The orbital motion of long-period comets is significantly influenced by non-gravitational accelerations arising from volatile sublimation. These effects are traditionally modelled using the Marsden formulation, introduced in the early 1970s and derived from empirical laws and limited and low-precision astrometric datasets. More recent studies have demonstrated that this model frequently breaks down at large heliocentric distances and poorly represents sublimation of volatiles other than water. This thesis aims to improve the modelling of non-gravitational accelerations in cometary orbits by addressing two closely related aspects: the physical formulation of the non-gravitational acceleration and the quality of the astrometric data used in orbit determination.
We preliminarily determine a tailored approach to cometary astrometric reductions, building on the zero-aperture extrapolation method. We also highlight the importance of high quality astrometric data in the determination of non-gravitational effects, which are easily masked by data inaccuracies.
Using high-precision astrometric datasets, we propose two formulations for the non-gravitational acceleration, respectively representing sublimation of a single volatile and sublimation of multiple volatiles in a subsequent fashion. We demonstrate that the proposed formulations are effective in capturing the effects of the outgassing acceleration, and moreover allow us to retrieve physical characteristics of comets relying exclusively on their dynamical behaviour.
The results of this work highlight the critical importance of high-quality astrometric data and physically informed dynamical models for reliable comet orbit determination, contributing towards high-fidelity trajectory estimation and production of reliable observation forecasts.