From Gregg Easterbrook:
Before the 2006 hurricane season, both the government's National Hurricane Center and the media-favorite hurricane prediction center at Colorado State University forecast an exceptionally severe Atlantic hurricane season: Instead, the 2006 season was smack on the average of the past half-century. Before the 2007 hurricane season, the National Hurricane Center and Colorado State again predicted a harsh Atlantic season. In April 2007, Colorado State called for a "very active hurricane season," with nine hurricanes, five of them intense. (The average of the past half-century is six Atlantic Ocean hurricanes, two of them intense.) In May 2007, the National Hurricane Center forecast a "75 percent chance that the Atlantic Hurricane Season will be above normal," with up to 10 hurricanes, as many as five of them intense.
The National Hurricane Center and Colorado State weren't alone. AccuWeather's weightlifting Hurricane Center Chief Forecaster Joe Bastardi, who is "recognized for his astonishing ability to grasp the potential impact of severe weather patterns," warned that the U.S. Gulf Coast "is at much higher risk of destructive tropical weather" in 2007 than occurred in 2006. Tropical Storm Risk, a weather forecasting consultancy in London, predicted the 2007 season would see nine hurricanes, as many as four of them intense. The forecast came from Mark Saunders of University College London, whose bio calls him a Professor of Climate Prediction. (Go here and click Saunders' bio under the "staff" link for hilarious puffery in the English vein, including that he received an award "at the Royal Albert Hall" and a claim Saunders "has published nearly 300 scientific research papers." What this means is Saunders' name has appeared in author lists on papers with dozens of authors. (No human being has ever done 300 papers' worth of original research.)
So what happened? The first half of the 2007 season was quiet, with no hurricanes. In early August, the National Hurricane Center announced it was standing by "expectations for an above-normal season. As we enter the peak months, August through October, of the Atlantic hurricane season, NOAA scientists are predicting an 85 percent chance of an above-normal season, with the likelihood of … seven to nine hurricanes, of which three to five could become major hurricanes." About the same time, William Gray of Colorado State -- the media's favorite hurricane forecaster, perhaps because he always predicts disaster -- declared, "Though the 2007 hurricane season has gotten off to a slow start, my colleague Phil Klotzbach and I still anticipate another active season with above average numbers of major hurricanes." A slow start. Come on you hurricanes, let's get cracking, form and destroy something! Instead, the bottom line on the 2007 Atlantic hurricane season was six hurricanes, two of them intense -- smack on the average of the past half-century.
This did not discourage Colorado State: "Based on our analysis of fall parameters, the 2008 Atlantic hurricane season is expected to be more active than the average 1950-2000 season," the school's forecasters said in December 2007. The initial Colorado State forecast for 2008: seven Atlantic hurricanes, three of them intense. Hurricane predictions bonus: In November 2006, after his 2006 Colorado State forecasts proved wrong, Gray declared, "We are improving our skill in seasonal prediction with an improved level of understanding." He predicted his predictions would get better -- and was wrong about that, too.
Anyway, scary as hurricanes seem, especially after Katrina, as a kid growing up in Tornado Alley TAE leared that hurricanes historically are pound for pound the least deadly of all natural weather phenomena. Earthquakes, tsunamis, tornados, and blizzards all occur with generally less than 12 hours warning, and therefore are virtually impossible to escape through planned evacuation. But hurricanes are almost always predictable to within 90 miles 48 hours in advance, giving residents of coastal regions plenty of opportunity to move inland. A good example is the 2004 Indian Ocean Tsunami, which killed an estimated 230,000 people. Hurricane Katrina, which caused a relatively similar amount of flooding and surface damage, claimed about 2,500 people (includes missing persons). In terms of lives lost, Hurricane Katrina, the fifth deadliest hurricane of all time, was only 1/100th as deadly as the 2004 Tsunami.