This past weekend a swarm of devastating tornadoes ripped through parts of Alabama, Florida and Georgia. According to Accuweather.com, "This was the deadliest tornado outbreak in the U.S. since May 2013, when a massive and devastating tornado killed 24 people in Moore, Oklahoma." In one case, the National Weather Service even issued a rare tornado emergency because of expected damage from a "confirmed large and destructive tornado." At least 23 fatalities have been confirmed, and I fear this number may increase as recovery efforts continue. As a meteorologist and a scientist who pays close attention to weather and the risk messaging, there are four things about this outbreak that I noticed.
The forecast. I went back to examine the National Weather Service (NWS) Storm Prediction Center (SPC) outlook two days before the outbreak (graphic below). It was clear meteorologists knew that region was going to be threatened Sunday afternoon. The NWS SPC mesoscale discussion on the morning of March 3rd was also aggressively discussing potential tornadic activity in a region that included southeastern Alabama, south-central Georgia, and northern Florida. Here is the text from that mesoscale discussion at 9:50 am CST:
In summary, as moderate buoyancy and strong/veering flow through the profile combine with moist low levels, the threat for strong low-level mesocyclones associated with the discrete storms will increase, along with a
corresponding risk for tornadoes of which a few may be strong.
In other words, the atmosphere was very favorable for the formation of rotating storms (graphic below). Matthew Cappuci wrote a great meteorological breakdown in the Washington Post Capital Weather Gang.
The technology works. The models sniffed out likely tornado regions well. Doppler and dual-polarization radars functioned properly. Doppler radar detects changes in the frequency of a pulse of energy reflected back from the cloud volume. Slight shifts in frequency can be used to determine if rotation exists in the cloud. Tornado warnings are often issued based on radar-indicated rotation. On March 3rd, I looked at all of the rotation signatures on radar and said to my wife, "this is going to be bad."
The dual-polarization aspect of the radar allows meteorologists to see something called the "tornado debris signature or ball." While this signal is not available until the tornado is traversing the ground lofting debris into the air, it is quite useful for warning other locations within the pathway of the tornado. The graphic below shows the hook echo (upper left), rotation couplet (upper right) and a debris ball (lower right) from the large tornado in Lee County, Alabama. The movie loop at this Twitter link shows how the storms were evolving from the perspective of our weather satellites.
A good forecast is not good if it is not received and acted upon. Even as meteorologists point out how good the forecast was, the sad reality is that people still died. We are in the business of saving lives and property; not self-affirmation. While cliche "It Came Without Warning Headlines" are hyperbole, an individual that did not receive or understand the message may feel exactly that way. What happens once the warning information leaves the National Weather Service, your Phone Alert App, or the TV meteorologist break-in?
There is a percentage of people who never get the warnings. I am very weather-aware, but many people are not. They are simply battling the challenges and routines of daily life. How do we reach them? There is a percentage of people who complain about TV shows being interrupted. James Spann, the legendary Birmingham-based meteorologist, has written about this in the past. How do we overcome this notion that if a person is not affected then there is no emergency ("survivor bias") or no need to interrupt my show? There is a percentage of people who got the message and were likely sufficiently alarmed but were not certain what to do. These points highlight continued focus on the social science aspects (communication, message interpretation, perception, psychology, color schemes, words, tendencies) of the weather forecast continuum. Ari Gerstman of the University Corporation for Atmospheric Research tweeted it very bluntly,
Maybe we need more funding for social science and actually integrate it into prediction and forecasting research?
Vulnerable structures. Many of the fatalities and injuries from the outbreak were associated with mobile homes. Sadly, this is a reality that has long been of concern and is well-studied. Professors Stephen Strader of Villanova University and Walker Ashley of Northern Illinois University published a recent study in the journal Weather, Climate and Society entitled "Fine-scale assessment of mobile home tornado vulnerability in the central and southeast U.S." They write in the abstract:
Tornado-mobile home impact potential is 4.5 times (350%) greater in Alabama than in Kansas because Alabama, in comparison to Kansas, is represented by 1) a greater number of mobile homes and 2) a more sprawling mobile home distribution. Findings reveal that Southeast’s mobile home residents are one of the most socioeconomically and demographically marginalized populations in the U.S. and are more susceptible to tornado impact and death than illustrated in prior research.
Dr. Daphne Ladue, a senior research scientist at the University of Oklahoma, further highlighted this challenge by tweeting the graphic below. She points out that in the affected parts of Alabama, there are relatively few tornado shelters. Research by Strader, Ladue, Ashley, and others can be useful in saving future lives if it is utilized by decision-makers.
The tragedy of this outbreak is multi-faceted, familiar, and heartbreaking. As a professor and scientist at the flagship University of Georgia, it was particularly difficult to watch parts of my state ravaged by Hurricane Michael take another hit from this tornado swarm. All of this just motivates my research and messaging even more.
