X-ray reverberation around accreting black holes: fitting complex GPU-based models to large datasets

黑洞吸积流是X射线的强大来源。这些X射线照亮吸积盘，并产生包含强烈发射特征的反向散射光谱。此外，X射线源是高度可变的。可变性、光源和光盘之间的光传播时间以及反向散射光谱的组合导致不同X射线能量带中的光变曲线之间的一组复杂的时间滞后。观察到时间滞后取决于变化的频率，因此该分析在傅立叶域中进行。为了模拟这种行为，我们利用GPU来整合从X射线源到光盘的光子路径，考虑到所有的相对论效应，并允许不同的X射线源几何形状。该项目的目标是开发技术，使这个复杂的，多维的模型进行评估，并适合大量的观察是在X射线数据档案。

Black hole accretion flows are powerful sources of X-rays. These X-rays illuminate the accretion disc and produce a back-scattered spectrum that contains strong emission features. In addition, the X-ray source is highly variable. A combination of the variability, light travel time between the source and disc, and back-scattered spectrum result in a complex set of time lags between the light curves in different X-ray energy bands. The time lags are observed to depend on the frequency of variability, so this analysis is done in the Fourier domain. To model this behaviour we make use of GPUs to integrate photon paths from the X-ray source to the disc, taking into account all of the relativistic effects, and allowing for different X-ray source geometries. The goal of this project is to develop techniques that allow this complex, multi-dimensional model to be evaluated and fit to the large number of observations that are in the X-ray data archives.