CHARLOTTE, N.C. -- Many will not forget the early morning hours of Saturday March 3, 2012 in the Reedy Creek neighborhood. Around 2:34 a.m. a EF-2 tornado with winds of around 135 mph touched down with a path of destruction 200 yards wide and 3.2 miles long. A total of 192 homes were damaged or destroyed in Mecklenburg and Cabarrus counties.
Approaching Storm system:
It was a long day and night of tracking the Tornado Outbreak to our Northwest and West Friday. I was sending out updates all day and night and around 11:35 p.m. after the news I made note of this radar image seen below. It shows that all the cells heading into the Western Carolinas currently had tornado warnings with them. This didn’t mean that they would continue too, but that these cells would have had a history of procuring tornadoes warnings as they approached.
The 2 super cells in western North Carolina had me worried at first but the ones in Georgia weren’t far from my mind. The 2 cells in Western North Carolina weakened as they moved northeast into the cold wedge over the foothills. The 2 cells in Georgia starting heading right up I-85 towards Charlotte.
Once they got to the upstate of South Carolina they started bowing out becoming “bow echoes” which made me concerned about straight-line wind damage. Below was the view of the first storm as it approached the Charlotte metro through York County.
The 1st dying supercell:
Thankfully the above storm just produced lots of lightning and thunder but very little winds around 40 mph. What this first storm did though it appears it move the wedge front slightly further north from the SC/NC line to areas just north of Uptown Charlotte. If you look at the hourly observations at the Charlotte airport you can see how temperatures and dew points went from the low 50s into the upper 50s. The wind also shifted from NW or calm to out of the SW meaning the wedge front shifted north. Placing the eastern part of Mecklenburg and western Cabarrus county in the warm unstable air mass. Or more likely right on the wedge front itself. Which might would have increased the storm relative helicity or wind shear as the 2nd storm moved in from the west. The increased SRH could cause a sudden rotation within the 2nd cell.
The 2nd Storm:
This storm right away looked more serious that storm #1 but not tornadic at first. This was a classic bow echo which meant it looked like a bow from a bow and arrow and was likely going to produce damaging winds of 60mph or higher. In fact the Doppler velocities where showing severe winds with this cell. This was the cell as it approached Uptown Charlotte.
Here’s where I know we are going to get damage this storm is bowing out with 60-70 mph winds at the apex of the bow. At this time I know we should have been under a severe thunderstorm warning at the least and we might have a developing tornado. Though that doesn’t become clear until the next volume scan of the radar.
The next volume scan of the radar I see this!
I know we might have a tornado forming or on the ground based on this image and I tweet this.
The whole time I’m using our Ustream account which I have had running since after the 11pm news and I tell those watching this looks bad. Then as the stream is going I get the warning, but it’s for Stanly County which isn’t even where the tornado is? At the same time my Twitter feed is exploding with people reporting power outages in the University, Harrisburg and Mint Hill areas. This is when I know we have damage occurring even if its just trees and power lines. The radar at this point clearly shows a tornado and I detect what appears to be a small debris ball which tells me we have damage occurring.
The scanner starts blowing up along with my Twitter and Facebook pages while I’m on TV showing the warning for Stanly County. All along I’m telling people in Cabarrus county to seek shelter because that’s where I know the actual tornado is. I track the storm until about 3:17am when I know the rotation is gone over northern Stanly County. The warning was supposed to go until 3:30am but is canceled early at 3:22am.
It would be easy to say we knew around 4-5am but not until close to 6am when our crews were seeing the damage and the sun came up did we know the scope. AirStar 36 started showing the view from the air and I knew we had at least an EF-2 damage from a tornado.
Why No Warning?
It’s really easy to blame the National Weather Service for not issuing a Tornado warning for Cabarrus County. I don’t think there was time to warn for Mecklenburg due to how fast the tornado spun up. Clearly they are second guessing the process down at the NWS in GSP. I will explain why this happened in a minute, but to me there was no excuse why there shouldn’t have been a Severe Thunderstorm warning in place for Charlotte and Mecklenburg County.
I’ve seen warnings for less around here and the impact on a major metro would make me error on the side of caution. There was clearly 58mph or higher winds with this storm and this warning might have alerted people to the threat of dangerous weather, even without the tornado warning. I don’t like getting into a blame game and in some ways there are many factors that go into the error of the warning that morning. One thing is for sure no one should have been un aware that severe weather was possible overnight. We just lacked the final step in the process of getting a timely warning out via the NWS.
So what happened?
Many factors come into play with the lack of a warning. There are 2 main reasons why I think it happened, both very complicated..
This storm produced a tornado very quickly. Which means there was no sign of a tornado forming until it actually touched down. This meant that Mecklenburg Co. was never going to get a tornado warning but clearly should have been under a severe thunderstorm warning.
This tornado formed between volume scans of the Doppler radar. People may think Doppler radar is “LIVE” it really isn’t. It takes one min for the radar to rotating around 360° or 1 RPM. Then depending on the mode it is in it will do that scan at 1-14 different levels of the atmosphere to get what we call a volume scan. So even in a ideal situation the radar operator only gets the data for part of the area every 5-6 minutes. Plus since it’s both nighttime and 2:30 a.m. spotters are not useful.
Human Error & Logistics:
Charlotte is on the border of 3 different NWS forecast offices which I will get into the reasons why in later. For now it is what it is and it leads to warning and watch issues all the time in the Charlotte metro.
Think about how warnings are issued for the Charlotte region and metro. When the people responsible for them live at distances from Uptown Charlotte of Greer, South Carolina 79 miles away, Raleigh, NC 130 miles and Columbia, SC 84 miles away. This also means the Doppler radars being used to scan the skies for tornadoes across our area are looking at the very limits of their capabilities.
This means that Charlotte, NC is the biggest city in the United states without a Doppler radar located within the acceptable range of the radar. We do have a Terminal Doppler radar located near Mountain Island Lake owned by the FAA to help cover the CLT airport. It has limitations & means the NWS must flip between 2 sources of radar data.
The lack of a radar and the area being split between offices makes for a coordination nightmare. All three of these offices have their own Doppler radar which they primarily look at. All of them are too far away to see the lowest levels of storms around Charlotte. So they flip between the TDWR near Charlotte which has limitations of its own but at least it’s here.
Lucky for us we at NewsChannel 36 have our own c-band radar in Dallas, NC to see storms around Charlotte. This helps me see things and you but I can’t issue tornado warnings that make weather radios, texts, emails and sirens go off. Which in this case being the middle of the night was needed. Unless you were awake watching TV or on-line with me you had no idea this was coming.
So how did we end up with Charlotte not only not having a local NWS office but also a Doppler radar? Politics? Maybe?
The National Weather Service used to have many more offices including one right here in Charlotte. There also was an early radar here but not the modern 88-WRD that are in use today. In the late 1980’s and early 1990’s the weather service started to modernize.
The modern Doppler radars were coming out and the Federal government wanted to consolidate offices. This was necessary to save some money but also because the modern technology allowed for covering larger areas. So the offices went from covering 10-20 counties to 40-80 counties in some cases. GSP went from a 10 county area to 46. In most of the country that meant the offices were consolidated into new offices close to large metro regions to maximize the coverage of higher population densities.
Except for here in Charlotte which should have been the home of the new consolidated office. It might not have been at the airport but in one of the surrounding counties maybe even Rock Hill to cover both North and South Carolina. Like all local projects politics likely played a role in the office being moved to Greer, SC. Just like senators and congress persons fight over which military bases get shut down or where other federal money goes. The stronger more influential politician usually wins. In this case in 1995 the winning location for the new consolidated office ended up being Greer, SC and not anywhere near Charlotte. In 1995 South Carolina had one of the most power senators around….. Strom Thurmond who just so happen to be born and raised in Edgefield, SC in the upstate of South Carolina.
How do we fix the problem?
I for one don’t think you are going to get a NWS office here in Charlotte, that doesn’t make sense. The county watch areas do need to be adjusted so that metro counties are not split between 3 offices. I also strongly feel like we need a stand alone National Weather Service Doppler radar built here around Charlotte. It would be really easy to replace the TDWR or upgrade that location with full modern Dual-Pol Doppler radar that everyone would have access to. This in my opinion would help solve not only the tornado warning issues but a whole bunch of forecasting issues in Western North Carolina.
What do you think? Please feel free to leave your ideas in the comment section below.