Indigo Brownhall, University College London; Miles Lifson, The Aerospace Corporation; Giovanni Lavezzi, Massachusetts Institute of Technology; Enrico Zucchelli, Massachusetts Institute of Technology; Mark Moretto, North Carolina State University; Santosh Bhattarai, University College London; Richard Linares, Massachusetts Institute of Technology
Keywords: Space Debris Modelling, Sustainability, LEO, Space Policy
Abstract:
The increasing amount of debris in Low Earth Orbit (LEO), combined with the sustained growth for the demand for LEO orbits, has heightened the need for new research in transdisciplinary policy and regulatory approaches. The growing inactive population raises significant concerns about the long-term sustainability of the orbital environment, as well as the burden it imposes on Space Situational Awareness (SSA) infrastructure. In practical terms, SSA efforts such as collision avoidance, catalog development and maintenance, could all be hampered by both the sheer volume and the uncontrolled growth of space objects. Beyond these technical challenges, there is also a pressing risk to darkand-quiet night skies, where the brightness and frequency of satellite passes will continue to adversely affect terrestrial astronomy. Within this context, the concept of orbital “carrying capacity” has gained traction, where policymakers, satellite operators and researchers alike want to determine how many objects in LEO can safely be accommodated without escalating collision risk to unsustainable levels. As new constellations are planned, debris models are used to understand the impact on the potential long-term sustainability of LEO. In addition, there is a need for transdisciplinary approaches, including models such as Integrated Assessment Models (IAMs)- which couple socio-economic theory to physical systems- to research new space policy and understand the second-order effects and actors respond to incentives and requirements imposed by policy.
In this paper, we ultimately compare two models in the MIT Orbital Capacity Assessment Tool (MOCAT) family, the Monte-Carlo (MC) with a Source Sink Evolutionary Model (SSEM). We introduce (i) a cross-fidelity comparison metric that combines four per-species components—magnitude agreement over time, final-year difference, time-series shape, and altitude-distribution similarity—into a single scenario score, and (ii) an (a,e) flux model propagator for SSEMbased on Brouwer–Lyddane averaged drag rates. We show that the flux model reproduces MC total-population trajectories when modelled with 100×100 bins, and R2 = 0.98 while running in ∼0.4 seconds versus 7.2 seconds for MC(about 18×faster). Furthermore, we then show how the flux model performs as well with as little as 9 bins, which is more realistic when integrated into a SSEM.
Next, we model 650 SSEM runs varying model type (circular vs. fragment-spread), number of shells (15–40), and number of species, and compare against MC using the new metric. Accuracy improves smoothly with shell count and shows diminishing returns beyond ∼30–35 shells; fragment spreading provides a modest additional gain at the same configuration. Species-level Pareto fronts exhibit clear “knees,” where most attainable accuracy is achieved at modest CPU. Type-II ANOVA attributes variance in the overall score primarily to shells (∼50%) and species granularity (∼40%), while CPU time is driven mainly by species (∼55–60%) with shells secondary; model type contributes only a few percent to either. The best configuration found (circular, 30 shells, 12 species) tracks MC closely across groups with per-scenario runtimes < 250 CPU seconds.
These results provide concrete guidance for credible, fast studies. We show that for under launch scenarios of the order 104 per year, MOCAT-SSEM and MOCAT-MC can be used interchangeably, supporting rapid policy analyses and integration within Integrated Assessment Model workflows.
Date of Conference: September 16-19, 2025
Track: Space Debris