A tornado damaged several houses and uprooted numerous trees over northeastern Rowan County and southeastern Davie County on 16 April 2011. The tornado was the first in a record outbreak of tornadoes across North Carolina on that day. Image from the National Weather Service damage survey.

Author's Note: The following report has not been subjected to the scientific peer review process.

1.  Introduction

An outbreak of severe thunderstorms occurred across the Mid-Atlantic region in 
the afternoon of Saturday, 16 April 2011, with numerous tornadoes over the 
central and eastern portions of the Carolinas (Fig. 1).  Across the state of 
North Carolina, at least 30 tornadoes were confirmed, resulting in numerous 
injuries and 24 fatalities (Parker et al. 2012).  Some of the tornadoes across 
central North Carolina were particularly destructive and long lived, including 
the Sanford-Raleigh tornado and the Fayetteville-Smithfield tornado.  An 
additional eight tornadoes were confirmed over the coastal plain of South 
Carolina.  The death toll and number of tornadoes makes the event of 16 April 
the most deadly tornado outbreak across North Carolina and South Carolina since 
the infamous "Carolinas Outbreak" of 28 March 1984 (Fujita and Stiegler, 1985; 
Gyakum and Barker, 1988).

Click on following links to view event summaries from National Weather Service
(NWS) offices in Sterling, Virginia; Wakefield, Virginia; Morehead City, North 
Carolina;, and Wilmington, North Carolina.

Figure 1.  Preliminary reports of large hail, wind damage, and tornadoes
compiled at the Storm Prediction Center for the 24 hour period ending at
1200 UTC on 17 April.  Click on image to enlarge.

The outbreak actually began in the early part of the afternoon over the 
western Piedmont of North Carolina with two relatively weak tornadoes in the 
Greenville-Spartanburg (GSP) county warning area (CWA).  The first tornado 
touched down at 1641 UTC about 4.5 miles north northwest of Salisbury, North
Carolina, near the Franklin community, in Rowan County (Fig. 2).  The tornado 
was on the ground for approximately four miles and crossed into the 
southeastern part of Davie County, North Carolina, before lifting at 1646 UTC.  
The damage associated with this tornado was rated as high as EF-1 on the 
Enhanced Fujita Scale while in Rowan County, but the rating dropped to EF-0 
over Davie County.  A second tornado touched down at 1700 UTC about three 
miles north of Monroe in Union County, North Carolina (Fig. 3).  The tornado 
was on the ground along an intermittent 2.5 mile path and lifted at 1703 UTC.  
The damage for this tornado was rated at EF-0 intensity.

Figure 2.  Approximate track of the tornado near Franklin and Cooleemee, North 
Carolina, on 16 April 2011.  The tornado path is shown by the thick light blue 
line.  Click on image to enlarge.

Figure 3.  Approximate track of the tornado near Monroe, North Carolina, on 
16 April 2011.  The tornado path is shown by the thick light blue line.  
Click on image to enlarge.

2.  Synoptic Features and Pre-Storm Environment

The potential for severe thunderstorms and tornadoes was anticipated by the 
Storm Prediction Center (SPC), which placed the area east of the Blue Ridge 
in a Slight Risk, and the area just to the east of Charlotte and Salisbury 
in a Moderate Risk on the Day 1 Convective Outlook issued at 0522 UTC. 
Numerous strong to severe thunderstorms developed across western and central
Georgia ahead of an approaching cold front during the early morning hours on
16 April 2011.  The strongest storms were located south of Atlanta around 
0900 UTC.  The SPC expected the potential for supercell thunderstorms to 
shift to the northeast around daybreak.  All signs pointed to an outbreak of 
severe thunderstorms and tornadoes across the central and eastern parts of 
the Carolinas and southern Virginia in the afternoon and evening.  At 1000 UTC, 
a rare early morning Tornado Watch (#149) was issued for northeast Georgia, 
western South Carolina, and part of western North Carolina.

Click here to view a 13 frame Java loop of a mosaic of composite reflectivity 
from 0000 UTC to 1158 UTC on 16 April 2011.

At 1200 UTC on 16 April, the cold front was moving into the western part of
the Carolinas while a warm front lifted north across the Piedmont of North 
Carolina.  A vertically stacked upper low was centered over northern Illinois 
at 500 mb, with an upper trough axis east of the Mississippi River (Fig. 4).  
A mid-level jet streak of 90 kt over central Mississippi and Alabama was 
associated with a short wave trough rounding the base of the upper trough.  
Water vapor satellite imagery showed a cyclonically curved dry slot across 
Arkansas, northern Mississippi, northern Alabama, and eastern Tennessee as 
further proof of the strong mid-level jet streak (Fig. 5).  At 850 mb, a 
low level jet of 65 kt was observed over central North Carolina (Fig. 6).  
The eastward movement of the short wave trough was expected to keep the low 
level winds backed over the central and eastern part of the Carolinas in the 
afternoon as the low level jet translated slowly eastward.  Meanwhile, the 
nose of the mid-level jet was expected to move overhead and increase the 
environmental deep layer shear during the time of maximum heating.

Click here to view a 17 frame Java loop of GOES-13 water vapor channel 
satellite imagery from 0645 UTC to 2245 UTC on 16 April 2011.

Figure 4.  SPC objective analysis of 500 mb geopotential height (dm; dark 
gray contours), temperature (oC; dashed red contours), and wind (kt; barbs) 
at 1200 UTC on 16 April 2011.  Click on image to enlarge.

Figure 5.  GOES-13 water vapor satellite imagery at 1145 UTC on 16 April 2011.  
Brightness temperatures are given by the color scale at the bottom of the figure.
Click to enlarge.

Figure 6.  SPC objective analysis of 850 mb geopotential height (dm; dark 
gray contours), temperature (oC; dashed red contours), dewpoint (greater 
than 8 oC; green contours), and wind (kt; barbs) at 1200 UTC on 16 April 2011.  
Click on image to enlarge.

The morning upper air analysis did little to change the thinking about the 
potential for severe thunderstorms and tornadoes.  As the cold front moved 
east of the Appalachians and over the western Piedmont around 1500 UTC (Fig. 7),
surface observations immediately ahead of the cold front in North Carolina 
showed temperatures in the middle to upper 60s and dewpoints in the lower 
to middle 60s.  The convection began to intensify and consolidate along the 
front as the warm sector expanded northward into southwest Virginia at this 
time.  Based on a rapidly destabilizing atmosphere ahead of the cold front, 
the SPC fortuitously coordinated an expansion of the existing Tornado Watch 
(#149) northward across the northwest Piedmont of North Carolina.

Figure 7.  Hydrometeorological Prediction Center (HPC) surface analysis of 
sea level pressure (mb; black contours) and fronts (traditional symbols) at 
1500 UTC on 16 April 2011.  Click to enlarge.

A special upper air observation taken at Greensboro, North Carolina (GSO, 
Fig. 8), at 1600 UTC sampled the pre-storm environment well, as the 
developing convective line was less than 50 miles to the west.  The sounding 
showed moderate instability with a most unstable convective available 
potential energy (CAPE) of greater than 1000 J kg-1.  Very strong low level 
shear (50-60 kt) and storm relative helicity (SRH, greater than 600 m2s-2) 
were present in the layers from the surface to 1 km AGL and from the surface 
to 3 km AGL.  The lifting condensation level (LCL) of less than 300 m was 
very low.

Figure 8.  Skew-T, log P, diagram and hodograph for the upper air observation 
taken at Greensboro, North Carolina (GSO), at 1600 UTC on 16 April 2011.  The 
temperature sounding is shown by the red line, the dewpoint sounding is 
shown by the green line, and the hypothetical path of a parcel lifted from 
the most unstable layer is shown by the brown dashed line.  A table of 
convective parameters is given at the bottom.  Click to enlarge.

The SPC objective mesoscale analysis confirmed the extent of the favorable 
pre-frontal air mass (Fig. 9).  The three-hour change of surface-based CAPE 
showed a plume of buoyancy developing from the coast of South Carolina to 
central North Carolina, coincident with a zone of 0-1km SRH in the range 
from 300 to 700 m2s-2 and an LCL of less than 750 m.  Objective indices such 
as the supercell composite parameter and significant tornado parameter 
(Thompson et al. 2004), energy-helicity index, and vorticity generation 
parameter pointed to a very high potential for the development of tornadic 
supercell thunderstorms across central North Carolina.  The sharp west to 
east gradient seen in the indices along a zone from Columbia to Charlotte 
and Statesville suggested that the line of convection might quickly become 
severe as it moved east across the gradient and into the more favorable 
environment.

Click here for an explanation of the severe weather indices listed above.

Figure 9.  SPC Objective mesoanalysis of (a) surface based CAPE 3-hr change 
(J kg-1; red contours increasing, dashed blue contours decreasing) and surface 
based convective inhibition (CIN, J kg-1; color fill), (b) 100 mb mean LCL 
height (m; dashed green contours) and 0-1 km SRH (m2s-2; blue contours), 
(c) Significant Tornado Parameter (effective layer; contours) and mixed 
layer CIN (J kg-1; color fill), (d) Supercell Composite Parameter (effective 
layer; blue contours) and Bunkers storm motion (kt; brown barbs), (e) 3 km 
vorticity generation parameter (VGP) and shear vector (kt; barbs), and 
(f) 3 km Energy Helicity Index (EHI; contours) at 1600 UTC on 16 April.  
Click to enlarge.

All of these factors strongly favored the development of tornadic supercell 
thunderstorms in the pre-frontal air mass and signaled the potential for an 
outbreak, including the possibility of violent long-track tornadoes.  As a 
result, the SPC issued a "Particularly Dangerous Situation" Tornado Watch for 
the counties immediately to the east of the GSP forecast area and raised the 
threat of severe thunderstorms to a High Risk for the central and eastern part 
of North Carolina.

Click here to view a 12 frame Java loop of GOES-13 visible satellite imagery 
from 1132 UTC to 2245 UTC on 16 April 2011.

3.  Radar observations

A Special Weather Statement was issued for much of the western Piedmont of 
North Carolina at 1531 UTC to heighten the awareness of thunderstorms that 
were moving toward an area favorable for strengthening.  The detection of 
an area of radial velocity directed toward the Terminal Doppler Weather Radar
located north of the Charlotte-Douglas International Airport (the TCLT radar) 
at greater than 40 kt on the 0.2 degree scan, toward the front of the 
convective band in eastern Gaston County at 1550 UTC (Fig. 10), prompted the 
issuance of a Severe Thunderstorm Warning (Fig. 11).  The intense part of 
the convective line crossed northern Mecklenburg County and southern Iredell 
County and produced wind damage near Huntersville as it passed, but this 
information was not discovered until well after the fact.  As the convective 
band reached western Rowan County and Cabarrus County, around the expiration 
time of the first warning, the outbound radial velocity detected by the TCLT 
radar remained nearly 45 kt associated with the leading edge of the 
reflectivity.  This prompted the issuance of a second Severe Thunderstorm 
Warning for parts of Cabarrus, Rowan, and Davie counties at 1630 UTC (Fig. 12).

Click here to view a 12 frame Java loop of a mosaic of composite reflectivity 
from 1158 UTC to 2257 UTC on 16 April 2011.

Figure 10.  Base reflectivity (a) and radial velocity (b) on the 0.2 degree 
scan from the TCLT radar at 1550 UTC on 16 April.  Note the area of velocity 
directed toward the radar at greater than 40 kt to the west northwest of 
Mount Holly.  The radar location is shown at the upper right of each image.  
Click on image to enlarge.

Figure 11.  Composite reflectivity from KGSP radar with outline of warning 
polygon for Severe Thunderstorm Warning #0096 issued at 1550 UTC on 16 April 
2011.  Click on image to enlarge.

Figure 12.  Composite reflectivity from KGSP radar with outline of warning 
polygon for Severe Thunderstorm Warning #0097 issued at 1630 UTC on 16 April 
2011.  Click on image to enlarge.

At this point, the height of the beam on the 0.5 degree scan from the NWS
Doppler Weather Radar (WSR-88D) at the Greenville-Spartanburg Aiport (the
KGSP radar) was between 9,000 and 11,000 feet above ground level (AGL) across 
the western half of Rowan County.  The failure to see the lower quarter of 
the storms rendered the KGSP radar not nearly as effective as the TCLT radar.  
In fact, the KGSP radar did not detect any significant rotation or 
reflectivity features as the line of storms moved across Rowan County.  As a 
result, the TCLT radar was used as the basis for all subsequent warnings.

A notable increase in storm relative motion and coherence of an inbound-
outbound velocity couplet was seen on the 2.4 degree scan at 1625 UTC west 
of Salisbury at an elevation of approximately 5,500 feet AGL (Fig. 13).  
However, rotational shear and velocity remained below thresholds normally 
associated with developing tornado mesocyclones (Fig. 14).  The increase 
in rotation was detected next on the 1.0 degree scan northwest of Salisbury 
at 1637 UTC at an elevation of approximately 3,500 feet AGL, indicating the 
tornado mesocyclone was developing downward (Fig. 15).  The 1637 UTC volume 
scan was the last before tornadogenesis.  At that time, the 0.2 degree base 
reflectivity showed little structure and the base velocity showed only a 
broad inbound/outbound couplet (Fig. 16).  A Tornado Warning was issued for 
north central Rowan County and southeastern Davie County at 1640 UTC, just 
one minute before a tornado touched down north-northwest of Salisbury 
(Fig. 17).  After the tornado was already on the ground, the TCLT radar 
detected the tornado misocyclone on the 1.0 degree scan at 1643 UTC over 
northeast Rowan County (Fig. 18). 

Click here to view an 11 frame Java loop of 0.2 degree base reflectivity 
from the TCLT radar from 1601 UTC to 1701 UTC on 16 April 2011.

Click here to view an 11 frame Java loop of 0.2 degree radial velocity 
from the TCLT radar from 1601 UTC to 1701 UTC on 16 April 2011.

Figure 13.  Storm relative motion on the 2.4 degree scan from the TCLT radar 
at (a) 1622 UTC, (b) 1625 UTC, (c) 1628 UTC, and (d) 1631 UTC.  Red shades 
indicate motion away from the radar and green shades indicate motion toward 
the radar, as given by the color table in the upper left corner of each image.  
The radar is located off the bottom left corner of each image.  Click on 
image to enlarge.

Figure 14.  Rotational velocity and shear calculated from maximum inbound 
and outbound velocity detected on the lowest three elevation scans from the 
TCLT radar from 1619 UTC to 1719 UTC on 16 April 2011.  The pink vertical 
bars represent the times of the Rowan-Davie county tornado and the lavender 
bar is the time of the Union County, North Carolina, tornado.  The “X” 
denotes the values of rotational velocity and shear from the only 0.2 degree 
scan available for post-analysis, at 1700 UTC.  Click on image to enlarge.

Figure 15.  Storm relative motion on the 1.0 degree scan from the TCLT radar 
at (a) 1619 UTC, (b) 1625 UTC, (c) 1631 UTC, and (d) 1637 UTC.  Red shades 
indicate motion away from the radar and green shades indicate motion toward 
the radar, as given by the color table in the upper left corner of each image.  
The radar is located off the bottom left corner of each image.  Click on 
image to enlarge.

Figure 16.  Base reflectivity (a) and base velocity (b) on the 0.2 degree scan 
from the TCLT radar at 1637 UTC on 16 April 2011.  The point where the tornado 
first touched down is indicated by the dot northwest of Salisbury.  The radar 
is located off the bottom left corner of the images.  Click on image to enlarge.

Figure 17.  Composite reflectivity from the KGSP radar with outline of polygon 
for Tornado Warning #0018 issued at 1640 UTC on 16 April.  Click on image to 
enlarge.

Figure 18.  Base Reflectivity (a) and storm relative motion (b) on the 1.0 
degree scan from the TCLT radar at 1643 UTC on 16 April.  The radar is 
located off the bottom left corner of each image.  Click on image to enlarge.

While the tornadic storm crossed northeast Rowan County, the broken line of 
convection had moved into the western part of Union County, North Carolina.  
Once again, the KGSP radar could not view the lower part of the storms as 
the height of the 0.5 degree beam was at least 8,000 feet AGL.  However, 
the volume products from the KGSP radar could be used to compare storms at 
similar distances.  The cells in the line across Union County were not as 
tall or as reflective as the storm that produced the tornado over Rowan and 
Davie counties.  In spite of the unimpressive reflectivity features, a 
couplet of inbound and outbound storm relative motion was apparent on the 
1.0 degree and 2.4 degree scans on the TCLT radar beginning at 1643 UTC 
over southwest Union County (Figs. 19 and 20).  The rotational velocity 
quickly reached a maximum on these two scans at 1649 UTC (Fig. 14).  The 
rotational shear on the 1.0 degree scan at this time peaked at 0.02 s-1, 
just below the "tornado probable" threshold on the rotational shear nomogram 
(Falk and Parker, 1998).  A Tornado Warning was issued for northeastern 
Union County based on the persistent storm relative velocity couplet at 
1700 UTC, which was concurrent with a tornado touching down north of Monroe 
(Fig. 21).  The shear weakened briefly after the 1649 UTC scan and did not 
increase again until the tornado was already on the ground to the north of 
Monroe.

Click here to view an 11 frame Java loop of 0.2 degree base reflectivity 
from the TCLT radar from 1631 UTC to 1731 UTC on 16 April 2011.

Click here to view an 11 frame Java loop of 0.2 degree radial velocity 
from the TCLT radar from 1631 UTC to 1731 UTC on 16 April 2011.

Figure 19.  Storm relative motion on the 1.0 degree scan from the TCLT radar 
at (a) 1637 UTC, (b) 1643 UTC, (c) 1649 UTC, and (d) 1655 UTC.  Red shades 
indicate motion away from the radar and green shades indicate motion toward 
the radar, as given by the color table in the upper left corner of each 
image.  The radar is located off the upper left corner of each image.  Click 
on image to enlarge.

Figure 20.  As in Fig. 16, except for the 2.4 degree scan from TCLT.  Click 
on image to enlarge.

Figure 21.  Base reflectivity (a) and storm relative motion (b) on the 
0.2 degree scan from the TCLT radar at 1701 UTC on 16 April, with 
(c) composite reflectivity from the KGSP radar and polygon outline for 
Tornado Warning #0019.  Click on image to enlarge.

4.  Summary 

Situational awareness of the potential for a severe weather outbreak was 
high.  The possibility of severe thunderstorms was raised in the Area 
Forecast Discussion and Hazardous Weather Outlook (HWO) beginning with the 
early morning issuance on Wednesday, 13 April.  The SPC included the extreme 
eastern part of the CWA in the Slight Risk area with the Day 3 Convective 
Outlook issued early in the morning of 14 April.  The midnight shift 
forecaster correctly upgraded the wording in the HWO to mention the 
possibility of tornadoes with the early morning issuance on 14 April.  
The tornado wording was included in every subsequent HWO through 16 April.  
The Slight Risk area was expanded across more of the CWA in the Day 2 
Convective Outlooks issued by the SPC on 15 April.  The Day 1 Convective 
Outlooks issued on 16 April ultimately included the eastern fringe of the 
CWA in a Moderate Risk in the last issuance before the outbreak commenced.

As is sometimes the case, the first few severe storms in the outbreak
were difficult to warn.  Although both tornadoes occurred within a warning,
the lead time was short.  Neither storm had grown large enough to exhibit
the classic supercell structure before producing a tornado.  In fact, after
the two relatively weak tornadoes over the western Piedmont, it took an
hour for the storms to strengthen and organize as supercells and finally 
produce additional tornadoes.  

Damage Pictures

Images of the damage from the Rowan-Davie Tornado on 16 April 2011, from the 
National Weather Service damage survay.  Click on images to enlarge.

Images of the damage from the Monroe Tornado on 16 April 2011, from the 
National Weather Service damage survay.  Click on images to enlarge.

References

Fujita, T. T., and D. Stiegler, 1985:  Detailed analysis of the tornado 
    outbreak in the Carolinas by using radar, satellite, and aerial survey 
    data, Preprints, 14th Conf. on Severe Local Storms, Indianapolis, Amer. 
    Meteor. Soc., 271-274.
    
Gyakum, J. R., and E. S. Barker, 1988:  A case study of explosive subsynoptic-
    scale cyclogenesis. Mon. Wea. Rev., 116, 2225–2253.

Parker, W., and K. Falk, 1998:  Rotational Shear Nomogram for Tornadoes.  
    Preprints, 19th Conf. on Severe Local Storms, Minneapolis, MN, Amer. 
    Meteor. Soc., 733-735.

Parker, M., J. Blaes, G. Lackmann, and S. Yuter, 2012:  Central North 
    Carolina tornadoes from the 16 April 2011 outbreak.  Preprints, Special 
    Symposium on the Tornado Disasters of 2011.  New Orleans, LA, Amer.
    Meteor. Soc., 6 pp.


Thompson, R. L., R. Edwards, and C. M. Mead, 2004:  An update to the Supercell 
    Composite and Significant Tornado Parameters.  Preprints, 22nd Conf. on 
    Severe Local Storms, Hyannis, MA, Amer. Meteor. Soc., P8.1.

Acknowledgements

The author wishes to thank Jonathan Blaes (NWS Raleigh) for providing the SPC 
mesoanalysis sector images.  The damage surveys were conducted by Larry Gabric, 
Tony Sturey, and Justin Lane.  Tornado track images were made using Delorme 
Street Atlas USA.  The upper air analyses and sounding plots 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.  The
radar loops from the TCLT radar were made using the NOAA Weather and Climate
Toolkit.  The warning polygon images were obtained from Iowa State University.  
Larry Lee (NWS Greenville-Spartanburg) provided a critical review of the 
manuscript.