Due to its accuracy and generality, Monte Carlo radiative transfer (MCRT) has
emerged as the prevalent method for Ly$\alpha$ radiative transfer in arbitrary
geometries. The standard MCRT encounters a significant efficiency barrier in
the high optical depth, diffusion regime. Multiple acceleration schemes have
been developed to improve the efficiency of MCRT but the noise from photon
packet discretization remains a challenge. The discrete diffusion Monte Carlo
(DDMC) scheme has been successfully applied in state-of-the-art radiation
hydrodynamics (RHD) simulations. Still, the established framework is not
optimal for resonant line transfer. Inspired by the DDMC paradigm, we present a
novel extension to resonant DDMC in which diffusion in space and frequency are
treated on equal footing. We explore the robustness of our new method and
demonstrate a level of performance that justifies incorporating the method into
existing Ly$\alpha$ codes. We present computational speedups of $\sim
10^2$-$10^6$ relative to contemporary MCRT implementations with aggressive
core-skipping. This is because the resonant DDMC runtime scales with the
spatial and frequency resolution rather than the number of scatterings - the
latter is typically $\tau_0$ for static media, or $(a
\tau_0)^2/3$ with core-skipping. We anticipate new frontiers in which
on-the-fly Ly$\alpha$ radiative transfer calculations are feasible in 3D RHD.
More generally, resonant DDMC is transferable to any computationally demanding
problem amenable to a Fokker-Planck approximation of frequency redistribution.
Description
[1709.10187] Discrete diffusion Lyman-alpha radiative transfer
%0 Generic
%1 smith2017discrete
%A Smith, Aaron
%A Tsang, Benny T. H.
%A Bromm, Volker
%A Milosavljevic, Milos
%D 2017
%K Lya algorithm radiative transfer
%T Discrete diffusion Lyman-alpha radiative transfer
%U http://arxiv.org/abs/1709.10187
%X Due to its accuracy and generality, Monte Carlo radiative transfer (MCRT) has
emerged as the prevalent method for Ly$\alpha$ radiative transfer in arbitrary
geometries. The standard MCRT encounters a significant efficiency barrier in
the high optical depth, diffusion regime. Multiple acceleration schemes have
been developed to improve the efficiency of MCRT but the noise from photon
packet discretization remains a challenge. The discrete diffusion Monte Carlo
(DDMC) scheme has been successfully applied in state-of-the-art radiation
hydrodynamics (RHD) simulations. Still, the established framework is not
optimal for resonant line transfer. Inspired by the DDMC paradigm, we present a
novel extension to resonant DDMC in which diffusion in space and frequency are
treated on equal footing. We explore the robustness of our new method and
demonstrate a level of performance that justifies incorporating the method into
existing Ly$\alpha$ codes. We present computational speedups of $\sim
10^2$-$10^6$ relative to contemporary MCRT implementations with aggressive
core-skipping. This is because the resonant DDMC runtime scales with the
spatial and frequency resolution rather than the number of scatterings - the
latter is typically $\tau_0$ for static media, or $(a
\tau_0)^2/3$ with core-skipping. We anticipate new frontiers in which
on-the-fly Ly$\alpha$ radiative transfer calculations are feasible in 3D RHD.
More generally, resonant DDMC is transferable to any computationally demanding
problem amenable to a Fokker-Planck approximation of frequency redistribution.
@misc{smith2017discrete,
abstract = {Due to its accuracy and generality, Monte Carlo radiative transfer (MCRT) has
emerged as the prevalent method for Ly$\alpha$ radiative transfer in arbitrary
geometries. The standard MCRT encounters a significant efficiency barrier in
the high optical depth, diffusion regime. Multiple acceleration schemes have
been developed to improve the efficiency of MCRT but the noise from photon
packet discretization remains a challenge. The discrete diffusion Monte Carlo
(DDMC) scheme has been successfully applied in state-of-the-art radiation
hydrodynamics (RHD) simulations. Still, the established framework is not
optimal for resonant line transfer. Inspired by the DDMC paradigm, we present a
novel extension to resonant DDMC in which diffusion in space and frequency are
treated on equal footing. We explore the robustness of our new method and
demonstrate a level of performance that justifies incorporating the method into
existing Ly$\alpha$ codes. We present computational speedups of $\sim
10^2$-$10^6$ relative to contemporary MCRT implementations with aggressive
core-skipping. This is because the resonant DDMC runtime scales with the
spatial and frequency resolution rather than the number of scatterings - the
latter is typically $\propto \tau_0$ for static media, or $\propto (a
\tau_0)^{2/3}$ with core-skipping. We anticipate new frontiers in which
on-the-fly Ly$\alpha$ radiative transfer calculations are feasible in 3D RHD.
More generally, resonant DDMC is transferable to any computationally demanding
problem amenable to a Fokker-Planck approximation of frequency redistribution.},
added-at = {2017-10-02T10:39:51.000+0200},
author = {Smith, Aaron and Tsang, Benny T. H. and Bromm, Volker and Milosavljevic, Milos},
biburl = {https://www.bibsonomy.org/bibtex/2dc7080895bc9b07b05c1cb3f1c464f29/miki},
description = {[1709.10187] Discrete diffusion Lyman-alpha radiative transfer},
interhash = {07d680302b6c7afd0b3917036379aa55},
intrahash = {dc7080895bc9b07b05c1cb3f1c464f29},
keywords = {Lya algorithm radiative transfer},
note = {cite arxiv:1709.10187Comment: 13 pages, 12 figures, MNRAS, submitted},
timestamp = {2017-10-02T10:39:51.000+0200},
title = {Discrete diffusion Lyman-alpha radiative transfer},
url = {http://arxiv.org/abs/1709.10187},
year = 2017
}