April 3, 2013 | Can something as gentle as sea spray influence a hurricane’s might? That’s an important question for atmospheric scientists as they work toward improving forecasts of hurricanes and other powerful storms over the water.
But it’s no easy matter to figure out how millions of saltwater droplets moving chaotically between an enormous expanse of ocean and a rotating storm that reaches to the stratosphere are affecting the course of nature.
At a special March 14 seminar, David Richter laid out the challenges of this type of research. A postdoctoral fellow at NCAR, Richter is creating simulations of the droplets by drawing on the vast power of the new Yellowstone supercomputer—the pioneering petaflop system that promises to usher in a new era of discovery in the geosciences. He believes that such simulations will shed light on the influence of spray on transferring additional heat and moisture from the ocean surface to the atmosphere, thereby increasing the strength of tropical cyclones.
The research can help researchers learn about on other problems related to the movement of particles, such as landslides of sand beneath the ocean that can rupture underwater fiber optic cables or damage offshore drilling platforms.
The March 14 seminar illustrated the diversity of science that is already being performed on Yellowstone by scientists at NCAR and across the university community. Supported by the National Science Foundation and housed at the new NCAR-Wyoming Supercomputing Center in Cheyenne, Yellowstone began operations at the end of September 2012.
The seminar featured 3 of the 11 projects whose computing was carried out at the NWSC through the Accelerated Scientific Discovery initiative. ASD allowed a select group of computing-intensive projects to be carried out during the first several months of Yellowstone’s operations. Scientists are now beginning to analyze those results. (See our roundup of ASD projects.)
While Richter is focusing on the motion of tiny droplets that can be measured in millimeters, NCAR scientist Justin Small is looking at atmospheric processes that span tens to hundreds of kilometers. He is using Yellowstone to run the powerful Community Earth System Model (CESM), simulating atmospheric events in more detail than is typical for global climate studies.
Thanks to the computational power of Yellowstone, Small was able to simulate the world’s climate at a resolution as fine as roughly 15 miles (25 kilometers) between points on a three-dimensional grid. By depicting global climate at this level of detail, scientists can begin to capture smaller-scale atmospheric events such as tropical cyclones, strong midlatitude fronts, and intense storm systems. These storms, moreover, interact with a more realistic ocean environment. Since ocean conditions influence storm intensity, this is an important feature that is not possible with coarser-scale simulations.
Small demonstrated a new CESM visualization of atmospheric processes, including fronts and tropical cyclones, moving across the globe—a possible preview of the next generation of climate simulations that will begin to look more like weather models.
NCAR’s Gabriele Pfister is drawing on Yellowstone’s power to answer a very different question: what will U.S. air quality be like in mid-century, based on current projections in pollution emissions and climate?
Some of Pfister’s early computer runs indicate the extent to which warming temperatures are likely to worsen incidents of ground-level ozone and other types of pollutants. On the other hand, her initial runs also indicate that the nation may be able to reduce the number of times that air pollution levels exceed public health standards if progress continues in reducing some emissions.
This research also has climate implications. Some of the initial computer runs indicate that local pollutants such as certain gases or aerosols (tiny airborne particles or droplets) can nudge temperatures up or down by about 2-3 degrees Fahrenheit over some regions.
Pfister noted that her research, which involved creating realistic simulations of atmospheric chemistry that interacted with a detailed climate model, would have taken far too long with Bluefire. “I would have been planning to analyze the data by the time I retired,” she said.
Other ASD projects seek new understandings of earthquakes, solar storms, and weather forecasting. While six of the projects were led by NCAR principal investigators, the other five were led by researchers at the University of Colorado Boulder, University of Southern California, Cornell University, University of Delaware, and the Center for Ocean-Land-Atmosphere Studies. Researchers hope to begin formally writing up their results and submitting them for peer review as soon as the end of this year.
NCAR-Wyoming Supercomputing Center opens (October 15, 2012)