The Dependence of Large Flare Occurrence on the Magnetic Structure of Sunspots

We have studied 8 yr of active region observations from the United States Air Force/Mount Wilson data set, supplied by the NOAA World Data Center, to confirm the relation between δ spots and large flares. We found that after correcting some errors we were able to describe relationships among active region size, peak flare soft X-ray (SXR) flux (measured by GOES 1-8 Å flux), and magnetic classification. We found the Solar Optical Observing Network magnetic classification to be reasonably accurate but its area measures to be inaccurate for many of the regions. This is due partly to transcription errors and partly to wrong correction for limb foreshortening. Errors could, however, be repaired by intercomparison of multiple observations. We confirm Künzel's original idea that regions classified βγδ produce many more large flares than other regions of comparable size. Almost all substantial flares occurred in regions classified βγδ by the Air Force sites. Each region larger than 1000 μh and classified βγδ had nearly 40% probability of producing flares classified X1 or greater. Yet only a half-dozen of those, showing the "island delta" configuration, produced great activity. There is a general trend for large regions to produce large flares, but it is less significant than the dependence on magnetic class.