Hurricane Predictions Improved Just Before Sandy
Data from flying planes directly into storms, plus some algorithm changes, have improved the U.S.' storm predictions every year over the past few years, IEEE Spectrum reported. With the 2012 hurricane season, scientists improved their predictions for the intensity of storms for the first time in 20 years.
The improvement, which went into effect in June, came in time to figure into predictions for Hurricane Sandy, which TechNewsDaily sister site Livescience reported were "on track."
Just this year, the U.S. Global Forecast System has improved estimates of hurricanes' tracks by 20 percent and intensity by 10 to 20 percent, Robert Gall, technical director for hurricane forecast improvement at the National Oceanic and Atmospheric Administration, told Spectrum. The changes are part of the National Weather Service's larger goals to sharpen track and intensity predictions by 20 percent by 2014, and by 50 percent by 2019, Spectrum reported.
Nevertheless, Sandy was a difficult storm to understand at first. The National Weather Service uses several computer models to predict a hurricane's path. When the models agree, predictions are more certain; when they have different outcomes, forecasters are unsure what will happen. In Sandy's case, the models initially disagreed about whether the storm would veer off to sea or make for land. By Oct. 25, however, the differences began clearing and by Oct. 27, all the models agreed that Sandy was heading toward the U.S. mainland.
The Spectrum story, which was published Oct. 28, ends with Gall's reporting that Sandy was expected to reach the U.S. as "barely a Category 1 storm, the lowest in the intensity rankings, but will come with a big storm surge and huge amounts of rain." Sandy made landfall in North America earlier than expected as a post-tropical cyclone with hurricane-force winds. The resulting storm surges were higher than predicted, causing severe flooding in New Jersey and New York.
Source: IEEE Spectrum
Follow TechNewsDaily on Twitter @TechNewsDaily, or on Facebook.





