Morphological variations of solar granules in the presence of magnetic fields

TL;DR

This study quantifies how solar granules morphologically respond to magnetic fields using high-resolution SST observations of NOAA 11768. By segmenting granules and extracting area, eccentricity, and orientation, and by inverting CRISP data with VFISV to obtain magnetic-field vectors and velocities, the authors show that granule area and brightness decrease with stronger magnetic fields, indicating suppressed convective transport. They introduce two custom eccentricity measures and analyze granule–field alignment, revealing that horizontal fields induce elongation and azimuthal alignment of granules, particularly in flux-emergence regions. The work advances understanding of magnetoconvection at the photosphere and suggests directions for future Lagrangian analyses of granule lifetimes and horizontal motions.

Abstract

Solar granulation consists of dynamic convective plasma cells that rise from the solar interior to the surface. The interaction between these plasma cells and the Sun's magnetic field provides valuable insights into plasma dynamics near the solar surface and how they evolve in the presence of magnetic fields. This study analyses the morphological characteristics of solar convective cells, investigating the relationship between magnetic field properties and granule dynamics - specifically how granule area, shape, and brightness vary under different magnetic field conditions. Observations of the active region NOAA 11768 were taken with the Swedish 1-m Solar Telescope (SST). A segmentation algorithm was applied to continuum intensity images to identify individual granules and determine their sizes, shapes, and mean brightness. The magnetic field vector and line-of-sight velocity were derived from CRISP spectropolarimetric data to investigate their role in shaping granule properties. We find that granular area decreases systematically with increasing magnetic field strength, with the largest granules occurring in non-magnetic regions and a mean granule area of approximately 1.58 arcsec$^2$ (effective diameter of 1.42 arcseconds). Both mean continuum intensity and granule size decrease with stronger fields, confirming the suppression of convective energy transport in magnetised regions. No correlation was found between mean granule brightness and mean up-flow velocity. Highly elongated granules appear in both magnetic and non-magnetic regions, while near-circular granules are exclusive to non-magnetic areas. An alignment between granule major axes and magnetic field azimuth is observed where the horizontal field component is strong, confirming that granules are highly sensitive to magnetic fields, which inhibit the lateral expansion of convective cells.

Figures (9)

• Figure 1: Left: Full-disk image context from Helioseismic and Magnetic Imager (HMI) on board of Solar Dynamics Observatory (SDO). The red rectangle shows the region of interest observed by the 1-m SST. Right: Four images grid displaying maps obtained from the SST data; a) blue-continuum intensity image, b) strength of the magnetic field, c) magnetic field inclination and d) magnetic field azimuth.
• Figure 2: Segmentation process of solar photospheric granules. a) Original solar surface region. b) Initial granule segmentation using local minima and watershed technique. c) Refined contour selection after parameter optimization. d) Final granule morphology after contour-based erosion.
• Figure 3: Granular area distributions. Top panel: raw distribution of granular areas including outliers, with a fixed bin width of $\sim0.15$ arcsec$^2$. Bottom panel: Histogram $\log(PDF)$ vs Area, where $PDF$ is the normalised probability density of the areas distribution. The dashed red line represents in each plot the fitting of the distribution with the found para-meters.
• Figure 4: 2D--$\log$(KDE) plots showing the distribution of the areas and three eccentricity metrics: (a) raw eccentricity (ellipse fitting), (b) convex eccentricity (computed on convex hulls), and (c) custom eccentricity (based on extreme points through the centre of mass of the segments). Each plot highlights the most probable eccentricity-area combinations and reveals the trends.
• Figure 5: 2D--$\log$(KDE) plots depicting the dependence of: (a) granular area versus granule eccentricity, (b) granular area versus granule continuum intensity, (c) granular area versus granule LOS velocities, and (d) granular continuum intensity versus granules LOS velocities. Bottom row shows the dependence of granule (e) area, (f) continuum intensity, (g) LOS velocity, and (h) eccentricity on magnetic field strength. Colour-coding follows the same scheme as in Fig. \ref{['kde_eccs']}.