Volume 636, April 2020
|Number of page(s)||5|
|Section||Numerical methods and codes|
|Published online||08 April 2020|
Near optimal angular quadratures for polarised radiative transfer⋆
Astronomical Institute ASCR, v.v.i., Ondřejov, Czech Republic
2 Instituto de Astrofísica de Canarias, Vía Láctea s/n, 38205 La Laguna, Tenerife, Spain
e-mail: firstname.lastname@example.org; email@example.com
3 Departamento de Astrofísica, Universidad de La Laguna (ULL), 38206 La Laguna, Tenerife, Spain
4 Consejo Superior de Investigaciones Científicas, Spain
Accepted: 27 February 2020
In three-dimensional (3D) radiative transfer (RT) problems, the tensor product quadratures are generally not optimal in terms of the number of discrete ray directions needed for a given accuracy of the angular integration of the radiation field. In this paper, we derive a new set of angular quadrature rules that are more suitable for solving 3D RT problems with the short- and long-characteristics formal solvers. These quadratures are more suitable than the currently used ones for the numerical calculation of the radiation field tensors that are relevant in the problem of the generation and transfer of polarised radiation without assuming local thermodynamical equilibrium (non-LTE). We show that our new quadratures can save up to about 30% of computing time with respect to the Gaussian-trapezoidal product quadratures with the same accuracy.
Key words: methods: numerical / polarization / radiative transfer
The tables mentioned in Appendix A are only available at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (188.8.131.52) or via http://cdsarc.u-strasbg.fr/viz-bin/cat/J/A+A/636/A24
© ESO 2020
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.