📢 TipTop v1.5.0, P3 v1.6.0 and MASTSEL v1.5.0 released!
TipTop v1.5.0, P3 v1.6.0 and MASTSEL v1.5.0 are now available! 🎉
This release focuses on improving the numerical robustness and spatial consistency of TipTop's analytical engine. These developments are particularly important for operational applications, where PSFs must remain reliable over a broad range of observing conditions, detector samplings, and wavelength ranges.
Highlights include improved multi-wavelength PSF generation, robust handling of odd and even Fourier grids, physically consistent atmospheric refraction modelling, and enhanced spatial resampling of PSFs.
✨ What's new
🌍 Rigorous atmospheric refraction modelling
TipTop now includes an improved wavelength-dependent atmospheric refraction model.
The implementation combines:
- Ciddor (1996) formulation at visible and near-infrared wavelengths,
- Mathar (2006) formulation in the infrared.
while accounting for atmospheric pressure, temperature, and humidity.
This provides a more realistic description of differential atmospheric refraction between guide-star and science wavelengths, particularly for infrared observations and broadband simulations.
🖼️ Consistent multi-wavelength PSF generation
Several improvements have been introduced to ensure consistent PSF generation across multiple wavelengths.
The internal pixel scale is now defined uniquely from the shortest simulated wavelength and propagated consistently throughout the computation. PSFs computed at different wavelengths are automatically remapped onto a common output grid while preserving their total flux.
The generation pipeline also determines the optimal internal grid size automatically, preventing field-of-view truncation at short wavelengths and ensuring consistent spatial sampling across all spectral channels.
📐 Robust Fourier handling for odd and even sampling grids
The internal Fourier conventions have been revised to correctly support both odd and even computational grids.
These changes eliminate the sub-pixel registration offsets that can otherwise appear when switching between parity conventions, ensuring that generated PSFs remain accurately centred regardless of the chosen detector sampling.
Dedicated unit tests now verify the consistency of the Fourier transforms for both grid types.
🌠 Improved PSF extrapolation and spatial resampling
The PSF resampling pipeline has also been refined.
PSFs generated on different internal grids are spatially remapped onto the requested output sampling, while a new radial profile extrapolation algorithm extends the PSF halo beyond the computed field of view using a physically motivated power-law model.
These improvements reduce truncation artefacts and provide a more realistic description of the PSF wings for large fields of view.
🚀 Why this matters
Together, these developments significantly improve the numerical stability and robustness of TipTop across a broad range of observing configurations.
They provide the foundations required for demanding applications such as exposure-time calculators, observation preparation, multi-wavelength simulations, and PSF reconstruction, where reliable PSF generation is essential.
🚀 Update now
pip install --upgrade astro-tiptop
Check it out on PyPI 👉