Dr. Christopher Brierley, a Yale postdoctoral associate in the Geology and Geophysics department, recently published work discussing the role of tropical cyclones in global climate. Dr. Brierley works with Professor Alexey Fedorov, who studies the past and present behavior of El Niño, with the aim of predicting its response to global warming. Fedorov has been trying to understand how similar climate forcings in the Pliocene Epoch can produce considerably different temperatures from today.
The Pliocene Paradox and Permanent El Niño
Scientists have proven that the climate of the Pliocene Epoch, the period between 5.3 and 2.6 million years ago, was very similar to that of today. The most recent and closest analog to today’s climate, the weather conditions present during the Pliocene Epoch was such that warm climate was sustained for periods on the order of millions of years. A few years ago, however, Brierley’s team found evidence suggesting that the tropical Pacific was even warmer during the Pliocene than anyone had expected.
The results were found in sediment cores drilled from the ocean floor. By analyzing the chemical properties of the sediment, researchers were able to determine the ocean’s temperature in the past. Their findings showed that a huge pool of warm water covered the vast majority of the Pacific and that the temperature gradient of sea surface was smaller than previously predicted—that is, the warm pool in the central Pacific was larger and more uniform in temperature. This warm pool is similar to the effects of the periodic El Niño phenomenon, which causes warming of the Pacific near South America. However, because the Pliocene pool was sustained, it has been dubbed a “permanent El Niño.”
Surprisingly, model projections of the future do not show significant changes in the pattern of sea surface temperatures in the tropical Pacific. Many geological factors, including the location of the continents, atmospheric carbon dioxide levels, and the intensity of sunlight, were all similar to those in present day. Yet, the water off the coast of California was considerably warmer than today’s. How could two similar sets of geological inputs produce two different outcomes in temperature distribution? Researchers could not explain the discrepancy.
To resolve the discrepancy, researchers used models. The results highlighted an inconsistency between the predicted and the actual, leading to speculations about the validity of current models. Brierley explains, “They suggested some disconnect between current models of climate and observations of the past. Maybe something important was wrong with our models.”
The team then set about searching for an explanation that could explain why current models were not predicting this warm pool. Then, about two years ago, Dr. Kerry Emanuel, Professor of Meteorology at the Massachusetts Institute of Technology and renowned hurricane expert, gave a talk at Yale about why tropical cyclones contribute to ocean mixing. Emanuel, who had speculated about the role of tropical cyclones played in global warming in the past, inspired Brierley’s team to consider hurricanes as the missing link: “We were wondering how to make this warm pool happen. And when Kerry came down, we realized that tropical cyclones were the phenomenon we were looking for.”
Hurricanes and Models
Hurricanes are not included in climate models due to their small size. According to Brierley, “It’s purely a question of scale. Hurricane centers are about five miles across… to resolve that on a grid, you’d need grid point resolution of five miles. Current resolution is about seventy miles.” Though hurricane simulation is possible in the short term, computers are not powerful enough to simulate hurricanes across climate time scales, so models often ignore them: “When they simulate something for a weather forecast, they only need to run the model for ten days. We need to run it for thousands of years.”
What effects could a hurricane have on global climate? Hurricanes can alter the temperature distribution of a body of water through a process known as vertical mixing. In oceans, sunlight causes water closer to the surface to be much warmer. High surface winds can perturb such a temperature gradient by mixing the water. When surface winds increase, the interior ocean is mixed as well—the warmer water at the top is pushed downward and colder water is brought to the surface. Thus, tropical cyclones can inject heat into the inner ocean with their high wind speeds.
Of course, no one has hurricane records from the Pliocene Era. Thus, to investigate whether hurricanes could be responsible for the homogenized warm ocean surface in the Pliocene, Brierley had to simulate the hurricanes and then feedback the effects of the hurricanes into an ocean model.
The first step was to model the hurricanes of the Pliocene Era. Previously, modeling hurricanes on climate time scales was computationally too expensive. But that has been changing with improvements in both hardware and software. In the years following Hurricane Katrina, the effect of global warming on hurricane intensity drew much attention, thus spurring development in computational methodology for hurricane simulation. When asked why no one earlier had tried this, Brierley simply explains, “Most people were concerned about what cyclones would do in the future. No one had thought to apply this to the past.”
Brierley’s team reconstructed the sea surface temperatures of the Pliocene and then added a global atmosphere model on top of that. “That’s all you need to say things about hurricanes: the ocean surface and the atmosphere,” says Brierley. To perform the calculations and optimize the simulation process, Brierley’s team used a downscaling model. Atmospheric conditions were extracted from the global model and interpolated onto a fine-scale grid capable of resolving a hurricane. An initial disturbance was then implanted. The process was repeated thousands of times to create a storm distribution, as many conditions do not favor hurricane growth.
Their simulation provided the hurricane distribution of the Pliocene era. What they saw was “a lot of hurricanes everywhere.” Compared to present day, there were twice as many hurricanes, each of them “a little stronger, lasting for two or three days more.” These results were not unexpected. Brierley explains that several factors contribute to cyclone growth: water temperature and friendly variations in wind speed increase as altitude increases. Brierley clarifies that “warm water provides energy for the hurricane, and constancy in wind speed is needed, otherwise you’d shear the cyclone in half.” Lastly, the cyclone needs to form away from the equator to allow for the Coriolis effect to provide the necessary rotation. Given the massive pool of warm water during the Pliocene Era, tropical cyclones formed everywhere.
Researchers examined more than just the geographical distribution of hurricanes. They also analyzed graphs of power dissipation index (PDI), a measure related to the cube of maximum wind speed. PDI is a useful measurement because it correlates with the degree of mixing in ocean waters. Brierley’s lab found that the PDI was more uniformly distributed as well, consistent with the wider distribution of hurricanes and more homogenized warm pool. Having modeled the hurricane distribution during the Pliocene Era, they were ready to include the impact of the hurricane activity in models.
Explaining the Permanent El Niño
Brierley represented the hurricanes in the model by including additional mixing in regions with more predicted hurricanes. Such a model is more representative of the global climate state of the Pliocene. The additional mixing from the hurricanes altered the results considerably, as the new model predicted a “strong reduction of the strong temperature gradient near the equator”, i.e., the existence of the warm pool.
Brierley explains that the results can be understood by considering the effect of hurricanes on global ocean circulation. The cold water off the coast of Ecuador has its source in the cold region in the central extra-tropical Pacific. It travels at depth to reach the Equator, avoiding the major regions of cyclone activity as it does so.
This leads to a positive feedback loop that could explain the permanent El Niño. The warm pool causes increased tropical cyclone activity in the central extra-tropical Pacific. Powerful surface winds stir the ocean, pumping warmer surface water to the bottom and drawing colder water upward. The heat pumped downwards warms the ocean currents. The ocean currents well up towards the equator, leading to warmer ocean surface temperatures in the El Niño region. These warmer surface conditions alter atmospheric circulation, causing greater hurricane activity in the central Pacific. This system, caught in a positive feedback loop, contributes to the sustained El Niño.
The implications of Brierley’s work are significant to our understanding of climate change. El Niño events occur periodically every three to eight years and cause significant changes to temperature and precipitation around the globe, producing both floods and droughts. The discovery of this new positive feedback loop induced from hurricane activity means that global warming may push the climate state toward a sustained El Niño similar to that of the Pliocene Era. Brierley comments, “I think it’s fair to say I wouldn’t like to live in a world with twice as many tropical cyclones.”
Brierley notes that this is only preliminary work. “There are still a fair few uncertainties. This was more of a hypothesis,” cautions Brierley. Researchers still need to constrain the magnitude of the feedback from hurricanes and work out the specifics of how they affect circulation. “Finally,” he adds, “if this turns out to be important, then we need to include it in other models.” Application of this approach to other time periods besides the Pliocene, especially those that experienced colder climates, is another topic of exploration.
Though they have only published preliminary results, Brierley’s research has received a fair amount of attention. Indeed, in light of the recent skepticism toward climate change, these findings show that we might want to reconsider our current models. Brierley imparts, “I’m pleased that someone hasn’t taken this and run with it and interpreted it as ‘climate change is wrong.’ If there’s an error in the predictions, it’s that they’re too conservative.”
About the author: Sherwin Yu is a sophomore in Morse College. He is majoring in Molecular Biophysics and Biochemistry and Computer Science and Mathematics.
Acknowledgements: “The author would like to thank Dr. Christopher Brierley for help in writing this article.”
Sources / Additional reading
Fedorov, AV, Brierley CM, & Emanuel K (2010). Tropical cyclones and permanent El Nino in the early Pliocene epoch. Nature, 463, 1066-70.
Sriver RL (2010). Climate change: Tropical cyclones in the mix. Nature, 463, 1032-33.
Fedorov AV, Dekends PS, McCarthy M, Ravelo AC, deMenocal PB, Barreiro M, Pacanowski RC, & Philander SG (2006). The Pliocene Paradox (Mechanisms for a permanent El Nino). Science. 312(5779), 1485-89.