May 31, 2007

Research finds evidence tropical cyclones have climate-control role

Average cyclone-induced surface cooling Download graphic
caption below

These findings support a 2001 theory by Kerry Emanuel, a professor of atmospheric science at Massachusetts Institute of Technology, and suggest that there is an additional factor to be included in climate models that may change predictions of future climate scenarios.

"It was thought that hurricanes occurred over too short of a time period and over too small of an area to affect the global system," said Matthew Huber, the Purdue University professor of earth and atmospheric sciences who led the research group. "This research provides evidence that hurricanes play an important role and may be one of the missing pieces in the climate modeling puzzle."

The research also showed that hurricanes cool the tropics, forming in response to higher temperatures and acting as a thermostat for the area, Huber said.

"Warm water fuels hurricanes, which have been shown to leave cold water in their wake," said Huber, who also is a member of the Purdue Climate Change Research Center at Discovery Park.

"I like to say the good news is that hurricanes function like a thermostat for the tropics, and the bad news is that hurricanes function like a thermostat for the tropics. The logical conclusion of this finding, taking into account past research into the impact of rising temperatures on cyclone and hurricane intensity, is that as the world and the tropics warm, there will be an increase in the integrated intensity of hurricanes."

Movies such as "The Day After Tomorrow" brought into the spotlight information about the ocean conveyer belt and its impact on climate. The upper part of the conveyer belt travels from the south to the north, passing through the Pacific Ocean and Indian oceans and past warmer latitudes warming the water brought to North America and Europe, Huber said.

In the tropical oceans, this pattern must be reversed; warm, buoyant water must be mixed downward, and cold, dense water must be mixed upward. This process, called vertical mixing, plays an important role in the conveyer belt's circulation. It was known that this mixing occurred, but the cause was not well-understood, said Ryan Sriver, the paper's lead author and a Purdue graduate student.

"Climate models today use what is called 'background mixing' to solve this problem," he said. "They represent the mixing as an average of the total amount that is needed and apply it over these regions consistently. However, we believe this mixing is not consistent; it is not everywhere all of the time. It is sporadic and happens over a small area for a limited amount of time."

In some areas of the world, such as the equator, there are no cyclones, and no mixing occurs.

"If cyclones were added to models in place of the background mixing, there would be zero mixing at the equator," Huber said. "This is very important because it is well-known that to get El Niño right in a climate model, the background mixing at the equator must be greatly reduced. Our data has a beautiful no-mixing zone right where there should be no mixing."

This explains some of the mystery of the observed temperatures from the distant past during a greenhouse climate. The poles were much warmer than today, about 82 degrees Fahrenheit, but the tropics were not much warmer than the present, he said.

"Using the best, most comprehensive models in existence, we could not obtain results that matched this past climate that we know existed," Huber said. "We knew a basic, fundamental process that cooled the tropics was missing from the models."

The results of the study, being published in the May 31 issue of Nature, are consistent with providing all of the mixing necessary to match what is needed in climate models.

"Our results suggest that this is the missing mixing and it is a vital part of ocean circulation," Huber said.

Steven Jayne, an assistant scientist at Woods Hole Oceanographic Institution in Massachusetts, said Huber and Sriver present strong evidence for a cyclone-driven heat pump.

"It is remarkable how closely the amount of mixing generated by the cyclones and the location of this mixing matches what appears to be needed to improve climate models," Jayne said. "People suspected these connections, but no one had done the necessary detailed calculations. It means there may be another feedback loop in the climate system, and that is significant."

Huber and Sriver studied the cooling effects of hurricanes from 1981 to the present using the cold wakes that follow a hurricane.

"These cold wakes can be easily observed," Sriver said. "The typical size is about 200 kilometers across and about 1,000 kilometers long, or about as big as the Eastern Seaboard."

The researchers used surface temperature data during the cold wakes to obtain an estimate of the cooling in the tropics due to cyclones and hurricanes. The data analyzed was provided by the National Center for Atmospheric Research, the National Oceanic and Atmospheric Administration, and the National Aeronautics and Space Administration. The team then examined the process that leads to this cooling and evaluated the ocean water mixing.

 "Multiple studies have shown that tropical cyclones are an excellent source of inertial oscillations, or internal waves that cause mixing in the upper layer of the ocean," Huber said. "It is like putting sugar in a cup of coffee. The sweetened coffee is more dense and will form a layer at the bottom of the cup. It needs to be stirred or agitated somehow to bring the sweet layer up to mix with the rest. The same thing is needed to mix ocean water. Dense water hangs out at bottom unless something stirs it up. Cyclones stir it up in addition to other processes."

Tropical cyclones cause waves below the surface of the ocean that break, just like what can be seen on the beach shore. When the waves break, the top layer of water curls into the bottom layer and water of different densities and temperatures mix, Huber said.

"Warm, fresh water is less dense than cold, salty water, so the cold water sinks, and this drives the conveyer belt," Huber said. "However, cold, salty water rises in the Pacific, and there has been no complete explanation for this. Cyclones and hurricanes appear to pump warm water down and bring cold water to the surface. Mixing down buoyant, warm water lessens the density of the cold water and allows it to rise "

The study did not examine deep ocean mixing, but it is reasonable to speculate that warm water pumped down joins the ocean circulation and becomes a part of the upper limb of the conveyer belt where dense water makes it up to the surface, Huber said.

Huber and Sriver plan to incorporate their findings into a climate model for further testing.

"Current predictions are based on tropical ocean mixing remaining constant or decreasing with warmer temps," Huber said. "This evidence suggests the opposite is true, and upper ocean tropical mixing increases with warmer temperatures. This has major implications for oceanography and climate as a new factor that had not been included in previous predictions."

The National Science Foundation and the Purdue Research Foundation funded this research. The Purdue Cyber Center and the Office of Information Technology at Purdue provided computational resources and support.

Writer: Elizabeth K. Gardner, (765) 494-2081, ekgardner@purdue.edu

Source: Matthew Huber, (765) 494-9531, huberm@purdue.edu

Ryan Sriver, rsriver@purdue.edu

Steven Jayne, (508) 289-3520, sjayne@whoi.edu

Purdue News Service: (765) 494-2096; purduenews@purdue.edu

GRAPHIC CAPTION:
This map illustrates the average cyclone-induced surface cooling of the upper ocean. Matthew Huber, a Purdue professor of earth and atmospheric sciences, and Ryan Sriver, a Purdue graduate student, found evidence that tropical cyclones play an important role in ocean circulation that transports heat and maintains the climate of North America and Europe. Their research suggests that tropical cyclones and hurricanes cool the tropics and act as a thermostat for the area. (Purdue graphic/Huber laboratory)

Ocean mixing affects global climate and the marine biosphere because it is linked to the ocean's ability to store and transport heat and nutrients. Observations have constrained the magnitude of upper ocean mixing associated with certain processes, but mixing rates measured directly are significantly lower than those inferred from budget analyses, suggesting that other processes may play an important role. The winds associated with tropical cyclones are known to lead to localized mixing of the upper ocean, but the hypothesis that tropical cyclones are important mixing agents on the global scale has not been tested. Here we calculate the effect of tropical cyclones on surface ocean temperatures using reanalysis data, and use these results to calculate the vertical mixing induced by tropical cyclone activity. Our results indicate that tropical cyclones are responsible for significant surface ocean cooling and vertical mixing in tropical regions. Assuming that all heat mixed downward must be balanced by poleward heat transport, the calculations suggest that approximately 15% of the total peak ocean heat transport may be associated with this mixing. Furthermore, our analyses show that the magnitude of tropical cyclone-induced vertical mixing is strongly related to sea surface temperature, indicating that future changes in tropical sea surface temperature may have significant effects on ocean circulation and ocean heat transport that are not currently accounted for in climate models.

To the News Service home page