Measuring Cometary Nuclei Behind Bright Comae: PSF Delta Decomposition with Bicubic Resampling and an Application to Interstellar Comet 3I/ATLAS C/2025 N1
Toni Scarmato
TL;DR
Measuring the sizes of unresolved cometary nuclei is hindered by bright inner comae. The paper introduces a PSF-delta decomposition approach that explicitly models the nucleus as a 2D Dirac delta plus a $K\rho^{-1}$ coma term and performs forward convolution with the PSF, enabling stable nucleus flux extraction after bicubic resampling. By fitting the azimuthally averaged profile with a two-amplitude model $I_{\rm obs}(\rho)=A P(\rho)+B (\rho^{-1} \ast P)(\rho)$ and converting the nucleus flux to a radius via $D=1329\times 10^{-H/5} p_V^{-1/2}$, the method provides nucleus size estimates with quantified uncertainties. Application to the interstellar comet 3I/ATLAS (C/2025 N1) yields a radius range of roughly $0.16$–$2.8$ km for $p_V\approx0.04$, consistent with HST upper limits, and demonstrates a robust, survey-ready workflow. The approach, being lightweight and compatible with Rubin LSST pipelines, offers a practical path to nucleus measurements across diverse datasets, with the dominant uncertainties arising from albedo and PSF/ coma systematics.
Abstract
Measuring cometary nuclei is notoriously difficult because they are usually unresolved and embedded within bright comae, which hampers direct size measurements even with space telescopes. We present a practical, instrumental method that, stabilises the inner core through bicubic resampling, performs forward point-spread function PSF+convolution, and separates the unresolved nucleus from the inner-coma profile via an explicit Dirac Delta function added to a Rho^-1 surface brightness law. The method yields the nucleus flux by fitting an azimuthal averaged profile with two amplitudes only PSF core and convolved coma, with transparent residual diagnostics. As a case study, we apply the workflow to the interstellar comet 3I/ATLAS alias C/2025 N1, incorporating Hubble Space Telescope constraints on the nucleus size. We find that radius solutions consistent with 0.16 <= Rn <= 2.8 km for Pv = 0.04 are naturally recovered, in line with the most recent HST upper limits. The approach is well-suited for survey pipelines Rubin LSST and targeted follow up.
