Western Kentucky University
Predicting the occurrence of solar flares using spacecraft videos of the Sun
Grade Level at Time of Presentation
Senior
Major
Physics
KY House District #
83
KY Senate District #
15
Faculty Advisor/ Mentor
Dr. Paolo Massa; Dr. Gordon Emslie
Department
Department of Physics & Astronomy
Abstract
Solar flares are massive explosions that occur on the Sun’s surface and can release the same amount of energy as a billion megaton nuclear bombs. They can also trigger solar storms, which can impact the Earth and cause damage to satellites, including those used for GPS communications, and, in extreme cases, electricity blackouts. Predicting the occurrence of flares is therefore crucial for mitigating the effects of solar storms on the Earth.
We have been developing a flare forecasting machine to support the NASA Sounding Rocket Campaign, during which three rockets with newly-developed scientific instrumentation on board will be launched almost simultaneously from the NASA Research Range in Poker Flat, Alaska. The goal is to launch the rockets just before the start of a flare so that the instruments will be able to observe it in its rising phase. Since these suborbital rocket flights only observe for about 5 minutes, the launch has to be based on a reliable prediction. Our aim is to develop a precise predictive capability that can be used to alert the launch team about ten minutes in advance.
Our method is based on the analysis of videos of the Sun recorded from the NASA Solar Dynamics Observatory in different colors, each color corresponding to a different temperature in the Sun’s hot outer atmosphere, from a few million degrees to more than ten million degrees. We are developing mathematical techniques to find patterns in these videos which accurately predict the onset of a flare. Analysis of this massive amount of data (more than a thousand trillion data points) is being carried out at the Lipscomb Compute Cluster (LCC) facility at the University of Kentucky. The project is supported by a NASA EPSCoR grant, with a generous match from the Commonwealth of Kentucky.
Predicting the occurrence of solar flares using spacecraft videos of the Sun
Solar flares are massive explosions that occur on the Sun’s surface and can release the same amount of energy as a billion megaton nuclear bombs. They can also trigger solar storms, which can impact the Earth and cause damage to satellites, including those used for GPS communications, and, in extreme cases, electricity blackouts. Predicting the occurrence of flares is therefore crucial for mitigating the effects of solar storms on the Earth.
We have been developing a flare forecasting machine to support the NASA Sounding Rocket Campaign, during which three rockets with newly-developed scientific instrumentation on board will be launched almost simultaneously from the NASA Research Range in Poker Flat, Alaska. The goal is to launch the rockets just before the start of a flare so that the instruments will be able to observe it in its rising phase. Since these suborbital rocket flights only observe for about 5 minutes, the launch has to be based on a reliable prediction. Our aim is to develop a precise predictive capability that can be used to alert the launch team about ten minutes in advance.
Our method is based on the analysis of videos of the Sun recorded from the NASA Solar Dynamics Observatory in different colors, each color corresponding to a different temperature in the Sun’s hot outer atmosphere, from a few million degrees to more than ten million degrees. We are developing mathematical techniques to find patterns in these videos which accurately predict the onset of a flare. Analysis of this massive amount of data (more than a thousand trillion data points) is being carried out at the Lipscomb Compute Cluster (LCC) facility at the University of Kentucky. The project is supported by a NASA EPSCoR grant, with a generous match from the Commonwealth of Kentucky.