Abstracts Details

ABSTRACT DETAILS PAGE

 
First Name * : Donald
Last Name * : Schmit
Affiliation * : Catholic University
Abstract Type * : Poster
Title * : Labeling Known Unknowns in the Chromosphere Through Forward Modeling
Author(s) * : Donald Schmit, Juan Martinez-Sykora, Tiago Pereira
Abstract Session * : Chromospheric dynamics
Abstract * : In this work, we study chromospheric line formation using a forward modeling approach. We use the Bifrost stellar atmosphere simulation to generate 340000 profiles pairs in two related transitions: Ca II 8542A and Mg II 2803A. This synthetic dataset is a powerful tool because we know both the properties of the emitting plasma (the unknown in remote sensing) and the emergent profile (the observable). We use this dataset to address two questions: 1) Which kinds of profiles occur most commonly in the simulation? 2) How much information overlap exists between the 8542A and 2803A profiles? Our research into the first question suggests that a substantial degree of degeneracy exists for chromospheric profiles: a wide range of atmospheres can generate similar profiles in a 2D or 3D hydrodynamic system. If a single profile does not provide a satisfactorily narrow distribution of atmospheric properties then a logical progression is to use more than one line. We use Bayes theorem to address to what degree the more broadly-affected 2803A spectral profile can predict the spectral profile of the 8542A line. We find that a Bayesian forecasting method based on 2803A profiles only reproduces about 30% of intensity variations in the 8542A line. The forecast has no predictive power for 8542A line core Doppler shift. This implies that multiple chromospheric lines, even if they