It’s Time To Stop Issuing Tornado And Severe Thunderstorm Watch Boxes

No, I don’t want the Storm Prediction Center to stop issuing Tornado and Severe Thunderstorm Watches. I just want them to stop issuing the boxes.

If you’re confused, I’ll explain.

If the Storm Prediction Center feels there is an organized threat for severe or tornadic thunderstorms, they will issue a Severe Thunderstorm or Tornado Watch. But what exactly do they issue? Years ago, they would draw a parallelogram (like the one pictured below), and other text bulletins:

may19-oldwatch

If you were in the box, you were in the watch. If you were out of the box, you weren’t in the watch. It was that simple.

Nowadays, the box is issued, but so is a list of counties in the watch. First the box comes down:

aug18-watchtext

Then the county list comes down (this example is a status update, not the initial update):

aug18-watchcounties

No problem, right? You see the box, you see the counties, and you know whether you are in the watch or not. Right? Not always.

Sometimes the watch and the counties don’t match up. Here’s a watch from earlier this year:

aug18-watchboxexample1

Is Charleston, West Virginia in the watch or not? Charleston is outside of the box, but is in a highlighted county. How about another example? Suppose you’re watching TV in Rapid City, South Dakota. Are you in this watch?

aug18-watchboxexample2

If the TV meteorologist on Channel A shows only the box, you’re “out” of the watch. If the TV meteorologist on Channel B shows only the counties, you’re “in” the watch. If the TV meteorologist on Channel C shows both, you’re “in” and “out” of the watch.

Here’s another tricky one:

aug18-watchboxexample3

What if you were in the southeastern part of the county just east of Colorado Springs? This area is in the box, but not in a highlighted county.

How about a watch where a few counties west and north of the box aren’t even close to the edge of the box?

aug18-watchboxexample4

Because the box “must” be a parallelogram, SPC color outside of the lines. Or perhaps the local National Weather Service forecast office wanted to include areas in the watch, and the SPC agreed. Storms rarely fit in parallelograms.

It gets more confusing. Suppose the threat for severe storms is high, and a Particularly Dangerous Situation Tornado Watch is issued:

aug18-watchboxexample5

This Particularly Dangerous Situation is particularly confusing. If you’re in Fort Campbell, Kentucky, are you in this watch? This is a bad time to be unsure. There were several strong tornadoes in this watch.

What about when several counties in the box are not in the watch? It has happened.

 

aug18-watchboxexample6

So let’s go back a second…was Charleston, West Virginia in the Severe Thunderstorm Watch above? The answer is yes. How about Rapid City, South Dakota? They were. Fort Campbell, Kentucky? Yes. The box doesn’t define the watch; the county list does. In other words, if your county is on the list, you’re in the watch. The box doesn’t matter.

Suppose a pair of watches are put up side by side, and you’re on the edge. One watch is in effect through 3am, and the other is in effect through 6am. There’s a big difference there. That’s a lot of lost sleep if you think you’re in the second watch and actually in the first.

This brings me to my point: it’s time to stop issuing Severe Thunderstorm and Tornado Watch boxes. Being in the box or out of the box means nothing, so why issue it? The box is a legacy product that is outdated and only confuses those reviewing the watch. The box idea worked in the 1960s, but it doesn’t work now. The only reason I can think the box is around is the Internet and media. Intellicast.com will show Tornado and Severe Thunderstorm Watches in box form. Some TV meteorologists will, too. I’m not sure why they do. But change is slow, so I suppose it’s easy to not make a change.

I’ll argue, however, this change needs to be made. Severe Thunderstorm and Tornado Watches are not life or death, but they are likely the most – at least least one of the – important types of watches issued. Many people have died in these watches. The last thing I want is for someone to get hurt or killed because they weren’t sure if they were in the watch.

Confirmed Tornado In Sardinia

At approximately 11:16am Wednesday morning, a tornado damaged homes in the city of Sardinia, Ohio (of north central Brown) County. The National Weather Service confirmed the tornado at 3pm this afternoon, but radar actually confirmed the tornado shortly after it occurred.

Here was the snapshot of radar reflectivity (the shower and storm mode) from the National Weather Service radar at 11:16am Wednesday:

aug17-reflectivity

The Tornado Warning is outlined in red. Note the weaker echoes between stronger echoes in the image above compared to the radar velocity (the “Doppler” part of Doppler radar) below:

aug17-velocity

In the highlighted area, the wind on the west side of the storm is moving away from the radar to the north of the storm (the red area), and the wind on the east side of the storms is moving towards the radar to the north of the storm (the green area immediately right of the red area). This is the circulation associated with the tornado.

How was it possible to confirm this tornado in real-time? See the correlation coefficient image from the National Weather Service’s radar at the same time:

aug17-cc

That “cool” colored spot in the middle of red colors is a debris signature. It is very close to the rotation on radar and the hook seen in the reflectivity image. Correlation coefficient shows the correlation in shape and size of objects (raindrops, hailstones, etc) in each pixel. A high correlation (red colors) between objects suggests objects the radar samples are relatively the same size and same shape. A low correlation (cooler colors) suggests objects are of different shapes and sizes. In this case, the objects are pieces of trees, someone’s home, or crops being lofted into the air and being sampled by the radar. As a meteorologist, you hope you don’t see this. This signature (really, all three of them) confirmed the damaging tornado shortly after it caused damage.

See the comparison of the products:

aug17-radarloop

The tornado had a damage path that was 1 mile long and up to 100 yards wide. The maximum wind was 75mph. This was the first confirmed tornado in Brown County since March 2, 2012; the 2012 tornado also went through Moscow, one of the deadliest Tri-State tornadoes in recent history.