To Hype or Not To Hype: Weather and The Media

As you know, one of the things that I like to talk about frequently on my blog is "Media Accuracy" in regards to the weather (I have a whole Category for it at right).

A couple weeks ago, I vehemently defended myself against attacks that accused me of hyping the weather (I argued that doing the opposite equates to censoring potential storms -- and providing a boring repetitive forecast). Today I want to take a look at a couple related videos from our meteorologists regarding weather and how the media handles it.

First, our C.O.O. Evan Myers talks with AccuWeather.com Broadcaster Katie Fehlinger about the hype of Hurricane Katrina, and what the role is for the media and meteorologists in warning people in advance:

Mainly at issue here was Hurricane Katrina. I wish that Evan had talked more about the recent snow storm that missed the big cities in the Northeast - I think that's what sparked this interview.

Now I'll admit, the clip above was kind of a softball interview; if you want to see a more passionate and animated version of AccuWeather employees talking about this topic... did you see last Thursday's Point-CounterPoint video? It's under "Weather Experts" in our our Video On Demand Player (PREMIUM | PRO). This video pits two meteorologists who disagree (or sometimes agree) about upcoming weather situations, but this time Ken Reeves was on fire about the NWS (the U.S. government agency responsible for issuing official weather advisories) over-issuing warnings during the recent wind event.

I talked last week about how it's silly to issue Thunderstorm warnings without lightning, and pointed out how the lightning didn't match the warnings in this blog and this earlier one, and I maintain that posture. That aside, I also think that he's right that the highest winds were not with the squall line in many places. Here in State College, Pennsylvania, home of AccuWeather HQ [Google Map], our highest wind gust as the squall line passed was perhaps 20 mph but we gusted to 56 mph the following morning.

Ken's point, and he is correct, is that the NWS already has High Wind Warnings, which fit this situation better than Severe Thunderstorm or Tornado Warnings, so why issue all three (except to get people's attention, as they were accused of during a hurricane a few years ago). I also agree with Ken that Tornado Warnings (or Watches) have no place if the storms are not going to be convective enough to produce significant lightning or tornadoes.

P.S.: Before you rip me in the Comments, we're not questioning whether the NWS should be issuing warnings, we're just looking for them to be more clear and consistent with the type of warnings issued, so the public knows what to expect.

Comments (6): Dan:

This argument is completely ridiculous. The goal of the National Weather Service is to protect lives and property. When forecasters are receiving reports of damage to property, the wind speeds are obviously worthy of warning criteria. Just because the speed is 'unknown' doesn't mean that the wind speed doesn't meet severe criteria. Given the nature of the event (QLCS) it is foolish to think a high wind warning should accompany these events. This event was a severe wind event created by convective processes, and should be warranted by a severe thunderstorm warning. I agree with the point about the tornado warnings as the environment was unfavorable, but the severe thunderstorm argument is ridiculous. You are basically saying that if a derecho is screaming through the upper-midwest and not accompanied by lightning, that a forecaster should issue a high wind warning.

FROM JESSE: To the public, high wind is high wind, I would think they would understand "high wind" quicker than "severe thunderstorm" in this situation - remember they are not issuing warnings for meteorologists or weather weenies, they are issuing them for the public. Ken's point was that damage can occur at almost any wind speed, trees can easily be downed by winds less than 58 mph. If the NWS felt that any damage was qualifying, they shouldn't have specified the 58 mph cutoff.

Posted by Dan | February 18, 2009 11:56 AM Steve:

I've come to believe that the only way to present the weather is through a confidence-level scheme. Jesse will probably disagree with me, but as I said to Henry Margusity, Accuweather lost a lot of credibility in the forecasting disaster earlier this month.

The main issue with the February 4 storm was not ignoring some of the possibilities, but instead failing to communicate the likelihood of all the scenarios. A forecast that there's a "20% chance of whiteout conditions" and an "80% chance of flurries" allows me to make an informed decision about whether to travel, rather than suddenly getting stranded on the Pennsylvania Turnpike for hours in blizzard conditions, as I did.

In the case above, Accuweather posted a map that said there would be "1-3 inches" of snow, when it would really have been more useful to say that there was a 90% chance of 1 inch and a 10% chance of a foot. That way, people would not have been entirely surprised when a foot actually did fall. The Accuweather forecast made it sound as if a foot was completely out of the question.

The NWS warnings similarly fail to communicate the severity or lack of severity of a situation. Meteorologists there use some criteria to determine whether a high wind warning should be issued, and then issue it without stating the confidence level they have in the forecast. They obviously know that confidence level, so why not publish it?

Posted by Steve | February 17, 2009 9:51 AM Allen:

OMG he is soooooo right to be upset. Especially because I know through years of following everything from NWS covered outbreaks to commentary by officials to forecast discussions. The NWS has always been intent on getting the warnings right for the precise reasons he's upset about. I have taken notice of several times here in North Jersey over the past few years of non convective Severe warnings -- at least void of any lightning -- that were associated with wind or high wind type advisories and it didn't even end up raining a couple of those times. YOU CANNOT DESENSITIZE THE PUBLIC TO THESE MOST IMPORTANT OF WARNINGS PERIOD. THE CRITERIA MUST BE STRICTLY FOLLOWED. Complacency of these warnings will result in a backwards movement in favor of human suffering.

Posted by Allen | February 17, 2009 9:16 AM Kevin:

I think people get upset when they get hit with something more severe than called for, rather than less severe than called for. I'll bet you get a lot of calls when a chance for hail is put out by the people in the 10 mile stripe who get pounded, then a few from the others who may have missed that particular storm all together. Which is not particularly fair until technology catches up to that level of accuracy for 5 miles on either side of a pop-up summer thunderstorm. The only sticking point I have is broadcasting a snowstorm model 6 and 7 days in advance - it should be given with the caveat "we have about 2% confidence in this model" for those who aren't aware of the accuracy curve. Its when a 6-12 inch snowstorm is forecast for the next day when I go to bed and I wake up to rain - that's a problem.

Posted by Kevin | February 17, 2009 7:59 AM Calvin Penner:

The problem seems to be that all the illustrations and graphics on accuweather always create hype. That is the purpose of them, for people who are meteorologically impaired to easily understand the nations weather with a quick glance at the computer screen.

Posted by Calvin Penner | February 16, 2009 7:17 PM Patricia W.:

I'd rather be overwarned than underwarned.

FROM JESSE: True. Here's the problem though. The NWS seems to think that people have become complacent to High Wind & Hurricane Warnings, so they've started issuing Tornado Warnings. What will happen when those become so common-place that we're complacent to Tornado Warnings, will they have to invent something else?

Posted by Patricia W. | February 16, 2009 5:31 PM