Advanced PSF Fitting and Extrapolation Using TipTop for Adaptive Optics Systems
Last update: October 10, 2025
Context and Motivation
Adaptive Optics (AO) systems are crucial for high-resolution astronomical observations, compensating for atmospheric turbulence. The Point Spread Function (PSF) characterizes the response of an imaging system to a point source and is essential for accurate scientific analysis. However, PSFs vary across the field of view due to anisoplanatism and atmospheric effects, making their accurate modeling and extrapolation a key challenge for astronomy.
The TipTop framework provides advanced tools for simulating, fitting, and extrapolating AO PSFs under various conditions. Recently, methods based on TipTop were successfully applied to galaxy science with the Large Binocular Telescope (LBT), demonstrating the potential of PSF extrapolation for improving galaxy image analysis.
Objectives
- Develop and implement advanced PSF fitting techniques using the TipTop framework to model PSFs from AO observational data.
- Extrapolate PSFs across the field of view, accounting for spatial variations, atmospheric turbulence, and instrument-specific effects.
- Apply and validate these methods on real observational datasets from ERIS and MUSE, with access already secured to specific fields observed for PSF testing.
- Optionally, the internship could lead to a continuation as a PhD project.
Methodology
- Literature Review: Study PSF fitting and extrapolation methods in AO systems, focusing on galaxy imaging applications.
- PSF Fitting: Use TipTop to fit parametric or hybrid models to the observed PSFs. Frame the task as minimizing a suitable loss function. Depending on the formulation, optimization may rely on gradient-based methods or gradient-free approaches; compare strategies and evaluate trade-offs between accuracy, robustness, and efficiency.
- Extrapolation: Develop algorithms to predict PSFs in unobserved regions, including anisoplanatic effects.
- Validation: Compare extrapolated PSFs with observational data to quantify accuracy and assess suitability for science applications.
Expected Outcomes
- Accurate PSF models capturing spatial variations across AO-corrected fields.
- Reliable extrapolation algorithms applicable to multiple AO instruments.
- Validation framework for PSF-based science analysis (galaxy imaging, high-resolution studies).
Skills and Tools
- Programming: Python, with experience in scientific libraries (NumPy, SciPy, Astropy).
- Mathematical Modeling: ability to define and work with loss functions, and to apply optimization strategies (gradient-based or gradient-free) depending on the problem.
- TipTop Framework: familiarity with AO PSF simulation, fitting, and analysis.
- Astronomical Imaging: knowledge of AO systems, PSF behavior, and observational data analysis.
- Data Analysis: statistical modeling, validation, and error analysis.
- Desired level: Master’s degree student (final year) or equivalent engineering school level.
Practical Information
- Duration: 4–6 months.
- Preferred start: between February and May 2026.
- Location: Laboratoire d’Astrophysique de Marseille.
- Application deadline: late November 2025.
Supervision and Collaboration
The intern will be supervised by Lisa-Marie Mazzolo and Benoit Neichel. Collaboration with the TipTop development team and other researchers working on AO systems and galaxy science is encouraged.