Deep probing of the photospheric sunspot penumbra: no evidence for magnetic field-free gaps

TL;DR

This study uses high-resolution spectropolarimetric data and advanced inversion techniques to investigate the magnetic field structure in sunspot penumbrae, finding no evidence for weak-field, field-free regions in the deepest photospheric layers.

Contribution

It provides the most reliable evidence to date that sunspot penumbrae lack magnetic field-free gaps, using Fe I lines sensitive to deeper layers and robust data correction methods.

Findings

No evidence for weak ($B<500$ Gauss) magnetic fields at $ au_5=0$ in penumbrae.

Results are consistent across different inversion parameters and data correction methods.

Supports models with continuous magnetic fields in sunspot penumbrae.

Abstract

Some models for the topology of the magnetic field in sunspot penumbrae predict the existence of field-free or dynamically weak-field regions in the deep Photosphere. To confirm or rule out the existence of weak-field regions in the deepest photospheric layers of the penumbra. The magnetic field at $\log\tau_5=0$ is investigated by means of inversions of spectropolarimetric data of two different sunspots located very close to disk center with a spatial resolution of approximately 0.4-0.45 arcsec. The data have been recorded using the GRIS instrument attached to the 1.5-meters GREGOR solar telescope at El Teide observatory. It includes three Fe I lines around 1565 nm, whose sensitivity to the magnetic field peaks at half a pressure-scale-height deeper than the sensitivity of the widely used Fe I spectral line pair at 630 nm. Prior to the inversion, the data is corrected for the effects of scattered light using a deconvolution method with several point spread functions. At $\log\tau_5=0$ we find no evidence for the existence of regions with dynamically weak ($B<500$ Gauss) magnetic fields in sunspot penumbrae. This result is much more reliable than previous investigations done with Fe I lines at 630 nm. Moreover, the result is independent of the number of nodes employed in the inversion, and also independent of the point spread function used to deconvolve the data, and does not depend on the amount of straylight (i.e. wide-angle scattered light) considered.