Speaker
Description
The performance of gravitational-wave interferometers is critically limited by the surface quality of their mirrors, which directly affects optical losses and increases the higher-order mode content. As future detectors such as the Einstein Telescope (ET) aim for unprecedented sensitivity, defining realistic and reliable mirror specifications becomes a central challenge.
We address this challenge by presenting results from a recently submitted paper currently under peer review, in which we develop a framework for generating virtual mirror maps that statistically reproduce the spatial properties of real optical surfaces. Our method combines Zernike polynomial decomposition and spatial frequency analysis via power spectral density (PSD), enabling the controlled synthesis of mirror surfaces that preserve both low-order aberrations and high-frequency roughness.
We validate this approach using metrology data from Advanced Virgo mirrors and apply it to ET-scale configurations through numerical optical simulations. The robustness of the framework is demonstrated through statistical validation based on optical figures of merit, including power loss and spectral consistency.
Our results show that different generation methods significantly impact optical performance. Our hybrid approach delivers the most realistic agreement with measured data across all spatial frequencies, and provides a powerful and flexible tool to bridge current experimental knowledge and future detector design, enabling a quantitative assessment of mirror requirements for next-generation gravitational-wave observatories.