Galaxy evolution in 3 dimensions

星系是数十亿颗恒星的巨大集合，代表了宇宙历史的丰富挂毯。 星系的形成和演化的线索被编码在它的光中：一个独特的指纹，包含了它的星星形成的过去集合历史，可以用来推断许多物理性质，如它的质量，它制造新恒星的速度和它的化学性质。 这些和其他物理特性可以从星系的光谱中测量，即通过将光作为波长的函数进行传播，就像彩虹一样。正如在许多研究分支中一样，需要大量的数据样本，以便提取有统计意义的结果，定义适当的控制实验并最大限度地减少随机效应。通过收集数十万个星系的样本，天文学家在过去十年中对星系形成和演化的理解取得了巨大进展。 传统上，这些大型光谱调查获得每个星系的综合光谱，将整个星系的所有光线加在一起，并使用单一光谱来推断全球性质。 然而，一场革命正在向我们走来。近年来，天文摄谱仪已经发展起来，可以获得空间分辨的光谱，即同时在每个星系的表面上拍摄数百或数千个光谱。 它相当于从单个像素到完整图像的过程，具有巨大的潜力。 这种技术被称为积分场光谱学（IFS），它使我们超越了每个星系的单一综合指标，并使我们能够制作星星形成，化学和质量的完整地图。 至关重要的是，IFS的第一次大型调查刚刚上线。 具体来说，MaNGA调查正在获得10，000个星系的光谱图，所有数据都可以通过电子方式获得。我提议的研究将使用丰富的MaNGA数据集来解决当代河外天文学中的几个紧迫问题。 我将研究两个星系之间的相互作用在哪里以及如何导致新恒星的形成，该星星形成的产物如何分布以及星系周围的外层气体储层如何受到影响。 观测数据将得到最先进的星系模拟的补充，这将使我们能够测试理论模型并改进我们的星系识别算法，并将开发复杂的机器学习技术来优化处理MaNGA数据集中的数百万个数据点。 以三管齐下的方法解决这些研究问题，结合观察，计算机模拟和人工神经网络，在加拿大和国际前沿都是独一无二的，为年轻科学家提供了特殊的培训机会。

Galaxies are vast collections of billions of stars, and represent the rich tapestry of cosmic history. The clues to a galaxy's formation and evolution are encoded in its light: a unique fingerprint that contains the past ensemble history of its star formation, and can be used to infer numerous physical properties, such as its mass, the rate at which it is making new stars and its chemistry. These, and other, physical characteristics can be measured from the galaxy's spectrum, i.e. by spreading out the light as a function of wavelength, much like in a rainbow. As in many branches of research, large samples of data are required in order to extract statistically meaningful results, define adequate control experiments and minimize stochastic effects. By collecting samples of hundreds of thousands of galaxies, astronomers have made huge progress in their understanding of galaxy formation and evolution in the last decade. Traditionally, these large spectroscopic surveys obtain an integrated spectrum of each galaxy all of the light from the entire galaxy is added together and a single spectrum is used to infer global properties. However, a revolution is upon us.In recent years, astronomical spectrographs have been developed that obtain spatially resolved spectra, i.e. hundreds or thousands of spectra are taken simultaneously across the face of each galaxy. It is the equivalent of progressing from a single pixel to a full image, and holds enormous potential. This technique, known as integral field spectroscopy (IFS), takes us beyond a single integrated metric for each galaxy, and allows us to make full maps of star formation, chemistry and mass. Critically, the first large IFS surveys are just coming online. Specifically, the MaNGA survey is obtaining spectroscopic maps for 10,000 galaxies, and all of the data are being made publically available. My proposed research will use the rich MaNGA dataset to tackle several pressing questions in contemporary extragalactic astronomy. I will study where and how the interaction between two galaxies causes new stars to be formed, how the products of that star formation are distributed and how the outer gas reservoir around galaxies is affected. The observational data will be complemented by state-of-the-art galaxy simulations that will allow us to test theoretical models and improve our galaxy identification algorithms, and sophisticated machine learning techniques will be developed to optimally process the millions of data points in the MaNGA dataset. Tackling these research questions with a three pronged approach, that combines observations, computer simulations and artificial neural networks, is both unique in Canada and at the international forefront, offering exceptional training opportunities for young scientists.