Diagnostics of comprehensive simulations of the chromosphere
by Patrick Alexander Ondratschek
Date of Examination:2025-03-13
Date of issue:2025-11-20
Advisor:Prof. Dr. Sami K. Solanki
Referee:Prof. Dr. Laurent Gizon
Referee:Prof. Dr. Sami K. Solanki
Persistent Address:
http://dx.doi.org/10.53846/goediss-11635
Files in this item
Name:eDiss_ondratschek_phd_thesis.pdf
Size:28.6Mb
Format:PDF
The following license files are associated with this item:
Abstract
English
The solar chromosphere is a highly dynamic and complex region of the solar atmosphere. It plays an important role in the formation of many of the observed solar features, such as, prominences, spicules, and flares. In addition, the chromosphere provides the mass for the solar wind. The chromosphere can be seen in many aspects as a region of transitions. The plasma, which is mostly neutral in the photosphere, becomes partly ionized in the chromosphere. The temperature starts to increase to values that are higher than in the photosphere. The gas density decreases approximately exponentially. Filling almost the entire space in the chromosphere, the magnetic field becomes dynamically important. Many approximations that are valid in the photosphere such as local thermodynamic equilibrium break down in the chromosphere. The change in physical conditions within a relatively small height difference (typically 1–3 Mm) leads to a highly corrugated region. Interpreting observations of the chromosphere is thus challenging. Therefore, the chromosphere is probably the least understood part of the solar atmosphere. Nowadays the chromosphere can be observed at high spatial and spectral resolution either from the ground with for example the Swedish-Solar-Telescope, from space for example with the Interface-Region-Imaging-Spectrograph, or from stratospheric balloons with for example the SUNRISE observatory. Interpreting the observations is difficult due to the complex nonequilibrium (NE) and nonlocal formation of chromospheric spectral lines. Numerical simulations of the solar atmosphere can be a valuable tool to tackle this difficulty. Within the last two decades, it became possible to include a comprehensive set of physics in such models to treat the chromosphere with an unprecedented degree of realism. Until recently, Bifrost was the only simulation code that was capable of simulating the chromosphere. While present models of the chromosphere can produce features that are very similar in structure and appearance to those seen in observations, there are still discrepancies. For example, the line widths of simulated chromospheric spectral lines are typically too narrow and their intensities are too faint compared with observations. In this work, a simulation of the chromosphere is used that was computed with the “Max Planck Institute for Solar System Research/University of Chicago Radiation Magneto-hydrodynamics (MURaM)” code. MURaM is an radiaton-MHD (rMHD) code and was recently upgraded to treat the physics of the chromosphere under non-LTE (NLTE) and NE conditions. The new version is named chromospheric extension of MURaM (MURaM-ChE). An important step in validating a numerical model of the solar chromosphere is the forward modeling of spectral lines. To this end, important chromo- spheric spectral lines from the simulation have been synthesized and were compared with observations. In the first project, a simulation of an enhanced network (EN) region was used. Under the approximation of 1.5D radiative transfer (RT), the Mg ii h&k lines from this simulation have been synthesized. It was found that the line width is significantly increased compared to previous models. This is attributed to the magnitude of velocity variations along the line-of-sight (LOS), which is larger in the MURaM-ChE simulation compared to a similar simulation computed with the previously used Bifrost code. The spatially averaged spectra from the MURaM-ChE simulation compare well in terms of line width to the observations, but the peak intensities of the Mg ii h&k lines appear a bit too strong. The higher peak intensity in the simulation is partly due to a higher magnetic flux density in the simulation than in the observation but also due to the 1.5D RT approximation, which is known to overestimate the peak intensities. In the second project, the effects of full 3D RT on the Mg ii h&k spectra have been studied as well as correlations between spectral line properties and the underlying atmosphere in the same simulation as in the first project. 3D RT is computationally more expensive than 1.5D RT. The synthesis of a spectral line in a single snapshot from the simulation can be as expensive as the rMHD simulation itself. The computations presented in this project confirmed, however, that 3D RT effects play an important role in the synthesis of the Mg ii h&k lines and must be taken into account. In particular, the differences between 3D and 1.5D RT are larger in MURaM-ChE than in the Bifrost model. Additionally, it was found that correlations between spectral line properties and the underlying atmosphere are valid, which were found by similar studies with the Bifrost model. The scatter in these correlations is, however, larger in the more dynamic MURaM-ChE simulation. A general finding of the studies on the Mg ii h&k lines is that in the MURaM-ChE model, a reasonably good match of the line profiles with the observations can be achieved at a relatively moderate spatial resolution of the simulation of 23.46 km (horizontal) and 20 km (vertical). In the third project, the Ca ii 𝜆854.2 nm spectral line has been synthesized. This line forms in the lower to middle chromosphere and is widely used for magnetic field measurements in the chromosphere. The line itself has an asymmetric shape whose origin is under debate in the literature. On the one hand, the dynamic motions of the chromosphere might lead to the asymmetry, on the other hand, it could be due to isotopic splitting as there are six stable isotopes of calcium in the solar atmosphere. The width of this line could not be reproduced in previous models, which typically resulted in too-narrow line profiles. It was found that both the velocities and the isotopic splitting contribute significantly to the observed asymmetric line shape. The spatially averaged line profile of Ca ii 𝜆854.2 nm computed from the MURaM-ChE simulation shows a relatively close match with an observed line profile both in line width and asymmetry when the effect of isotopic splitting is taken into account. The close match of the line width is, similar to the findings for the Mg ii h&k lines, a result of the dynamic motions in the MURaM-ChE simulation. This work contributes to the scientific literature in that it presents the first forward modeled spectra of the new chromospheric extension of MURaM. The improved match with observations helps to interpret complex chromospheric spectral lines. Additionally, the here presented findings demonstrate the need to properly model the large velocities in the solar chromosphere, to account for horizontal RT effects in strong lines with considerable scattering, as well as to take the isotopic splitting into account for elements for which multiple isotopes are present in the solar atmosphere.
Keywords: Solar atmosphere: (QB528), Solar chromosphere: (QB528), Radiative transfer (175.25.R3), Simulation methods (541.15.S5), MURaM-ChE code
