Posted by Tropical Storm Fay is bearing down on Cuba and the Florida Keys as I write this and is on the cusp of hurricane strength winds. A new study from the University of Miami’s Rosenstiel School of Marine and Atmospheric Science just published in Geophysical Research Letters looks again at the contentious issue of whether a warmer globe means more or worse hurricanes. The proposition that global warming mean more and fiercer hurricanes is derived from large computer simulations that have relatively low resolution for local weather events like hurricanes. The computing power needed to get better detail is immense, but if there are variables of particular interest for specific events we can do better than what is usually possible from models that are more general. This kind of trade-off is typical in science. In epidemiology we always want more subjects to study, but when we do that we typically incorporate less detail about each individual. In the new hurricane study, UM researchers David S. Nolan and Eric D. Rappin, working with MIT’s Kerry Emanuel, used extra detail about processes important to tropical hurricane formation (sea surface temperatures and the change in wind speed and direction with altitude [wind shear]):
“We designed the computer simulations to show that as the ocean temperature increased, hurricanes would form more rapidly and easily, even in the presence of wind shear,” says Nolan, associate professor of Meteorology at the Rosenstiel School. “Instead, we got exactly the opposite result. As the water temperature increased, the effectiveness of the wind shear in suppressing hurricane formation actually became greater.”
The simulations show that if they do form, hurricanes become stronger in the warmer environments. Together, these results suggest that in a global warming world, there would be less hurricanes, but those that do form could become stronger. The same prediction has recently been made by other studies using global climate models, and the similarity of the two predictions enhances confidence in the results. (Eurekalert)
Here’s an example of one of the computer runs:
Figure shows an example of a hurricane computer simulation generated by the Rosenstiel School team. The colors indicate water vapor in a vertical column of the atmosphere, where the dark red areas would indicate extremely heavy rainfall. The small size of each pixel, 3 km x 3 km provides remarkably accurate detail in the storm. In comparison, the number of pixels in an image used to represent storms in global climate models are typically 100 km x 100 km, at best. (Credit: UM/RSMAS)
The hurricane bottom line seems to be “fewer but worse.” Whether this will turn out to be the case or not and if it does whether it has to do with human activity will no doubt be the subject of vigorous debate. Unfortunately not all parties in the debate are equally well armed. If you don’t think global warming will affect hurricanes, then it would seem you would have to say what you think is wrong with the scientific reasoning that went into this work. I’m guessing that the attacks on it won’t be about the science but about the meta-science (e.g., “You can’t believe computer models,” or “You can make a computer model say anything you want it to”).
There’s no way to stop that kind of response. Meanwhile, this is the best guess we have yet as to what the consequences of our current policies might be.
August 18, 2008
August 18, 2008
August 18, 2008
August 18, 2008
August 18, 2008
August 19, 2008
August 19, 2008
August 20, 2008
August 20, 2008
August 20, 2008
August 21, 2008
August 24, 2008
