Cross-tagging is a catalogue identity swap: observations of one satellite get attributed to another flying nearby, so each object's catalogue number ends up carrying the other's orbit. It plagues freshly launched satellite batches, whose members fly almost identical orbits for days.
Understanding Cross-Tagging
Why launch day breaks correlation
Correlation assumes each object owns a unique neighbourhood of orbit space — an assumption batch launches demolish. Sixty satellites released from one vehicle share energy, plane and phase to within the errors of their fresh, hastily fitted elsets, whose uncertainty exceeds the spacing between objects. In that regime assignment is close to guesswork, and early catalogue identities for a new batch are best treated as provisional labels. Amateur observers know the drill: for the first days after a big deployment, predictions for individual batch members are unreliable even when the train as a whole appears exactly on schedule — the group's orbit is right; the name-tags within it may not be.
Detecting a swap after the fact
Cross-tags reveal themselves through discontinuities. An object's history suddenly shows a period change with no plausible manoeuvre; drag behaviour flips character (the "satellite" starts decaying like the neighbouring rocket body it was confused with); an operator reports their craft healthy on a different orbit than the catalogue claims. Analysts untangle the braid by re-fitting archived observations against both candidate orbits and reassigning tracks wholesale — after which both objects' histories are quietly amended. For anyone doing fine-grained historical analysis, this is the caution: identity in the catalogue is a maintained hypothesis, and around crowded deployments it is occasionally revised.