A winter storm is coming to Indiana and will likely impact travel Thursday night and Friday, December 22 and 23.

Some areas could get a lot of snow. Some could get mostly rain.

As of this morning, the expert meteorologists at the National Weather Service’s Northern Indiana Office can’t forecast where the heaviest snow will fall.

The biggest reason is that meteorological science isn’t capable yet of forecasting the exact track of a low pressure system four days in advance.

I’m not a meteorologist, but I’ve spent enough time collaborating with meteorologists to develop a fair understanding of their work. So, I’ll try to explain here what I know about the coming storm, based on what the pros have published.

First, it helps to know how low pressure systems affect winter weather. Typically, the heaviest snow falls to the north and northwest of a low pressure system. Areas to the east and northeast usually get mostly rain. Areas to the southeast can get clear skies with little or no precipitation.

The path a low pressure system takes therefore has a major impact on where the biggest winter storm impacts will occur.

The main way meteorologists predict a system’s path is by monitoring the output of special computer programs known as numerical weather prediction models. These programs run on supercomputers, because they ingest massive amounts of weather observation data and apply complex equations in their attempts to create reliable, mathematical simulations of the atmosphere. Several such programs exist, all written differently. It’s not unusual for the various models to produce different solutions, which typically come out every 12 hours.

Meteorologist Megan Dodson at the Northern Indiana NWS office wrote in an Area Forecast Discussion this morning that they’re watching the outputs of at least four different models. They’re paying attention to how much consensus exists between the solutions of the various models and how consistent each model’s output is with previous output from the same model.

As of this morning’s forecast discussion, these models indicated that the low pressure system would move from the southwest to the northeast and that the path it follows would be farther west than previous model runs indicated. The image below, published by the Indianapolis NWS office, shows three possible paths for the system. One model, known as the Global Forecast System (GFS), is an outlier, forecasting the most westward path. If it’s right, and the center of the low stays west of Indiana, the Hoosier state won’t get much snow. Other models forecast more easterly paths.

All the models could still be wrong!

Numerical weather prediction is not perfect, but the closer the low pressure system gets to Indiana, the more the models will align and the better the forecasts will get.

As I write this, the system is still over the Pacific Ocean, so it should be understandable that it’s too early for the models to accurately predict how it will track over the Midwest.

For now, it’s important to understand that a winter storm will very likely occur somewhere, but as much as you’d like to know exactly where the most snow will fall (and where the most travel impacts will be), it’s just too early to know.

That said, because it’s also too early to rule out big impacts in any part of northern Indiana, it’s wise to prepare for such impacts. For example, you might shop for Christmas dinner ingredients a day earlier than you originally planned, just in case.

It’s also important to realize that some people who are not meteorologists post official-looking but unscientific forecasts on social media. Often, they choose the single model output that looks the most dramatic (even though it’s an outlier), rather than create legitimate forecasts based on a full understanding of model limitations and model consensus. Often, the main goal of such posts is to create bigger audiences for advertisements. Keep that in mind the next time you see a dramatic snow forecast for any time that’s more than a few days away.