Cancelation of small-scale magnetic features a

Abstract

Aims. We investigate small-scale flux cancelations in a young active region observed with the high-resolution imaging magnetograph IMaX on the SUNRISE balloon-borne solar observatory. Methods. The observed Stokes profiles of the photospheric Fea I 5250.2 Å line were inverted using the SPINOR code to obtain the atmospheric parameters, including magnetic field vector and the line-of-sight velocity. We then identified 11 opposite-polarity canceling pairs using an automatic detection code, studied their evolution in detail, and derived their statistical properties. We classified the cancelations into two groups. Class I events are those for which cancelation happens between a pre-existing large magnetic feature of one polarity and a smaller feature of the other polarity that emerged or appeared nearby. For Class II events cancelations occur between two pre-existing, previously unconnected features that converge. Results. All studied events have an apparent cancelation time of less than ten minutes and display a significant transient linear polarization signal along the polarity inversion line. The cancelation events are characterized by a flux decay rate of about 10 15 Mx s -1 . For Class I events, the Doppler velocity of the disappearing patch gradually switches from blueshift during the initial phase of cancelation to redshift toward the end of the cancelation. For class II events, the Doppler velocity is consistently redshifted. Horizontal convergence speeds of Class II pairs fall between 0.3 and 1.22 km s -1 . The elements often do not converge directly toward each other, so that the proper motion speeds of the individual elements is higher, in the range of 1-2.7 km s -1 . Conclusions. We propose that these cancelation events result from either field-line submergence (Class I), or reconnection followed by submergence (Class II and/or Class I). Ohmic dissipation of magnetic energy could also play a role for both classes. The dynamics and evolution of these events are influenced by neighboring granular motions. We also propose that, at least for the Class II events, the granular motions could possibly be driving magnetic reconnection, rather than the supergranular motions proposed for the larger cancelation events studied earlier. Specific flux cancelation rates of the Class II events seem to indicate that they belong to somewhat different category of cancelations when compared with those studied in SOT/Hinode and MDI/SOHO data.