2023
DOI: 10.1016/j.jcp.2023.112013
|View full text |Cite
|
Sign up to set email alerts
|

Scalable matrix-free solver for 3D transfer of polarized radiation in stellar atmospheres

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
6
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
6

Relationship

3
3

Authors

Journals

citations
Cited by 9 publications
(6 citation statements)
references
References 69 publications
0
6
0
Order By: Relevance
“…The problem is thus characterized by axial symmetry, which allows us to keep computational costs down to a manageable level. It can be expected that the impact of the general AD treatment of PRD effects will be greater when the radiation field scattered by the atom has a more complex angular dependence, as happens when the complex 3D geometrical structure of the solar chromospheric plasma is taken into account (e.g., Anusha 2023; Benedusi et al 2023, and references therein).…”
Section: D Semiempirical Atmospheric Modelmentioning
confidence: 99%
“…The problem is thus characterized by axial symmetry, which allows us to keep computational costs down to a manageable level. It can be expected that the impact of the general AD treatment of PRD effects will be greater when the radiation field scattered by the atom has a more complex angular dependence, as happens when the complex 3D geometrical structure of the solar chromospheric plasma is taken into account (e.g., Anusha 2023; Benedusi et al 2023, and references therein).…”
Section: D Semiempirical Atmospheric Modelmentioning
confidence: 99%
“…Finally, this investigation has allowed us to assess the suitability of the solution strategy and implementation for dynamic scenarios. This facilitates the ongoing development of a software framework for solving the non-LTE RT problem for polarized radiation in realistic 3D models of the solar atmosphere, while taking AD PRD effects into account (Benedusi et al 2023).…”
Section: Discussionmentioning
confidence: 99%
“…( 6). More details on the algorithm for calculating the contribution of R II to the emissivity can be found in Benedusi et al (2023). In order to speed up the calculation of the emissivity, the scattering inte-gral is evaluated in the comoving reference frame (i.e.…”
Section: Methodsmentioning
confidence: 99%
“…However, stationary iterative methods have been largely superseded by Krylov subspace methods (e.g., Ipsen & Meyer 1998;Saad 2003;Meurant & Duintjer Tebbens 2020;Pearson & Pestana 2020), which also gained popularity in the radiative transfer context (e.g., Warsa et al 2003;Anusha et al 2009;Ren et al 2019;Badri et al 2018Badri et al , 2019Benedusi et al 2021). Lately, Benedusi et al (2021Benedusi et al ( , 2022Benedusi et al ( , 2023 analyzed the application of Krylov methods to linear radiative transfer problems of polarized radiation, showing that they outperform standard stationary iterative methods in terms of convergence rate, time-to-solution, and are also favorable in terms of robustness with respect to the problem size. Moreover, Krylov methods can be applied in a matrix-free context, making them highly effective solution strategies for large linear systems.…”
Section: Matrix-free Krylov Methodsmentioning
confidence: 99%
“…A correct theoretical interpretation of such signals requires solving the radiative transfer problem for polarized radiation, taking partial frequency redistribution (PRD) effects into account (e.g., Stenflo 1994;Bommier 1997aBommier ,b, 2017Belluzzi & Trujillo Bueno 2014;Casini et al 2014Casini et al , 2017a. Recently, Janett et al (2021a) pointed out that a relevant class of radiative transfer problems involving scattering polarization can be reframed, once discretized, as linear systems through a convenient physical assumption (see also Trujillo Bueno & Manso Sainz 1999;Štěpán 2008;Janett et al 2021b;Benedusi et al 2022Benedusi et al , 2023. Such linear problems can then be solved iteratively by exploiting powerful tools from numerical linear algebra, such as Krylov iterative methods.…”
Section: Introductionmentioning
confidence: 99%