MOCAT-SSEM is an open-source source–sink evolutionary population model that predicts the low-Earth-orbit (LEO) space population. It divides resident space objects into families using a predefined set of parameters that determine change in population. However, these parameters are highly sensitive to initial conditions, and the SSEM estimate quickly diverges from higher-fidelity, but computationally expensive, Monte Carlo (MOCAT–MC) estimates. By treating these parameters as stochastic variables, the high-fidelity data were merged according to filtering procedures, obtaining a new set of parameters for the SSEM for fast and more accurate forecasting. This paper provides a new set of parameters for the low-fidelity SSEM model that adheres better with the Monte Carlo datasets. The numerical results show that the SSEM propagation with the trained set of parameters better describes the LEO population evolution, as shown by the high-fidelity model.
Ashley et al. (Sun,) studied this question.