Multi-scaling Properties of EUV Intensity Fluctuations and Models for Impulsive Heating
There is wide agreement on the importance of impulsive processes ("nanoflares") to explain coronal heating. Diagnostics of observational data are necessary to uncover signatures of the underlying mechanisms, and, by comparing to those of simulated data, to determine whether a model explains the observations. We have investigated the multi-scaling properties that characterize the intermittency of AIA/SDO radiance fluctuations. Lags between pairs of wavebands were used to distinguish coronal from transition region (TR) signals. Noise degrades the signals, so the 171Å emission, with the highest signal-to-noise ratio, provides the best information. In an active region core, for both loops and for diffuse emission, the probability distribution functions (PDFs) of the increments of both TR and coronal signals are "quasi-Gaussian" for large temporal scales and "leptokurtic" (peaked with heavy tails) for small time increments, as expected for turbulent systems. Multifractal Detrended Fluctuation Analysis (MF-DFA) shows that the variance of the fluctuations obeys a power law as a function of temporal scales in the range 15-45 min. The value of the scaling exponent indicates that, on average, the time series are nonstationary and anti-persistent with small fluctuations following large fluctuations and vice versa. Other moments of the fluctuations obey corresponding power laws and the multi-scaling exponents quantify the degree of the intermittency in the context of multifractality. The variation in the scaling exponents results from long term correlation in the time series. The multiscaling of the EUV data agrees qualitatively with simulated intensity from a simple model of impulsive bursts plus noise, and also with the ohmic dissipation in a Reduced Magnetohydrodynamic (RMHD) model for coronal loop heating. However, the observational data were found to disagree with the modeled PDFs of increments. There is indication that the multifractal properties in the observations could be related to a modified multiplicative cascade model characterized by two parameters.