Jing Hu et al J. Stat. Mech. (2009) P02066 doi:10.1088/1742-5468/2009/02/P02066
Jing Hu1,2, Jianbo Gao1 and Xingsong Wang3
Show affiliationsMultifractal theory provides an elegant statistical characterization of many complex dynamical variations in Nature and engineering. It is conceivable that it may enrich characterization of the sun's magnetic activity and its dynamical modeling. Recently, on applying Fourier truncation to remove the 11-year cycle and carrying out multifractal detrended fluctuation analysis of the filtered sunspot time series, Movahed et al reported that sunspot data are characterized by multifractal scaling laws with the exponent for the second-order moment, h(2), being 1.12. Moreover, they think the filtered sunspot data are like a fractional Brownian motion process with anti-persistent long-range correlations characterized by the Hurst parameter H = h(2)−1 = 0.12. By designing an adaptive detrending algorithm and critically assessing the effectiveness of Fourier truncation in eliminating the 11-year cycle, we show that the values of the fractal scaling exponents obtained by Movahed et al are artifacts of the filtering algorithm that they used. Instead, sunspot data with the 11-year cycle properly filtered are characterized by a different type of multifractal with persistent long-range correlations characterized by H≈0.74.
96.60.qd Sun spots, Solar cycles
Issue 02 (February 2009)
Received 6 August 2008, accepted for publication 27 January 2009
Published 27 February 2009
Jing Hu et al J. Stat. Mech. (2009) P02066
Peter Tass et al 2010 J. Neural Eng. 7 016009
Laurent Boué et al J. Stat. Mech. (2009) P11010
Hui Zhai and Yong-Shi Wu J. Stat. Mech. (2005) P07003
Gérard Ben Arous et al J. Stat. Mech. (2008) L04003
Johannes Fuchs et al J. Stat. Mech. (2008) P04015
N Sterer et al 2009 J. Breath Res. 3 016006
Emre Erdem et al 2007 Phys. Scr. 2007 12
Delai Chen et al 2009 New J. Phys. 11 075017
Lei-Han Tang and Qing-Hu Chen J. Stat. Mech. (2008) P04003