论文标题
使用自洽边界条件的新型解决方案
A Novel Solution for Resonant Scattering Using Self-Consistent Boundary Conditions
论文作者
论文摘要
我们向球形几何形状中的Lyman $α$(LY $α$)的半分析溶液提出了两个新颖的添加:(1)(1)实现稳定源的正确边界条件,以及(2)脉冲源的时间相关问题解决方案。对于稳态问题,可以将解决方案表示为两个术语的总和:平均强度$ j = 0 $在表面上的方程式的分析解,以及一种新颖的半分析溶液,可以在表面上执行零设置强度的正确边界条件。将该解决方案与蒙特卡洛法的解决方案进行比较,该方法在任意光学深度时有效。结果表明,当频谱峰接近多普勒核心并随着线中心的光学深度而缓慢减小时,校正的大小是阶的统一,特别是$(aτ_0)^{ - 1/3} $,这可以解释以前研究中看到的差异。对于冲动性问题,使用本征函数膨胀来表达溶液的时间,空间和频率依赖性,以表征逃逸时间分布和光子的新光谱。结果表明,最低的特征频率与在足够大的线中心光学深度上的蒙特卡洛逃逸时间分布中发现的衰减率很好。逃生时间分布的表征突出了蒙特卡洛加速法的潜力,该方法将从分布中采样光子逃生特性,而不是计算每个光子散射,从而减少计算需求。
We present two novel additions to the semi-analytic solution of Lyman $α$ (Ly$α$) radiative transfer in spherical geometry: (1) implementation of the correct boundary condition for a steady source, and (2) solution of the time-dependent problem for an impulsive source. For the steady-state problem, the solution can be represented as a sum of two terms: a previously-known analytic solution of the equation with mean intensity $J=0$ at the surface, and a novel, semi-analytic solution which enforces the correct boundary condition of zero-ingoing intensity at the surface. This solution is compared to that of the Monte Carlo method, which is valid at arbitrary optical depth. It is shown that the size of the correction is of order unity when the spectral peaks approach the Doppler core and decreases slowly with line center optical depth, specifically as $(a τ_0)^{-1/3}$, which may explain discrepancies seen in previous studies. For the impulsive problem, the time, spatial, and frequency dependence of the solution are expressed using an eigenfunction expansion in order to characterize the escape time distribution and emergent spectra of photons. It is shown that the lowest-order eigenfrequency agrees well with the decay rate found in the Monte Carlo escape time distribution at sufficiently large line-center optical depths. The characterization of the escape-time distribution highlights the potential for a Monte Carlo acceleration method, which would sample photon escape properties from distributions rather than calculating every photon scattering, thereby reducing computational demand.