# A new analysis method for solar surface magnetism

Standard algorithms for tracking small scale magnetic features suffer

from ambiguities due to the intrinsic complexity of the feature

interactions which are driven by photospheric convective motions.

In his Ph.D. work, Andrei Gorobets will work on an alternative concept

for tracking the evolution of photospheric magnetic features, which

sidesteps the uncertainties of conventional direct feature tracking

methods. The new method essentially considers the magnetic flux

evolution pixel by pixel in a time-ordered sequence of magnetograms: the

flux in each pixel above a noise threshold is tracked if it has a given

minimum lifetime. The change of flux over time is then monitored and

analysed for each pixel and its neighbors. The magnetograms are treated

as if they were snapshots of the instant microstate configuration of a

“pixel gas” in a continuous medium, meaning that statistics can be used

to analyse the random "pixel flux" variations.

In the framework of kinetic gas theory it can be shown that the pixel

fluxes obey a geometrical random walk model, from which master equations

and the Fokker-Planck equations for the pixel flux probability density

can be estimated. The applicability of the fluctuation theorem can be

proven using a thermodynamical approach which can then be used to study

irreversibility measures of the pixel flux stochastic dynamics.

A comparative study is planned to test the new method for small-scale

magnetic features of the quiet Sun registered by SDO/HMI, Sunrise/IMaX,

and various MHD simulations.