Strong Tornado -- Without Advance Warning -- Strikes North Carolina

Tornado Warning Cancer Strikes Again

This Time With an EF-3 Tornado

We've had yet another strong tornado go unwarned -- for at least ten minutes -- in northeast North Carolina. This was another straightforward warning situation. And, to make it even worse, the National Weather Service (NWS) knew there was a tornado doing damage for four minutes before they issued a tornado warning!

I'm not the only meteorologist who has noticed. 

"Concerning/disappointing" is to put it mildly. We know there are injuries and significant damage. We also know the NWS office Raleigh knew the tornado was doing damage for at least four minutes before issuing its first tornado warning of the day!

This is another example of poor tornado warning service; and they seem to be coming at least weekly these days. "Tornado warning cancer" has spread across the National Weather Service. The term "meteorological cancer" was coined by National Weather Service meteorologist Len Snellman in a peer-reviewed piece in the Bulletin of the American Meteorological Society in 1977. The term "cancer," which pertains to inability of the NWS to issue a tornado warning in these straightforward tornado situations, seems appropriate. The retirement of experienced meteorologists and their replacement by relatively untrained young, inexperienced meteorologists clearly leaves the organization weaker. 

Review of This Afternoon's Tornado Signals

At 11:42 am (all times Eastern), the NWS's Storm Prediction Center highlighted the area where the tornado later occurred, which should have given the local offices with warning responsibility a "heads up." One would have thought this notice would have caused the Raleigh WSR-88 radar, which was operating at five minute data intervals, to be put it into SAILS mode with updates at two minutes or faster. It didn't. The radar remained on five minute intervals which is too infrequent to properly monitor tornadoes. 

Let's pick up the radar at 12:06pm. This is the Raleigh NWS WSR-88D. There is rotation beginning as well as a suggestion that a hook might form with the strengthening thunderstorm northeast of Raleigh. 

Note: I would not have issued a tornado warning at this point. 

12:11pm

There are clearly signs of a tornado at this point. The rotation (right) has improved. On the left, there is a weaker echo (yellow) surrounding by stronger echo (orange) which suggests that a tornado may be centrifuging raindrops. Considering the SIGTOR (a measure of how conducive the atmosphere is to producing tornadoes) was a mediocre 1, I may have held off. I'm not sure. 

12:16pm

The image at left show what might be a "debris ball" -- lofted debris -- forming. The rotation, at right, is about the same as before but with the debris ball, I wouldn't have risked it. I would have sent a tornado warning to my clients. 

12:21pm

A tornado warning is unquestionably called for here. At left, a classic "hook echo" signature of a tornado and, at right, tightened rotation. All of this was confirmed by the Terminal Doppler Weather Radar's data (not shown). Unfortunately, the tornado warning wouldn't be issued for another 11 minutes!

However, between 12:21 and 12:25, NBC 12 in Raleigh was independenly warning its viewers. 

There are some good television meteorologists in Raleigh and I suspect the other stations sere doing the same or similar. 

12:25pm 

Wow. I'm just flabbergasted there is no tornado warning at this point. Everything a meteorologist working the radar would want to see is there. The rotation could be tighter but it is infrequent that everything lines up perfectly. 

12:27pm

Significant damage has occurred. 

12:31pm 

The reflectivity data tells us everything we need to know. There is a tornado (hook echo) on the ground lofting debris (pink pixels in the middle of the hook). 

Finally, a tornado warning is issued. 

But, incredibly, even though they know of damage and they know from the coincident 12:31pm radar image that the tornado is still on the ground (because it is lofting debris), they write "radar indicated rotation" instead of much stronger language. 

The warning hit Twitter at 12:32pm.

With an official tornado warning, WEA is activated on smartphones, weather radios go off, communities with sirens sound them, and television and radio stations often go to continuous coverage. Without the tornado warning, people in the path who are going about their business (there was no tornado watch to tip them off to monitor the weather) wouldn't know they were in danger. 

12:57pm

Twenty-six minutes later, at 12:57pm, the tornado evidently is still on the ground with quite a bit of debris in the air. 

Damage

Update: It was at least an EF-3

Pfizer has a large manufacturing plant that took a hit. No serious injuries reported. Local media is reporting that 50,000 pallets of medicine have been "affected" by the tornado. 

There are numerous homes damaged and reports of injuries.

The tornado crossed Interstate 95, which was closed for at least 30 minutes.

Comments

The crucial issue can be summed up with a single image. The rotation path (a proxy for the tornado's path but we will not know the exact path until the NWS does a field survey) was determined with radar. Most of it was unwarned. Raleigh is at lower left (click to enlarge).

I've literally said prayers that none of the people reported injured were seriously hurt. I would ask you to do the same.

The National Weather Service's utter failure to issue timely tornado warnings on major tornadoes is a major scandal. The agency cannot be depended on to provide protection to the American people in these circumstances. 

It is crucial that concerned citizens contact their congressional delegation. Only outside intervention will cause this to stop. I continue to call for the creation of a National Disaster Review Board modeled after the hugely successful National Transportation Safety Board.