Hurricane Florence is churning towards the US East Coast as a major storm. Further out in the Atlantic, Hurricanes Isaac and Helene are picking up steam as they move across waters with conditions favorable for strengthening. In the Caribbean, forecasters are watching a potential disturbance that could become a tropical storm within a few days.
When you're anticipating incoming storms, it's understandable to want to know exactly where that storm could have an impact. That's especially true for residents of places that have been battered by storms like Irma, Maria, or Harvey.
Forecasts from the National Hurricane Center (NHC) go five days out, yet people often want to know what'll happen a week ahead of time. Because of that, people seek out longer-term forecasts, which sometimes involve "spaghetti models."
These models are technical tools used by scientists who study weather to better understand storms and how they might behave. But, as you can see in the image below, they tend to look like a bunch of squiggly lines on a map, something that has led to quite a few jokes about forecasting.
Unfortunately, these spaghetti plots are a confusing mess for non-experts, since not every line has equal importance when it comes to predicting where a storm is going. In fact, most meteorologists wish people wouldn't look to them for guidance about how to prepare for a storm.
In general, when most of the lines tend on these charts are pointed in the same area, that tends to indicate some agreement between different weather forecasting models. But even then, there can be a lot of variation. Ensemble models like the one below show that using one forecasting system, there are still a number of possible paths — and other forecasting systems would introduce other potential storm routes.
The thing to understand is that these messes of squiggly lines are not the final forecast — they're potential paths that meteorologists use when creating forecasts.
Forecasters often aggregate data from multiple models, weighting ones that have proved most accurate. The paths aren't all equally likely to happen, so you can't just look at a spaghetti plot and easily interpret where the storm will go.
"One of the things that the Hurricane Center encourages people to do is not focus on specific track forecasts themselves," James Belanger, a senior meteorological scientist with The Weather Company (the group behind The Weather Channel and the Weather Underground forecasting service) told Business Insider last year.
According to Belanger, you can classify the slew of different lines you see on these charts into two camps. The first is statistical guidance, which is based on historical data of tropical storms. In some cases, you may even just see tracks of previous storms that were similar, but that doesn't mean any new storm will follow the same track.
The second type of plots are based on simulations that incorporate atmospheric physics into the prediction. These use the effects of things like ocean water temperature, air temperature, and wind shear to help calculate where a storm might go. The two main physics-based models are called GFS, which is primarily run in the US, and Euro models that come from Europe.
The NHC and other forecasting organizations tend to weigh these physics-based models as more important than the ones that are based on historical data.
All of these lines are then used by meteorologists to build a consensus forecast. This forecast takes into account different models, using them to estimate the intensity of the storm and develop a cone within which the storm is most likely to travel. But even then, the intensity of a storm can change rapidly, as Hurricane Florence has shown.
In other words, spaghetti models are useful, but primarily for meteorologists. For the rest of us, an official forecast is more helpful.
