Kraken will be officially retired and no longer accessible on August 27, 2014. For more information see Kraken Decommission FAQs.
Kraken will be officially retired and no longer accessible on August 27, 2014. For more information see Kraken Decommission FAQs.
The National Institute for Computational Sciences

The Complications of Climate Change

Heat waves, pollution, diseases to mushroom under current forecasts

by Gregory Scott Jones

As climate change has largely dominated the scientific conversation in the media over the past couple of decades, certain phenomena have come to represent the larger, macro-effects of global warming. Melting glaciers and rising sea levels in particular have disproportionately embodied the collection of numerous abnormalities that make up what we have come to know as climate change.

And while rising sea levels are certainly something to be concerned about, there are plenty of other factors that pose equally, if not greater, threats to our wellbeing in a warming world. The heat brings the heat, literally.

One of the greatest, and perhaps most obvious, threats posed by climate change are longer and more severe periods of extreme heat. While longer summers may not sound so bad to some, they carry serious consequences—hotter temperatures lead to greater air pollution transport and better conditions for vector-borne diseases to thrive. In other words, as temperatures rise, not only do glaciers melt, but air pollution travels farther and becomes more intense in certain areas, and conditions grow riper for maladies such as malaria and West Nile virus.

Thanks to supercomputers such as Oak Ridge National Laboratory’s Jaguar and the University of Tennessee’s (UT’s) Kraken, the connections and outcomes of these related yet separate phenomena are becoming easier to predict.

And thanks to the team of the University of Tennessee's Joshua Fu, policymakers will soon have a better picture of what climate change means to human health. Coupling climate change, heat waves, and the effects of those phenomena on vector-borne diseases is no easy task—the variables are many, the relationships subtle and complex. For this reason Jaguar and Kraken, two of the world’s fastest supercomputers, are extremely necessary. Fu is using their raw computing power in concert to predict the effects of numerous phenomena at different regional scales, down to 4 kilometers in some cases.

The findings are beginning to trickle in. One of the more important ones: the frequency, intensity, and duration of heat waves all increase in both the moderate and worst emission scenarios proposed by the United Nations’ Intergovernmental Panel on Climate Change (IPCC), the two similar scenarios the team is investigating.

Essentially, said Fu, this complex project has several aims: to the determine the current and future population exposures to ambient air pollution; to model the frequency and intensity of heat waves; and to model the influence of these heat waves on prototype disease vectors in the eastern United States, particularly Lyme disease. In addition, the project seeks to develop exposure-response relationships for each phenomenon and their primary health outcomes and map these outcomes in terms of general populations as well as susceptible subpopulations for use by community groups and local governments.

It all starts with the heat waves. It is this phenomenon that ultimately affects the pollution and disease vectors being investigated by Fu’s team. The escalating heat and sunlight can “cook” the compounds emitted into the air, i.e., pollution, and combine it with naturally occurring nitrogen oxide. The result is the smog so common in urban areas, made up of primarily of the ground-level ozone that is harmful to the human respiratory system.

The above map shows the potential for extended heat waves in the Eastern United States according to IPCC emissions scenario 8.5.

Specifically, when it comes to modeling the air pollution, Fu’s team is targeting ozone and particulate matter less than 2.5 micrometers (PM2.5), the size at which particulates such as soot are too small to be filtered by cilia in the nose and can enter the respiratory system. In other words, the stuff that is directly and seriously detrimental to human health.

Heat waves provide the perfect atmosphere for insects, which carry a host of diseases, to thrive. Freezes do a great deal to kill off insect populations, thus keeping them under control. Understandably, heat waves do just the opposite; they provide the perfect catalyst for populations to blossom. This also means that the diseases they carry, such as Lyme disease, likewise flourish, creating yet one more public health hazard as things heat up across the globe.

Ultimately, the data from Fu’s simulations, which has been published in the Earth System Grid, a Department of Energy-funded data portal, will produce physical and temporal maps of various resolutions so that policymakers can visualize the potential impact on human populations and the projected changes in the relationships and intensities of the phenomena.

“The great thing about this research is its practical,” said Fu. So practical, in fact, that funding is coming from five different federal agencies: the EPA, CDC, DOE, NASA, and the NSF are all chipping in to get their hands on the invaluable data. In fact, Fu’s research will definitely contribute to the IPCC’s landmark report on climate change known as Assessment Report 5, to be published in 2014.

Resolution Revolution

Fu’s team has a valuable starting point in this daunting quest: the Community Earth Systems Model (CESM), a climate mega-model that couples components of atmosphere, land, ocean, and ice to reflect their complex interactions and serves as a leading tool for predicting future climate scenarios.

The climate component of Fu’s research is especially important because the air pollution and disease vectors are directly dependent on exactly how hot things get in the future. The team started things rolling on Jaguar, a Cray XK6 managed by the Department of Energy and capable of a peak performance of 3.3 petaflops, making it among the most powerful computers in the world.

For starters, Fu’s team used the CESM to simulate the climate from 2001-2100 for the entire Eastern United States on a 100 kilometer by 100 kilometer grid. Next, the team used the data from the climate simulations and implemented chemistry, giving a fairly detailed scenario of potential ambient air pollution according to the two IPCC emission scenarios. Despite this enormous initial effort, however, they were just getting started.

Enter the Kraken, a Cray XT5 funded by the National Science Foundation but managed by UT’s National Institute for Computational Sciences capable of nearly 1.2 petaflops. Once the team wrapped up its CESM runs and gathered the data, they downscaled to a regional climate model known as the Weather Research and Forecasting model and a regional chemistry model known as the Community Multiscale Air Quality modeling system. What started as a 100 kilometer by 100 kilometer grid on Jaguar evolved into a 12 kilometer by 12 kilometer grid for the entire continental U.S. and a 4 kilometer by 4 kilometer grid for the eastern U.S., among the finest spatial resolutions in such a large simulation domain ever achieved in climate research. Furthermore, the temporal nature of the maps is unprecedented as well, down to three hours in some cases. Many climate models simply provide annual results—the finer timescale will give researchers and policymakers perhaps the clearest picture yet of the various evolutions of extended future heat waves and their consequences.

“This is the first time that high-resolution downscaling applies to such a large domain, which will widely benefit the climate community,” said Fu . . . “This dynamical downscaling takes advantage of local detailed topography and land use information, and significantly improves the prediction of local extreme climate events such as heat waves and extreme precipitation. These simulations are very useful for countries and agencies to plan adaption strategies. Governments can use this data to prepare for scenarios now.”

While the team is constantly revising and perfecting the individual simulations and the project as a whole, the data is already paying off. The team published numerous papers in 2011 and many more are in the pipeline. And besides the team’s involvement in the upcoming AR 5 report, its research was also heavily featured in the climate change transport area of the U.N.’s Economic Commission for Europe final report.

There is little doubt that the importance of the team’s work will continue to reverberate throughout numerous scientific communities, a testament to the importance of world-leading supercomputers such as Kraken and Jaguar in solving humanity’s most pressing problems.

"The large simulations could not have taken place without the computing power available at NICS and ORNL,” said Fu. “These systems allowed us to explore the science problems raised in the previous IPCC report (AR4) in the finer resolution, providing insights for policymakers to develop adaptation and mitigation strategies.”

About the research: This research was supported in part by the National Science Foundation through TeraGrid resources provided by National Institute for Computational Sciences (NICS) under grant number [TG-ATM110009 and UT-TENN0006]. This research also used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

About NICS: The National Institute for Computational Sciences (NICS) is a joint effort of the University of Tennessee and Oak Ridge National Laboratory that is funded by the National Science Foundation (NSF). Located on the campus of Oak Ridge National Laboratory, NICS is a major partner in NSF’s Extreme Science and Engineering Discovery Environment (XSEDE).