AI Modeling Magnetic Solar Patches (IMAGE)
Caption
SwRI scientists integrated three types of machine learning models to generate images of solar magnetic patches with physically realistic properties and used those as a query to find matching patches in real observations. These artificial intelligence techniques allow scientists to tease out hidden magnetic data from real data (left panel). They work by structuring (ordering) data in a way that allows scientists to change the properties of objects by moving a slider along pre-defined directions that correspond to certain physical parameters. This illustration shows how scientists can explore the effects of increasing the magnetic field (blue bar) to the point that they develop the complexity (red slider) to potentially drive space weather events (right image).
Credit
Southwest Research Institute
Usage Restrictions
Images/Videos may be used by the public for educational and informational purposes only.
License
Original content