Have you seen this forecast map? It was shown by a weather group and picked up by many snow hungry people. If I had a dollar for every question I got about this today, well… I’d have a few dollars. But I felt the need to expand on this as we enter the winter season, because there is a lot wrong with it. I am as big of a snow fan as anyone. This snowhound will sniff out any potential flakes for us, and get excited with you. Faith-in-the-Flakes*, right? But I can’t and will not show a snow forecast that is 336 hours away. That’s 14 days away! Two full weeks! My examples below compare, contrast, and show the different weather outlooks from models 10 days away. Heck, just this morning the rain was much less impressive than what I showed the night before. Computer models are guidance, not gospel. Especially when plotting the weather so far away. Even more important in this hyped El Nino year. More responsibility is needed when showing a map with snow.
I am purposely NOT mentioning the group that first shared this image, because I don’t believe in calling people out. I am just trying to extend proper practice, otherwise rumors go rampant in social media. We’ve seen that in prior winters. Two years ago a teenager showed a megastorm weeks away on model and didn’t understand what it really meant. The rumors got so bad, many TV newscasts between Philadelphia, Baltimore, and Washington were interrupted to explain the rumor and that the meteorologists were not ignoring an epic event… because there was none. I hope it doesn’t come to that again.
The Chaos Theory clearly states that farther in the future, predictability of anything has much lower odds. There are so many variables, and if you miss one, it could lead to a much different result. It’s the expression: If a butterfly flaps it’s wings in Bangkok, it will rain in New York. One small push could have an unforeseen chain reaction.
The best way to show you this is to look at the 240 hour (10 day) plot of three different computer models below. Notice that some features have sublet differences, the last model has a completely different outcome.
GFS Model: 10 Days Away
A cold front reaches the east coast, but gets hung up in Texas.
European Model: 10 Days Away
This time frame is a little earlier in the day, but the cold front in the Gulf coast is clear of Texas and near New Orleans. The GFS was trying to develop a storm in Texas at that time, while this model has it clear and therefore harmless down the road.
Canadian GEM Model: 10 Days Away
This model, my favorite in cold weather patterns (closer to an event), is completely opposite. It has our region clear and dry under High Pressure. Also, look at that potent storm off the west coast of Mexico. Is it trying to develop something tropical there? Could that eventually sent moisture into the southern US that the other models are missing, or is this bogus?
If anything, showing how a weather pattern might develop in a week or so would be best demonstrated using upper level maps of the jet stream. There is a better chance of seeing what the pattern might do in the future aloft, than fine details at the surface. That is why I like to wait until within a week to discuss a storm.
If you have followed my forecasts over any winter, then you know I also don’t give snow amounts until at the earliest 3 days away, and then only a low end confidence number. Having an early idea of how a storm will behave is important, but guess too early most often leads to being wrong.
This is my appeal to readers and weather bloggers (meteorologists and hobbyists): Please be careful what you show. Going out 14 days with a fun looking weather map, most of the time will not play out. There could be a miracle and it does, but more than 95% of the time it won’t. That only leads to the notion that weather people are wrong even though its not a forecast. Lots of people just see these maps, look of the headlines, and don’t see farther down the page. I hope we see less hype and more real snow this year. But that, still a wild card thanks in part to the record El Nino.
If this does verify, it might be a first for the GFS model, and I would be the first in line to point out my post here. I’d like to be wrong… but I still believe posting a surface map with wintry precip 2 weeks away should not be done. Even with my buddy the Canadian model 😉
It will get cold this weekend. We can live vicariously through Chicago and their storm. That will get more in the mood for Faith-in-the-Flakes* and we will get ours soon enough.
Please share your thoughts, best weather pics/video, or just keep in touch via social media
- Facebook: Justin Berk, Meteorologist
- Twitter: @JustinWeather
- Instagram: justinweather
Get the award winning Kid Weather App I made with my oldest son and support our love for science, weather, and technology. Our 3 year anniversary of the release and our contribution to STEM education is this November. It has been downloaded in 60 countries, and works in both temperature scales. With your support we can expand on the fun introduction to science and real weather.