Research On The Model Decomposition Of Sunspot Area Changes In The Northern And Southern Hemispheres Of The Sun Based On EMD

It is widely believed that activities on the sun exhibit complex spatial evolutionary behavior.The solar activity in the northern and southern hemispheres is highly synchronized,forming the famous "butterfly map." However,in recent research it has been found that solar activity is very asynchronous between the northern and southern hemispheres.The north-south asynchronous of dynamic processes is an important theme for understanding the origin and evolution of solar activity regions and their various manifestations in the sun.The sunspot area change signal is a one-dimensional representation of solar activity,a nonlinear,non-stationary solar activity signal,and a carrier of solar information.For nonlinear,non-stationary sunspot area change signals,the commonly used signal decomposition methods,Fourier transform,and wavelet transform have some limitations.They are based on band-pass filtering to achieve signal decomposition.The decomposition process is constrained by the size and length of the filtering window,resulting in frequency leakage and other problems.In summary,the signal components decomposed by the commonly used signal decomposition methods cannot accurately acquire the phase and amplitude characteristics of the signal,and lose the significance of further exploring the scientific connotation of the signal.Aiming at the problems existing in the signal decomposition method based on band-pass filtering,this thesis uses the empirical mode decomposition(EMD)to realize the decomposition of the sunspot area change signal.The EMD method decomposes the result to accurately represent the phase and amplitude information of the signal.The effects of different threshold choices on the EMD results were explored.An improved method for removing endpoint artifacts during EMD is proposed.The improved EMD method is applied to decompose the decomposition of the sunspot area change signal,and more accurate results are obtained.According to the set of Intrinsic Mode Functions(IMFs)obtained by the improved EMD,the north-south hemisphere asymmetry of the sunspot area change activity is verified from the three aspects of frequency,amplitude and phase.At the same time,the EMD method is applied to study the extraction and phase relationship analysis of the Rieger-type periodic feature of the solar north and south hemisphere.By extracting the period during the solar activity period,a more accurate Rieger-type period result is obtained.Phase correlation analysis found that the closer the periodic scale of the hemispherical IMF component is,the stronger their correlation is.Overall,the southern hemisphere of the solar activity Rieger-type periodic component is ahead of the northern hemisphere for 51 days.