A domestic flight across the United States never belongs to one air traffic controller. It is passed from hand to hand like a baton: from a tower, to a radar room near the airport, to a string of en-route centres, and back down again at the other end.

The scale is hard to picture. In fiscal year 2024 the Federal Aviation Administration handled an average of 44,360 flights a day, with around 5,500 aircraft in the sky at peak times. Every one of them is talking to someone on the ground, and every one is kept a fixed distance from its neighbours.

This is how air traffic control works, position by position: who you talk to, when the handoffs happen, what the separation rules are and how Europe's system differs from America's.

Clearance delivery and ground: control starts at the gate

Before an airliner flying under instrument flight rules can even start its engines, it needs a clearance. At a busy airport a dedicated clearance delivery position issues it: the destination or clearance limit, the route, the initial altitude, the departure frequency and a four-digit transponder code. Pilots read it back word for word.

At many large US airports that exchange no longer happens by voice. The FAA's Data Comm system sends departure clearances to the cockpit as text, and the agency has rolled it out to more than 60 airport towers. It cuts the read-back errors that come with spoken clearances.

Once the aircraft pushes back, ground control takes over. The ground controller manages every movement on the taxiways, issuing routes to and from the runway and keeping aircraft and vehicles apart. Close to the runway, the pilot is told to switch to the tower frequency.

Wendover Productions follows a flight through every stage of air traffic control, from clearance delivery to the en-route centres.

The tower: a five-mile world

The tower controller, often called local control, owns the runways. Nobody takes off, lands or crosses an active runway without the tower's clearance. The tower also handles light aircraft flying visually through the airport's control zone and sequences them around the airliners.

For decades, each flight in a tower or radar room has been tracked on a flight progress strip: a narrow slip carrying the callsign, aircraft type, route and altitude, moved between bays as the flight progresses. Many US towers still use paper. In February 2026 the US Department of Transportation said electronic flight strips, part of the FAA's Terminal Flight Data Manager programme, had been deployed at 15 airports, including Washington National.

Spacing on final approach is not only about collisions. Big aircraft shed powerful wake vortices, so controllers add extra distance behind heavy jets. Under FAA rules, a small aircraft landing behind a heavy one on the same runway needs 6 nautical miles.

The tower's airspace is surprisingly small. Once a departing aircraft is airborne and a few miles out, the tower hands it to a radar controller, and an icon flashes on the next controller's screen until it is accepted.

Departure and approach: inside the TRACON

Departure and approach control sit in a terminal radar approach control facility, or TRACON. The FAA runs 146 of them. Some serve one airport; others, such as those covering New York or Southern California, handle several major airports at once from a single dark room.

According to the FAA's Jennifer Ross, a TRACON's airspace typically reaches 20 to 40 miles from the airport and up to somewhere between 15,000 and 23,000 feet. Departure controllers climb aircraft out of the airport and feed them into the en-route system. Approach controllers do the reverse, merging streams of arrivals into a single line for each runway.

Close to the airport, aircraft can be packed tightly. Using terminal radar within 40 miles of the antenna, FAA controllers may separate aircraft by 3 nautical miles laterally, or by 1,000 feet vertically. When arrivals outnumber the runway's capacity, approach controllers use speed control, radar vectors and, at the busiest airports, holding stacks.

En route: the centres that run the sky

Above and beyond the TRACONs lies en-route airspace, run in the United States by 21 Air Route Traffic Control Centers, or ARTCCs. Each centre covers a huge block of the country and is divided into sectors, some stacked by altitude, each worked by its own controller or team on its own frequency.

Away from the airports the radar picture is built from long-range sensors, and the minimum gap is wider: 5 nautical miles below FL600 on the FAA's en-route automation system. Vertically, the 1,000-foot rule applies up to FL410 thanks to reduced vertical separation minima, with 2,000 feet above that.

Much of how air traffic control works in the cruise comes down to handoffs. A cross-country flight works through sector after sector, centre after centre. Chris Keyes, a controller at the FAA's Washington Center, described the routine on the FAA's podcast: as an aircraft nears the edge of his airspace, a series of keystrokes makes its data block flash on the next controller's screen, and the cycle continues.

Centre controllers also reach much lower than many people assume. Where small airports have no tower or no approach control, the centre can handle traffic all the way down to the ground.

Radar, ADS-B and datalink

Two kinds of radar built the system. Primary radar sees anything that reflects its signal, cooperative or not. Secondary surveillance radar interrogates each aircraft's transponder, which replies with its code and pressure altitude, the data that puts a callsign and flight level next to every blip.

Increasingly, aircraft announce themselves. With ADS-B, an aircraft broadcasts its own satellite-derived position, altitude and identity, which ground stations pass to controllers. The FAA has required ADS-B Out in most US controlled airspace since 1 January 2020.

Voice radio is also giving way to text. Controller-pilot datalink communications, or CPDLC, lets controllers send clearances and frequency changes as messages the crew accepts with a button. It began over the oceans, where radio is poor, and the FAA's Data Comm programme has been extending it from towers into the en-route centres.

The FAA's own explainer on how its towers, TRACONs and en-route centres pass a flight from one controller to the next.

Over the ocean

Far from land there is no conventional radar coverage, so oceanic control has traditionally been procedural. Aircraft fly assigned routes, altitudes and speeds, report their position at set points and are kept much further apart than over land.

The North Atlantic is the busiest example. Its eastern half, the Shanwick oceanic area, is run by NATS from Prestwick in Scotland, while Gander handles the western side. Today many aircraft report their position automatically by satellite using ADS-C and exchange messages with controllers by CPDLC, which has allowed the spacing to shrink.

NATS explains how its controllers at Prestwick manage the Shanwick oceanic area over the North Atlantic.

Europe: ACCs, UACs and Maastricht

Europe uses the same building blocks with different names. Towers and approach units work much as they do in America, but the en-route job falls to area control centres, or ACCs, run by national providers such as NATS in Britain, DFS in Germany, DSNA in France and ENAV in Italy. NATS runs the en-route airspace over England and Wales from Swanwick in Hampshire.

Some countries split the sky further, with upper area control centres for high-altitude traffic. The most unusual is Eurocontrol's Maastricht Upper Area Control Centre, which manages the upper airspace from 24,500 feet over Belgium, the Netherlands, Luxembourg and north-west Germany. It handled 1,829,242 flights in 2025.

Above the national providers, Eurocontrol's Network Manager balances demand across the continent, handing out departure slots when sectors are about to overload. That is why a European flight can sit at the gate on a sunny day: somewhere along its route, a sector is full.

The terminology differs, but the principle is the same on both sides of the Atlantic, and it is the simplest summary of how air traffic control works: a flight is never alone. It is always someone's traffic, and the handoff to the next controller is already being prepared.

Sources: Federal Aviation Administration (Air Traffic By The Numbers; FAA Order JO 7110.65 chapters 4 and 5; The Air Up There podcast, 'How Air Traffic Works', 2020), US Department of Transportation (electronic flight strips release, February 2026), FAA Safety Team (ADS-B Out mandate), Flying Magazine (Data Comm), Honeywell Aerospace (en-route CPDLC), EUROCONTROL (MUAC 2025 performance release), BBC (via AOL), NATS, NATCA, Wikipedia