Nothing happened (unless you count that tornado)

By: Josh Eachus

“You cannot predict squat! And you have successfully created the “Culture of Fear” in all your forecasts.”

Accusations of crying wolf, fear mongering and hyping were abound in South Louisiana last week. A difficult forecast met some resistance from the observed weather, but by a meteorologist’s scorecard it could be registered as a hit.

I have been forecasting in Baton Rouge for almost three years and until now, not seen such ire from our forecast users. Our weather department received a steady stream of derogatory emails, negative social media comments and demands to get it right, or get out. We’ll look at a few specific cases and then move on to some thoughts.


 

Forecasts are presented in unedited form from our 6am weather blog on wbrz.com. My on-air broadcasts from 5am-1pm conveyed the same message. User complaints are copied verbatim from my email inbox.

Wednesday Forecast: Today, expect highs in the upper 70s, with mostly cloudy skies, scattered showers and thunderstorms. There will be breaks in the precipitation. In fact, some locations could feasibly go without rain today; we’re forecasting 60% coverage. While a widespread severe weather event is not anticipated, a few storms could be strong, producing gusty wind or a brief tornado. As always, review your safety plan and stay in touch with the forecast in case any warnings are issued.

Wednesday Verification: One large thunderstorm developed in the northwestern third of our forecast area. This storm produced golf ball sized hail, primarily over rural areas. In reality, approximately 40% of the area saw measurable rain—still meeting the “scattered” qualifier.

Wednesday Complaint: What’s been up with y’all and the weather? Y’all have been so off on this weather it’s been unreal. We go off of y’all on whether we work or not bc of y’all saying it’s going to rain here or not. And the last two times I missed work bc “y’all say it’s going to rain” and it never did. Please figure out the problem and fix it.

I can understand this person’s frustration. Certainly, cancelling work because of our forecast costs money. The issue though is that this person seems to be interpreting our forecast for ONE specific location. I would like to think that at some point responsibility lies with the forecast user to understand the forecast, but we as a community can continue to improve by communicating forecast limitations—specifically those for point forecasts outside of a few hours.

 

160405_GTF

Image 1: WBRZ designated market area with accumulated rain, storm reports

Thursday Forecast: On Thursday, showers and thunderstorms will once again be isolated to scattered. Areas along and north of I-12 remain at risk for a few severe storms as the ingredients in place mean that any activity that does spark could quickly strengthen. Downpours, gusty wind and frequent lightning are the main threats. An isolated tornado can’t be ruled out. As always, stay safe by reviewing severe weather plans and stay in touch with the forecast.

Thursday Verification: An EF-1 tornado hit a rural area just north of I-12 around daybreak. Rain fell across the northern half of the forecast area between 4am – 8am with mostly cloudy skies for the remainder of the day.

 

Thursday Complaint: Another wonderful prediction?? Last Thursday, yesterday, today are just the most recent examples. You cannot predict squat! And you have successfully created the “Culture of Fear” in all your forecasts. You talk about Action Days, Storm Tem, Storm Tracker, and now every time there is a chance of rain, it is described in terms of “Risk of Severe Weather.” Really? I teach an accountability class in which part of it is how to deliver and receive appreciative and constructive feedback. Notice I did not say constructive criticism!! Why? Because when I say that, the only word you hear me say is criticism. So when you say a slight risk or a marginal risk of severe weather, what do we hear?  SEVERE WEATHER. I have very little faith any more in your ability to predict, and you have more tools. Quit trying to “sell” the weather, and just give me the data. I do not understand how YOU keep me safe. Just give me the temp forecast and the probability of rain (if you can) and quit talking to me as if I am a moron. Don’t tell me to wear a jacket, or a sweater, or carry an umbrella. I am smart enough to figure that out if you give me the data. If I am too stupid to adjust mu situation to the forecast, the accountability lies with me. Keeping me safe is on me once you give me the data. While I am giving you constructive feedback, the recommendation form whatever consultants all the stations apparently used to stand up and talk to me is in one word, AWKWARD. I do not feel like you are my friend. I could care less. I just want the facts. Report the news and let ME draw my own conclusions. I do not care what you think.

This note is a little more disturbing. This person referenced branding monikers sometimes used by television stations—something our station does not do, in an effort to avoid confusion. Next, we notice a misunderstanding of the word “risk” followed by a clear and common indictment of the severe weather outlook categories—they’re just confusing words to most, and mean nothing outside of severe weather. Third, there is a call to discontinue actionable items in a forecast. This of course, completely disregards all scholarly research on the need for threat and efficacy in warning messages. If this person is indeed an academic (as they claim), one would think they’d have a better grasp of creating public awareness for the sciences, not to mention some professional courtesy.  


 

On Friday the entire forecast area received several periods of rain with flash flooding occurring in adjacent southern parishes. Still, emails and comments of a less detailed but critical nature continued. I kept asking myself, “What is it about this particular event that has the masses so riled up?” I have a few hypotheses.

  1. Lead time: This event was broadcast well before onset with a day three outlook from the Storm Prediction Center. Just as our generous emailer noted—while it may have been a marginal/slight risk, repetition of the words “severe weather” may have mainstreamed forecast users to expect worse. Furthermore, the area is not far removed from a significant tornado outbreak and is hyper-sensitive to the term “severe weather.” Right now, attentiveness and preparedness to the weather may exceed usual standards, 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 issue.
  1. Transparency: Our field, and our weather department in particular, has made an effort to be very forthcoming about the challenging aspects of a forecast, uncertainty and even going as far as showing forecast verification—hits and misses. Less than one week before this scenario, we experienced a “bust.” Though fairly inconsequential–a brief band of showers was expected and nothing happened–this particular situation garnered a little extra attention. A gravity wave squashed showers as they approached our forecast area making for an interesting radar loop and a good opportunity to explain a unique scientific phenomenon. It is possible though that this coverage primed forecast users to associate no rain with a blown forecast.
  1. PoPs: Our field continues to research the extent to which forecast users understand the probability of precipitation—in short, not very well. In this instance though, we got an added wrinkle, the location of precipitation and the population. On Wednesday, the forecast called for a 60% chance of rain. Examining the observed rain totals, geographically speaking, about 40% of the area received measurable rainfall—not a slam-dunk forecast, but not a major bust either. But the population factor complicates the forecast’s verification. The areas that received rain have an approximate population of 47,000 in a forecast area of over 856,000. That means only about 5% of our viewers saw rain, making “bust” a seemingly much more appropriate adjective for the forecast. Expanding upon this, with a population over 440,000, East Baton Rouge Parish, including the city of Baton Rouge, contains half of our viewership that may interpret whether or not a forecast was validated. This large number of residents is in only about 10% of the geographical area to which WBRZ weather forecasts are directed (see Image 1). It’s a forecasting headache dealt with by meteorologists everywhere.

In retrospect, this event just about spared a majority of the population in our forecast area. After words of encouragement from colleagues and friends, I thoroughly understand the adage made popular by Ron White, but that doesn’t mean those folks should be ignored. We need to communicate to those people that every head and every home in a forecast area is important, not just those in the population center. We need to continue our path to a clearer, more user friendly and more SOCIAL forecast.

4 thoughts on “Nothing happened (unless you count that tornado)

  1. even tho they know it’s unlikely, some folks let themselves feel that an element of certainty is somehow possible. a surgeon can’t promise any outcome but hopefully she does the best she can with what she’s got, and the patient puts aside his fear insofar as he’s able and invests some trust in his surgeon. you make the best decision you can based on the best information you can get. that’s all anybody can do. there are no guarantees.

    Like

  2. The “Thursday” complainer doesn’t know what he is talking about. He says he just wants the “data.” Well, isn’t that what the risk categories essentially are? They represent probabilities and their inherent uncertainty. But he seems incapable of interpreting or using that information appropriately.

    Like

    • I agree Jim… there seems to be some level of malice in there as well with the “I don’t consider you my friend comment.” Would they like an angry tv ‘caster? Sheesh!

      Like

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s