Dynamical modelling of dwarf spheroidal galaxies

银河系的矮椭球（dSph）卫星是已知的最暗物质主导的恒星系统，因此被广泛认为是研究暗物质性质的有价值的实验室。对恒星运动学的观测已经产生了几千个恒星速度的数据集。然而，迄今为止，这些数据的动力学建模是基于非常严格的简化假设（球形，平衡）。我们将使用高斯过程仿真，使我们能够探索更一般的模型参数空间比已经可能使用更传统的建模方案（如马尔可夫链-蒙特-卡罗）。底层模型可以是静态的、基于分布函数的模型或N体模拟。

The dwarf spheroidal (dSph) satellites of the Milky Way are the most dark matter dominated stellar systems known and, consequently, are widely recognised as valuable laboratories in which to study the properties of dark matter. Observations of the stellar kinematics have yielded data sets of several thousand individual stellar velocities in several dSphs. However, dynamical modelling of these data has to date been based on very restrictive simplifying assumptions (sphericity, equilibrium).This project will develop a new approach to the modelling of dSphs. We will use Gaussian Process Emulation to enable us to explore more general model parameter spaces than have been possible using more traditional modelling schemes (such as Markov-Chain-Monte-Carlo). The underlying models may be static, distribution function based models or N-body simulations.