KMSP - I am sure you have heard by now that there is the potential for a winter storm at the tail end of this week. Social media is all abuzz with images of snowmaggedon and the worst case scenario apocalypse that I have come to dread whenever we have the potential for more than a dusting of snow. Remember, it’s the internet, so misinformation is EVERYWHERE. Well, I want to put your fears at ease. At this point, DO NOT believe any image you see on the interwebs about the potential snow totals for the end of the week. Notice that there isn’t one posted here because it is irresponsible to do so for a few reasons…
First: The atmosphere has hundreds of trillions of variables that we can’t measure and only a few we can. Any one of these unforeseen variables can change the forecast drastically. Therefore posting a snowfall forecast more than 5 days away is just illogical. It’s GOING to change. Do we know how it will change? Nope… that brings me to my next point…
Second: The forecasting computer models do their best to compute all of the variables we can measure to come up with a forecast. There are a few “long range” computer models we can look at (long range in the forecasting world typically means over 72 hours) that give us an idea of what may occur during its forecasting time period. EVERY model will show you something different. In some cases they are similar to each other, in other cases, not so much. So if these computer models change constantly and never seem to agree with one another, how can they be accurate? That brings me to my final point…
Third: The simple answer to that question is, they can’t. It takes a TRAINED meteorologist to decipher these different weather models, take their knowledge of the atmosphere and fluid dynamics, and come up with a forecast. We do this because the models are computers, but the calculations they make were derived by HUMANS. This means that there are some flaws in EVERY computer model, and it takes a meteorologist to know what those are. So what do we know at this point?
Here’s what we know: Record or near record breaking heat will continue through Wednesday and more typical February style weather will arrive by Friday. There will likely be a storm SOMEWHERE in the central US. Right now, it looks to give at least a little snow to some part of Minnesota.
Questions that still need answers but we probably won’t have those answers until Wednesday or Thursday: How much snow will it be? Where will it land? Will it be significant? What is the meaning of life? Ok, we may never know that one… I was just seeing if you are paying attention.
Bonus meteorology lesson:
In the grand scheme of things, 5 days is a blink of an eye, and yet we still can’t predict where a storm (big or small) will end up in just a few short days. Along with the reasons already stated, there is one key to our failure. Our knowledge of the atmosphere comes mostly from actual observations. Yes, we have several satellites orbiting the planet, but they still can’t accurately measure what’s going on in every corner of our atmosphere at every time of the day. We also use weather balloons to send instruments up into the atmosphere to record everything it senses all the way to the Tropopause. But these balloon launches are limited to about 50 locations nationwide and rarely done in other countries. So we still rely heavily on physical ground level observations… sensors placed at airports around the globe to measure temperature, wind, rain, snow, humidity… and a list of other variables. It’s these observations that get loaded into the weather computer models that give us forecasts for what may happen next.
But here’s the issue. We are limited to locations on LAND. Other than a few dozen buoys around the world, the vast majority of our surface and upper air observations are made from locations on land. This makes most of the weather over our oceans, 70% of the total surface area of our planet, nearly invisible to our computer forecasting models. Well, considering that North America gets the majority of its weather from the west, and the world’s largest ocean is west of the continent, than it can make for some tricky forecasting when these storms are nearly invisible to forecasting computer models until they hit the coast. We don’t usually get a better handle on these storms until they impact the West Coast, hit surface and upper air observations there, send that data into the supercomputer, and compute a better forecast. Unfortunately, that typically gives Minnesota about 36 hours (on average) advance notice on a “far more accurate” forecast, especially when it comes to big storms. That’s why we spend a lot of time talking about the potential for the track of the storm to “change”. The storm track doesn’t necessarily change since it hasn’t occurred yet, BUT because the storm is now onshore and hitting more data points, we get better and more information about the potential direction the storm will take and then adjust our forecast accordingly. See… there is a lot more to forecasting then you thought huh?