Review of uncertainty sources affecting the long-term predictions of space debris evolutionary models

Since the launch of Sputnik-I in 1957, the amount of space debris in Earth's orbit has increased continuously. Historically, besides abandoned intact objects (spacecraft and orbital stages), the primary sources of space debris in Earth's orbit were (i) accidental and intentional break-ups which produced long-lasting debris and (ii) debris released intentionally during the operation of launch vehicle orbital stages and spacecraft. In the future, fragments generated by collisions are expected to become a significant source as well. In this context, and from a purely mathematical point of view, the orbital debris population in Low Earth Orbit (LEO) should be intrinsically unstable, due to the physics of mutual collisions and the relative ineffectiveness of natural sink mechanisms above 700 km. Therefore, the real question should not be "if", but "when" the exponential growth of the space debris population is supposed to start. From a practical point of view, and in order to answer the previous question, since the end of the 1980's several sophisticated long-term debris evolutionary models have been developed. Unfortunately, the predictions performed with such models, in particular beyond a few decades, are affected by considerable uncertainty. Such uncertainty comes from a relative important number of variables that being either under the partial control or completely out of the control of modellers, introduce a variability on the long-term simulation of the space debris population which cannot be captured with standard Monte Carlo statistics. The objective of this paper is to present and discuss many of the uncertainty sources affecting the long-term predictions done with evolutionary models, in order to serve as a roadmap for the uncertainty and the statistical robustness analysis of the long-term evolution of the space debris population.