论文标题
质量的流量:大量中微子的多流体非线性扰动理论
Flows For The Masses: A multi-fluid non-linear perturbation theory for massive neutrinos
论文作者
论文摘要
大规模中微子的速度分散对非线性宇宙学扰动理论提出了艰巨的挑战。我们将中微子种群视为非线性流体的集合,每种液体具有均匀的初始动量,这是通过时间重归于扰动理论的扩展。我们采用了最近开发的快速傅立叶变换技术,我们通过超过两个数量级加速了非线性扰动理论,使其足够快地用于实际使用。 After verifying that the neutrino mode-coupling integrals and power spectra converge, we show that our perturbation theory agrees with N-body neutrino simulations to within 10% for neutrino fractions $Ω_{ν,0} h^2 \leq 0.005$ up to wave numbers of k = 1 h/Mpc, an accuracy consistent with 2.5% errors in the neutrino mass determination. Non-linear growth represents a >10% correction to the neutrino power spectrum even for density fractions as low as $Ω_{ν,0} h^2 = 0.001$, demonstrating the limits of linear theory for accurate neutrino power spectrum predictions.我们的代码FlowsForThemasses可以在github.com/upadhye/flowsforthemasses上在线上可用。
Velocity dispersion of the massive neutrinos presents a daunting challenge for non-linear cosmological perturbation theory. We consider the neutrino population as a collection of non-linear fluids, each with uniform initial momentum, through an extension of the Time Renormalization Group perturbation theory. Employing recently-developed Fast Fourier Transform techniques, we accelerate our non-linear perturbation theory by more than two orders of magnitude, making it quick enough for practical use. After verifying that the neutrino mode-coupling integrals and power spectra converge, we show that our perturbation theory agrees with N-body neutrino simulations to within 10% for neutrino fractions $Ω_{ν,0} h^2 \leq 0.005$ up to wave numbers of k = 1 h/Mpc, an accuracy consistent with 2.5% errors in the neutrino mass determination. Non-linear growth represents a >10% correction to the neutrino power spectrum even for density fractions as low as $Ω_{ν,0} h^2 = 0.001$, demonstrating the limits of linear theory for accurate neutrino power spectrum predictions. Our code FlowsForTheMasses is avaliable online at github.com/upadhye/FlowsForTheMasses .