Cyclone Catarina in the South Atlantic as photographed by astronauts on the International Space Station in 2004. Image credit: NASA via Wikimedia Commons // Public Domain

Weather forecasts are hardly ever perfect. Most of the forecasts you receive these days are pretty solid if you use the right source, but it’s not always a sure bet. Sometimes your friendly local meteorologist uses bad information to arrive at a bad forecast, and sometimes he or she just misses one. Predicting the future inherently involves some level of uncertainty, and some situations are more uncertain than others. Hurricanes are no different. When we look at a hurricane forecast map, the most striking feature you’ll see is the aptly named cone of uncertainty.

A hurricane forecast is conveyed to the public using a map that shows the forecast position of the very center of a tropical cyclone at seven different time steps up to five days out from the initial forecast. The cone of uncertainty, which looks like a quotation bubble emanating from the first point on a hurricane forecast map, is the historical margin of error in a meteorologist’s forecast track for the center of a tropical cyclone.

The forecast track and cone of uncertainty for Hurricane Igor in September 2010. The hurricane’s center stayed within the cone of uncertainty for almost its entire lifespan. Image: National Hurricane Center

At the end of every hurricane season, meteorologists at the National Hurricane Center look at all of the forecasts they made for storms in previous years and average out how far off their forecasts were from the actual location of the tropical cyclone. They average out and compile these errors for every time step in the forecast period all the way out to five days from the initial forecast. When you plot their average track errors on a map using a sequence of circles, the smoothed-out end result is the cone of uncertainty we’re all familiar with.

The center of a tropical cyclone typically stays within the cone of uncertainty 66 percent of the time, which also means that it strays outside of the cone 34 percent of the time. Tropical cyclone forecast tracks are much better today than they were in years past, but track error naturally grows with time as confidence in a forecast track decreases. The National Hurricane Center is only off by an average of about 35 miles when they predict the location of the center of a tropical cyclone 12 hours out. That average error grows to 97 miles two days out and goes all the way up to 273 miles five days out. To give you an idea of how wide the cone of uncertainty is five days out, it’s just roughly the distance you would cover traveling from Boston to Philadelphia.

The cone of uncertainty does not show you the confidence that forecasters have in that particular forecast. It shows you neither the size nor the total extent of the impacts a tropical cyclone will have. It doesn’t matter if it’s a 500-mile-wide hurricane or a 20-mile-wide tropical depression; the cone only applies to the very center of a hurricane, and it only tells you how far off forecasters were when they predicted the track of previous cyclones. You aren’t necessarily safe if your location falls outside of the cone of uncertainty, nor does it mean that areas inside the cone are definitely going to get blasted. Understanding the size and impacts of a tropical cyclone requires more context than one simple map can provide you.

Most tropical cyclones form and move along tracks that don’t stray too far from their predicted paths. A sudden jog to the west or east usually isn’t too much of an issue, but even a small deviation in course could be a big deal during complex or high-risk situations.

The forecast track and actual track of Hurricane Joaquin in October 2015. Map: Dennis Mersereau

The most common reasons storms deviate from their predicted paths are that they’re extremely weak (and subject to erratic movement) or they exist in a complicated setup with lots of moving parts. One of those latter situations occurred in 2015 during Hurricane Joaquin. The major hurricane formed over the Bahamas in a complex environment that would ultimately dictate whether it would make landfall in the United States or swirl harmlessly out to sea. On October 1, meteorologists issued a forecast that showed the storm approaching the Mid-Atlantic coast of the United States. Most of the models misread the weather pattern that week and the storm wound up trekking out to sea, causing Hurricane Joaquin’s ultimate path to fall almost entirely outside of the cone of uncertainty for that forecast.

If you find yourself in or near the cone of uncertainty four or five days out, that’s your cue to keep tabs on the situation and make preparations in case you have to secure your property or evacuate to safer ground. It’s time to act if you’re still in the cone a few days out. Watches and warnings will already be in effect by then and you’ll have plenty of information from the experts telling you what you should do in order to stay safe.