Zola in snow

My cousin’s dog (puppy) delighted in the snow; suburban Atlanta.

There’s lots of finger-pointing going on down south. Georgia officials on one side. National meteorologists on the other.

What is clear is that an unusual winter storm descended on most of Georgia and much of the southeast on Tuesday. Rather than play the blame game, what can we learn from this?

Take-aways

Most people don’t read weather advisory discussions, for a very good reason. It’s rare that a watch turns into a major storm. And they’re hard to read (ALL CAPS) and decipher (abbreviations left over from teletype days?). And vague, as pointed out below.

Moreover, the bracketed time periods … WINTER STORM WATCH IN EFFECT FROM TUESDAY MORNING THROUGH WEDNESDAY AFTERNOON… remind me of waiting for the cable installation guy. It’s no wonder that what people hear from the news is (probably) the more “precise” worries such as the one about Tuesday evening rush hour.

  1. It is simply not possible for a lay person to assess risk when a weather-related warning is defined as being the “potential” that an abnormal event (defined depending on the event) might happen. What does “potential” mean?
  2. It’s long past time for the NOAA/NWS to publish weather forecast “discussions” in a more readable form, particularly if the reports are going to be cited as reasons folks shouldn’t complain about said forecasts.
  3. If NOAA/NWS isn’t willing to engage in risk assessment, and nothing in the written record suggests that they are, then who will? It’s not the job of the TV meteorologists. And I don’t think it’s a post for a gubernatorial appointee. For sure, local public officials – elected and appointed – don’t have the domain knowledge needed to do this. Yes, there are “emergency managers” but They Manage — they don’t assess the risk of a specific emergency actually happening!
  4. As we experience more frequent and severe abnormal weather events, this question of who assesses risk and decides to shut down a city has to be addressed. Immediately top of mind: Katrina and the Challenger Explosion. Both fraught with poorly presented information and fears of being wrong. At the same time, we have to beware of “crying wolf” too often.

So what the hell happened in Georgia? And Alabama (et al)?

Winter weather in the south

I’m from Georgia. My dad lives in SW Georgia, and I get weather alerts on my phone for the farm. I have lots of family in Georgia, Alabama and Florida. And I followed the storm on Facebook, as family and friends shared pictures, video and commentary.

I know that the south is no stranger to snow and is very well acquainted with winter ice storms.

My senior year at UGA, I was driving to Atlanta from Athens to catch a plane to Chicago to staff the press booth for National 4-H Congress. It was the day after the Georgia/Georgia Tech football game; late November.

I was barely out of Athens when whoosh-swish-bang, I was no longer headed west on the highway. My little blue Toyota slid to the right, and then the rear quarter panel made a close encounter with a highway sign post. But the abrupt halt was better than cascading down a deep ditch. My first experience with black ice.

No memory of how I got to the airport.

But my flight was cancelled, and I do remember moving from gate area to gate area, trying to get a stand-by seat. And spending the night sleeping on the floor. I finally snagged a flight the next day; it was the wee hours of the morning, which had me walking into the convention Hilton about the time I was to start work. No orientation. And, more tragically, no free “night before” in ChiTown.

But better than spending the night in a car stranded on the I-285 perimeter. Maybe not as much fun as hanging with friends as a teen or tween and spending the night in a gym, especially if pizza was involved.

Looking at NOAA alerts

I knew a winter storm was en route to southwest Georgia, because my phone kept warning me. My focus was on the temps, which have already been brutally low. But I noted the freezing rain and ice.

See that “hazardous weather” note? I posted this to Facebook after midnight Tuesday morning.

Storm Warning

Monday night, there were storm alerts for southwest Georgia.

 

But what did the Atlanta-area alerts say?

First, let’s nail down the difference between a Winter Storm Watch and a Winter Storm Warning.

According to NOAA:

Winter Storm Watch
This product is issued by the National Weather Service when there is a potential for heavy snow or significant ice accumulations, usually at least 24 to 36 hours in advance. The criteria for this watch can vary from place to place.

Winter Storm Warning
This product is issued by the National Weather Service when a winter storm is producing or is forecast to produce heavy snow or significant ice accumulations. The criteria for this warning can vary from place to place.

LaGrange GA

In my mental model, a weather “watch” means something might happen but the odds of it happening where I am is less than 50-50. A weather “warning” dramatically shifts that probability, and not in my favor.

But that’s not quite in line with how the forecasters are using the terms.

And here’s the Georgia definitions (pdf):

Winter Storm Watch
Issued when the potential exists for 2 inches or more of snow in 12 hours, or 4 inches or more of snow in 24 hours. Also issued for potential of a quarter inch or more of freezing rain, or half an inch of sleet. In the North Georgia Mountains, the criteria are 3 inches in 12 hours or 4 inches in 24 hours.

Winter Storm Warning
Issued when a combination of snow, blowing snow, sleet, and/or freezing rain is likely to exceed warning criteria. Warning criteria are those detailed in the Winter Storm Watch.

What in the world does “potential” mean? Ditto “forecast”? A 20% chance? 50-50? 95%?

How can any normal human being make an intelligent risk assessment based on definitions that totally ignore probability?

We can’t.

What, then, does the forecast record reveal? (NOAA’s ALL CAPS has been converted for readability.)

Saturday afternoon

228 PM EST SAT JAN 25 2014

Main concern in the short term period is [chance] for light snow late Sun night and Monday morning over North GA.

Sunday morning

327 AM EST SUN JAN 26 2014

Cold airmass supports a mix of rain…sleet and snow… Most likely late Tuesday and Tuesday night with the precip favoring sleet and snow Tuesday night […] significant accumulations of winter type precip do not look to be in the cards.

In English: late Tuesday some light snow or sleet might happen in central Georgia. And it’s the weekend. People aren’t paying attention.

Sunday afternoon

307 PM EST SUN JAN 26 2014

Will cut straight to the chase for the long term as all eyes will focus in on Tuesday. Cold air will definately be in place on Tuesday […]

Regardless of amounts…The models are unanimous on wintry event shaping up across central GA. Have issued WSW (Winter Storm Watch) for areas generally south of I20 not including the core Metro ATL counties but will have to watch later runs for possible advisory or WSW upgrade depending on trends.

In English: it’s going to get cold (for Georgia) and a “wintery event” might happen south of Atlanta. Watch issued.

Monday morning

430 AM EST MON JAN 27 2014

Big picture…Confidence is increasing that at least a portion of the forecast area will see significant impacts from this winter storm […] Now have likely to categorical snow across entire Atlanta Metro and extend from Haralson County to Madison […] Most significant accumulations would be just south of the Metro with 2 to 4 inches from Greenville to Milledgeville.

In English: forecasters are more confident that a snow storm is coming and predict it will cover an area 100+ miles wide. But the snow forecast is considerably south of metro Atlanta.

Monday afternoon

322 PM EST MON JAN 27 2014

A wintry mix is still on tap for most of the CWFA for Tuesday afternoon through early Wednesday morning. Confidence is still low that the northern tip of the WSW headline will reach warning criteria.

In English: forecasters still think the brunt of the storm will hit south of the Metro Atlanta area which is still in a watch, not warning.

Tuesday morning

411 AM EST TUE JAN 28 2014

… expanded this [warning] to another tier of counties including ATL Metro area […] Final note…We remain concerned about impact with onset of precip around rush hour and school release.

In English: forecasters now think that the storm will hit Metro Atlanta but they think it will be late in the day and could impact evening rush hour.

Metro Atlanta residents went to bed thinking they were in a relatively low-risk position (watch) and woke up to a warning.

How did morning radio and TV weather folk report? I don’t know, but we do know that some web properties were explicit.

Mike Smith, senior vice president for AccuWeather Enterprise Solutions, says AccuWeather sounded the alarm at 3 p.m. Monday, including the following statement in a notice to clients:

The ice and snow amounts indicated here [on a map, with Atlanta along the northern edge of the hazardous zone], spread over a large area generally unaccustomed to seeing winter storms of this magnitude, will bring transport and logistical operations to a virtual standstill. Delays and cancellations will cripple air, ground, and rail travel for days.

Late Tuesday morning

1118 AM EST TUE JAN 28 2014

.UPDATE…
Added NW GA to the advisory. Getting reports of accumulating snow on the ground around an inch.

In English: only now concerned (again) about Northwest Georgia.

Georgia Storm

What a mess.

Preventable? No.

Could it have been mitigated? Yes. Not only could it have been, it’s imperative that we put steps in place to ensure that the next major storm — winter or spring — be understood at all levels of government and private enterprise.

KATHY GILL, Technology Policy Analyst
Click here for reuse options!
Copyright 2014 The Moderate Voice
  • epiphyte

    I think it’s pretty impressive that they were able to predict it at all. The jetstream is so far out of place that the models are in unfamiliar territory – and because the anomaly is over the central US, the air over atlanta has been following an abnormal track for a couple of thousand miles. Nailing the exact latitude where the precip changes from rain to snow is hard even under the most predictable conditions – & this doesn’t qualify!

  • Thanks, @epiphyte — the discussions go into a bit of detail about how the models differed.

  • Willwright

    I live in the Atlanta area and fortunately spent the day warm and cozy at home. The biggest risk in a storm like this here are the the drivers who don’t have a clue on how to drive on snow and ice. They were unprepared and the politicians of course saying it was an unpredictable event, meteorologists are hoping mad at the suggestion they got it wrong. Anyway the kids got a couple of days to play in the snow which doesn’t happen too often here.

  • Excellent analysis by PeachPundit — and good point RE the places where traffic made national TV are not in Atlanta (proper).

    I didn’t know about the briefings. That makes me more convinced we need someone doing risk assessment who is (a) not a politician, (b) not a political appointee and (c) has the domain knowledge (read that as science if you like) to make these kinds of assessments.