TailKit: An Italian In-Kind Software for Rubin LSST to Detect and Characterize Cometary Dust Activity

Abstract. TailKit is a new software suite developed as part of the Italian in-kind contribution to the Vera C. Rubin Observatory, designed to detect and characterize cometary dust environments using LSST data. TailKit generalizes and extends the capabilities of a modeling approach previously presented in [1,2], which has been successfully applied to fit the dust morphology of cometary comae and tails in a series of ground-based images. Developed in alignment with the recommendations of the LSST Solar System Science Collaboration, the software will be publicly available through Rubin platforms and integrated with alert brokers, enhancing LSST's scientific return and enabling rapid follow-up of dynamically new comets, including potential targets for ESA's Comet Interceptor mission.

Introduction. The Rubin Observatory LSST will provide an unprecedented view of the dynamic Solar System, with a particular impact on the detection and tracking of active small bodies. Our group, already engaged in supporting ESA's Comet Interceptor mission, is contributing a software infrastructure,TailKit,as part of Italy’s in-kind contribution to the LSST project. 

Core Modeling Framework and Implementation. TailKit is a dedicated software tool developed to analyze and characterize the dust environments of comets through quantitative modeling of their coma and tail morphology. At its core lies a statistical dust tail simulation model [1,2], which reconstructs the observed brightness distribution by considering the probabilistic behavior of individual dust particles ejected from the cometary nucleus. The model simulates the dynamical evolution of dust grains under the influence of solar radiation pressure and gravitational forces, incorporating a size-dependent β parameter that modulates the trajectory of each particle. The key innovation of the approach is the use of a minimal set of three free parameters, which are finely tuned to reproduce the observed dust features in order to match the observed isophotes. The parameters are:  i) Dust ejection velocity, potentially varying with time and heliocentric distance; ii) Dispersion of the β parameter, representing the spread in particle sizes and response to solar radiation; iii) Heliocentric distance dependence of the dust mass loss rate, typically expressed as a power-law, and linked to nucleus activity and size.

These parameters are physically motivated and correlated with the nucleus size, enabling an efficient and scalable fitting procedure. The model is built upon a supervolatile-driven activity hypothesis, in which the sublimation of volatiles such as CO, CO₂, and CH₄ at large heliocentric distances drives early cometary activity. This is especially relevant for dynamically new comets (DNCs) and long-period comets that will be observed by Rubin LSST at larger distances [3].

The suite fits coma and tail morphologies in LSST alert cut-out and survey images to derive: a) Tail orientation and extent; b) Afρ and  dust  optical depth; c) Time-dependent dust size distribution, emission velocity and loss rate. Where time-series images are available, the software tracks the evolution of activity. Results are output as science-ready tables suitable for ingestion by Rubin databases and alert brokers such as SNAPS.

Development status. TailKit has been developed to meet professional programming standards, with clear modular structure, comprehensive documentation, and test coverage. The simulation engine is designed for parallel execution, allowing the exploration of a large grid of parameter combinations over multiple cores or computational nodes. This enables rapid convergence toward the parameter set that minimizes residuals between model and observations, even when analyzing dozens of comets per night. Thanks to its flexibility, robustness, and integration with LSST alert systems and databases, TailKit represents a powerful tool for both automated large-scale surveys and targeted cometary science campaigns. Indeed, we are testing the tool on proprietary images of over 20 comets collected with the Telescopio Nazionale Galileo (TNG, La Palma), which happens to have a spatial scale similar to LSST (0.2”/pixel) [4].

Integration with LSST Infrastructure and Open Access. TailKit will be deployed via Rubin’s Science Platform and shared repositories, enabling community use and collaborative development. This contribution is developed in alignment with the LSST Informatics and Statistics Collaboration and fulfills the criteria for international data rights acquisition by Italy through in-kind work.

Conclusion. In this talk, I will present the TailKit software suite and its modeling core in detail, focusing on the methodology and computational implementation. I will also show the first results obtained from a fully automated model optimization on a large sample of comets, using images acquired at the TNG. These results demonstrate TailKit’s potential for deriving key dust environment properties with minimal human intervention, supporting statistical studies ahead of Rubin LSST's operational phase.

This project is supported by ASI agreement n. 2020-4-HH.0 (and Addendum). It is developed within the LSST Solar System and Informatics and Statistics Science Collaborations and the INAF/LSST data rights agreement.

References. [1] Fulle, M. et al. 2010, Astronomy & Astrophysics, 522, A63; [2] Fulle M. et al., 2022, MNRAS, 513, 5377; [3] Inno, L. et al. 2025, Icarus, Volume 429, id.116443 [4] Bertini I. et al. in preparation.