Easter Sunday East Tennessee Tornado Outbreak

At 11 PM EDT Sunday evening, shortly before the first tornado occurred, a highly favorable synoptic pattern was in place to promote severe thunderstorms, tornadoes, and flooding. The 300 mb analysis indicated a large trough over the middle of the United States which placed East Tennessee under very strong southwesterly flow aloft. These winds in the upper levels of the atmosphere were between 100 and 140 knots and were blowing in a divergent pattern. The strongest and most divergent of these upper level winds was occurring over two areas, with one centered over eastern Kentucky and another over Alabama and Georgia. This is called jet coupling and encourages deep lift within the atmosphere. Lift is one of the most important ingredients for severe thunderstorms to develop, and East Tennessee was located under the area of greatest lift beneath the coupled jet structure.

This upper level divergent flow and subsequent lift were also contributing to enhanced lower level winds. At 11 PM EDT, 850 mb winds, which is about a mile above the ground, were blowing from a southerly direction at 50 to 70 knots. This provided a tremendous amount of wind energy for thunderstorms to tap into and also advected warm and moist air northward from the Gulf of Mexico to fuel thunderstorms. The dark green colors are higher dew points that were being pulled into the southern valley by the strong southerly winds at 850 mb. These higher dew points signal a more moist and humid airmass.

Finally, at the surface, a strong low pressure system was centered over West Tennessee at 11 PM EDT and moving northeastward while also strengthening due to divergent upper level flow. Since the wind moves counterclockwise around a low pressure system, surface winds over East Tennessee were blowing out of the southeast. This created directional wind shear by significantly changing wind direction with height from southeasterly at the surface to southwesterly aloft. This is a key ingredient for tornadoes in addition to having strong winds aloft. These southeasterly surface winds also brought warm and humid air northward behind a surface warm front as seen by the green colors moving into the southern valley. As the low pressure system continued to move northeastward, a cold front moved in from the west further strengthening the lift over East Tennessee. The presence of significant upper and low level wind speeds, shear, moisture, and heat ahead of this front allowed thunderstorms to explode in intensity over portions of the southern valley where the greatest warmth and moisture near the ground worked together with the wind energy and shear aloft.

As we discussed earlier, a necessary ingredient for severe thunderstorms and tornadoes is wind shear. When both speed shear and directional shear are present, tornado potential drastically increases, which was the case on the night of April 12th. At 11 PM EDT, shear in the surface to 1 kilometer layer was 50 to 60 knots over the southern valley. Research has shown that only around 20 knots of shear in this layer is required for tornadoes.

The directional shear was equally impressive at 11 PM EDT. Meteorologists measure directional shear with a parameter called Storm Relative Helicity (SRH). The minimum values for tornadoes only need to be around 100 m2/s2, but in this case, SRH values were between 600 and 950 m2/s2 over East Tennessee. This created a highly favorable environment for supercell thunderstorms to produce strong rotation and potentially violent tornadoes. 

The warmer and more humid/moist airmass that was pulled into the southern valley allowed the atmosphere to become unstable. Instability occurs when cooler air aloft resides above warm and moist air at the surface. One method to quantify instability in the atmosphere is using a parameter called Convective Available Potential Energy (CAPE). Even when massive amounts of wind shear are in place, instability near the surface is needed in order for tornadoes to form. CAPE values at 11 PM EDT were between 100 and 250 Joules/Kilogram in the southern valley. While these values are not very high, they were enough to fuel strong tornadoes when combined with the extraordinary amounts of speed and directional wind shear.

Another important ingredient to consider when forecasting the potential for tornadoes is the height of the cloud base. If the base of the clouds are too high, supercell thunderstorms may strongly rotate, but never be able to produce a tornado. Cloud base heights are measured by the Lifted Condensation Level (LCL). At 11 PM EDT, LCL heights over the southern valley were around 500 meters. This is very favorable for tornadoes because research shows that values only need to be slightly below 1000 meters in most cases.

The surface warm front was a key ingredient to the development of tornadoes, and at 11 PM EDT, it was just nudging into the Southern Plateau and southern valley. That warm front locally enhanced the directional wind shear and also allowed instability to build near the surface as the warmer and more humid air flowed in behind it. Instability being rooted at the surface is necessary for tornadoes to form because it lowers the LCL heights. To the north of this warm front, all of the instability was elevated more than a mile above the ground, which led to very high LCL heights and no tornadoes, despite the presence of extreme wind shear. The warm front nudged just far enough into the southern valley for the strong tornadoes to occur.

The combination of the extremely high shear, very low LCL heights, and enough CAPE near the surface led to favorable values of the Significant Tornado Parameter (STP). Research shows that values greater than 1 are usually enough to support tornadoes, and in this case, STP values ranged between 2 and 4 at 11 PM EDT over the southern valley. This is another signal that several tornadoes were likely in the southern valley, including the potential for some to be long track and violent, as was the case with the Hamilton County EF-3 tornado.

Finally, the warm and humid air that was advected into the area combined with all of the strong lifting in the atmosphere led to very heavy rainfall and flooding. One way to measure the amount of water vapor in the atmosphere to support heavy rainfall is Precipitable Water (PW). At 11 PM EDT, PW values were 1.5 to 1.7 inches over much of East Tennessee. These values were significantly above the climatological average for mid April and supported torrential rainfall rates.

The 4-panel image shows various radar products from 11:15 p.m. to 11:52 p.m.

On the top left, reflectivity imagery is shown, which is useful for giving an overview of the basic structure of the storm. Reflectivity gives a measurement of the amount of energy reflected from a target back to the radar. Heavier rain, hail, or solid objects like tornado debris have high reflectivity values.

On the top right and bottom left, two types of velocity data are shown, which shows motion within the storm. The red colors show outbound velocity and the green colors represent inbound velocity. Notice how the two colors are coupled together south of I-75, heading to Collegedale and up to Cleveland. Where these couplets occur, tight rotation is present and this is associated with the tornadoes.

The bottom right image shows correlation coefficient (CC). CC can be used to detect non-meteorological signatures such as debris resulting from a tornado. This debris is called a tornado debris signature (TDS) when low CC is paired with a rotation couplet. A TDS is clearly visible near Collegedale, and again near Cleveland.

We'll take a closer look at the radar imagery for each of the four tornadoes that occurred on the night of April 12, 2020.

As a line of severe thunderstorms moved east, velocity signatures on radar began to increase in north Georgia as storms were moving east-northeast. A Severe Thunderstorm Warning was issued for Hamilton County at 11:03 PM EDT.

This line of severe thunderstorms had a history of producing QLCS tornadoes in northeast Alabama and north Georgia as the line progressed eastward, so a "Tornado Possible" tag was added to the Severe Thunderstorm Warning. Many of the tornadoes in northeast Alabama and north Georgia were preceded by a rear-inflow jet (RIJ) notch at about 7,000 feet. This RIJ notch signature appeared at about 11:08 PM EDT about 6 miles west of Fort Oglethorpe and raised the attention of forecasters that a tornado could quickly develop. While not imminent, it had forecasters' full attention. The Severe Thunderstorm Warning was updated at 11:10 PM EDT to adjust potential winds upward to 70 mph with a continuation of the "Tornado Possible" tag because of the approaching RIJ.

At 11:16 PM EDT, low-level rotation was not yet strong, but the low level signatures of a bowing out RIJ at the surface were evident in Walker County, GA as the storm moved east-northeast. A Tornado Warning was quickly issued for Hamilton County.

At 11:24 PM EDT, gate to gate rotation had quickly strengthened to 72 kt about 2 miles west of East Brainerd. The Tornado Warning was updated to include a "Considerable" tag meaning that a strong tornado is likely.

At 11:25 PM EDT, a tornadic debris signature (TDS) was observed on radar confirming that a tornado was occurring. The Tornado Warning was updated with a "Radar Observed" tag and also continued with the "Considerable" tag highlighting that this was a "Particularly Dangerous Situation" in the warning text.

At 11:26 PM EDT, gate to gate rotation had strengthened to 78 kt with a large defined TDS. It was quickly decided to update the warning with a "Tornado Emergency" and "Catastrophic" tag at 11:27 PM EDT. Based on radar data alone, forecasters knew it was very likely than an EF-3 or greater tornado was in progress. The image on the right is of radar data at the time the "Tornado Emergency" was issued.

As it became clear that this tornado was violent and likely to continue, a Tornado Warning with a "Considerable" tag was issued downstream across Bradley County, including Cleveland, at 11:28 PM EDT.

Low-level rotation weakened, but remained strong, as the tornado passed Ooltewah and approached the Bradley County line. A TDS continued to be observed as debris was lofted to around 10,000 feet.

Supplemental Adaptive Intra-Volume Low-Level Scans (SAILS) was very beneficial for diagnosing and warning for this tornado.

The Tornado Warning was continued with a "Considerable" tag as a TDS continued to be observed as the East Brainerd/Ooltewah tornado debris continued to be observed by radar and was co-located with the velocity rotation couplet. It's important to remember that a TDS does not mean a tornado is currently in progress; it simply means that a tornado has occurred and lofted debris high enough to be picked up by radar. In some instances, debris can be seen on radar for many scans after a tornado has dissipated. Since this storm was known to have formed a damaging tornado across Hamilton County, the decision was made to continue the "Observed" and "Considerable" warning into Bradley County since velocity signatures and the TDS continued to show the potential of a tornado in progress.

Rotation became weaker and the debris signature became more diffuse on radar, so the "Considerable" and "Observed" tags were dropped in the update at 11:42 PM EDT; however, the Tornado Warning remained in effect through 12:00 AM EDT for Bradley and Polk Counties.

At about 11:45 PM EDT, an EF-2 tornado occurred in southeast Cleveland. A brief TDS was noted but only lasted for two quick SAILS scans.

There are some important limitations to consider when analyzing radar for the Cleveland tornado. First, Cleveland is about 61 miles from the radar. This distance causes lower radial resolution from the radar beam (it gets wider the further you get from the radar). Wider radials mean there are more radar returns averaged into one pixel. There's less detail. In addition, the further away from the radar you are, the higher up the 0.5° radar tilt is because of the tilt and the curvature of the Earth. For example, at Cleveland, the 0.5° tilt is about 5600 feet AGL and in East Brainerd, the 0.5° tilt is about 3800 feet AGL. This makes it more challenging to differentiate small-scale rotation and debris signatures.

Lastly, we'll take a look at the interesting pair of EF-1 tornadoes in northeast Bradley County.

The image on the right shows the point at 11:50 PM EDT that two areas of rotation are noted. You can more clearly see them in the Normalized Rotation (NROT) field in the bottom right. Each of these areas of rotation are responsible for separate EF-1 tornadoes that converged and dissipated as they reached the Polk County line.

The southern EF-1 is a continuation of the rotation that spawned the EF-2 tornado in southeast Cleveland which was enhanced by the RIJ racing northeast. This EF-1 occurred about 11:50 PM EDT. The northern EF-1 tornado was beneath the comma head and formed as rotation increased due to the convergence of the RIJ wind field. This tornado is believed to have occurred about 11:55 PM EDT. It may have also been enhanced locally by increased low level rotation near the warm front which was near northern Bradley County and McMinn County. These tornadoes caused minor damage to several homes and damaged or destroyed several barns and farm outbuildings.

In real time, these were not recognized as two separate tornadoes. While surveying, our survey team came across two separate EF-1 tracks that were clearly from two different areas of rotation. When viewing the tracks with radar data, it became evident that two tornadoes had occurred with two distinct mesovortices.

Another TDS was observed with this area of rotation as it became one stronger area of rotation at 11:56 PM EDT, so a separate Tornado Warning was issued downstream for southeast McMinn and northwest Polk Counties. Because of the history of this storm and the height of the TDS (up to around 10000 ft), "Confirmed" and "Considerable" tags were added for a potential EF-2 or greater tornado. Both of these tags were dropped from the warning at 12:07 AM EDT as rotation weakened and the TDS dissipated.

No additional tornado paths were noted after the EF-1 tornadoes ended near the Polk County line.

This map depicts the data points collected by NWS staff to determine the path of both the Chattanooga/Ooltewah and Cleveland tornadoes. The points are colored to denote the severity of damage based on the Enhanced Fujita (EF) scale with the green points denoting non-tornado/thunderstorm-related wind damage. For the Chattanooga tornado, the line denotes the center of the path with polygons corresponding to the level of damage on the EF scale. The damage points, polygons, and path line all depict the true length of the path. While much of the damage observed was EF-0 to EF-2, severe damage observed where the path crossed East Brainerd Rd. met the criteria for EF-3 damage. For the Cleveland tornado, most damage was EF-1 or less with a small area reaching criteria for EF-2. 

Map legend accessible by clicking the legend icon in bottom left of map.