Joe’s Weather Blog: The overuse of severe thunderstorm warnings (SUN-8/26)
Good afternoon and for those of you who are reading this from the National Weather Association annual meeting…welcome to the weather blog. This will recap and add some additional information to what has been my passion project over the past several years. For my usual weather blog readers…you sort of may remember this project of mine. I’ve revisited it a couple of times since around 2015 or so…and each time I try to add other perspectives and/or respond to your thoughts and concerns in the comments section of the blog or via FB. Once again I’ll be doing that this year.
A little history first. For those of us within the Intergrated Warning Team (IWT for short)…in an effort to build better communications, for example meeting with and discussing various ways of improving what I/we do in the TV world with others who are not in TV…for example, emergency managers and various NWS and NOAA employees…we get together once or twice per year and have discussions. We listen to each other and IF there is a way to better things we attempt to do that. The Kansas City IWT was the 1st one in the country I think…and we roughly get about 100 folks come out for a day long session.
A number of years ago…I brought up the lack of color consistency with the little TV bugs that stations use in the KC TV market. Myself and a few others wondered out loud IF something can be coordinated. Magically, after numerous emails between TV stations…it happened and the results are still apparent. So changes can happen.
A few years ago I brought up the fact that, in my opinion, and I feel my viewers opinion as well…the usage of severe thunderstorm warnings had gotten to the point that it was essentially “noise” to not only myself but my customers as well. In essence, I felt that very few were paying attention to them anymore. Surprisingly there was more agreement to my statement than I thought there would be…which lead me down a road that I’ve paved for myself…and turned it into my passion project.
Passion projects are good things and bad. On the one hand…I think I’ve brought up something that many of us within the IWT think about on occasion. It’s lead me to do so much number crunching…staring at monitors and seeking other opinions that it’s been rewarding to what I do on a daily basis. On the negative side is that fact that, deep down, what I’m trying to do is change the government ways…good luck with that!
So far I’m a single blade of grass in my grassroots campaign. I’ve had so many within the IWT world say I’m absolutely right in the path I’m taking so I use that as motivation.
My colleagues in the TV world, that I’ve talked to can see my point as well. Today I’m talking to about 100 or so TV broadcasters from around the country and sharing with them my thoughts on this. Others will be in attendance as well…either they agree or disagree but all this is bound to at least generate discussion. Perhaps they will go back to their home markets and visit with their respective NWS office, who issue the warnings, and ask them about it…and generate further discussions…and maybe 10 years from now…things may change.
I’ll be about to retire by then. :)
So here goes…
Some previous blogs/podcasts…I think this was the 1st one in 2016
So that’s a lot already…the presentation that I did this afternoon in front of my colleagues is similar.
Let’s start with the basics…
IN NO WAY IS THIS A CRITICISM OF THE JOB MY COLLEAGUES AT THE NATIONAL WEATHER SERVICE DO DURING TIMES OF SEVERE WEATHER.
OK that’s out of the way. They are required, by mandate, to issue a severe thunderstorm warning when a storm has the potential or is indeed producing 58 MPH winds and/or 1″ sized (quarter-size) hail. This is the mandate and this is their responsibility.
My efforts are to change that criteria…and strengthen it…in other words I want to increase the thresholds to something that is more likely to “be something” and not be anything of significance. It’s the hundreds of warnings issued each year…that I feel my customers have essentially become numb too. If they blow them off…then the warnings are really not serving a lot of purpose.
So I did research…and I looked at a period of time from roughly 2009-a couple of weeks ago and counted each and every warning issued from the various NWS offices across the country. I focused my research on what the NWS office in Pleasant Hill and Topeka did each year.
I then drilled down to the years 2012-2016 and hyperfocused on the warnings issued and how many were issued for what I consider marginal situations.
So let’s start this out with a look at just how many severe thunderstorm warnings are issued each year. We’ll start nationwide…and go back to 2015. All the yellow coloring below represents at least parts of/or whole counties under a warning(s) in a year.
and finally where we are through mid August of 2018
You know what is fascinating about the slides above…that from 2015-2017…five or take a couple of hundred warnings…each year was around 18,500 warnings.
You can easily see how the warnings are issued with the fewest west of the Rockies (mostly) and the most from east of the Rockies.
I then wanted to break down the data…and focus on individual NWS offices with, again a hyperfocus on the two offices that cover the FOX 4 viewing area the most. This data goes back to April of 1009. The reason why I started there is that, at least for the offices in MO/KS…we changed our minimum hail criteria for a SVR (severe thunderstorm warning) to 1″ from 3/4″. This was done…why? To try and reduce the issuance of warnings for storms that did little to no damage based on the 3/4″ hail criteria.
I then looked at other offices in the nation starting in 2010. I believe that’s when they too changed their criteria to 1″ hail as a trigger for a SVR
My goodness…so many warnings. Look at Norman, OK. Now that office covers central and western OK…a lot of land in a part of the country where storms are common…but more than 5500 warnings is NOT doing anybody any favors…what is that some 600/year? While not as bad…Jackson, MS with a smaller geographic reach than Norman’s responsibility has issued almost 4500…some 500/year. When viewers, and granted it’s not the same viewers time after time, are bombarded with sooooo many warnings…why won’t they blow them off after awhile…why won’t they say to themselves…well odds are this storm is going to do this…
Which is not much. Tiny hail…scary looking clouds…a few tree limbs broken…and that’s IF it even does that. Oh and that’s IF it occurs in a place where somebody is actually affected by that “damage”. As my colleagues at the NWS tell me…one of the tough things about the warning process can be verifying IF the storm is actually “severe”…especially at night…especially in rural areas. THEY WANT TO BE RIGHT…I WANT THEM TO BE RIGHT…
So this lead my to my 1st blog in March of 2016…asking my usual blog readers…who are more weather savvy than perhaps my typical TV weather viewer (they’re reading my blog…they care about the weather)…their thoughts on where we were. I asked them a couple of questions…
I was delighted that they pay attention to the warning process when it comes to SVRs…GOOD
I was interested in their response to the work I was undertaking…and they voted that YES they would pay more attention IF the warning had more “meat” to it.
So let’s drill down…from the period of 2012-2016…here are all the SVR warnings issued from Pleasant Hill. Look at 2012…low number right…remember what happened in 2012…the drought.
So over 1000 warnings in that 5 year period.
I wanted to specifically look at the warnings issued for the minimum criteria…58/60 MPH winds…and or 1″ hail. That is the root of what I’m trying to change.
Let’s start with wind only…and for the sake of rounding values they way they issue the warnings 58/60 MPH are interchangeable really. This next slide is for SVRs issued based on min criteria winds ONLY and the percentage breakdown with the overall number in that year.
Let’s say, on average about 25%…so about 1 in 4 warnings were issued based on min criteria winds ONLY.
Now SVRs issued based on min criteria hail (1″/quarter size) ONLY
Let’s say about 1 in 10…not too many really.
Now the jackpot…warnings based on min criteria wind AND/OR min criteria hail. Often when a warning comes out…there will be two reasons why…60 MPH winds and 1″ hail potential. I’ve merged all these together and come up with the following slide which includes the ones above and the 60 MPH/1″ tag on the warning
Now we’re up to about 7 out of every 10 warnings…and hence the “noise” factor in my opinion. I then did a quick perusal of the SVRs issued in the last 1 1/2+…and the data for min criteria issuance really hasn’t changed that much from the Pleasant Hill office.
Farther west…towards the NWS Topeka office…it was about the same…although overall numbers were a bit different. For the sake of brevity (I’m at about 1600 words so far in this missive)…
Total warnings from 2012-2016
Warnings based on a combination of the min criteria…
Close to 60% based on some sort of combo criteria…
The recent trends (actually better)
The thing is…when I looked to see IF the storms were even verifying…in other words were there reports from the public…law enforcement…emergency managers…anyone…that the storms were actually doing what they were indicating on radar…it was at best so-so.
2014 in particular was a bad verification year…so many warnings…so many min criteria warnings…so few verifiable warnings…there has to be a better way.
AGAIN…NOT a criticism of the job at hand. Verifying warnings IS NOT EASY for the reasons mentioned earlier and this doesn’t necessarily mean that 60 MPH winds and/or 1″ hail didn’t happen somewhere.
I then looked at storms that SVRs were issued and thought to myself…well what about when a SVR is issued and then the storm strengthens…perhaps to 70 MPH and/or 1.5″ hail…in my little weather world…this could be the new warning criteria. So how many storms like that occurred…where a SVR would’ve been issued anyway.
A handful at best…
So let’s take a little journey into Joe’s Weather World…where snowstorms are perfectly predicted from days away…where rain amounts are always forecast correctly…and when it rains and/or snows at the exact time I say it’s going to occur.
OK it’s fantasy…but what am I trying to change?
How about a storm that has the ability to do this…
Something more meaty…how about 70 MPH winds and/or 1.5″ hail. In conversations with others…including insurance folks…this seems like a good starting point. I’ve had conversations with agricultural folks as well about this whole thing. They can actually buy special crop insurance IF they know a storm is coming a few hours out. SVRs though are not necessarily meant to do that…actually way back in the day they were mainly meant for aviation concerns…
What would happen to the number of warnings in the course of the year…
Well HELLO THERE!
Notice how, instead of hundreds per year…we’re down to 25-50 or so.
I have to believe (it’s my motivating point for all this) that the fewer the warning…the fewer times we say severe thunderstorm warning in effect for _____…the fewer times we light up the screen with warning information…the better the end result can be. It’s a fatigue factor really.
Back to a poll question…this time asked at our KC-IWT meeting before and after a presentation like this.
Look at #1…the KC-IWT felt that almost 70% of their customers DID NOT PAY ATTENTION to SVRs in their current state (that’s NOT good).
#2…almost 92% (after my talk) felt change was needed…92%. Let that sink in. I’m good but I’m not that good ;)
So what do we do?
I don’t like A…I like B…and I feel C might be the middle road. Many of my colleagues at NOAA and the NWS that I’ve talked to actually feel we should change the min criteria…which I’d love to do.
My concern is that there are things that I’m NOT thinking of…effects that I don’t know IF we were to change. I wonder IF C would make the most sense and be the easiest/quickest to work towards. It can’t be that hard to create a new header for a new product. It would be OK for me to have the ability of showing a part of a county on live TV that has a different color code to indicate that THIS part of THIS county has a higher likelihood of seeing something more significant!
I know this is going to shock you…but I’m getting to the end…
There are questions and concerns…
For those who don’t know…HAZsimp is an effort to cut back on the types of warnings/advisories issued. I think though sometimes addition can be a good thing if needed.
There are no doubt parts of the country that would want NO part in this. There was division when the transition was made from 3/4″>1″ hail criteria…it was messy. In the end though…the earth kept spinning around…and it was OK.
So maybe look deeper at these ideas for areas of the country that are in warning fatigue.
The Storm Prediction Center thing is a question that I don’t have an easy answer for. They base their severe thunderstorm watches on 60 MPH/1″ criteria storm chances…do they alter the way they issue watches based on the part of the country that they are being issued for? That’s a toughie and I need more talks with them for solutions on that one. With that said though THERE ARE DIFFERENT CRITERIA FOR WARNINGS/ADVISORIES already. A winter storm warning in Atlanta, GA differs in criteria from a winter storm warning in KC.
I keep coming back to this though…when it comes to damaging storms…
Back to my blog readers…
After presenting some of this information I asked them a couple of questions…
The answer to #2 was fascinating to me…84%.
Dr Labosier reached out to me a few months ago after hearing my thoughts on all this…he is in the process of trying to fund a bigger research project with more specific questions that would be released to the general public…we’ll see IF that goes anywhere.
So that’s it…2500 words of my passion project.
I hope you enjoyed my presentation today…and I hope you tell me what you think of the situation. IF you think I’m crazy…that’s OK! I want to hear what I’m NOT thinking about…or where I’m wrong…or hopefully where I’m right!