Garrett’s Blog: Winter Weather Wisdom 101
After forecasting winter weather in Arkansas and Oklahoma for the last 12 years I’ve noticed a lot of the same patterns in weather and human behaviors.
To be a great meteorologist, you have to make mistakes but you must be able to learn from them. You also learn how to communicate perception of risk in new ways. Your perception of risk growing up in Arkansas is different for those who grew up in Minnesota.
In our area, snow is a unique phenomena because it doesn’t happen often. It usually only occurs 2-3 months of the year and normally just a few inches within a handful of weather events.
Below is a list of 10 things that are difficult to communicate; yet directly affect your life when winter weather is in the forecast.
1) We don’t know the exact amounts of of snow you’ll see in your specific town.
The science of tornado forecasting has saved countless lives. With a high degree of certainty, I can tell you if a storm has violent rotation on your specific street.
With winter weather, I can give you a broad area with a range of totals. For example a 50 mile wide, 100 mile long area with a range of 1-3″ of snow. For some reason if I forecast 1-3″ in one area and 2-6″ in another, usually my entire audience perceives the highest snow number (6 inches) as what they’ll see in their backyard. It’s always the highest total. For this reason alone, I’m conservative with snow forecasts.
I’ve noticed the same thing with storm damage. If a resident has damage to their home, it *must* have been a tornado. Social science is a newly evolving area of meteorology and further research will lead us to better understand how society perceives risk; and as a result we’ll change the way we communicate.
2) You don’t need large amounts of bread and milk.
I’ve lived here nearly all of my life and survived ice storms in 2000 & 2009. I’ve never been stranded for more than a day. The bread and milk fascination is thought to have originated in the northeast in 1978 when a large blizzard blocked traffic for days and when the delivery trucks arrived at the store people rushed to buy the bread and the milk. Interestingly, hurricane research has shown that Pop-Tarts are more common when “sheltering in place” during severe weather.
You do not need to rush out to the store and empty the aisle of milk & bread. They’re both perishable. But, you DO need a disaster kit with enough food for 3 days along with a stash of bottled water. The Red Cross has an excellent kit available online. Or, you can build your own. A disaster kit is ease of mind and will come in handy if our area is ever hit by a tornado the size of Tuscaloosa or Joplin; or a major flooding event which requires massive evacuations. You should also have a plan for a structure fire. Your house is more likely to burn down than anything else.
So skip the bread and milk aisle and go buy a fresh set of batteries and a new flashlight or a battery powered radio/NOAA Weather Radio.
There are some things you should “stock up” on such as prescriptions and baby formula; enough of both for 3-days is a good rule of thumb.
3) “Can I drive to ______?”
Other than a rare severe blizzard, yes, you can drive wherever you want.
One of my biggest pet peeves is “If you don’t have to get out: stay home”. Well, the only time I leave my house is “if I have to”. I typically do not drive around in circles during the day for no reason. (With the exception of a crying child or a wife who is insistent on looking at real estate that we have no intention of buying.)
Meteorologist Joe Pennington drove to Springdale with 2 feet of snow and drifts to 5 feet. I’ve driven to Tulsa in a major ice storm before. Is it for everyone? No. If you don’t have the driving skills or you’re a teenager, it might be wise to delay the trip or plan for extra time. Most accidents can be easily avoided by simply slowing down. Never go the speed limit on icy roads. As a rule, I never drive faster than 30mph on ice covered roads, sometimes much slower. I also know roads and hills I need to avoid in winter weather travel.
Like you, I’m more concerned about another vehicle hitting me than I am driving myself into a ditch. One more important note: 4 wheel drive doesn’t matter. It will get you out of the ditch, but won’t prevent you from ending up in the ditch. You still lose traction.
4) Road Preps
The often overlooked angle is temperatures. Yes, we know bridges and overpasses freeze first because of airflow on top and on bottom, but did you know most roads only freeze up when the temperature is below 30 or the snowfall rates are exceptionally high?
Last week, Fort Smith received 4 inches of snow at 36 degrees. The worst thing you can do is rush out in the middle of the snow and try to buy bread and milk and speed home. Pay close attention to the temperature when it’s snowing or snow is in the forecast. Expect major road problems when the temperature is between 25-28 degrees, ice patches on roads from 29-30, and slush from 30-32.
UVA and UVB ultraviolet radiation from the sun helps to melt snow even when the clouds are thick. Snow during the day will lead to fewer problems than a snow in the middle of the night.
In most cases, a trace of snow at 25 degrees is worse than 6 inches of snow at 36 degrees.
Also, in freezing rain with near freezing temperatures the roads are usually fine. The road problems are severe and extreme in heavy snow and below freezing temperatures. Ice will accumulate on trees and powerlines but doesn’t accumulate on road surfaces as easily. The key is always the temperature and amount of sunshine.
I always keep a blanket, snacks, and bottled water in the trunk of my vehicle; you should too. It will come in handy if there are accidents and you’re stuck for hours in traffic in cold weather or traffic. Several years ago, I540 was blocked for hours due to accidents on high elevation bridges and overpasses stranding hundreds for 4 hours in 25 degree weather.
In our area, for an overnight snow, I’ve noticed most main roads are fine around 10 or 11am the next day. Don’t cancel your plans too far in advance, have a back up plan and be ready to adjust because on how conditions unfold.
5) Computer Models Forecasts
Computer model snow forecasts are a piece of the puzzle. They are not the picture.
Weather software companies sell TV stations computer models which show exact amounts of forecast snowfall and ice. It looks really sharp in a newscast, but the totals are usually never exact. Think I’m crazy? DVR the next snow event on all of the local news stations and then verify the forecast with the actual measurements.
The problem for the broadcast meteorologist is how to communicate uncertainty.
Several consultants and media researchers preach ‘certainty in a forecast’. I actually had a consultant tell me to eliminate the word “chance” because it made it sound like I wasn’t certain. (Um, that’s because I wasn’t certain). Research shows people watch meteorologists who give concrete forecasts and who are 100% accurate. While that makes sense, our weather is anything but concrete or 100% certain. Weather is an untamed science. We understand the atmosphere more every day but there are still some things that we don’t understand. For example, meteorologists still don’t know exactly why some storms produce tornadoes while other storms in exactly the same environment don’t. It’s the mystery of weather.
Computer models work by taking the current conditions from balloon launches and satellites and then putting those variables into mathematical algorithmic models which simulate what will happen. (Yes, it’s as confusing and as complicated as it sounds… and it’s a miracle they even work) Our closest balloon launches are in Little Rock, Springfield, & Norman; that’s a large chunk of land. Nationally, the computer models are looking at a large grid with several holes missing. To account for the data holes; the model interpolates. So if it’s 30 in Oklahoma City and 60 degrees in Little Rock, the model assumes it’s 45 in Fort Smith. See the problem?
Hurricane landfall forecasts have large margins of error several days out because there is limited data and balloon launches in the middle of the ocean. In some cases no data at all other than satellite.
With winter weather forecasts the vertical temperature profile of the atmosphere is extremely important but usually not known with a large degree of certainty until around 36-48 hours before the event starts. So if snow is in the forecast more than 2 days out… we really do not know exactly how the event will unfold until it actually starts.
6) Small Scale Weather Features
We call small, localized events “mesoscale” and they cannot be forecast several days in advance.
If I show an area 1/2 the size of Arkansas and say a tornado is possible here Wednesday and then a tornado occurs on Wednesday, I look like a weather genius.
If I show an area of the same size and say 2-6″ of snow is possible and an area receives 2-6″ yet the rest of the area doesn’t see snow: “I’m the only person alive that can be wrong 50% of the time and still keep my job”.
Both processes are similar. They’re small scale (mesoscale) features that must be forecast in real-time. We can pinpoint locations favorable for tornadoes but I can not tell you with 100% certainty that Washington Co will see a tornado 24 hours in advance, nor can I pinpoint the exact location of a mesoscale heavy snow band 1-2 days in advance.
Are the effects from 1″ of snow different than 3″ of snow? If the temperatures is the same for both: not really.
In general, I prefer the National Weather Service definition of a Winter Storm; 4 inches or more of snow. I realize media outlets across the county refer to trace and 1/2 inch accumulations as “winter storms” but in my opinion that desensitizes society to the really big storms.
8) The 10:1 Ratio
Another problem with snow totals is the liquid to snow ratio. Usually, you get 10″ of snow for every 1″ of water. This is typically how we derive snow total forecasts.
However, humidity in the snow growth region of clouds can vary. In the case of the 2011 Highway 412 snow storm the liquid to snow ratio was as high as 15:1. The liquid remained the same, but the snow totals were significantly higher because of the fluffy nature of the snowflake (combined with the mesoscale band). See how it messes with forecast totals?
What’s interesting as a meteorologist is the difference in perception. If rain is in the forecast and you get .15″ of rain and your friend down the road gets .60″ of rain… in both of your minds: it rained.
Take the same liquid and convert it to snow. Now you have 1.5″ of snow and your friend has 1/2 a foot. The perception is completely different even though the liquid equivalent is the same.
9) 3″ Of Dusting
I don’t say the word dusting much in forecasts. There are several instances where light snow accumulations are possible and in those cases I usually say “minor accumulations”. Even if 80% of the area only sees trace or dusting amounts, there will inevitably be a narrow localized band that yields totals of 1-3″.
I usually get phone calls about the “3” of dusting in my backyard”; so with weak systems I refer to the forecast amounts as minor accumulations and leave it vague. Generally, with any winter system there will be some localized variations of higher or lower amounts. The goal of the forecast is to communicate the most likely scenario for the most people while also showing the different possibilities that can occur for the few people who see far less or far more snow.
10) The 5 Types of Systems
There are generally 5 common types of winter set-ups that cause snow and ice in Arkansas: Upper lows, Surface lows, Shortwaves, Wrap Around, & Overrunning. Here’s a brief look at each one:
– Upper Lows: Snow is directly under the low’s track. Surface can be warm but yield very high snow totals because of moisture and cold air aloft. Path of snow is somewhat easier to forecast once we determine the low’s forecast track and with daytime snows the traffic problems are usually limited.
– Surface Lows: Snow forms north and northwest of the low’s track by around 100-200 miles. Can cause blizzard conditions depending on pressure gradient. Usually brings us the heaviest snows.
– Shortwaves: Surprise snows!!! Shortwaves are small spins in the atmosphere known as areas of ‘vorticity’. As spinning air moves, the air traveling through it stretches and contracts. The stretching causes uplift (known as vorticity advection) and the contracting causes downward motion and drying air (which is why Water Vapor is excellent to see them). When they’re embedded in a northwest flow they can move from Wyoming to Arkansas overnight and generate enough lift for a quick inch or two of snow. They’re not easily identified on computer models and are the most common culprit when you wake up to snow and it wasn’t in the forecast.
– Wrap-Around: After a surface low leaves, cold air is pulled behind the system and mixes with moisture. This causes some light snow. In our area, wrap-around snow is overplayed and models tend to generate more snow than will actually occur. Elevation plays a key role so Northwest Arkansas sees more wrap around snow than the River Valley. As a forecast rule of thumb I typically take the forecast amount of snow and then cut it in half. In the last wrap-around snowfall, 4″ was forecast. I cut the total down to 2″. In reality only 1/2 inch to an inch occurred. Forecasting snow in our area is a nightmare for meteorologists for who aren’t from this part of the country or meteorologists who are just starting out because of the small weird quirks of Arkansas that can only be gained for experience and paying close attention to what’s happening as it happens.
– Overrunning: Ahead of a surface low, warm air flows over the warm front on a slope or an incline. Because of this you have warmer air flowing over colder air. This is usually the cause of ice storms.
Thanks for taking the time to learn a little bit of insight into my brain and the different things that go into making a winter weather forecast. My philosophy with winter forecasting is to try and communicate the upcoming weather and how that weather will intersect with your life. I also try to communicate the uncertainty within the forecast so you know the range of conditions that are possible.
Looking forward to Spring!