When was the last time you heard about a hurricane hitting the state of Arizona? Can’t remember? Neither can I. But yet the best new research on predicting the frequency of hurricanes appears to come from a graduate student and his professor at the University of Arizona (UA).
The student, Kyle Davis, and his professor, Xubin Zeng, have developed a new model that improves the accuracy of seasonal hurricane forecasts for the North Atlantic and the Gulf of Mexico by 23 percent. The team’s research paper was published online in the journal Weather and Forecasting on March 25, 2015.
Hurricanes are a big deal. Tens of millions of people and countless dollars worth of property are threatened each year in this country by hurricanes. Investigators have estimated that damages from U.S. hurricanes from 1970 to 2002 were $57 billion in 2015 dollars — more than earthquakes and human-caused disasters combined for the time period.
Predicting the occurrence of hurricanes more accurately will help disaster response teams have a more educated guess about how to prepare. We all know what not being prepared means – think Hurricane Katrina (2005), Hurricane Ike (2008) and numerous others.
Maybe it’s just a technicality but the article first says the new model developed by this team improves accuracy of season hurricane forecasts by 23 percent. Then later it says that the new model has 23 percent less error. I’m not entirely sure if the two ways of reporting the results mean the same thing.
The team developed the new model by using data from the 1950 to 2013 hurricane seasons. They tested the new model by seeing if it could “hindcast” the number of hurricanes that occurred each season from 1900 to 1949.
“It performed really well in the period from 1949 to 1900,” Davis said. “That’s the most convincing test of our model.”
Here is a link to the article on the UA web site: http://uanews.org/story/a-better-method-for-forecasting-hurricane-season.
And here is a picture of Hurricane Katrina in all her menacing beauty.