New astronomy in the age of Big Data

大数据时代的新天文学

• 批准号: RGPIN-2018-05750
• 负责人: Hlozek, Renee
• 金额: $ 1.75万
• 依托单位: University of Toronto
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=713309
• 关键词: New; astronomy; age; Big; Data

What is the mysterious dark energy that drives the apparent acceleration of the universe? What could the similarly named but distinct dark matter in the universe be, and how does it change the galaxies, clusters of galaxies and overall structure in the universe around us? What is the nature of the bright-but-fleeting radio bursts that go off in the sky, and how can we separate those signals from local man-made radio signals that are not astrophysically interesting? These seemingly unconnected topics are linked by their requirement for big (cosmological) data. My work uses surprising solutions and techniques applied to these data to answer these questions about the cosmos. Our standard cosmological model of the universe is a simple one. However, the dark energy that is believed to drive cosmic acceleration remains a mystery. One way to probe dark energy is through observations of Type Ia Supernovae (SNe Ia), brilliant stars that explode with a roughly standard brightness throughout the universe, allowing them to act as cosmic beacons. We use SNe Ia to measure distances, revealing the universe's accelerating expansion, which occurs due to the negative pressure of the mysterious dark energy component. Using LSST, we can probe dark energy properties over cosmic time. The spectacular amount of data that LSST will generate will also contain other interlopers: bright objects that are not useful cosmologically. I am pioneering critical techniques to classify and separate these objects, and then to use the statistical (probabilistic) confidence that the object is a useful SNe Ia, to weight our cosmological constraints. A related problem is the riddle of what the dark matter could be. I theoretically model the axion, one promising dark matter candidate, physically motivated by string theory. Axions can actually mimic dark matter or dark energy, depending on their mass (expressed in units of electron volt or eV). Axions`wash out' structure in the universe by suppressing structure formation on small scales. As measurements of the CMB become more precise on small scales, our axion constraints increase by almost ten-fold. My team works to constrain axions and in so doing, constrain the 'dark sector' of the universe. In addition to the key cosmological problems of dark matter and dark energy, I am tackling the critical challenge of understanding the physics behind the new mysterious phenomenon of "fast radio bursts". We don't know if they are merging neutron stars, or something exotic like cosmic superradiance. Classifying them into different statistical groups is critical to uncover their underlying properties and nature. By applying novel classification techniques approaches directly to the steady stream of ARO data, we will build cutting-edge statistical classifiers for next-generation instruments. My work brings together theory, data and statistical tools to exploit the new data-driven cosmological epoch.

驱动宇宙明显加速的神秘暗能量是什么？ 宇宙中命名相似但截然不同的暗物质可能是什么，它如何改变我们周围宇宙中的星系、星系团和整体结构？ 天空中发出的明亮但转瞬即逝的射电爆发是什么性质，我们如何将这些信号与当地没有天体物理学意义的人造无线电信号区分开来？ 这些看似互不相关的话题因它们对大数据(宇宙学)的需求而联系在一起。我的工作使用了应用于这些数据的令人惊讶的解决方案和技术来回答这些关于宇宙的问题。 我们对宇宙的标准宇宙学模型很简单。然而，被认为推动宇宙加速的暗能量仍然是一个谜。探测暗能量的一种方法是通过观测Ia型超新星(SNE Ia)，这是一种明亮的恒星，以大致标准的亮度在整个宇宙中爆炸，使它们能够充当宇宙的灯塔。我们使用SneIa来测量距离，揭示了宇宙的加速膨胀，这是由于神秘的暗能量成分的负压而发生的。使用LSST，我们可以探测宇宙时间内的暗能量特性。LSST将产生的惊人数量的数据还将包含其他入侵者：在宇宙学上没有用处的明亮物体。我正在开拓关键技术，对这些天体进行分类和分离，然后使用统计(概率)置信度来衡量我们的宇宙学约束。 一个相关的问题是暗物质可能是什么的谜团。我在理论上模拟了轴子，这是一种很有前途的暗物质候选者，受到弦理论的物理激励。轴子实际上可以模拟暗物质或暗能量，这取决于它们的质量(以电子伏特或电子伏特为单位)。公理通过抑制小尺度上的结构形成来“洗掉”宇宙中的结构。随着对CMB的测量在小尺度上变得更加精确，我们的轴子约束几乎增加了十倍。我的团队致力于约束轴子，并在这样做的过程中，约束宇宙的“黑暗部分”。 除了暗物质和暗能量的关键宇宙学问题外，我还在解决关键的挑战，即理解“快速射电爆发”这一新的神秘现象背后的物理学。我们不知道它们是在合并中子星，还是像宇宙超辐射这样的异类。将它们归入不同的统计分组，对于揭示它们的基本属性和性质至关重要。通过将新的分类技术方法直接应用于稳定的ARO数据流，我们将为下一代仪器建立尖端的统计分类器。 我的工作将理论、数据和统计工具结合在一起，以探索数据驱动的新的宇宙学时代。