On the Prognostic Efficiency of Topological Descriptors for Magnetograms of Active Regions

Solar flare prediction remains an important practical task of space weather. An increase in the amount and quality of observational data and the development of machine-learning methods has led to an improvement in prediction techniques. Additional information has been retrieved from the vector magnetograms; these have been recently supplemented by traditional line-of-sight (LOS) magnetograms. In this work, the problem of the comparative prognostic efficiency of features obtained on the basis of vector data and LOS magnetograms is discussed. Invariants obtained from a topological analysis of LOS magnetograms are used as complexity characteristics of magnetic patterns. Alternatively, the so-called SHARP parameters were used; they were calculated by the data analysis group of the Stanford University Laboratory on the basis of HMI/SDO vector magnetograms and are available online at the website (http://jsoc.stanford.edu/) with the solar dynamics observatory (SDO) database for the entire history of SDO observations. It has been found that the efficiency of large-flare prediction based on topological descriptors of LOS magnetograms in epignosis mode is at least s no worse than the results of prognostic schemes based on vector features. The advantages of the use of topological invariants based on LOS data are discussed.

[1]  G. Barnes,et al.  Photospheric Magnetic Field Properties of Flaring versus Flare-quiet Active Regions. IV. A Statistically Significant Sample , 2007 .

[2]  Rami Qahwaji,et al.  Automatic Short-Term Solar Flare Prediction Using Machine Learning and Sunspot Associations , 2007 .

[3]  Monica G. Bobra,et al.  PREDICTING CORONAL MASS EJECTIONS USING MACHINE LEARNING METHODS , 2016, 1603.03775.

[4]  N. G. Makarenko,et al.  Comparison of the dynamics of active regions by methods of computational topology , 2015, Geomagnetism and Aeronomy.

[5]  Herbert Edelsbrunner,et al.  Computational Topology - an Introduction , 2009 .

[6]  Monica G. Bobra,et al.  SOLAR FLARE PREDICTION USING SDO/HMI VECTOR MAGNETIC FIELD DATA WITH A MACHINE-LEARNING ALGORITHM , 2014, 1411.1405.

[7]  Daria Malkova,et al.  Methods of Computational Topology for the Analysis of Dynamics of Active Regions of the Sun , 2014 .

[8]  G. Barnes,et al.  Photospheric Magnetic Field Properties of Flaring versus Flare-quiet Active Regions. II. Discriminant Analysis , 2003 .

[9]  D. S. Bloomfield,et al.  A COMPARISON OF FLARE FORECASTING METHODS. I. RESULTS FROM THE “ALL-CLEAR” WORKSHOP , 2016, 1608.06319.

[10]  Jean Serra,et al.  Image Analysis and Mathematical Morphology , 1983 .

[11]  R. Ghrist.  Barcodes: The persistent topology of data , 2007 .

[12]  Katharine Turner,et al.  Principal component analysis of persistent homology rank functions with case studies of spatial point patterns, sphere packing and colloids , 2015, 1507.01454.

[13]  Hans von Storch,et al.  Complexity and extreme events in geosciences: an overview , 2013 .

[14]  Murray Dryer,et al.  Predicting Activity Levels for Specific Locations within Solar Active Regions , 1972 .

[15]  Gunnar E. Carlsson,et al.  Topology and data , 2009 .

[16]  L. Boucheron,et al.  PREDICTION OF SOLAR FLARE SIZE AND TIME-TO-FLARE USING SUPPORT VECTOR MACHINE REGRESSION , 2015, 1511.01941.

[17]  D. S. Bloomfield,et al.  TOWARD RELIABLE BENCHMARKING OF SOLAR FLARE FORECASTING METHODS , 2012, 1202.5995.

[18]  Yuliy Baryshnikov,et al.  Euler integration over definable functions , 2009, Proceedings of the National Academy of Sciences.

[19]  J. T. Hoeksema,et al.  The Helioseismic and Magnetic Imager (HMI) Vector Magnetic Field Pipeline: SHARPs – Space-Weather HMI Active Region Patches , 2014, 1404.1879.

[20]  N. G. Makarenko,et al.  Power law distribution in statistics of failures in operation of spacecraft onboard equipment , 2011 .