Galactic and Extragalactic Cosmic Rays, Neutrinos and Gamma Rays

宇宙中能量最高的粒子信使是超高能宇宙射线（UHECR）和甚高能（VHE）中微子和伽马射线。 发现它们的起源可以说是高能天体物理学当前最重要的问题。 该项目正在四个不同的方面努力实现这一目标：1）从观测到的UHECR光谱和成分以及天体物理中微子和VHE伽马射线光谱中限制源特性; 2）从各向异性中限制UHECR源和VHE银河宇宙射线源的密度; 3）深度机器学习从UHECR观测中提取更多信息; 4）改进多信使到达方向研究。 事件重建的新概率方法将对模式识别问题产生更广泛的影响。 PI指导学生和初级研究人员，并为女高中生举办夏季研究集体，在夏季和学年期间亲自会面，细节只有随着工作的进展才能知道，但方法是明确的。 1)来自观测到的UHECR光谱和成分以及天体物理中微子和VHE伽马射线光谱的组合多信使信息有力地约束了源环境的几个属性，并且可能排除了许多候选源类别。 2)改进磁偏转的处理和考虑各向异性将更好地测试银河系瞬变假设，并解决密度的UHECR源和VHE银河宇宙射线的来源。 3)深度机器学习将从UHECR观测中提取更多信息，减少解释这些数据的困难和复杂性。 对于珍贵的10%的事件，有空气荧光数据，在这10%上训练卷积神经网络将允许提取嵌入在90%的地面信号中的当前不可访问的信息。 4)改进的多信使到达方向研究将使用一个新的最先进的银河磁场模型。 UHECR源方向将与中性中微子和伽马射线信使相结合，产生一个多信使天体物理数据集，这将大大提高相对于仅使用中微子和伽马射线的源发现能力。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The highest energy particle-messengers in the Universe are Ultrahigh Energy Cosmic Rays (UHECR) and Very High Energy (VHE) neutrinos and gamma rays. Discovering their origin is arguably the most important current problem in high energy astrophysics. This project is working towards that goal on four different fronts: 1) constraining source properties from the observed UHECR spectra and composition, and the astrophysical neutrino and VHE gamma ray spectra; 2) density of UHECR sources and source of VHE Galactic cosmic rays, from anisotropies; 3) deep machine learning to extract more information from UHECR observations; and 4) improved multi-messenger arrival direction studies. The new probabilistic methods of event reconstruction will have implications for pattern recognition problems more broadly. The PI mentors students and junior researchers, and runs a summer research collective for women high-school students that meets in person in summer and virtually during the academic year.The details will only be known as the work progresses, but the approaches are clear. 1)The combined multi-messenger information from the observed UHECR spectra and composition, and the astrophysical neutrino and VHE gamma ray spectra, powerfully constrains several properties of the source environment, and likely excludes many candidate source classes. 2)Improving the treatment of magnetic deflections and considering the anisotropies will better test the Galactic transient hypothesis and address the density of UHECR sources and the source of VHE Galactic cosmic rays. 3)Deep machine learning will extract more information from UHECR observations, reducing some of the difficulties and complexity of interpreting those data. For a precious 10% of the events there are air fluorescence data, and training Convolutional Neural Nets on this 10% will allow extraction of currently inaccessible information embedded in the ground signals of the 90%. 4)Improved multi-messenger arrival direction studies will use a new state-of-the-art Galactic magnetic field model. The UHECR source directions will be combined with the neutral neutrino and gamma ray messengers to produce a multi-messenger astrophysics dataset, which will greatly enhance the source-finding-power relative to using neutrinos and gamma-rays only.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.