What’s wrong with this picture? A lot!


By: Gary Szatkowski

Who thinks this is strictly a physical science problem? Hopefully, you did not raise your hand.

I started chatting on Twitter about a potential East Coast snowstorm on Sunday, January 17th. The National Weather Service (NWS) office in Mount Holly, NJ issued its first briefing package for the potential upcoming event on Monday, January 18th. Five more briefing packages would be issued before the first snow started to fall on Friday afternoon, January 22nd. Well before the first flake of snow fell, NWS offices all across the affected area issued Blizzard Watches and Winter Storm Watches, and then upgraded them to Blizzard Warnings and Winter Storm Warnings.   Media and other parts of the weather enterprise reacted similarly, talking about the impacts of the storm multiple days in advance of the first snow.

And despite all that advance notice and all that advance information, two tractor trailers jackknifed on the Pennsylvania Turnpike around 800 PM Friday evening, only several hours after the storm had begun.  The accident occurred in one of the most remote and hilly stretches of the PA Turnpike, with roughly 40 miles between exits.   Stranded motorists spent over 24 hours on the PA Turnpike, with wind chills in the single digits and near-blizzard conditions.


Photo credit: Michael Watkins, Associated Press

Continuing a pattern for me that became painfully obvious during Hurricane/Superstorm Sandy, we now have the capability to often correctly identify and warn about very high impact weather events 5+ days in advance.   But what has been a lagging indicator has been the societal response to this information. Some people over-react, some people under-react (sometimes with fatal consequences), and I’m left with the growing conviction that an increasing amount of excellent weather information is being left on table, not being fully leveraged by our partners and by the general public.  

Is there anyone who wants to argue that what happened in the picture above would NOT have happened if I had simply started to tweet 24 hours earlier? Or that briefing packages had started a day earlier? Or watches & warnings had been issued earlier? Or media and private sector partners had started talking about the storm a day or two earlier? Once again, hopefully not.

This is not, at a fundamental level, a physical science issue. Yes, we will continue to get better at weather forecasting, and yes, accuracy rates will improve and lead times will grow. But that doesn’t address the problem above.

And it is not just hurricanes or nor’easters. The problem stretches across the weather spectrum. When I first started working for the NWS over 35 years ago, tornado leadtimes were under two minutes. And please don’t ask about the accuracy rate.   When your tornado warning leadtime is two minutes, and you are in the tornado warning business, your goal is simple, double the leadtime. And when you do that, double it again. And when you do that, double it again. But eventually there comes a time when doubling the leadtime (16 minutes, 32 minutes, 64 minutes, etc.) is no longer enough. No longer even close to being enough. When people have that much leadtime for a tornado warning, or a growing awareness of an upcoming tornado event (or hurricane event or coastal storm event), a whole range of societal behaviors becomes possible, often with outcomes we neither intended nor desired.    

We are in great need of better language and better communication regarding the societal impacts from these very high impact weather events. And that brings us here, to ‘The Weather Social.’   For right now, the gap between our scientific weather messages and society is growing.   It’s a compliment, as it’s a reflection of our growing technical skill. And it’s a challenge, as our goal is for people to survive events that are clearly survivable, and lessen damage from weather events that can clearly be mitigated.  

Let the discussion here at ‘The Weather Social’ begin. We have much to accomplish.

6 thoughts on “What’s wrong with this picture? A lot!

  1. […] This statement by Gary Szatkowski, then Meteorologist-in-Charge at the National Weather Service Office in Mount Holly, New Jersey, refers to a significant contributing factor in nearly 500 vehicles being left stranded on the turnpike for more than 24 hours during the record-setting January 2016 mid-Atlantic snowstorm. Here is a link to Gary’s full article: https://thewxsocial.com/2016/02/09/whats-wrong-with-this-picture-a-lot […]


  2. We’re a free society. All you can do is put the information out there about the risks and the uncertainties, and let people make up their own minds about how much personal risk they are willing to take.


  3. […] altering plans and inconveniencing many which would explain irritation in a false alarm. Much like Gary Szatkowski pointed out in February, this is more a social science issue and less a physical science […]


  4. Steve Szyszkiewicz

    You dont have any control over how other people live their lives. Your office issues forecasts that reach millions of INDIVIDUALS. You are not responsible for their behaviour. Life is controlled chaos. So many interactions.


  5. Gary, I seriously doubt that many of the people caught “unaware” of the storm were actually unaware at all. Big business often requires employees to go when they shouldn’t and to push all limits. There are also others out there that just have the attitude it won’t be them getting caught out there.

    Of coarse, due to the occasional swing and misses with the forecasts, we generally have a reluctance to close businesses before the worst of the storm sets in. In search of the $$, sacrafices are made, whether right or wrong.


  6. I don’t know the details behind the PA Turnpike event, but I think in many cases, knowledge about the upcoming weather event still isn’t going to offset demands of employers. Especially in the ongoing recession, businesses continue to be more concerned with company goals and success then with employee safety when it comes to weather events. I don’t know that the disconnect is with the dissemination of information so much as it continues to be potential risk vs known risk.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.