Dr André Silva

Dr André M. Silva holds a Master's degree in Engineering Physics (2019) and a PhD in Astrophysics (2024) from the University of Porto (UP). His doctoral thesis on "New Paradigm for the Estimation of Precise Stellar Radial Velocities" won in 2024 the IAU PhD prize of Division B. Since 2024, he is a postdoctoral researcher at the Instituto de Astrofísica e Ciências do Espaço (IA) and an Invited Assistant Professor at FCUP. His research focuses on the extraction of precise radial velocities (RVs) and identification of systematic biases in this process. He contributes to the development of the observational and control software for PoET, a portuguese-built solar telescope that will be connected to the ESPRESSO spectrograph. He is actively engaged in public outreach initiatives in Portugal.

One of the most powerful techniques for exoplanet detection and characterization is the radial velocity (RV) method, which measures variations in a star's motion along our line of sight. Dr Silva developed and analysed methodologies to extract ultra precise RVs from state-of-the-art stellar spectra, achieving breakthroughs on different avenues.

He developed a semi-Bayesian approach to deriving radial velocities (s-BART), built around the assumption of achromaticity, expected from planetary signals. The s-BART pipeline was released as an open source tool, leading to its adoption by multiple international consortia and surveys. He identified a previously unknown bias affecting template-based analysis, traced to artefacts in the stellar template: contamination by micro-telluric lines and detector-frame features induce the apparent RV drift. This bias has a significant impact on RV campaigns, asteroseismology, or transit and atmospheric characterization by transmission spectroscopy. Understanding such biases will play a critical role in the exploitation of data from the current and future generation of spectrographs. His thesis lays the groundwork to overcome a fundamental limitation of template matching: the assumption of a time-invariant stellar model. In his Bayesian approach the stellar spectrum and telluric absorption are fit simultaneously with the RV, providing a pathway to directly characterize planetary signals on the stellar spectra. This novel direction could overcome limitations of fixed templates and further improve RV precision by explicitly accounting for stellar variability. In summary, the work of Dr Silva has advanced the state of the art in RV exoplanet detection. 

The work was conducted at the University of Porto and Instituto de Astrofísica e Ciências do Espaço (Portugal).