Isn't winter fun in Georgia? Maybe some of you haven't really tracked winter storms here before, at least not in the detail that I track them, and you have now experienced the roller coaster ride that brought us our winter weather on Friday and Saturday.  As you have seen, it certainly comes with it's share of highs and lows, and it sometimes ends up well and sometimes not at all.  Those that saw snow are happy and those that didn't are very disappointed or even depressed, and I feel your pain.  So here's my look at what went wrong. These are simply my thoughts and opinions about what happened and what could have been done differently. This is NOT a dissection of the storm itself, but instead a look at what could have been done to better communicate timely changes to the forecast. I do know that after the fact armchair quarterbacking is much easier than coming up with the actual forecast, and I do not claim to be able to do better... the NWS guys are the experts and they had their reasons for what they forecasted and why.  I just thought that I would look at this from an outsiders point of view and try to make some sense of the communication process that led up to the storm. I'm hoping that the NWS will have more to add as they review what happened. 

First, and I want to emphasize this, I do not in anyway place any blame on the meteorologists at the Atlanta NWS office. These ladies and gentlemen are among the best in their field and we are very lucky to have them working in our NWS office. They are thorough and focused on their mission to provide the finest weather forecast anywhere. My hat goes off to them for the hard work that went into this storm event and the long shifts (almost left out the "f" :-) ) they had to endure. I've said this over and over but I'll say it again, forecasting snow in the southeast is one of the most difficult and demanding jobs there is, and I don't envy anyone having to make that forecast. 

Based on comments from people like you, it seems as though you can summarize the thoughts about how you feel down to a few words and phrases: "Bust", "Hype", "Cry wolf", and "won't believe next time". Some people are even angry. As a meteorologist, I'm sure those words have to sting a little, and it's not something that they ever want to hear, but I also know they will be going back and analyze and re-evalute their own performance and what they could have done better from their perspective.

So why do people feel the way they do, and what could have been done differently to help prevent all of the anguish to start with?

The atmosphere is very moist as indicated by the small amount of separation between the air temperature (red line) and the dew point (blue line). Even though the air temperature increases a few hundred feet above the ground (a temperature inversion) the air temperature, throughout the entire atmosphere, remains below freezing. So, when precipitation begins, it will be in the form of snow and will remain frozen as snowflakes reaching the ground.

As with the previous sounding, the atmosphere is very moist. So much so, the air temperature and dew point are the same from near 900 millibars (3,000 ft. / 1,000 m) to a little above 700 millibars (10,000 ft. / 3,000 m).

At the surface, an arctic cold front had moved south of the observation station with an air temperature well below freezing. The air temperature begins to decrease with height (which is normal) dropping from 23°F to 12°F (-5°C to -11°C).

However, the density of the arctic air is such that it lays close to the ground with its vertical extent fairly small, in this case only about 3,000 feet (1,000 meters) deep. Above 900 millibars (3,000 ft. / 1,000 m) the air becomes considerably warmer. This area is called an inversion, where temperature change with height is 'inverted' as it increases with height instead of typically decreasing with height. This inversion is often also referred as a 'warm nose'.

Eventually, the temperature of the atmosphere will return to the typical decrease with height (near 800 mb) and will continue to cool until it falls to below freezing again (about 720 mb).​

While there may be some precipitation forming as rain in the warm 'nose' region where the air temperature is above freezing, the vast majority of precipitation will form as snow in the colder below freezing air above the inversion.
As snow falls into the 'warm nose', it melts into a liquid drop/rain. Then the liquid drops fall back into the arctic air mass (near the ground) that is cold enough and deep enough for the liquid to freeze into ice pellets before reaching the ground.

The basic pattern for freezing rain is similar to ice pellets. The main difference is the cold, dense air near the surface is very shallow and/or the 'warm nose' is large or very warm or both. The melted snow does not have sufficient time to freeze before is reaches the ground.

Therefore, precipitation falls as rain but then it freezes upon contact with a surface such as a tree, power line, automobile or bridge. As a general rule, elevated surfaces will ice first because the ground cannot keep them warm, allowing them to cool to the air temperature quickly. This allows the rain to freeze to ice on contact with these surfaces.​

Elevated surfaces may be capable of accumulating ice as soon as the air temperature falls below 32°F (0°C). Road surfaces in contact with the ground will generally begin to ice when the air temperature falls to 28°F (-2°C).

I'd like to first look at the forecast timeline and see how the forecast evolved from start to finish.  This slideshow starts on the 4th and ends on the 6th. Even at 10:34 PM while the storm was underway, the official forecast graphic was still showing snow for most people, when it was already apparent that the snow wasn't coming for many.

The 1 PM sounding on Friday the 6th, showed a significant warm nose that appeared to start at roughly 3000 feet and extend all the way to almost 10,000 feet, and even below 3000 feet, many of us were still above freezing. This sounding was taken after the first round of precip came though.

In addition to the warm nose, the flow of air around the mountains created a thermal ridge which helped keep warm air in place longer than expected. 

At 1 PM, the Atlanta NWS office lowered snow totals and produced this graphic.

This next two graphics were included in the 4 pm weather briefing. This is just my opinion... in the snow forecast image, too much emphasis was placed on "Significant" snowfall", "higher amounts", and "2-6" with locally higher amounts" and not enough on the decreasing chance for snow and the increasing chance for freezing rain or even nothing at all. At this point, everything had changed and it was becoming very apparent that was a change for the worse, at least in terms of the original forecast.  Optimisim was running high at this point and these slides helped to pertetuate the optimisim.  Again, just my opinion, but some harse reality needed to be expressed to the general population to prepare them for what wasn't coming. 

Finally, around 10:34 pm, the new graphics were posted, but it still didn't match the reality of what was going to happen.  In my opinion, this is where the second mistake was made. By 10:34 pm, many people had already gone to bed, expecting to wake up to snow on the ground.  While this map depicted pretty much what was going to happen, it was posted entirely too late to be of any real benifit for most people.  I would have thought by this point that the snow totals for central Georgia would have been removed entirely but it stayed on the map.

There was discussion about the warm nose as early as 10 am on Friday, Knowing how winter weather in the south goes, I would have thought that the forecasted snow amounts would have decreased more rapidly than they did.  The lingering snow forecast led to snow total expectations that would never materialize for many people, and I think that's the rub.  

​When you have a snow forecast for 4 million+ people and it doesn't happen, you have a lot of people who start to dismiss the talk of any winter weather, which can end up being a dangerous way to think.  This storm could have easily gone the other way and we would have been looking at a very large and dangerous mess, fortunately, most of us were spared.  And that's the problem, forecasting snow in the southeast is more often a crap shoot at best, and a few degrees is the fine line we sometimes walk. 

So here are some great weather lessons that can be taken from this storm... 

• If the snow forecast is dependent on the timing of cold air, it is a low confidence forecast.  
• If the cold air is already in place, much higher confidence in the forecast.  
• Miller A storm tracks are not always great for us
• ALWAYS pay a lot of attention to the air temps at ALL levels of the atmosphere, those temps are just as important as the ground temps. 

On to the next storm.... :-)

Now (hourly) through Saturday (mid) morning