Two machines keep the unradared ocean separated: the clearance that fixes each flight's route, speed and level before it commits, and the communications that carry every message across four thousand kilometres of water. This lesson learns both, with one honesty note upfront: the syllabus and a generation of exams are built on the oceanic clearance regime the manual's earlier editions describe, and the current edition has retired it. The classic machinery comes first, because the examination asks it; what replaced it closes the account.
The oceanic clearance
An oceanic clearance was built from three elements: route, speed and level, and the trio was no accident, for the elements served the three basic elements of separation: lateral, longitudinal and vertical. Clearances were required for all flights in NAT controlled airspace at or above FL 60, requested from the ATC responsible for the first oceanic control area, recommended at least 40 minutes before the boundary estimate, 20 minutes when entering Reykjavik from the Scottish or Stavanger areas, and delivered by VHF clearance delivery, by HF through the aeradio station, through domestic ATC or by data link. The clearance applied only from the entry point, however early it arrives.
The request carried planning gold: the pilot was to notify the oceanic centre of the maximum acceptable flight level possible at the boundary, because the climb to the assigned oceanic level had to be achieved before entering oceanic airspace, normally while still inside radar coverage, and a track request named the next preferred alternative. Some geography forced the paperwork earlier still: at aerodromes situated close to oceanic boundaries the clearance had to be obtained before departure, which on the eastern side applied to all Irish airfields, UK airfields west of 2 degrees 30 minutes west, and French airfields west of the zero meridian. And an aircraft that lost RVSM or NAT HLA capability to an equipment failure on the way to the ocean was to advise ATC at initial contact when requesting the clearance.
Between clearance and entry
Longitudinal spacing over the ocean was planned solely on the estimates for the entry fix, so the estimate is guarded: after obtaining and reading back the clearance, the pilot monitors the forward estimate for oceanic entry and, unless position reports flow via ADS-C, passes a revised one to ATC whenever it changes by 3 minutes or more. Failure to do so may cost the flight a re-clearance to a less economical track or level for the whole crossing, and may penalise the aircraft behind.
Two traps live at the boundary. First, the issued clearance may differ from the flight plan, and pilots should pay particular attention when it does: a significant proportion of navigation errors investigated in the NAT involve an aircraft that followed its flight plan rather than its differing clearance. Second, if the cleared entry point differs from the one requested, or the oceanic level differs from the current one, the pilot is responsible for requesting and obtaining the domestic re-clearance that brings the flight to its oceanic clearance at the boundary. The abbreviated clearance of the full-length organised track, destination, Track plus letter, level and Mach, leaned on the previous lesson's TMI: quoting the identification in the readback replaced reading the coordinates, while any doubt about the TMI put the full coordinate readback straight back on the frequency, and random-route clearances always demanded it.
What the current edition changed
The current edition of the manual retires the gate: en-route aircraft now enter oceanic airspace in accordance with their existing ATC clearance, and no oceanic clearance is required. In its place stands the RCL, a voice or data link message to Bodo, Gander, Santa Maria and Shanwick, not required for New York Oceanic East or Reykjavik, giving the estimated time over the entry point, the requested flight level, the speed and, mandatorily, the maximum flight level; the manual is explicit that the RCL is not a clearance. Flights fly cost-index speed unless ATC assigns a fixed Mach, rarely more than 0.01 faster or 0.02 slower than requested, and ATC must be advised when the speed changes by plus or minus Mach 0.02 or more. The TMI readback survives where it always made sense: an abbreviated re-clearance along a whole organised track is confirmed by quoting the current identification. The old regime's logic still explains the new one, which is why the exam's questions remain worth answering well.
Talking across the ocean
Who is on the other end of the radio matters. Routine air-ground ATS voice communications in the NAT are conducted via aeronautical radio stations staffed by operators who have no executive ATC authority: messages are relayed between the ground station and the air traffic controllers of the relevant oceanic centre, whether the medium is HF, general-purpose VHF or SATVOICE. Six stations share the ocean: Bodo, Gander, Iceland, New York, Santa Maria and Shanwick Radio.
Even in the data link era a significant volume of NAT air-ground communication rides single sideband HF voice. Twenty-four HF frequencies are allocated to the region, in bands from 2.8 to 18 MHz, organised into six families, A to F, each spanning frequencies from every band and shared across stations that cooperate as a network. The ionosphere sets the schedule: its refractive layers vary with the sun, so frequencies from the lower bands serve the night and the higher bands the day, in the North Atlantic roughly below 6 MHz by night and above 5 MHz by day. Each flight gets a primary and a secondary frequency before the boundary, and because the radio operators keep watch on several frequencies at once, the pilot should state the HF frequency in use when initiating contact with the station.