Every satellite operator makes manoeuvre decisions based on conjunction data messages that arrive hours before a close approach. That is reactive, not predictive. What national space programmes need is a density forecast: a probabilistic map of where debris concentrations are building, which altitude bands are approaching critical flux, and when the next fragmentation event will statistically trigger a cascade. Without this foresight, operators burn fuel on unnecessary avoidance manoeuvres, waste ground-station contact windows, and — worse — miss the genuinely dangerous geometry because the alert arrived too late to act.
The satellite stack that makes forecasting possible combines in-situ flux sensors on a LEO constellation with ground-based radar and optical correlation. Miniaturised impact-detection arrays on 6U to 12U cubesats flying at 400–600 km measure sub-centimetre particle flux directly — data that no ground telescope can produce. That in-situ signal is fused with Two-Line Element (TLE) propagation and atmospheric-drag models driven by real-time solar flux telemetry. Machine-learning ensemble models then project density evolution over 14-day windows at 25 km altitude-band resolution, updated every six hours as new observations arrive.
The operational outcome is a shift from fire-fighting to scheduling. A national civil space agency can advise its own satellite operators — and its military space command — which orbital slots to avoid for the next fortnight, time launches to thread through low-density windows, and build a sovereign picture of LEO health that is not filtered through a foreign conjunction service. Nations that operate the forecasting layer hold the authoritative view of their own orbital neighbourhood; everyone else is reading someone else's summary.