Space Weather Prediction Gets a Supercomputing Boost – HPCwire

Solar winds are a hot topic in the HPC world right now, with supercomputer-powered research spanning from the Princeton Plasma Physics Laboratory (which used Oak Ridges Titan system) to University College London (which used resources from the DiRAC HPC facility). One of the larger efforts is a $3.2 million initiative led by the University of Alabama in Huntsville (UAH), which last September began a three-year effort to develop space weather prediction software in partnership with an array of supercomputer resources. Now, the teams efforts are bearing fruit.

Space weather is dangerous to spaceborne electronics including satellites and to infrastructure like communications networks on Earth. Space weather requires a real-time product so we can predict impacts before an event, not just afterward, said Nikolai Pogorelov, principal investigator for the project and a professor of space science at UAH, in an interview with Aaron Dubrow of the Texas Advanced Computing Center (TACC). We dont think about it, but electrical communication, GPS and everyday gadgets can be affected by extreme space weather effects.

The research is being assisted by supercomputers from the NASA Advanced Supercomputing Facility, the San Diego Supercomputer Center (SDSC) and TACC. At TACC, the researchers have been using Frontera, which delivers 23.5 Linpack petaflops and places in the top ten supercomputers on the Top500.

These resources are being levied to study backstreaming ions, which are carried away from the sun by the magnetized plasma of the solar winds. The researchers find these particles useful in predicting the time and size of coronal mass ejections (major solar weather events). These simulations, run primarily on Frontera, are recreating the phenomenon and comparing the results to observations from the Voyager 1 and 2 probes. Some non-thermal particles can be further accelerated to create solar energetic particles that are particularly important for space weather conditions on Earth and for people in space, he said.

Fifteen years ago, we didnt know that much about the interstellar medium or solar wind properties, Pogorelov said. We have so many observations available today, which allow us to validate our codes and make them much more reliable.

This research, blending intricate science, advanced computing and exciting observations, will advance our understanding of how the sun drives space weather and its effects on Earth, added Mangala Sharma, director of the Space Weather program at the National Science Foundation (NSF). The work will help scientists predict space weather events and build our nations resilience against these potential natural hazards.

For Pogorelov, this research is just the beginning of a surge in newly high-resolution research on space weather, enabled by data from instruments like the Solar Wind Electrons, Protons and Alphas (SWEAP) instrument aboard the Parker Solar Probe.

No doubt, in years to come, the quality of data from the photosphere and solar corona will be improved dramatically, both because of new data available and new, more sophisticated ways to work with data, he said. Were trying to build software in a way that if a user comes up with better boundary conditions from new science missions, it will be easier for them to integrate that information.

To learn more, read the article by TACCs Aaron Dubrow here.