Super Outbreak 5 years later: My story and the meteorology (Part 4 – the day everything changed)
As I have blogged all this week, the historic tornado outbreak 5 years ago was forever etched in the minds of millions of people, destroying countless communities and taking too many lives. It was as it decades of hard work to improve safety and forecasting in tornadoes was wasted. In addition, to add a personal twist I was about to go on a long-planned Florida vacation, and we were driving this time. It just happened everything would coincide. After it was over, my future was changed, as I will explain more on Friday. This is Part 4 of the 6-part series, and it is a very long post to describe what was a very long day. I have divided the story into four sections here, since it was truly that kind of a day.
Early morning of April 27: Nocturnal double trouble
Wednesday, April 27 was a busy day that started right at the crack of midnight, like it was New Year’s Eve, and there was no rest for already-weary meteorologists. It must be remembered that this outbreak was remarkably continuous for well over 72 hours, and had many rounds of activity. At midnight, Round 3B was underway over Arkansas, Louisiana and Mississippi as the supercells from Round 3A had congealed into a squall line. Normally in those situations, straight-line winds are the main threat. However, shear was increasing rapidly and many tornadoes were able to form. At midnight, there had been 98 tornadoes in the outbreak, a number which would rise much higher in the pre-dawn hours. The presence of multiple low-pressure areas and backing cold air allowed very rich warm air to penetrate northward, maintaining high energy for severe weather even in the overnight hours.
The squall line was very efficient in producing mesoscale vortices and numerous tornadoes. Unusually for such a situation, some were strong and long-tracked, with two EF-3 and several other EF-2 tornadoes reported across Mississippi before that squall line weakened and eventually dissipated. At least two people died during that activity, and multiple other injuries were reported. However, with the air mass still highly volatile, weak cells farther west were also able to congeal into a second squall line, which I consider to be Round 4 of the outbreak. It developed over Mississippi and intensified entering Alabama where the previous line dissipated about 90 minutes earlier, eventually overtaking the previous weakening QLCS.
The new squall line rapidly intensified over eastern Mississippi and western Alabama through middle Tennessee. Dozens of tornadoes were reported from this activity between 4:00 am and 9:00 am, and like with the previous line, some were strong and long-tracked including three EF-3 tornadoes and numerous EF-2 tornadoes. One efficient mesoscale vortex in Marshall County, Alabama produced a whopping 16 tornadoes in a very small area! These tornadoes also killed at least five people and the associated tornadoes and straight-line winds knocked out power to hundreds of thousands of people, which hindered abilities to get warnings later. That has been suspected in playing a big role later. In fact, several communities that were hit by the early morning tornadoes were also hit again later in the day! Combined, the two squall lines produced 86 tornadoes, which would be a very impressive tornado outbreak in itself.
Late morning of April 27: The hits keep on coming
By breakfast, the outbreak had produced 184 tornadoes. As expected, the SPC upgraded the risk level for today from moderate to high, noting that the parameters were all there for a major, possibly historic, tornado outbreak in the South. Also notice that the risk level was enhanced all the way up to the eastern Great Lakes, despite relatively modest tornado probabilities (in retrospect, probably too low). The placement was just north of where a previous tornado outbreak occurred on April 15, but this one would blow that out of the water. For Thursday, uncertainty on whether activity would be able to redevelop behind the squall lines before the cold front reached the coast was still present, but the Carolina coast was upgraded to an enhanced risk (using current standards). That area was already still recovering from a major outbreak on April 16 which killed 26 people.
After a brief lull in activity, yet another QLCS was developing over the middle Mississippi Valley around 9:30 am. Normally, secondary lines would quickly die as the previous line had used up all the energy. However, with so much reinforcing energy and with some of the most impressive dynamics ever seen, it managed to thrive and produce seven more tornadoes and more wind damage over northern Alabama shortly before lunchtime, and a few more weak tornadoes in east Tennessee during the early afternoon in what I consider to be Round 5 of the activity. It would only add insult to injury on what lied ahead. It may have also helped create a more conducive environment for later supercells through additional boundaries and gravity waves, which is contrary to conventional thinking that early MCS’s would stabilize the atmosphere. That QLCS didn’t last particularly long but may have also fed into activity farther north later in the day.
Afternoon of April 27: History and tragedy
Despite the morning activity, the atmosphere was explosively unstable and only expected to become more so, with some of the most remarkable dynamics ever recorded. The SPC first issued two extremely strongly worded PDS tornado watches, which is the highest level of watch that exists, and they covered much of Mississippi, Alabama and nearby parts of other states. Additionally, for a region from roughly Starkville, MS to Huntsville, AL, the tornado threat was increased beyond the high risk threshold to 45%. All indications were that violent tornadoes were likely. It was up to individuals to protect themselves and save lives if necessary. Meteorologists could only do so much. After 191 tornadoes in 48 hours, meteorologists could have probably used a break, but really the adrenaline was only going to kick up another notch.
In addition, the threat farther north was increasing and multiple tornado watches were issued for the eastern Great Lakes and most of the mid-Atlantic. In addition, the moderate risk was eventually expanded into the middle and upper Ohio Valley, but that area largely busted as it remained behind the frontal activity. In retrospect, that should have been placed roughly along and just west of the I-81 corridor.
While Round 5 was still going on in northern Alabama, the atmosphere was recharging at warp speed in Mississippi. The first supercells developed shortly after noon roughly along the I-55 corridor, which began Round 6 of the outbreak. This would be the most prolific part, producing many violent tornadoes and countless tragedies, and would last through the afternoon and evening. By mid-afternoon, the energy-helicity index (EHI) and significant tornado parameter (STP) were both over 10 for a large area, which is outrageously high. Values of 1 or 2 for both are sufficient for strong tornadoes. The first major tornadoes were shortly after 2:00 pm across the region and they quickly flourished across much of Alabama and Mississippi. The first violent tornado, an EF-5, touched down near Philadelphia, MS at 2:30 pm local time. It was followed by more and more.
Between 2:30 pm and 8:30 pm, there wasn’t a single half-hour period without a violent EF-4 or EF-5 tornado on the ground. Taking advantage of a practically perfect environment, boundaries left behind by the previous squall lines and even ground terrain, tornadic supercells continued to develop frequently, and after a tornado dissipated, another one quickly formed. Tragically, these tornadoes were going right over populated areas, and the earlier storms knocked out power to large areas of northern Alabama, hindering the ability to receive warnings. Some of the cities and towns devastated include Cordova, Cullman, Chilton, Hackleburg, Phil Campbell, Tanner, Smithville, Rainsville, Dadeville and many others. But most of the coverage focused on Tuscaloosa and Birmingham, two of the larger cities in Alabama which were devastated.
This coverage from James Spann, a highly respected meteorologist at ABC 33/40 in Birmingham, says it all:
Sadly, many people lost their lives during those storms that afternoon and early evening. Some people took shelter, but there was nowhere that they could survive (basements and reinforced storm shelters, which are pretty much a must to survive the core of a violent tornado, are rare in that part of the world). Others were unable to hear the warnings due to power outages or their personal situations, or did not respond properly. They happened in solid frame houses, mobile homes, public buildings, outdoors and in vehicles. 99% of the time, you can survive a tornado in an interior closet or room in the center of a solid house (basically, anything up to an EF-3). Many of these tornadoes represented the 1% of the time where such was not the case. Four of these tornadoes were rated EF-5 while another seven were rated EF-4 (and that was just before sunset). In Alabama alone, over 250 people died (mostly from these supercells which killed well over 220), while in Mississippi, 32 people died (of which about 25 from this round of activity). And no, it was not over. Not by a longshot.
Farther north, the atmosphere was remarkably unstable for the mid-Atlantic and Northeast, especially for late April. While the headlines were in the South, a fairly impressive tornado outbreak developed from Virginia through Maryland, Pennsylvania, New York and even into Canada starting around 2:00 pm Eastern. Supercells and linear convective activity began to form in the afternoon roughly along the I-81 corridor in response to the secondary low pressure area. Those storms continued to fire throughout the afternoon and into the night, possibly enhanced by energy from the previous rounds of activity in the South. That activity would continue into the night, as explained in the next section.
Evening of April 27: Tornado myths busted
The sun was starting to set over the South around 7:00 pm Central/8:00 pm Eastern (times from here on are in Eastern time). The supercells were still highly active and producing destructive tornadoes in eastern Alabama but were spreading into Georgia and east Tennessee, with a few more scattered but intense supercells also forming well south of I-20 where the cap had largely held activity to a minimum until now. Those were now producing EF-3 and EF-4 tornadoes which resulted in numerous fatalities. With wind shear literally off the charts and the energy slow to weaken, it took a long time for the storms to temper – they kept firing and producing intense tornadoes long after dark. It wasn’t until past midnight before they started to weaken.
The hardest hit areas in the evening storms were along the western foothills of the Appalachians and even into the Great Smoky Mountains. Some of the tornadoes went over large bodies of water as well including lakes and rivers. Those completely busted some well-believed but false myths about tornadoes (including violent ones). They included an EF-4 (almost EF-5) tornado which completely devastated Shoal Creek and Ringgold (on Friday, I will publish a more detailed account and story from those communities) and an EF-4 tornado directly inside Great Smoky Mountains National Park which flattened electrical towers. Even in far northeastern Tennessee and southwestern Virginia, four deadly tornadoes resulted in at least 13 deaths as the supercells from AL and MS tracked eastward into the region producing strong tornadoes in an area that doesn’t receive them very often. In total, at least 40 more people died that evening from tornadoes across five states.
Farther north and east, the secondary activity was intensifying, with more and more tornadoes touching down as well across the Mid-Atlantic ahead of the secondary low and with the help of the moisture in the area, plus remnants of previous QLCS’s that reformed into supercells in the region. One person died in central Virginia as a result of an EF-2 tornado, while others were injured across several states. Fortunately, none of the tornadoes from this activity were stronger than EF-2, which prevented a double-barrel area of destruction like the 1974 Super Outbreak which spread from the Midwest and Great Lakes into the Deep South.
Still, it was a day for the record books and a day of destruction for many areas. By midnight, there had been a whopping 215 tornadoes in just 24 hours, which was a new one-day world record (exceeding the 148 in the 1974 Super Outbreak, although I believe that is a low estimate as the technology did not exist then to capture every weak tornado, it is likely many F0 and F1 tornadoes were missed). The overall outbreak total increased from 98 to 313, which alone would have set a new record for tornadoes in all of April in the US. The most telling numbers were the human toll. That day, over 320 people lost their lives, most of which were as a result of tornadoes. That kind of death toll was unheard of in the modern era. More on that tomorrow and what may have gone wrong…but I was sick to my stomach that night.
While all that is happening…
I barely got any sleep the previous night. I was tracking the cells for most of the night and watching as the QLCS’s were producing tornadoes down south. However, it was travel day for me. We knew the weather situation but were also aware that we were hamstrung by the large-scale situation and long-held plans. It was a stormy start to the day, as we were in the warm front most of the morning, which slowed the driving down. Rain was pelting for the first couple hours. We agreed to stick to the script though. It was an enhanced risk for the most part, pushing at times into a moderate risk.
By lunchtime, temperatures had risen from the 50s to the mid-70s with fairly high dew points. I had my phone but wasn’t checking in too often from elsewhere (I wouldn’t know of the catastrophe down south until evening). At that time, not surprisingly, I found out we were under a tornado watch for most of the rest of the trip that day. As a result, thanks to the phone plan which I enhanced the previous couple days, I knew the next few hours would be monitoring the weather as well as our positions, and if I needed to I would place the plans into action. Low and behold, cells were developing to our southwest starting around 1:00 pm.
Watching every radar screen and every move, we continued onward. It was around 2:00 pm when the cell strengthened into a rotating supercell, although no tornado warning had been issued yet. We agreed on a stopping place at a freeway service area in a town that we knew about around 2:30 pm, which was shortly before I expected it to cross our route. My parents were at first not in agreement – they felt it wasn’t too bad and we should continue (my father, especially, who was driving). I tried to warn them of the situation and eventually we agreed to pause at that point. The heavy rain was another danger on the highways – we had experience with that and it is blinding especially in heavy traffic. This was all from the early part of the secondary severe weather area.
It was 2:33 pm when we arrived and the skies were darkening, right on cue. At the same time, a tornado warning was issued for the area as stronger rotation was detected. There were many other people taking shelter there or simply taking an afternoon break unaware of what lied ahead. I talked to a manager and encouraged everyone to get under tables or behind closets when the time came. By 2:38, it had started raining outside and the wind was picking up. What would happen next?
A couple minutes later, winds would swirl rapidly. By then, I said “everyone, under tables now!” loud. Most people followed suit as the windows were shattering and debris flying. Somehow the power stayed on, but there was a lot of debris moving around the building. I quickly realized a tornado had struck, with my findings confirmed when I walked outside when safe to do so. It was determined through postings and words I said that the exact time was 2:41 pm when it occurred. Most other people were startled, but the building was largely intact with no visible damage (it is made of bricks, and no windows shattered). No one was injured. However, the nearby gas station lost part of its siding, and trees were down, signs were blown out and outhouses thrown. People could have been injured or killed if they were outdoors at that time. It was later determined that it was a weak tornado, although rated without a survey based on my photos (my best guess looking at the tree damage was that it was an EF-2 but I was unable to take a photo of that).
Shaken but determined, knowing the mess that today already brought, we continued driving about an hour later after we knew conditions were safe. We wanted to wait out until the rain stopped and any damage may have been picked up. As it was, the tornado was likely quite narrow as there was no visible damage right on the highway – probably no more than 50 yards wide. We shifted the route somewhat west to get behind the cold front and avoid more problems, knowing that the boundary was slow moving and the threat remained. We got into our hotel around 6:30 pm, which was when I started to become aware of the catastrophe unfolding down south. I followed that all night (primarily by watching either James Spann or Jason Simpson – both very well respected and beloved meteorologists in Alabama), realizing what I had gone through was nothing compared to that. But I sure learned a lot of lessons, which I will mention in Parts 5 and 6.
What can I say? That was a day that will forever be remembered in so many ways. I had to really think a lot and realized that vacation can be memorable in other ways. Also my human side realized that maybe it wasn’t as important since I was also watching catastrophe unfold. On Thursday and Friday, we were going to be driving directly through areas that were devastated badly. What would I find when I saw them? We agreed to keep the route the same through Thursday, but my mind was clearly not in tune. Suddenly, I didn’t care too much that I was going to meet Mickey Mouse and go to the beach. I was talking to professional meteorologists for advice as well, and they said to stick to the route. But that was for another day. It was a very long, tragic day, but I was thankful to still be in one piece.
Forecaster Craig Ceecee
Remainder of the series: