I’m sick of seeing snow totals maps. Perhaps more than anything else in the industry, they perpetuate a perception that forecasters don’t know what they’re doing. Yet we continue to indulge.
Nothing but maps of snow totals. Why? Sets up for potential backlash. Impacts! This storm will be high impact whether 6 or 12″ of snow fall.
— Josh Eachus (@JoshEachus) March 13, 2017
I did a quick and dirty postmortem of social media in Philadelphia one day before the “Blizzard of 2017.” I looked at Facebook and Twitter messages from the four major local television stations and the National Weather Service, Mount Holly. I categorized each message (Facebook post or Tweet) based on the accompanying picture, or text if no picture was included. Messages were labeled as follows:
- Snow Totals – forecast amounts or model projections
- Threats – meteorological hazards or watches/warnings
- Impacts – what could occur as a result of the weather (treacherous roads, power outages)
- Uncertainty – probabilities, forecast challenges with this scenario
- Other – extraneous content that doesn’t fit into the other categories
*Why did I choose pictures only? Given that humans are now argued to have their lowest attention spans in history—only seconds, if a picture isn’t captivating, it may be glazed over. Many I have talked to admit that this is how they use social media.
The statistics showed a trend that I expected to find. With a high impact event only hours away, very little focus was placed on what impacts may occur. Instead, emphasis was placed on snow amounts, warnings, watches, wind speeds, temperatures and varying other information. To me, this implies that we believe the forecast user is capable of interpreting the meteorological variables into possible impacts.
On 13 March, one television station provided 4 impacts tweets and 1 Facebook post. NWS did not provide any.
*I think it is really important not to confuse threats and impacts in this conversation. A threat is heavy snow and high wind, an impact is what could result from that threat such as downed tree limbs or power outages.
— Gina Eosco (@WxComm) March 14, 2017
Certainly we cannot forecast flight delays or school closures. But we can offer possible consequences of the weather that is anticipated. Studies that focus on microblogging and website posts have found that there is a tendency in the meteorological world to highlight the imposing weather phenomena and not what societal issues may arise or the actions people can take to mitigate. I’m not sure if the reason for this has been determined yet. Perhaps it comes from an abundance of enthusiasm for the subject or a comfortable topic when adrenaline and stress are elevated.
*Days before the event, impact messaging was included on social media. But when attention was highest, one day before, it was sorely lacking.
— Susan (@CommDocPA) March 14, 2017
Instead of talking to people, we are talking the language of scientists and meteorologists. I am thoroughly impressed with the fact that 10 miles could be the difference between 6 and 12 inches of snow—but the average working parent is not. They ask, “Do I have to send my kids to school or will it be closed?”
So, why are we so focused on these amounts? In speaking with friends and family, it has become clear to me that the spread of such information is often like a game of whisper down the line. To continue the 6 to 12 inches example, people seem to latch on to that larger number. The regurgitation is, “they said we may get up to a foot of snow.” That second person then hangs on to “foot of snow.” So when 5, 6 or even 7 inches of snow fall, which would be well within the forecast range, what is the response? “They were way off!”
Outside of record keeping, to the average user, what does it matter if 6 or 12 inches of snow fall? The post snowstorm backlash can be anticipated verbatim on many social media platforms. What is everyone so mad at?
- Maybe, deep down, they wanted a big snowstorm—doesn’t the kid in us all? Perhaps there is a vested interest in seeing the higher amount come to fruition for a day off of work or school. When that doesn’t happen, it is easy to deflect displeasure with the forecaster.
- Maybe a fresh forecast hadn’t been retrieved for a day or two. Obviously, things change. Saying a snow total forecast has to be right 24 hours or more earlier would be like calling a baseball game in the 7th inning, but the onus continues to lie with forecasters. This uncertainty and likelihood of change needs to be emphasized from the first mention.
- Maybe Monday Morning quarterbacking weather forecasts is just as much of a thing as sports games.
- Maybe there is a perception that only weather people get to be wrong and have a job…
(Reader) “This freakin’ guy is going to talk about impacts again, isn’t he?” You’re damn right! Somebody I know in the Philadelphia area said to me, “we only got 5 inches, and they said 6 to 12 inches, so it looks like they missed again!”
I asked, “Did roads close?
“Were flights cancelled?”
“Did work call off?”
“Were schools closed?”
When the storm is examined that way, impacts were the same no matter the number value assigned to the snow. Perhaps if this was the primary message prior to the storm, there would be less of a bust perception from forecast users. Did the snow totals work out exactly as expected? No. Were certain areas faced with the issues that stem from a high impact storm? (In most cases) Yes.
I am NOT implying the impact messaging was completely absent. I can say with great confidence that television meteorologists across all four stations in Philadelphia and the National Weather Service made mention of the uncertainty, potential causes for shifting snow totals and what societal impacts may arise. I AM implying that such messages were mainly on “old guard” platforms that no longer reach the entire population.
I am NOT implying we totally do away with snow totals map–after all, describing “how much” is in the job description. But the current way in which people consume forecasts does not allow for the disclaimer that must accompany a snow totals forecast given the current science. I AM implying that snow totals maps become secondary in the arsenal of snow forecast products. Much like it has become taboo to share forecast models days in advance, I believe it has reached the same point with snow totals maps. They just don’t convey the right message.
Here are some considerations. I asked friends and family how they got their forecast before the event.
How people get snow forecasts:
-Facebook – algorithm creates all kinds of issues
-Twitter – not a mass following of weather accounts, one or two preferred
-Website – maybe via a video thumbnail, only shows snow totals
-App. – hourlies, dailies that don’t give full details
Ways we’d prefer they get snow forecasts:
-Watch a full TV News weathercast
-Read the NWS Forecast discussion, briefing packages
-Follow credible weather sources on social media
-Read local television meteorologist blogs
That’s the thing—we want them to come to us, and the honest truth is that they all aren’t doing that anymore. We need to make sure to take all of every message to every platform. I believe there is a valid argument to say that was not done in this case.
The more conservative reader might argue that I’m an over-controlling, over-evaluating wacko. “Just give me the information and let me make the decision myself.” But did you ever have this conversation as a teenager?
“You’re not driving today!”
“Mom the roads are fine.” This, after several inches of snow had fallen.
I would argue that we still have that stubbornness within as adults. We still think the roads are fine. We do want the information and we do want to make the decision ourselves. However, research tells us and examples have proven that humans are poor assessors of risk. Heck, sometimes meteorologists even stay outside and finish cutting the grass after hearing thunder!
After all of this, I spoke to my brother who was in a heated debate with associates over the abilities of weather forecasters. His description of how they receive a weather forecast is telling.
“It’s easy to blame weather forecasters for getting it wrong because the average consumer only looks at the slideshows posted on twitter or on a website—they seem to never hear the explanation about the intricacies of particular events. In our most recent storm, I listened to vivid explanation about storm tracks, and the potential 10-15 mile shift of the freezing line that could dramatically affect the snowfall totals the night before the storm hit on my local news channel. Now, I know that anyone watching the news heard the same thing I did, but chances are that those complaining either a) got their news off of social media b) heard someone say “they’re saying we might get a foot and a half of snow” and assumed that included the area they live in (when it may only pertain to part of the viewing area) or c) they watched the news to SEE a snowfall forecast map and paid selective attention (they didn’t understand what the meteorologist was saying SO they chose not to hear/retain it.”
Maybe we need to give something that people can retain. Our good friend Gary Szatkowski got into it with 6ABC Meteorologist Chris Sowers prior to the storm. Withholding opinion on that debate, one of Sowers’ viewers did offer at least one alternative to framing the forecast.
Impacts…bottles of wine…whatever…I’m just sick of seeing snow totals maps.
Surely, you have something to say about this. CLICK HERE to jump in on the topic.