sized hail was reported across Iredell, Rowan, and Davie counties. A similar thunderstorm developed behind and just to the south of the track of the first storm. In many ways, the radar presentation of the second storm was more impressive, but this storm did not produce any severe weather.
The thunderstorm responsible for the two tornadoes across Cleveland and Lincoln counties displayed several characteristics associated with supercell storms. It had a low level reflectivity appendage and inflow notch, a weak echo region, and a mesocyclone through at least half the depth of the storm that persisted for over one hour. An amateur cellular phone video also showed a wall cloud feature. That it was a miniature supercell made radar interpretation more challenging. While the mesocyclone was regarded as weak to minimal, it was possible that it might have been small enough to prevent proper sampling by the radars. This supercell was another example of how the traditional mesocyclone strength nomogram (Andra 1997) might be misleading when applied to situations other than the classic Southern Plains supercell. The rotational shear nomogram of Falk and Parker (1998) (Fig. 23) shows promise in that regard. The horizontal shear peaked at 0.012 s-1 at 2353 UTC, about ten minutes prior to the Double Shoals - Fallston tornado, which fell in the "tornado possible" category on the nomogram.
Figure 23. The rotational shear nomogram of Falk and Parker (1998). Click to enlarge.
The high reflectivity core did not extend high enough above the freezing level to allow the storm to produce cloud-to-ground lightning until after the last tornado lifted. While unusual, other examples of low cloud-to-ground flash rates in tornadic supercells have been noted (Markowski and Straka 2000, McCaul et. al. 2002, Perez et. al. 1997). In the end, the presence of cloud-to-ground lightning is not a requirement or a reliable precursor to tornado formation, especially in a thunderstorm with supercell characteristics.
Selected images of damage from the Double Shoals - Fallston tornado and the west Lincolnton tornado. Images were taken by Vince DiCarlo (NWS) during a storm survey. Click to enlarge.
Andra, Jr., D. L., 1997: The origin and evolution of the WSR-88D mesocyclone recognition nomogram. Preprints, 28th Conf. on Radar Meteorology, Austin, TX, Amer. Meteor. Soc., 364-365. Doswell, C. A., and D. W. Burgess, 1993: Tornadoes and tornadic storms: A review of conceptual models. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, Geophys. Monogr., No. 79, Amer. Geophys. Union, 161-172. Falk, K., and W. Parker, 1998: Rotational shear nomogram for tornadoes. Preprints, 19th Conf. on Severe Local Storms, Minneapolis, MN, Amer. Meteor. Soc., 733-735. Markowski, P. M., and J. W. Straka, 2000: Some observations of rotating updrafts in a low-buoyancy, highly sheared environment. Mon. Wea. Rev., 128, 449-461. McCaul, E. W., Jr., D. E. Buechler, S. Hodanish, and S. J. Goodman, 2002: The Almena, Kansas, tornadic storm of 3 June 1999: A long-lived supercell with very little cloud-to-ground lightning. Mon. Wea. Rev., 130, 407-415. Perez, A. H., L. J. Wicker, and R. E. Orville, 1997: Characteristics of cloud-to-ground lightning associated with violent tornadoes. Wea. Forecasting, 12, 428-437.
Thanks are given to Steve Udelson and Lindsay Varner of WSOC-TV in Charlotte, who supplied the images of the funnel cloud over Cleveland County. The upper air analyses, mesoscale analyses, and sounding were obtained from the Storm Prediction Center. The surface analyses were obtained from the Hydrometeorological Prediction Center. Radar mosaics and satellite imagery were obtained from the University Corporation for Atmospheric Research. Radar data images were created using the Java Nexrad viewer from the National Climatic Data Center and the GR2Analyst software package.