In late March, a team at the University of Arizona developed a better model for predicting the number of hurricanes in the upcoming season.
The new model, developed by graduate student Kyle Davis and atmospheric science professors Xubin Zeng and Elizabeth A. Ritchie, makes predictions that are 23 percent more accurate than previous models.
With the improved predictions, areas along the Atlantic and Gulf Coast can better prepare for catastrophic storms.
“Tens of millions of people are threatened by Atlantic hurricanes,” Davis told the University of Arizona. “It affects their properties; it affects their lives.”
According to the National Aeronautics and Space Administration (NASA), hurricanes are tropical cyclones that form over the Atlantic Ocean or eastern Pacific Ocean. They only form over warm ocean waters near the equator.
Wind blowing westward across the Atlantic helps the warm surface water evaporate. As it rises, the water vapor cools and condenses, forming cumulonimbus clouds.
As the pattern of evaporation and condensation continues, wind circulates around a center. If the moving column of air and clouds encounters other clouds, it forms a cluster known as a tropical disturbance.
While the thunderstorm builds, air at the top cools and sinks. The heat energy released from the cooling creates a high-pressure zone. Winds move away from the area, creating low-surface pressure, where the winds return and rise to contribute to the growing thunderstorms. Winds within the storm clouds whip around in a circular motion.
If wind speeds reach 74 miles per hour, then the National Weather Service classifies it as a tropical cyclone. These storms are at least 50,000 feet high and 125 miles across.
Hurricanes are among the most destructive natural disasters in the United States, with some investigators estimating that damages from U.S. hurricanes from 1970 to 2002 cost $57 billion by today’s standards.
That is more than the damages caused by earthquakes and human-caused disasters combined during the same time period. For a season that lasts from June 1 to November 30, preparation is vital.
The previous model for hurricane prediction was effective from the 1980s until the late 1990s. This period had a steady average of roughly six hurricanes per year.
The 21st century saw an increase in variability of hurricanes per season: there were 15 hurricanes in 2005, but only two in 2013. The old model, which made its predictions based on the state of the El Nino climate cycle in that particular year, no longer made consistent predictions.
The new model adds two important variables: the state of the Atlantic Multidecadal Oscillation (AMO) and the force of the wind on the ocean.
The AMO is a longer-term climate cycle than the El Nino. It affects ocean temperatures, shifting from colder to warmer to colder again over the course of 40 to 70 years. Warmer sea surface temperatures tend to spawn more hurricanes.
In the late 1990s, the AMO started to cycle back toward warm, which could explain the increased variability in the number of hurricanes since that time.
Along with the AMO, the team incorporated the force of ocean winds into their statistical model. Strong winds mix ocean layers, bringing cooler water from the depths to the surface.
The team used data from the 1950 to 2013 hurricane seasons. To test the model, they created a “hindcast” of the number of hurricanes that occurred from the 1900 to 1949 hurricane seasons.
“It performed really well in the period from 1949 to 1900 [sic]. That’s the most convincing test of our model,” Davis said of the results.
“Our model is better at predicting the number of seasonal hurricanes in the Atlantic than the other existing models.”
