The Social Side of Communicating Weather

[Randy Jennings]

I ran across a fairly new blog called The Weather Social: The Social Side of Communicating Weather at http://thewxsocial.com/. It describes it self as "a collaboration of weather and communication experts focused on narrowing the gap between scientific messages and society. With unique backgrounds and experiences, this group hopes to inspire a progressive discussion contributing to long-term initiatives in the broad and growing fields of meteorology and climatology."

The most recent blog entry is from Mark Fox, NWS Fort Worth WCM. It is entitled "We didn't know it was coming" and it is about the Dec. 26, 2015 tornados in the DFW area.

This was one of the most commonly used phrases by the victims of the December 26, 2015, tornado outbreak. Twelve tornadoes impacted parts of North Texas and the Dallas-Fort Worth Metroplex, killing thirteen. For several days leading up to the event, nearly every forecaster and meteorologist were talking about severe weather in North Texas on the day after Christmas 2015. The Storm Prediction Center had highlighted the area for days in advance. Our office had been talking about active weather for that weekend for over a week.

And yet, “we didn’t know it was coming”.

The forecasts for this event were good and the meteorological set-up for the event was fascinating. I won’t go into that here, as I would rather expand on something equally as fascinating revealed from conversations with survivors during the damage surveys: a gap between what the weather community thought it was saying and what our North Texas communities were hearing.

You can read Mark's full post at: http://thewxsocial.com/2016/03/18/we-didnt-know-it-was-coming/ . It is a similar but shorter version of his presentation at TESSA two weeks ago. In that presentation he talked about the importance of spotter/chaser reports of damage/impacts and how it allows them to issue warnings and the IWT to inform the public in ways that get people to act.

 

[Jeff Duda]

their main three points about ways this event went wrong:

• “unusual” time of year
• uncertain geographic recognition of the threat
• non-personalization of the threats

And a testimonial:

The family next door told us that while they were watching television that night, they assumed they were fine as the broadcast meteorologist kept mentioning Red Oak. (Red Oak is a bigger city on I-35E about 10 miles away, which was mentioned in NWSChat several times as a reference point). That assumption quickly changed when the trees started breaking and the house began to fall apart.

It seems like these types of findings have come up before in events past. So why does it seem like nothing has been done about it? There is one S-band WSR-88D, two C-band TDWRs, and at least 4 working CASA radars serving the DFW metroplex as of that event. How is it possible that some broadcast meteorologist in such a large market with so many tools (that many other markets don't have) kept using an unhelpful or incorrect frame of reference? I'm sure all of the weather centers in each news media station in DFW has a GIS database of every single suburb, and probably many neighborhoods. Seems like someone made an error that didn't need to be made here. Are operational meteorologists (especially the ones that provide warnings) not reading service assessments from past events? The one from the Joplin tornado highlighted the 'non-personalization of the threat' issue pretty strongly.

It really seems to me that there have been enough recent similar events with investigation by social scientists that has been published that these types of tragedies should not continue to occur. It even seems like broadcasters should point out the three bullets above when they're wall-to-wall for the warnings.

Days before: "We want to make sure you are aware that it looks like there may be severe weather in our area on XXXday (N days from now), including the possibility of tornadoes. This is not typical weather for this time of year, which means you need to pay even closer attention to this than you would normally.
Fewer days before: "It is looking very likely that severe weather, including tornadoes, will threaten our area on XXXday (N days from now). Start making your 'day-of' preparations now, by making a plan of knowing where you'll be between 12 PM and 6 PM that day and where you'll need to go if severe weather threatens your location."
The night before and/or the morning of: "The severe weather threat for today is very high. Treat it like a typical springtime/summertime situation. The possibility for particularly bad severe weather, including strong tornadoes, is there. This is a potentially serious situation. Get ready to activate a severe weather plan at a moment's notice should severe weather threaten and a warning be issued today. This is one of those days where you will want to take any warnings seriously."

Yes, many people are still not going to see any of that because they still probably won't be watching the weather forecast on the 6 o'clock news. But I sure hope the above dialog is being included in regular weather broadcasts during 5, 6, and 10 o'clock shows during the days leading up to the event, when most people are likely to be watching. It should also be on their website and probably out on social media, too, to reach as many people as possible.

 

[Gregg Schulz]

There is a gap between actual vs perceived danger in regards to warnings given to the public. It often takes several key indicators for the public to take action. Sometimes they won't take action until the neighbor's home across the street with blowing away. As Jeff eludes to, this is the problem we all face as part of the warning decision process and something the NWS and Emergency Managers will need to deal with for many years to come. All we can do is give the most accurate information we can to those who can warn the public and warm them ourselves when we can.

 

[Paul Knightley]

Being on the operational meteorologist side of things (not in the US, I admit) communication of the 'story' to the public is still something which is not always easy to achieve, especially these days with the proliferation of weather apps, Facebook pages, etc - I think people don't always know from whom they're getting their info. There is also an expectation in society these days that everyone should be warned about everything. Of course, on the flip-side, if you're providing a warning service then you should expect to get some tough questions!

Having watched a lot of US severe weather coverage on TV (both from afar via online, and whilst there chasing), the overwhelming feature I find is that it can be very confusing at times. Of course, being a met. and storm chaser, I can pretty quickly 'get' the situation from the radar imagery they show, etc, but for a non-met just trying to work out if they need to take shelter, it sometimes seems extremely confusing. On the El Reno day, we had the News 9 feed streaming in-car, just for info - and it sounded terrifying. I don't know what residents would have thought, or known, watching it, but you can bet that they thought the world was about to end.

What's the answer? I don't know! But I think the overwhelming amount of coverage perhaps makes people *think* they will be fully informed, but when you really watch it closely, it's not so easy to see how they can know what's going on. When all is said and done, people are responsible for themselves - but the huge amount of coverage seems, at least to me, to cause some people to absolve themselves of personal responsibility.

 

[Michael Snyder]

An example: I'm an aviation forecaster in Seattle. I will send out a forecast for a specific area, SeaTac airport in this example. I then send out my forecast to a large number of people. I have to be quite careful in what I say, how I say it, and even the order I present the information. For instance I found that the order in which you present certain aspects of the forecast has a huge effect on what get absorbed by the users.

I usually end up sending out the summary and highlighting main points at the top of the email, followed by a basic timeline of when events are going to occur, then followed by a more technical discussion of what is going to occur.
Take into account that it's tough to communicate details to people who do this for a living, and it's not really a surprise that the general public always has a percentage of people that are caught off guard.

To me the combination of most people never even seeing a tornado in most areas vs. the actual number of people affected by Severe weather vs. the time between experiencing these events will always have some people unaware, even in the biggest events.

 

[Steve Holmes]

Human nature, folks. Good luck cracking that.

1) One of my weird interests is reading NWS post-tornado service assessments (one reason I’m not invited to a lot of parties). It’s strange what trips people up.

The 2011 Alabama assessment said most people don’t know what a large tornado looks like. The same response came in the 2008 Picher (OK) assessment: “The tornado did not look like a tornado.” Think about it. All the graphics and drawings of tornadoes show curvy, tapered funnels, not a wedge. They may have been looking straight at a tornado, and not registered it as one.


2) The Optimism Bias, the belief that bad things happen to other people, not to us. A neuroscientist says our brains are wired this way (https://www.ted.com/speakers/tali_sharot).


On WeatherBrains podcasts, one guest said people shop around for forecasts until they find one they like. Another guest said they can see shades of grey in a 100% forecast. So if nine radio or TV stations are saying, “It’s coming your way,” and the tenth is running Gilligan’s Island, people let their guard down.


The Optimism Bias is not all bad, though. People in Ottawa, Illinois and Perryville, Missouri who believe their recoveries will be remarkable are better equipped for what’s to come than are the pessimists.


3) The National Institutes of Standards and Technology did a post-mortem on Joplin. It found that the sheer number of tornadoes in the area (12.5 a year within 80 miles of Joplin since 1950) made people *less* likely to take tornadoes seriously. “They hit all around, but not here.” It ties in to that notion that folks live in a bubble, protected by a river or a hill or a Harry Potter spell.


A friend works at a restaurant that became a meeting place in the weeks following Joplin’s tornado. His eyes got wide when he talked about the number of people who “didn’t believe it could happen here.” He believed. He was in the Newton County 2008 tornado (just south of Joplin) that killed 21 people in OK & MO. People around here should also remember the 2003 outbreak that killed 25 north and south of Joplin.


4) I keep hearing about how the social sciences are going to change the way meteorologists communicate, but I have yet to hear any specific examples. Is the language of Impact-Based Warning their doing? If not, what is? As Chuck Doswell asked on WeatherBrains, paraphrasing here, “When are we going to have something we can do something with?”


5) Inviting in the social sciences is not a new idea. The NWS assessment after the 1965 Palm Sunday tornadoes recommended getting behavioral-science groups involved. Did that not happen until that last few years?


6) Just curious. What do they teach about communication in college-meteorology programs? Have any schools recommended social-science classes?