论文标题

部分可观测时空混沌系统的无模型预测

First constraints on axion-like particles from Galactic sub-PeV gamma rays

论文作者

Eckner, Christopher, Calore, Francesca

论文摘要

储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。

Experimental refinements and technical innovations in the field of extensive air shower telescopes have enabled measurements of Galactic cosmic-ray interactions in the sub-PeV range, providing new avenues for the search for new physics and dark matter. For the first time, we exploit sub-PeV (1 TeV -- 1 PeV) observations of Galactic diffuse gamma rays by HAWC and Tibet AS$γ$ to search for an axion-like-particle (ALP) induced gamma-ray signal directly linked to the origin of the IceCube extragalactic high-energy neutrino flux. Indeed, the production of high-energy neutrinos in extragalactic sources implies the concomitant production of gamma rays at comparable energies. Within the magnetic field of the neutrino emitting sources, gamma rays may efficiently convert into ALPs, escape their host galaxy un-attenuated, propagate through intergalactic space, and reconvert into gamma rays in the magnetic field of the Milky Way. Such a scenario creates an all-sky diffuse high-energy gamma-ray signal in the sub-PeV range. Accounting for the guaranteed Galactic astrophysical gamma-ray contributions from cosmic-ray interactions with gas and radiation and from sub-threshold sources, we set competitive upper limits on the photon-ALP coupling constant $g_{aγγ}$. We find $g_{aγγ} < 2.1\times10^{-11}$ GeV$^{-1}$ for ALP masses $m_a \leq 2\times10^{-7}$ eV at a 95\% confidence level. Our results are comparable to previous limits on ALPs derived from the TeV gamma-ray domain and progressively close the mass gap towards ADMX limits. The code and data to reproduce the results of this study are available on GitHub \url{https://github.com/ceckner/subPeVALPs}.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源