Automatic detection of chromospheric swirls based on a morphological approach
I. Dakanalis, G. Tsiropoula, K. Tziotziou, K. Koutroumbas
High-resolution observations from modern ground-based and space-borne telescopes have revealed the existence of ubiquitous photospheric vortical motions that occur in both active and quiet Sun regions. These motions twist the overlying magnetic field lines creating vortex-like structures that extend throughout the solar atmospheric layers and up to the low corona. In observations of lines, such as the Hα and Ca II lines, that map different heights of the chromosphere, they manifest themselves as circular and spiral-shaped dark structures called chromospheric swirls. They maintain their vortical motion for considerable time and are believed to actively contribute to the energy transfer from the sub-photospheric layers to the transition region and corona. Consequently, accurate statistical information is very important in order to acquire understanding of the nature and physical properties of chromospheric swirls and to estimate the relevant upwards energy transfer. To this end, visual inspection methods applied in observations obtained in the chromospheric Hα and Ca II lines have provided some preliminary results towards this direction. However, obtaining information from a wide statistical sample requires the application of an automated detection technique. Such techniques currently depend on the derivation of the velocity fields of such vortical motions through Local Correlation Tracking (LCT) and mainly focus on lines mapping the photospheric layer. However, the effectiveness of LCT is ambiguous in the case of chromospheric lines due to the complexity of both relevant line formation and the chromospheric layer itself. To overcome this drawback, we developed an automated method of chromospheric swirl detection based on the morphological characteristics of these structures that nicely complement previous LCT-related approaches. The designed algorithm was tested on observations obtained in the line center of the Hα line with the CRisp Imaging SpectroPolarimeter (CRISP) mounted on the Swedish 1-m Solar Telescope (SST). A brief description of the entire procedure will be presented along with representative results of this method to the selected dataset.