National Institute of Standards & Technology (NIST) Technical Investigation of Joplin, MO Tornado (492 page PDF File) #mowx
There’s no doubt that the Joplin, MO tornado of 22 May 2011 was a watershed event in 21st century USA weather history. A triple digit death toll had not been seen in the USA since 1953. Since then, spotter networks, improvements in radar and warning technology had reduced the overall death tolls dramatically. In light of the 3 May 1999 Bridge Creek/Moore/OKC tornado, much speculation was generated about the potential death tolls that could result from a violent, long-track EF-4 or EF-5 moving through a large metropolitan area such as Dallas/Ft. Worth. Many people, including yours truly, assumed it would take a direct hit for a triple digit death toll to occur in contemporary society. Only then, the death tolls could possibly top 100 or more. The Joplin, MO event was a dramatic wake up call that it didn’t take a large metropolitan area suffering a direct hit from a violent tornado for a 100+ fatality death toll to occur. The Joplin metro has a population of roughly 50,000 yet had a death toll of 161 from a violent EF-5 that moved through the city during a late Sunday afternoon.
The NIST has released a technical investigation of the Joplin tornado event. While rather technical, I highly recommend it to anyone interested in the atmospheric or earth sciences. A word of caution; this is a large 492 page PDF file. The download time, depending on your internet connection and computer, may take some time. Regardless, it’s a worthwhile read with a wealth of information.
What will likely be one of the largest (if not the largest since record keeping began in 1950) March tornado outbreaks took place on 2 March 2012. The forecast, synoptic setup, and SPC product details can be found at this link. What I have for this post are a few examples of base reflectivity (BR) and storm relative velocity (SRV) that I captured during the event. At the height of the outbreak, it was very difficult to keep up with the multiple warnings, special weather statements, spotter reports, and multiple radar sites simultaneously. In fact, my Twitter feed on Hootsuite was probably posting an incredible 150-200 ‘tweets’ per minute making it almost impossible to keep up with the flow of info. During the outbreak, I was fortunate enough to capture a few radar images and two tornado warnings.
This scan, taken at 2019Z shows two tornadic supercells back-to-back. The supercell on the right produced the Henryville, IN tornado which resulted in significant damage and several fatalities. A classic ‘hook echo’ configuration can be seen as well as what’s often referred to as a ‘debris ball.’ This results from the radar beam hitting a large amount of airborne debris that results in a high reflectivity signature.
This scan, taken at 2023Z shows the same two tornadic supercells a few minutes later. Two tornadoes were still in progress and substantial damage was taking place.
The above scan taken at 2023Z shows the Storm Relative Velocity product for the two tornadic IN supercells. Two substantial couplets can be seen at the location of the mesocyclones. On the eastern supercell, high velocity data to the right of the couplet (pink colors) may be indicative of a very strong Rear Flank Downdraft (RFD) which, according to some research, instrumental in tornado genesis.
The Severe Weather Statement issued by the Louisville, KY NWSFO expresses the nature of the situation with a Tornado Emergency.
The same sense of urgency is expressed in this Tornado Warning issued by the Wilmington, OH NWSFO for the same tornadic supercells as they moved further east.
This reflectivity scan, taken at 2301Z, shows the supercell that produced a tornado that damaged a large portion of West Liberty, KY. The scan shows a classic supercell reflectivity image with a well defined hook echo and a “debris ball” just east of West Liberty.
Once again we see very strong wording in a tornado warning…this one being from the NWSFO in Jackson, KY with emphasis on a Tornado Emergency for the warned area.
The second base reflectivity image is from 0001Z and shows a more pronounced debris ball with 70dBZ at the precipitation core as well as high values at the center of the mesocyclone.
Tornado outbreaks in March are not unheard of and there are several events in the past that have made the “top ten” lists of March events including the Tri-State tornado of 1925. This event was certainly no exception and is likely to remain in the record books for some time to come. Fortunately, with doppler radar and spotter networks, the storms were well observed which no doubt kept the death toll from being much higher. Unfortunately, a significant portion of the fatalities occurred in mobile homes…deaths that could have easily been prevented…but only when a full understanding of the dangers of remaining in fragile and cheaply built structures is finally comprehended by the public.
During the afternoon & evening of April 3-4, 1974, approximately 148 tornadoes occurred in an unprecedented outbreak. Tornadoes formed in thirteen states and killed approximately 330 people. Many records were set during this event including the number of tornadoes during a 24 hour period and the number of F-5 (using the old Fujita scale) tornadoes that formed in one event. It was truly a remarkable event in many ways. Information devoted to that event can be found here and here. An event that would be equal to or surpass the Superoutbreak was inevitable. That event took place on April 27, 2011 as part of a three day episode of severe weather that began in TX & OK and spread over the next two days to the east coast.
The image above is the Day 1 Outlook Tornado Probabilities map issued by the Storm Prediction Center (SPC) at 10:29 a.m. on April 27, 2011. The map shows the likelihood of a tornado within 25 miles of a point. The hatched area, running from MS & AL to KY, shows the risk of a EF-2 tornado or higher within 25 miles of a point. As you can see, the probability of a tornado was very high in a widespread area. The area with the greatest threat at the time this product was issued was over east-central MS through central AL. For all of the forecast products issued this day, the SPC was ‘spot on’ in every instance.
Moving ahead a few hours, the visible satellite image above shows numerous supercells from MS and AL northward to KY. Cumuliform overshooting tops can be seen over the most intense updrafts and large anvils have been spread downstream by the upper level winds. Most of the storms were moving at speeds up to 50 mph. In some cases, the storms even exceeded this speed. Storms moving at such speeds leave people little time to react if they’re in the path and need to take shelter.
Now let’s take a look at several doppler radar images, both in base reflectivity (BR) and storm relative velocity (SRV) from several sites during the height of the outbreak.
The first image was a scan from the Birmingham, AL (KBMX) National Weather Service doppler radar taken at 2215 UTC. The supercell structure is clearly evident. At this time, the tornado has just passed through the Tuscaloosa, AL metro. In the hook echo, a significant “debris ball” can be seen (purple area) as the tornado lofts tremendous amounts of debris and vegetation thousands of feet in altitude. There is a significant inflow notch visible and the flanking like has developed enough to become a substantial thunderstorm. At this point in the outbreak, there were 29 tornado warnings in effect.
The next image is from a few minutes later. The tornado has moved further to the northeast at a speed of at least 55 kts. The debris ball is still visible on the BR image. The Storm Relative Velocity (SRV) scan shows a significant couplet with velocities of +97kts & -117kts. A couplet this substantial can almost pinpoint the location of the tornado in spite of low clouds and/or precipitation that may prevent visual confirmation of the existence of the condensation vortex.
By 2228, we can see the supercell has maintained a well defined hook echo. The SRV couplet is still striking in appearance. The tornado was likely at EF-4 intensity at this point and was closing in on the Birmingham metro.
A few minutes later, the tornado is northeast of Brookwood, AL and has a very well defined ‘debris ball’ on the BR image. The SRV couplet is still very pronounced with +112 kts/-110 kts of shear.
A wider view at 2224 UTC shows the overall storm structure and very strong couplet moving to the northeast. The next series of scans show the storms progress towards the Birmingham, AL metro where the same tornado did a considerable amount of EF-4 damage.
In the above scan, the ‘debris ball’ is still very clear as the tornado has maintained its intensity upon approaching Birmingham.
Taking a wider view, there’s an astounding number of tornado warnings in effect. At least four distinct couplets can be seen in the SRV scan.
By 2309, the tornado was making its way through the northern parts of the Birmingham metro. A strong couplet and hook echo are still visible.
As the Birmingham tornado continued to the northeast, it maintained strength and a very strong SRV couplet. When the supercell reached a point southwest of Centre, AL, the storm structure had the appearance of the textbook supercell thunderstorm with a large precipitation core and a very pronounced hook echo.
The next three scans show BR & SRV from KBMX as another tornadic supercell approached the Birmingham metro at approximately 2323 UTC. Both the BR & SRV show tornadic characteristics.
In the last scan from KBMX, it’s interesting to note that the structure is more difficult to discern in the BR scan the closer the supercell moves to the doppler radar site. Fortunately, with doppler radar, the couplet can still be clearly seen.
The last scan is from the Atlanta, GA doppler radar site of a tornadic supercell near Brooks, GA. This was one of several tornadic supercells that formed in GA & the Carolinas as part of this outbreak.
It’s obvious to see that the advancements made in doppler radar have made the tornado warning process more accurate. From a scientific perspective, it’s also interesting to observe the structural changes and levels of intensity some storms can maintain for very long periods of time.
In this post, I’ve included a variety of supercell structures from more than one location. Regardless of the locale, each of these storms presented a significant threat to life and property.
In our first image, we have a supercell scanned from the Dyess AFB, TX doppler radar (KDYX) at 0238 UTC on 24 April 2011. A very clean inflow notch and hook echo can be seen.
The Storm Relative Velocity (SRV) product shows a very strong couplet indicating vigorous rotation with the storm. It’s very likely that a tornado was forming or already in progress.
This Base Reflectivity (BR) product is a supercell scanned from the Fort Worth, TX doppler radar (KFWS) at 2045 UTC on 25 April 2011. The image shows more of the classic supercell structure with a textbook hook echo a few miles south of Cleburne, TX. A tornado was reported at the time of the BR scan.
The above SRV scan is from a supercell near Little Rock, AR (KLZK) on 26 April 2011 at 0006 UTC. A very strong couplet can be seen near Mayflower, AR along with a strong Rear Flank Downdraft (RFD) passing over Maumelle, AR. At the time of this scan, a tornado was in progress.
The storms during this event were moving very rapidly, almost 50 mph. As you can see in this BR scan, the center of circulation has moved northeast of Mayflower, AR in just nine minutes. A large tornado was reported and the purple reflectivity is indicative of a substantial amount of airborne debris. This is often referred to as a “debris ball.”
The SRV product from KLZK at the same time as the above BV image shows a well defined couplet. Just south of the couplet is a hint of a RFD and/or gust front associated with the storm just to the south.
Approximately one hour later, the storms had consolidated into more of a linear structure. North of the KLZK radar site, a “comma head” echo had taken shape. This is frequently seen at the northern part of a particularly intense squall line. The northern portion of the storm structure begins to wrap around itself and almost becomes a mesoscale low pressure area. It’s also not uncommon for tornadoes to occur in this region.
As displayed by the KZLK SRV scan at 0124 UTC, a broad area of rotation exists with the comma head echo. This is especially evident by the strong inbound (green) winds. A substantial bow echo has also developed to the south.
During the afternoon of the same day, storms redeveloped over portions of AR including this supercell observed by the Fort Smith, AR doppler radar (KSRX) which, in the BR scan, shows an unusual circle in the mesocyclone. This is likely a result of a very heavy rain curtain being drawn around the south side of the mesocyclone creating what storm chasers refer to as a “bears cage.” A tornado was reported with this storm at the time of this scan.
The KSRX SRV scan at the same time shows a significant couplet. A confirmed tornado was in progress at the time of the scan. The inflow winds on the couplet are particularly strong.
Later that evening at 2307 UTC, a supercell with strong rotation approached the Memphis, TN metro as seen in this BV product from the KNQA doppler radar. This image shows structure that is more characteristic of a High Precipitation (HP) supercell. HP supercells can be particularly dangerous since they often hide the tornado in very heavy rain. The location of the mesocyclone circulation would be just a few miles southwest of Marion, AR.
As we can see from the KNQA SRV product at 2311 UTC, the advantage of doppler radar is obvious. A very strong couplet can be observed just southwest of Marion, AR moving to the northeast. By just using visual observations, the area of rotation would likely have been obscured or impossible to see due to the heavy precipitation wrapped around the mesocyclone. Damage was reported with this storm, but was very minor compared to the record breaking outbreak that would occur the next day.