讲座时间：2020-04-06 09:00
讲座地点： ZOOM 直播
主讲人：冯发波（Fabo Feng）
主办单位：李政道研究所
联系人：石井花 (leeastronomy@sjtu.edu.cn)




Fabo got his bachelor's degree from Wuhan University and master's degree from Nanjing University. He obtained his PhD from the Max Planck Institute for Astronomy in Heidelberg, Germany. From 2015 to 2018, he conducted his postdoctoral research on exoplanets at the University of Hertfordshire in England. Since November of 2018, he has been working with Dr. Paul Butler to continue exoplanet hunt at the Carnegie Institution for Science in Washington, DC, USA. He has developed many algorithms and methods to detect exoplanets using the radial velocity method. He has discovered more than 20 exoplanets such as super-Earths around tau Ceti and has helped to discover planets around nearby stars such as Barnard's star and Proxima Centauri. His recent work is focused on developing the PEXO software to analyze high precision data in order to find Earth twins.
自从1995年在类太阳恒星周围发现第一颗系外行星以来，在过去的几十年里已经发现了数千颗系外行星。然而，它们都不像我们的地球，地球是一个温和的岩石行星，围绕着一颗G型恒星运行。尽管最近设计了一些高精度的光谱仪来检测由地球孪星引起的每秒一米以下的径向速度（RV）信号，但目前的RV模型还不够精确，无法探测这样小的信号。首先，以往的噪声建模并没有通过模型比较来选择最优的噪声模型，存在过拟合或欠拟合问题。其次，用于RV数据还原的重心校正忽略了地球运动与恒星反应运动之间的耦合，忽略了目标系统中的色差和相对论效应，存在着天体测量误差和地球星历误差。我们开发了Agatha和PEXO算法，通过在贝叶斯框架下选择最优噪声模型来减轻恒星活动、仪器和大气引起的相关噪声，并对地球和目标系统的运动以及其中的相对论效应进行综合建模。PEXO能够模拟到纳秒的时间、到μas的天体测量和到μm/s的RV，它将用于合成天体测量、RV、时间和光度数据，以便探测地球孪星和系外卫星。PEXO还能够测试双星的广义相对论，并在脉冲星计时数据中探测引力波。
Since the discovery of the first exoplanet around a Sun-like star in 1995, thousands of exoplanets have been detected in the past decades. However, none of them are like our Earth, which is a temperate rocky planet orbiting around a G star. Although a few high precision spectrographs were recently designed to detect sub-m/s radial velocity (RV) signals caused by Earth twins, the current RV modeling is not precise enough for the detection of such small signals.
Since the discovery of the first exoplanet around a Sun-like star in 1995, thousands of exoplanets have been detected in the past decades. However, none of them are like our Earth, which is a temperate rocky planet orbiting around a G star. Although a few high precision spectrographs were recently designed to detect sub-m/s radial velocity (RV) signals caused by Earth twins, the current RV modeling is not precise enough for the detection of such small signals. First, previous noise modeling does not choose optimal noise models through model comparison and thus suffer from overfitting or underfitting problems. Second, barycentric correction used in RV data reduction ignores the coupling between the Earth's motion and the stellar reflex motion, ignores the chromatic aberration and relativistic effects in the target system, and suffers from astrometric errors and errors in Earth's ephemeris. To avoid these problems, I have developed the Agatha and PEXO algorithms to mitigate correlated noise caused by stellar activity, instrument and atmosphere by choosing optimal noise models in the Bayesian framework, and to model the motions of the Earth and the target system as well as relativistic effects in them comprehensively. PEXO is able to model timing to nanosecond, astrometry to μas, and RV to μm/s. It will be used to synthesize astrometric, RV, timing and photometric data in order to detect Earth twins and exomoons. PEXO is also able to test general relativity in binary stars and to detect gravitational waves in pulsar timing data.
Zoom 链接：https://zoom.com.cn/j/580729099 会议号: 580729099; 密码: 289948
欢迎各位参与