Winter storms are probably one of the most difficult things to forecast and predict in the field of meteorology. It takes just the right ingredients to produce snow. If one thing is just slightly off, like the temperature profile of the atmosphere, it can change a forecast from seeing snow to just a cold rain. We know some of you were not happy about the results of the January 6-7, 2017 winter storm. What happened and why were so many upset that a storm didn’t produce the snowfall accumulations they were hoping to see? We analyze it all in this post. Here’s a look at the hype, the timeline of events leading up to the storm, and understand the science on what went right and what mostly went wrong.
January 1-3, 2017
Winter is the season where weather weenies love to look at computer models hoping to find a major winter storm that will impact millions. In recent years, weather models that were previously only available to professional meteorologists have become easily accessible to the public. This has prompted the growth of weather blogs, forums, and social media sites (hey, you’re reading one!). Weather weenies love to talk about weather and share their passion with others on these websites. They’re a great place for people to enjoy and love the weather.
I’m a big fan of weather blogs and forums. When I was a teenager and in my early college career, I would always go to weather forums for fun. I also fully support sharing your love for the weather on social media. However, it’s important to remember that weather forums often consist of various members with differing backgrounds. Some have meteorology degrees. Many others do not, with varying levels of forecasting experience. It’s important to keep this in mind the next time you see a forecast in a forum, in a blog, or on social media. Is it from a legitimate source?
Weather models are usually fairly accurate within five days. After about a week into the future, models become more and more inaccurate. The American model (GFS) is notorious for showing major storms 8+ days out. As a meteorologist, I’ve grown to ignore these bogus storms as they hardly ever actually happen. Model forecasts that far out just aren’t that good. If I shared these images, you’d stop following us because that big storm a week and a half out would almost certainly not happen. Unfortunately, these far-out model images often do get shared in forums and on social media. If you see a far-out European model graphic floating out in social media with 6 inches of snow for north Georgia, recognize that this is not an accurate forecast.
With that said, 2017 began with buzz about a winter storm in the Southeast 5-7 days away. The image below, in particular, went viral at the beginning of January, prompting lots of questions from people wanting to know if it was real.
It does no service to hype up a storm that won’t even begin forming for another 5-7 days. If they had simply written that the Southeast would have the potential to see a winter storm, then it would have been okay. Instead, I saw several sources calling for a blizzard or a major winter storm in the Atlanta area. It’s even more frustrating that none of the models, even those multiple days out, ever showed a blizzard for the Southeast. All of a sudden, people shared these posts and they went viral. I wonder if this was the beginning of why people were upset at the final outcome of this storm? Were expectations for snow already set high from an illegitimate source?
Forecasting is difficult. Winter weather forecasting in Georgia is extremely difficult. A few degrees or a few hours can make all the difference in the outcome. Winter storms forecasts can change drastically even several days out, and the 48 hours before an event is critical. Because of these difficulties, your weather app’s forecast for winter weather will likely not be accurate regarding accumulation totals and precipitation type. For the best winter weather forecast, you must depend on an actual meteorologist.
Two Days Before the Event: January 4, 2017
By January 3-4, 2017, confidence was increasing that snow could be in the forecast for January 6-7, 2017. The European and GFS models were hinting at the potential for snow in North Georgia. The GFS was more aggressive with the snow, but the European was hinting it would only be for North Georgia and locations along and north of I-85.
The National Weather Service at Peachtree City tweeted:
It is time to stay weather aware! The potential for accumulating wintry weather is increasing for the weekend. Here is the latest… pic.twitter.com/yjMOxuefXu
— NWS Atlanta (@NWSAtlanta) January 4, 2017
We had model support for a storm, but the models continued to show different snow totals. At the time, the models were showing a pretty solid column of cold air which would result in snow as the dominant precipitation type. Late that night and the following morning the National Weather Service at Peachtree City issued a Winter Storm Watch for all of North Georgia.
One Day Before the Event: January 5, 2017
Within the 24 hours before the winter storm. Our short-term models were now in play (NAM/RPM/WRF), and we were getting a slightly better idea of what could happen. The last thing a meteorologist wants to do is make a bad forecast. It is why we find it very important to talk about scenarios that could happen and state uncertainty. There is never a sure bet when it comes to Mother Nature. She loves to throw surprises at us.
With that in mind, I wrote this excerpt in the evening post on Thursday, January 5, 2017:
With the storm only 24 hours away, it was time for the National Weather Service to push the warnings out so you and the GA DOT could prepare for the potential of treacherous road conditions due to ice and snow.
— NWS Atlanta (@NWSAtlanta) January 5, 2017
Naturally, one will assume snow is going to happen when you see the words “It’s Going to Snow!”. You’ll also feel pretty confident it will snow since you see “high confidence for accumulating snowfall across this area”. I was not particularly a fan of the graphic saying “it’s going to snow!” because a few models still didn’t completely lock onto this solution. However, it was the right action by our NWS to issue the warning so people could prepare for the worse. Better safe than sorry, right?
The models were aligning and all indicated moisture pushing into a layer of cold air which would allow for accumulating snow in Atlanta and Athens. We decided to take a blend of the models and give Athens 1-3 inches of snow. We stated Thursday that the best chance to see snow would be early Saturday morning. We knew that rain would be the dominant precipitation type for most of Friday. At this point in the game, none of the models showed the potential for freezing rain. It simply did not look like that would be an issue.
Here’s our first snowfall map that was posted on the 5th of January:
January 6, 2017
We knew the storm would evolve and take place on Friday afternoon, January 6, 2017. We were already seeing reports of snow and sleet developing in Mississippi and Alabama. We were also watching the position of the area of low pressure forming in the Gulf and the push of cold, Arctic air coming in from the northwest. We knew the timing and phasing of these two systems would be vital on determining what precipitation we would see.
We had models and radar, but we needed more real-time information to get a better handle on the situation.
One way to analyze the atmosphere is to release weather balloons. NWS Peachtree City launched two extra radiosondes (the weather instrument attached to weather balloons) into the atmosphere Friday afternoon to get a better idea on what was developing. We wanted to know how cold the air was and if there was a nose of warm air in the lower atmosphere.
Here’s what they found:
The image above is called a Skew-T diagram. It’s a way to analyze the atmosphere data that comes from the weather balloon as you go from the surface (bottom of the image) to the top of the atmosphere. The balloon found that there was a dominate warm nose about 1-2 miles above the surface. Meanwhile, the surface was cooling to below freezing. This sounding is a strong idication for freezing rain as the main precipitation type. The cold air was not deep enough to produce all snow, especially in and around Atlanta, and a surge of cold air that would be needed to push out the warm nose and produce snow didn’t appear to be happening. We knew then that the original forecast for snow totals was going to be a bust.
By that afternoon, NWS Peachtree City began to increase their ice accumulations for the Atlanta and points to the south. They were also lowering snowfall totals based on trends.
— NWS Atlanta (@NWSAtlanta) January 7, 2017
We were watching the trends from our short-term models and were becoming unimpressed with the snow chances for Athens. All along we knew the greatest chance for snow would be north of I-85. If the air was colder and further southeast, we could have easily seen 2-4 inches of snow in Athens. We were also watching cold air chasing all of the moisture. By the time the cold air arrived, the moisture was almost out of here. Had the moisture started 3-5 hours later, it may have changed the forecast.
— NWS Atlanta (@NWSAtlanta) January 7, 2017
The image below is showing the area of low pressure traveling right along the Gulf Coast into Southeast Georgia and up the East Coast. The low was too far north and likely brought in that warm nose which created a freezing rain event across Atlanta. If the low was further south, it would have given us a better opportunity to see snow.
Chris Davis wrote a great article several years ago that explains our best setups to see snowfall in Athens. Chris wrote:
No two winter storms are the same. The amount of cold air available, the amount of moisture, and the track of the low pressure system will determine what kind of winter storm impacts the Athens area. A minute change in either of these characteristics can lead to a drastically different final result.
January 7, 2017
The combination of warm air aloft and faster transport of moisture resulted in very little snow in Athens. Once the cold air was in place Saturday morning, we were only able to squeeze a dusting at best as dry air was pushing in and moisture was leaving.
Philippe Papin believes that a thermal ridge formed thanks to the Appalachian Mountains could have resulted in our warmer temperatures:
— Philippe Papin (@pppapin) January 6, 2017
As cold, dry air was pushing into Georgia, it was meeting up with the remaining moisture in place. The atmosphere acted like a sponge and squeezed out all of the available moisture by giving us snow showers Saturday morning. The HRRR model, which updates hourly, did a pretty good job of showing a few snow bands forming around 7 a.m. EST Saturday morning producing a few bursts of snow in the Athens area. It was enough to cover the ground with a light dusting of snow, but that was about it.
Here’s the final snow/ice results from the NWS Peachtree City:
You can compare that with our final snow map created on the evening of January 6, 2017:
We were off on the snow totals for Atlanta and points to the south, but had the right idea on where the freezing rain would be located. We also did a good job adjusting the forecast to show very little snow accumulation in Athens with higher totals north and west of I-85. We were also accurate on the higher snow totals as you go along and north of I-85 as you go into South Carolina. Greenville, SC saw 3-6 inches of snow from this storm.
Here’s the entire event showing the freezing rain/sleet (pink) persisting across the Metro Atlanta area and snow the dominant precipitation type north of I-85. Video via 13WMAZ Weather:
Dr. Janet Frick, Associate Professor of Psychology at the University of Georgia, provided us her thoughts on why people were so upset with the outcome of this storm.
When plans are changed — schools let out early, events are canceled, “winter storm warnings” are issued, I think subjectively people feel like it’s more of a “sure thing.” It’s complicated to say / hear / understand “our models currently, at this moment, show a reasonable likelihood of this possible high-impact event, so we’re going to take some proactive steps because such events are rare in the south and it’s probably better to be safe than sorry, and if we wait until it’s certain to make any decisions, then we have a chance of something like snowjam.”
I’m a little disappointed in rude comments following the event. We had comments that we should quit our jobs. Another comment said we were doing this for ratings. I’d like to point out that we provide this blog without pay. We maintain it on our own, on our own time, and with money out of our own pockets and donations from readers. Finally, we aren’t on TV. We don’t care about “ratings.” What we care about is providing you readers with the best, most accurate forecast for the Athens area. If you’re upset about the lack of snow, imagine how it feels to meteorologists. The last thing we want is a busted forecast, after spending days watching the models, analyzing the data, and making forecasts. Most models update and run four times a day, so you can bet we are always looking at them when they are posted. We spend hours writing posts and responding to social media. This is a difficult and humbling job, but I’d like to put out there that we do a great job forecasting for the Athens area most of the time.
People will never remember the 90+% of the time we make a great forecast. They just remember the 5-10% of times we get it wrong. Please remember that we are essentially predicting the future, albeit with physics and thermodynamics. While meteorology is a science, weather is chaotic, and forecasting will never be perfect. It’s actually amazing that we are able to do what we do, and we thank you for following us as we do it.
Finally, you were prepared. Prepare for the worse, and hope for the best. Meteorologists saved lives during this event. If warnings had never been issued, people would have been out on the roads while ice was forming. It could have resulted in accidents, injuries, and even fatalities. While we’re all sad that the snow didn’t pan out, we’re thankful that you didn’t go unwarned. Had the weather behaved a little differently, the Athens area could have easily seen several inches of snow.