SCIENCE
$17 million grant supports predictive science and supernovae research
By simulating and studying supernovae, a new center at the University of Michigan aims to advance predictive science. 
The Center for Radiative Shock Hydrodynamics is funded primarily by a $17-million, five-year cooperative agreement from the National Nuclear Security Administration's (NNSA) Office of Advanced Simulation and Computing. This is one of five new Predictive Science Academic Alliance Program centers that the NNSA announced today. The grant will also fund a doctoral program for predictive science and engineering at U-M.
Predictive science focuses on the behavior of complex systems, especially systems where full scale experiments or prototypes are impossible. The center will work to quantify uncertainty in simulation results, understand the sources of those uncertainties and use that information to make better predictions. Researchers will do this by studying radiative shock waves.
In nature, radiative shock waves occur in supernovae, the most dramatic explosions in the universe. The shock waves that ripple from the demise of massive stars are so hot and fast that they emit radiation. These radiative shocks, in turn, change the structure and behavior of the exploding material, making the system difficult to simulate accurately with computers. That's why radiative shocks provide a great test case for research to improve predictive science.
Researchers at the new center will create radiative shocks in experiments at large laser facilities. They will study the shock waves, simulate them with computers and then close the gap between prediction and reality.
"When significant decisions are based upon simulations, one needs reliable, defensible assessments of simulation accuracy. This assessment of predictive capability involves quantifying the expected differences between the simulation and the actual event," said R. Paul Drake, director of the new center and a professor of Atmospheric, Oceanic and Space Sciences and of Applied Physics.
The NNSA's primary mission is to certify the safety of the U.S. nuclear weapons stockpile. The agency relies heavily on large-scale computer simulations to understand a wide range of physical phenomena, including materials behavior, fluid flow and basic nuclear properties.
"The NNSA's need for well-trained computational scientists was a big impetus to their interest in funding the kind of basic science research that our center will undertake," said James Paul Holloway, a co-principal investigator at the center and a professor in the Department of Nuclear Engineering and Radiological Sciences. "Our program will educate students in how to carry out large complex simulations and in new methods of predictive science."
The methods developed by the center will be applicable to a wide range of problems, including nuclear security, climate change and efficient manufacturing processes.
"The research at this center has the potential to contribute to solving major problems facing humanity," said David Munson, the Robert J. Vlasic Dean of Engineering. "The work there will improve scientists' abilities to understand and model the world and the universe around us."
The computer simulations the center will conduct are believed to be the largest ever developed at U-M. In the second year, the researchers expect to be able to run complex simulations utilizing 10,000 processors for as long as two weeks. That's the equivalent of using one desktop computer for about 380 years.
The center's co-principle investigators are: Kenneth Powell, an Arthur F. Thurnau Professor of Aerospace Engineering and the director of the Center for Advanced Computing; James Holloway, an Arthur F. Thurnau Professor of Nuclear Engineering and Radiological Sciences and associate dean for undergraduate education at the College of Engineering; and Quentin Stout, professor of computer science and engineering, professor of atmospheric, oceanic and space sciences, and co-director of the Center for Space Environment Modeling. The research team also includes professors from Texas A&M University
The Center for Radiative Shock Hydrodynamics is funded primarily by a $17-million, five-year cooperative agreement from the National Nuclear Security Administration's (NNSA) Office of Advanced Simulation and Computing. This is one of five new Predictive Science Academic Alliance Program centers that the NNSA announced today. The grant will also fund a doctoral program for predictive science and engineering at U-M.
Predictive science focuses on the behavior of complex systems, especially systems where full scale experiments or prototypes are impossible. The center will work to quantify uncertainty in simulation results, understand the sources of those uncertainties and use that information to make better predictions. Researchers will do this by studying radiative shock waves.
In nature, radiative shock waves occur in supernovae, the most dramatic explosions in the universe. The shock waves that ripple from the demise of massive stars are so hot and fast that they emit radiation. These radiative shocks, in turn, change the structure and behavior of the exploding material, making the system difficult to simulate accurately with computers. That's why radiative shocks provide a great test case for research to improve predictive science.
Researchers at the new center will create radiative shocks in experiments at large laser facilities. They will study the shock waves, simulate them with computers and then close the gap between prediction and reality.
"When significant decisions are based upon simulations, one needs reliable, defensible assessments of simulation accuracy. This assessment of predictive capability involves quantifying the expected differences between the simulation and the actual event," said R. Paul Drake, director of the new center and a professor of Atmospheric, Oceanic and Space Sciences and of Applied Physics.
The NNSA's primary mission is to certify the safety of the U.S. nuclear weapons stockpile. The agency relies heavily on large-scale computer simulations to understand a wide range of physical phenomena, including materials behavior, fluid flow and basic nuclear properties.
"The NNSA's need for well-trained computational scientists was a big impetus to their interest in funding the kind of basic science research that our center will undertake," said James Paul Holloway, a co-principal investigator at the center and a professor in the Department of Nuclear Engineering and Radiological Sciences. "Our program will educate students in how to carry out large complex simulations and in new methods of predictive science."
The methods developed by the center will be applicable to a wide range of problems, including nuclear security, climate change and efficient manufacturing processes.
"The research at this center has the potential to contribute to solving major problems facing humanity," said David Munson, the Robert J. Vlasic Dean of Engineering. "The work there will improve scientists' abilities to understand and model the world and the universe around us."
The computer simulations the center will conduct are believed to be the largest ever developed at U-M. In the second year, the researchers expect to be able to run complex simulations utilizing 10,000 processors for as long as two weeks. That's the equivalent of using one desktop computer for about 380 years.
The center's co-principle investigators are: Kenneth Powell, an Arthur F. Thurnau Professor of Aerospace Engineering and the director of the Center for Advanced Computing; James Holloway, an Arthur F. Thurnau Professor of Nuclear Engineering and Radiological Sciences and associate dean for undergraduate education at the College of Engineering; and Quentin Stout, professor of computer science and engineering, professor of atmospheric, oceanic and space sciences, and co-director of the Center for Space Environment Modeling. The research team also includes professors from Texas A&M University 