Automated Sunspot Classification and Tracking Using SDO/HMI Imagery
Technical Report,01 May 2014,24 Mar 2016
AIR FORCE INSTITUTE OF TECHNOLOGY WRIGHT-PATTERSON AFB OH WRIGHT-PATTERSON AFB United States
Pagination or Media Count:
Verification of an automated sunspot detection and classification algorithm is conducted utilizing two years of solar imagery from NASAs Solar Dynamics Observatory SDO satellite. Automated McIntosh classifications are compared against sunspot reports from the National Oceanic and Atmospheric Administrations NOAA Space Weather Prediction Center SWPC using a three-tiered comparison metric. Statistical confidence is demonstrated for algorithm performance and consistency when compared against the SWPC data set, suggesting future applications of the algorithm will perform similarly. A sunspot tracking algorithm is added to the existing code and demonstrates reliable feature tracking for time periods out to four days between consecutive images. Finally, an empirical Mount Wilson Magnetic Classification algorithm is generated with early testing exhibiting a direct match of 79.78 with SWPC magnetic classifications.