aircraft.json). Source operates from █████████████ and asked only that the receiver geography not be specified. Veracity rated high.
Three contacts in the AD-001 feed, broadcasting consistent identifiers and squawk codes, are moving in straight-line tracks at speeds and altitudes inconsistent with any biological migration but consistent with operator-controlled relocation. The contacts are not aircraft. Their broadcast identifiers carry Avianic callsigns matching the operator typology established in TEARLINE-2026-01.
Their bearings, projected forward along great-circle paths, terminate at three publicly known submarine cable landing sites distributed across the North Atlantic, Northwest European, and South Atlantic theatres.
The contacts in question carry ICAO identifiers in the a1b1cN series and broadcast callsigns prefixed with the Avianic morphemes TA-KA, HU-MA, and KA-RU — all of which appear independently in the FC-001 decoded MAC sequence. The match is not coincidental. The same operators are transmitting on both the 802.11 layer (locally, at perimeters) and the 1090 MHz layer (in transit between perimeters).
For each migrating contact, the source recovered:
Forward projection along a great-circle bearing from the reported position, at the reported heading, terminates with one-degree precision at a specific city in each case. The cities, when overlaid against publicly catalogued submarine cable landings (TeleGeography and equivalent open sources), are confirmed as cable-landing endpoints.
The operators are not migrating randomly. Their tracks are aimed at infrastructure. This is the first piece of evidence in this archive establishing that APT-AVIAN has a targeting model, not merely an existence.
The targeting model itself — what APT-AVIAN intends to do at the landings once they arrive — is not resolvable from AD-001 alone. The subsequent assessment in TEARLINE-2026-04 proposes a hypothesis: the operators are positioning themselves near landing-station infrastructure not because they can tap the fiber directly (they cannot; see █████) but because the data has already been decoded back into packet form by the time it reaches the landing, and is therefore harvestable downstream.
The three migration contacts in AD-001 are not the only contacts in the feed. There are also stationary contacts at low altitude, a coherent rotating cluster over Tokyo Bay, an intermittent ghost contact at altitude 1947 ft squawking an invalid hexadecimal code, and approximately forty contacts whose broadcasts are entirely consistent with normal civilian aviation traffic. Distinguishing the meaningful from the benign requires reading the data.
The recognition signal is the callsign. Any contact broadcasting a callsign that uses an Avianic morpheme (a hyphenated two-syllable token recovered from the FC-001 decode) is a candidate for closer attention. Other contacts may also be of interest for different reasons (squawk anomalies, altitude anomalies, persistent stationarity in unusual locations) but those signals are addressed in separate tearlines.
This tearline drops one Request for Information against the AD-001 feed.
I publish AD-001 in its raw form. Open it in any tooling you prefer — jq, your browser's developer console, the flockaware tracker built into this archive. The flockaware view will let you click any moving contact, draw its forward great-circle bearing on the map, and overlay a list of catalogued cable landings drawn from public sources. This is the geospatial analyst's toolchain reproduced inside a browser tab. It is not a substitute for actually understanding what a great-circle bearing means.
If you derive the three landings, submit the city names to RFI-γ on the Challenges page. The verification happens locally in your browser; nothing is reported back to me except the eventual handle, if you choose to claim credit.