Context. The sunspot penumbra comprises numerous thin, radially elongated filaments that are central for heat transport within the penumbra, but whose structure is still not clear. Aims. We aim to investigate the fine-scale structure of these penumbral filaments. Methods. We perform a depth-dependent inversion of spectropolarimetric data of a sunspot very close to solar disk center obtained by Solar Optical Telescope/Spectropolarimeter onboard the Hinode spacecraft. We have used a recently developed, spatially coupled 2D inversion scheme, which allows us to analyze the fine structure of individual penumbral filaments up to the diffraction limit of the telescope.Results. Filaments of different sizes in all parts of the penumbra display very similar magnetic field strengths, inclinations, and velocity patterns. The temperature structure is also similar, although the filaments in the inner penumbra have cooler tails than those in the outer penumbra. The similarities allowed us to average all these filaments and to subsequently extract the physical properties common to all of them. This average filament shows upflows associated with an upward-pointing field at its inner, umbral end (head) and along its axis, as well as downflows along the lateral edge and strong downflows in the outer end (tail) associated with a nearly vertical, strong, and downward-pointing field. The upflowing plasma is significantly, i.e., up to 800 K, hotter than the downflowing plasma. The hot, tear-shaped head of the averaged filament can be associated with a penumbral grain. The central part of the filament shows nearly horizontal fields with strengths in the range of 1 kG. The field above the filament converges, whereas a diverging trend is seen in the deepest layers near the head of the filament. The fluctuations in the physical parameters along and across the filament increase rapidly with depth. Conclusions. We put forward a unified observational picture of a sunspot penumbral filament. It is consistent with such a filament being a magneto-convective cell, in line with recent magnetohydrodynamic simulations. The uniformity of its properties over the penumbra sets constraints on penumbral models and simulations. The complex and inhomogeneous structure of the filament provides a natural explanation for a number of long-running controversies in the literature.

Context. Sunspot penumbrae show high-velocity patches along the periphery. Aims. The high-velocity downflow patches are believed to be the return channels of the Evershed flow. We aim to investigate their structure in detail using Hinode SOT/SP observations. Methods. We employ Fourier interpolation in combination with spatially coupled height dependent LTE inversions of Stokes profiles to produce high-resolution, height-dependent maps of atmospheric parameters of these downflows and investigate their properties. Results. High-speed downflows are observed over a wide range of viewing angles. They have supersonic line-of-sight velocities, some in excess of 20 km s −1 , and very high magnetic field strengths, reaching values of over 7 kG. A relation between the downflow velocities and the magnetic field strength is found, in good agreement with MHD simulations. Conclusions. The coupled inversion at high resolution allows for the accurate determination of small-scale structures. The recovered atmospheric structure indicates that regions with very high downflow velocities contain some of the strongest magnetic fields that have ever been measured on the Sun.

The Atacama Large Millimeter/submillimeter Array (ALMA) is a new powerful tool for observing the Sun at high spatial, temporal, and spectral resolution. These capabilities can address a broad range of fundamental scientific questions in solar physics. The radiation observed by ALMA originates mostly from the chromosphere -a complex and dynamic region between the photosphere and corona, which plays a crucial role in the transport of energy and matter and, ultimately, the heating of the outer layers of the solar atmosphere. Based on first solar test observations, strategies for regular solar campaigns are currently being developed. State-of-the-art numerical simulations of the solar atmosphere and modeling of instrumental effects can help constrain and optimize future observing modes for ALMA. Here we present a short technical description of ALMA and an overview of past efforts and future possibilities for solar observations at submillimeter and millimeter wavelengths. In addition, selected numerical simulations and observations at other wavelengths demonstrate ALMA's scientific potential for studying the Sun for a large range of science cases.

Recent observations show that the buildup and triggering of minifilament eruptions that drive coronal jets result from magnetic flux cancelation at the neutral line between merging majority-and minority-polarity magnetic flux patches. We investigate the magnetic setting of ten on-disk small-scale UV/EUV jets (jetlets, smaller than coronal X-ray jets but larger than chromospheric spicules) in a coronal hole by using IRIS UV images and SDO/AIA EUV images and line-of-sight magnetograms from SDO/HMI. We observe recurring jetlets at the edges of magnetic network flux lanes in the coronal hole. From magnetograms co-aligned with the IRIS and AIA images, we find, clearly visible in nine cases, that the jetlets stem from sites of flux cancelation proceeding at an average rate of ∼1.5 × 10 18 Mx hr −1 , and show brightenings at their bases reminiscent of the base brightenings in larger-scale coronal jets. We find that jetlets happen at many locations along the edges of network lanes (not limited to the base of plumes) with average lifetimes of 3 min and speeds of 70km s −1 . The average jetlet-base width (4000 km) is three to four times smaller than for coronal jets (∼18,000 km). Based on these observations of ten obvious jetlets, and our previous observations of larger-scale coronal jets in quiet regions and coronal holes, we infer that flux cancelation is an essential process in the buildup and triggering of jetlets. Our observations suggest that network jetlet eruptions might be small-scale analogs of both larger-scale coronal jets and the still-larger-scale eruptions producing CMEs.