If the busiest shell were simply walked away from

Imagine the company running the busiest shell in orbit went bankrupt tomorrow — a hypothetical, but one no law currently prevents: nothing anywhere obliges anyone to tidy up a fleet on the way out. Six thousand satellites would just… stay, drifting, un-steered, in the lane where they were parked.

In plain terms: this explorer compares two endings for the same fleet. Walking away leaves the shell to grind itself into shrapnel for a decade. Tidying up on the way out — steering each satellite down before the lights go off — leaves the shell roughly thirty times cleaner five years on (median of the model's runs). Same shutdown; the entire difference is whether anyone is still at the wheel. Slide the years to watch the two futures separate.

Years after the shutdown

Walked away — pieces of trackable shrapnel in the shell

Tidied on the way out — same shutdown, steered down

The full 50-year picture, with the model's uncertainty (tap to open)

MODELED Precise terms for the labels above: "pieces of trackable shrapnel" = tracked debris objects (≥10 cm class) in the 465–495 km trigger band; "the model's runs" = N=100,000 parameter draws, engine v1.2, seed 20260716, paired draws; medians shown, p05–p95 bands in this chart. The spread is parameter uncertainty, not a forecast interval; every P-style statement is belief over assumed knobs (F15). Disposal compliance in the tidy branch is ASSUMED 0.85–0.97, untested at fleet scale. Dots are the engine's sampled years (0, 5, 10, 25, 50); connecting lines are guides. All magnitudes conditional on f_imp. Source: qb_engine_results.json; if a number here and a number in that file disagree, the file wins.

AS FILED · CLAIMED The official voice: the newest filings commit to disposal reliability ≥0.99 per satellite (Orbital Compute ODAR) and >0.9 (Blue Origin) — at or above the entire 0.85–0.97 band the tidy branch assumes here. If those commitments held at fleet scale, the blue ending would be tidier than drawn. Filed commitments; untested at fleet scale; the same docket also contains both new entrants asking off the deployment clock. Sources: the filings (primary), paper §7.4.

In the paper: §4.4 (the exit asymmetry) · §5.4 (the continuity gap) · stakes rows: kessler_stakes_table_v1.json