Alex didn’t wait for June, the official start of hurricane season, or even for the end of winter.
Instead, Hurricane Alex formed in January, becoming the first Atlantic hurricane in that month since 1938, and only the fourth since 1851. The rare winter storm developed near the African coast, churned a path across the Atlantic Ocean and hit the Azores Islands near Portugal.
“Hurricane Alex defied the traditional storm season,” said Jenni Evans, a Penn State meteorology professor and associate in the Earth and Environmental Systems Institute.
While Alex may have come early, for Evans hurricane season never truly ends.
Throughout her career, Evans has studied the powerful, unpredictable nature of hurricanes. Her research group has developed innovative techniques to better understand how these storms evolve, with an eye toward improving forecasting accuracy and warning times.
“Imagine you are a forecaster, and you get dozens of potential forecasts of where this storm might go,” Evans said. “You have an hour to tell emergency mangers and all these people living in an area whether or not they’ve got to get out. It’s a great challenge.”
The scenario played out this fall in the United States, when millions along the East Coast nervously watched the forecast to see whether Hurricane Joaquin would make landfall. The storm had already hit the tropics hard, including the Bahamas, Antilles and Bermuda.
As many in the U.S. were bracing for the worst, the storm veered out into the Atlantic Ocean, sparing us a direct hit but still contributing to serious flooding in North and South Carolina.
To study how storms like Hurricane Joaquin behave, Evans turns to a technique called ensemble forecasting. The approach uses computer models to run a single forecast many times with varying initial conditions. Changing the initial conditions takes into account uncertainties that exist in our observations of the atmosphere and oceans.
“In this way, we aim to find all of the possible future states of the weather, starting from the way things are today,” she said.
Clustering forecasts
Evans has helped introduce another technique, called path clustering, to take all the potential forecasts and narrow them into a handful of ways the storm might evolve.
Path clustering uses statistics to analyze many forecasts, making it easier to quickly and accurately assess where storms are headed and how strong they’ll be.
“Say we’ve got 120 forecasts, but we know there aren’t 120 fundamentally different ways the atmosphere works,” Evans said. “We boil it down to three, four or five possible “clusters” of storm forecasts; each cluster is forecasting very different ways the atmosphere could evolve.
“By considering the differences between the forecasts based on current theories in concert with advanced observations, and informed by historical events, the number of clusters can be further reduced,” she said. “Then we can examine the forecasts in each remaining cluster and see what each tells us about the possible ways the storm might evolve.”
Evans’ research group developed a framework for summarizing hurricane evolution that is used today by the National Hurricane Center. And her work was recently featured by NBC Learn, the educational arm of NBC News, in its series "When Nature Strikes: Science of Natural Hazards." NBC Learn partnered with the National Science Foundation and the Weather Channel on the online videos, designed to educate students about natural disasters. The NSF also funds Evans’ research.
Evans is particularly interested in storms that leave the tropics and enter higher, cooler latitudes, where they can become powerful, unpredictable hurricanes — called an extratropical transition events. Notable examples include Hurricane Joaquin in 2015 and Hurricane Sandy in 2012.
“When you take a tropical cyclone, which is a warm, moist atmospheric energy ‘bomb’, and you bring it up to these latitudes, it’s a big instability,” Evans said. “In the environment outside the tropics, the storms are really juiced up compared to typical storms in the region, so they have a lot more potential to cause damage, and they also move a lot faster. The energy a tropical cyclone brings to a region can also cause new, damaging storms to grow.”
Unfortunately, these storms have also been difficult to forecast — with factors like the jet stream, other weather systems or water temperatures causing storms to veer off into the Atlantic Ocean or to turn and strike the East Coast, making ensemble forecasting even more important.
“It’s important we get these forecasts right, because people are in the way,” Evans said. “They are in the way in terms of shipping, fishing, vacationing and just living."