The cause of spatial structure in solar He i 1083 nm multiplet images

Context. The He i 1083 nm is a powerful diagnostic for inferring properties of the upper solar chromosphere, in particular for the magnetic field. The basic formation of the line in one-dimensional models is well understood, but the influence of the complex three-dimensional structure of the chromosphere and corona has however never been investigated. This structure must play an essential role because images taken in He i 1083 nm show structures with widths down to 100 km. Aims. We aim to understand the effect of the three-dimensional temperature and density structure in the solar atmosphere on the formation of the He i 1083 nm line. Methods. We solved the non-LTE radiative transfer problem assuming statistical equilibrium for a simple nine-level helium atom that nevertheless captures all essential physics. As a model atmosphere we used a snapshot from a 3D radiation-MHD simulation computed with the Bifrost code. Ionising radiation from the corona was self-consistently taken into account. Results. The emergent intensity in the He i 1083 nm is set by the source function and the opacity in the upper chromosphere. The former is dominated by scattering of photospheric radiation and does not vary much with spatial location. The latter is determined by the photonionisation rate in the He i ground state continuum, as well as the electron density in the chromosphere. The spatial variation of the flux of ionising radiation is caused by the spatially-structured emissivity of the ionising photons from material at T ≈ 100 kK in the transition region. The hotter coronal material produces more ionising photons, but the resulting radiation field is smooth and does not lead to small-scale variation of the UV flux. The corrugation of the transition region further increases the spatial variation of the amount of UV radiation in the chromosphere. Finally we find that variations in the chromospheric electron density also cause strong variation in He i 1083 nm opacity. We compare our findings to observations using SST, IRIS and SDO/AIA data.