GIGO In Action

USA Today presents us with today's object lesson in the concept of GIGO, or 'Garbage In, Garbage Out'. Reporting on a new study that claims to have found a direct link between global warming and a rise in the number of hurricanes.

The researchers found that average hurricane numbers jumped sharply during the 20th century, from 3.5 per year in the first 30 years to 8.4 in the earliest years of the 21st century. Over that time, Atlantic Ocean surface temperatures increased .65 degrees, which experts call a significant increase.

This study also shows that years with more hurricanes didn't coincide with changes in the way storms are measured, says hurricane researcher Kerry Emanuel of Massachusetts Institute of Technology, who was not part of the study. "This makes it very unlikely that these upward jumps are owing to changing measurements and suggests that they are real."

But buried way down at the end, USA presents this:

The new study drew criticism from experts who dispute the merits of combining data from the late 19th and early 20th centuries, when hurricane-tracking satellites didn't exist, with statistics gleaned from more modern technology.

"Looking for trends in noisy count data is fraught with problems," says researcher James Elsner of Florida State University in Tallahassee. "I agree with the message, but cannot recommend the science."

"They're saying there's a long, upward trend of the last 100 years in tropical storms. All the data I have looked at show that's not the case," says scientist William Gray, head of the Tropical Meteorological Project at Colorado State University.

Gray, a critic of the view that human-induced greenhouse gases drive climate and hurricanes, says 19th-century data "is just not that good."

Reuters report on the same study has this:

Skeptics say hurricane data from the early decades of the 20th century are not reliable because cyclones likely formed and died in mid-ocean, where no one knew they existed.

More reliable data became available in 1944 when researchers had airplane observations, and from 1970 when satellites came into use.

Despite the confident claims, consider for a moment: Until the mid 1940s, there were no reliably consistent way to track the majority of storms. Unless a ship sailed through the storms itself - and reported that to someone – there would be no records. You can look at the compiled hurricane data yourself and notice that hurricane reporting from the 19th and early 20th century very often show no activity out in the middle of the Atlantic. At all. Does anyone in their right mind believe that? Also, how many ships gathered temperature data through the 19th and 20th centuries? Probably the most significant criticism noted above is the one from James Elsner of Florida State University. He believes that warming increases hurricanes,, but also points out that the claims of the two researchers are based on nothing.

UPDATE: The AP headline is actually the worst one yet, but their story also has the most strongly worded slap at these two charlatans:

Atlantic tropical storms have doubled

Chris Landsea, science and operations officer at the National Oceanic and Atmospheric Administration's National Hurricane Center, said the study is inconsistent in its use of data.

The work, he said, is "sloppy science that neglects the fact that better monitoring by satellites allows us to observe storms and hurricanes that were simply missed earlier. The doubling in the number of storms and hurricanes in 100 years that they found in their paper is just an artifact of technology, not climate change."

I wouldn't even call it sloppy. Fraudulent fits better.

UPDATE: Others: Wizbang, Iowa Voice, Flopping Aces, Iconic Midwest, Wizbang has a really excellent graphic up that shows exactly how much data collection has changed over the time period in question. Paul did a really good job putting that post together. So did Rich Horton over at Iconic Midwest.

• 

  By Rich Horton, July 30, 2007 @ 6:58 am

  Hear, hear. I’ve been arguing against these fraudulent “studies” for a long time now. (My latest tirade is here.)

  I wrote that last night based on the Reuters article…which reminds me of soemthing since you quoted the egregious bit:

  Skeptics say hurricane data from the early decades of the 20th century are not reliable because cyclones likely formed and died in mid-ocean, where no one knew they existed.

  Skeptics? It is these researcher who are outside the mainstream of the scientific literature. By every normal use of the word THEY are the skeptics.

  This type of crapola really gets me going.

  *grumble*

• 

  By Neo, July 30, 2007 @ 7:36 am

  Equally credible study:

  MP3 damage to human hearing up after Beatles come to America.

  data: 1960 no reported damage
  1964 Beatles come to America
  2007 at least one

• 

  By FedUp, July 30, 2007 @ 11:07 am

  Oh Dear! I guess next they’ll be telling us the world isn’t flat!

Other Links to this Post

1. Global Warming, Hurricanes, and the Anti-Science Right - Liberal Values - Defending Liberty and Enlightened Thought — July 30, 2007 @ 12:29 pm

2. Don Singleton — July 30, 2007 @ 3:23 pm