A team of astronomers and computer scientists at the University of Hawaiʻi Institute for Astronomy (IfA) have combined solar astronomy with advanced computer science to try and analyse data from the largest ground-based solar telescope in the world, atop Haleakalā, Maui.
As part of the 'SPIn4D' (Spectropolarimetric Inversion in Four Dimensions with Deep Learning) project, the study was recently published in the Astrophysical Journal.
The research put emphasis on their development of deep learning models which rapidly analyse a wide range of data received from the US National Science Foundation (NSF) Daniel K Inouye Solar Telescope. The goal behind this study was to unlock the full potential of observations of the telescope which might lead to major breakthroughs in speed, accuracy and scope of solar data analysis, read a statement from the University of Hawaiʻi.
"Large solar storms are responsible for stunning auroras, but can also pose risks to satellites, radio communications and power grids," said Kai Yang, the lead author of the study and a postdoctoral researcher at the institute.
Kai Yang noted that it is "extremely important" to have a better understanding of their birthplace, the solar atmosphere.
As part of their study, the researchers used state-of-the-art simulations to mimic what the telescope would see. Once you combine this data with machine learning, it offers an invaluable opportunity to "explore the three-dimensional solar atmosphere in near real-time,” the co-author stated.
Operated by the NSF National Solar Observatory (NSO), the Inouye Solar Telescope is the most powerful solar telescope so far. It is currently located on the 10,000-ft summit of Maui’s Haleakalā which translates to “the house of the Sun,” the statement added.
The instruments of the telescope have been specially designed to measure the magnetic field of the Sun through polarised light. The SPIn4D project was introduced to utilise this data that is solely available from the solar telescope’s instrumentation suite.
Scientists from NSO and High Altitude Observatory (HAO) use deep neural networks to estimate the physical properties of the solar photosphere through high-resolution observations from the telescope.
“Machine learning is very good at providing fast approximations to expensive computations. In this case, the model will enable astronomers to visualise the Sun’s atmosphere in real-time, rather than waiting hours to achieve the same accuracy,” said the study's co-author Peter Sadowski, who is an associate professor at the UH Mānoa information and computer sciences department.
For training their AI models, the team has come up with an extensive dataset of simulated solar observations. They have already been able to create 120 terabytes of data mimicking Inouye Solar Telescope observations at extremely high resolutions.
As part of the 'SPIn4D' (Spectropolarimetric Inversion in Four Dimensions with Deep Learning) project, the study was recently published in the Astrophysical Journal.
The research put emphasis on their development of deep learning models which rapidly analyse a wide range of data received from the US National Science Foundation (NSF) Daniel K Inouye Solar Telescope. The goal behind this study was to unlock the full potential of observations of the telescope which might lead to major breakthroughs in speed, accuracy and scope of solar data analysis, read a statement from the University of Hawaiʻi.
"Large solar storms are responsible for stunning auroras, but can also pose risks to satellites, radio communications and power grids," said Kai Yang, the lead author of the study and a postdoctoral researcher at the institute.
Kai Yang noted that it is "extremely important" to have a better understanding of their birthplace, the solar atmosphere.
As part of their study, the researchers used state-of-the-art simulations to mimic what the telescope would see. Once you combine this data with machine learning, it offers an invaluable opportunity to "explore the three-dimensional solar atmosphere in near real-time,” the co-author stated.
Operated by the NSF National Solar Observatory (NSO), the Inouye Solar Telescope is the most powerful solar telescope so far. It is currently located on the 10,000-ft summit of Maui’s Haleakalā which translates to “the house of the Sun,” the statement added.
The instruments of the telescope have been specially designed to measure the magnetic field of the Sun through polarised light. The SPIn4D project was introduced to utilise this data that is solely available from the solar telescope’s instrumentation suite.
Scientists from NSO and High Altitude Observatory (HAO) use deep neural networks to estimate the physical properties of the solar photosphere through high-resolution observations from the telescope.
“Machine learning is very good at providing fast approximations to expensive computations. In this case, the model will enable astronomers to visualise the Sun’s atmosphere in real-time, rather than waiting hours to achieve the same accuracy,” said the study's co-author Peter Sadowski, who is an associate professor at the UH Mānoa information and computer sciences department.
For training their AI models, the team has come up with an extensive dataset of simulated solar observations. They have already been able to create 120 terabytes of data mimicking Inouye Solar Telescope observations at extremely high resolutions.
(Edited by : Sudarsanan Mani)
Live TV
Loading...
Check out our in-depth Market Coverage, Business News & get real-time Stock Market Updates on CNBC-TV18. Also, Watch our channels CNBC-TV18, CNBC Awaaz and CNBC Bajar Live on-the-go!