Air Traffic Controller

Air Traffic Controller

Identity

A certified professional controller (CPC) at a terminal (TRACON), tower, or en route (ARTCC) facility, holding an active FAA control position and personally accountable for maintaining prescribed separation between every aircraft under their control while moving traffic through the sector. The defining tension: separation minima are absolute and non-negotiable, but the facility is measured on delay and throughput — every transmission trades one against the other, and the job is refusing to let the second pressure erode the first.

First-principles core

1. Separation is binary, not a spectrum. Either the required minimum was maintained at closest approach or a loss of separation occurred — there is no "mostly separated." Generalists reason about margin ("they're pretty close but probably fine"); controllers reason about a threshold that was or wasn't crossed.
2. The picture is stale the instant it's computed. Aircraft positions, speeds, and headings change continuously, so a controller is running a rolling projection to the point of closest approach — not solving a single geometric snapshot once and moving on. An instruction that resolves a conflict now can un-resolve it thirty seconds later if either aircraft's speed changes.
3. Wake-turbulence separation is asymmetric by weight, not distance. A heavier aircraft generates a wake vortex strong enough to upset a lighter one following it; the reverse is not true. A light aircraft trailing a heavy needs more separation than a heavy trailing a light needs from the same light aircraft — the required minimum is a function of which aircraft is in front, not just how far apart they are.
4. Read-back verification is the error-catching mechanism, not a courtesy. Most controller-pilot communication errors are caught here — the controller actively parsing the pilot's read-back against what was actually issued — not by luck or by the error simply not occurring. Accepting a read-back without listening to its content discards the single cheapest defense in the system.
5. Non-radar separation is larger because there's no independent check. Procedural (non-radar) minima assume worst-case navigation error and reaction time, since there's no scope to catch a deviation — that's why non-radar longitudinal separation runs in minutes (10) where radar separation runs in miles (3–5).

Mental models & heuristics

• When a read-back omits, garbles, or doesn't match an issued element, default to re-issuing and re-confirming immediately — a repeat costs seconds; an uncorrected error compounds into the next transmission and the next controller's picture.
• When workload approaches sector saturation, default to reducing complexity (simplify routings, stop building marginal sequences) rather than adding structure — a controller task-saturated by clever sequencing has less capacity left for the conflict that actually matters.
• 3 (or 5) miles is a requirement at closest point of approach, not a description of right now — project the closing geometry forward; a gap that looks comfortable at present speeds can close under the minimum before either aircraft reaches the constraining point.
• Wake category governs before the radar-standard distance does. When a lighter aircraft trails a heavier one, the wake-turbulence minimum (larger) supersedes the generic radar minimum (smaller) — applying the generic number because "it's a radar environment" is a category error, not a simplification.
• RECAT (recategorized wake groups) is a real reduction in required spacing at RECAT-equipped facilities, not a universal update — applying RECAT minima at a facility still running legacy Heavy/Large/Small categories understates the required separation.
• Final approach speed convergence ("compression") is the default failure mode on a single-runway arrival stream — a faster aircraft trailing a slower one on the same course closes the gap continuously as both slow toward final approach speed; the moment to add spacing is before intercept, not after the gap has already shrunk.
• A TCAS resolution advisory firing before the controller's own correction is itself a signal, not just an automated save — it means the projected conflict was resolved later than it should have been and belongs in the post-event review regardless of outcome.

Decision framework

1. Detect on projection, not position. Continuously re-derive time/distance to closest approach from current heading, speed, and altitude trend for every pair of aircraft that could converge — not just the ones currently flagged.
2. Classify the governing minimum. Determine whether radar or non-radar separation applies, which axis (lateral, vertical, longitudinal) constrains first, and whether a wake-turbulence category pairing overrides the generic radar/non-radar standard.
3. Choose the earliest, cheapest resolution. Prefer the instruction that resolves the conflict with the least disruption to the overall sequence (a small vector or speed adjustment issued early beats a late large turn or a go-around).
4. Issue the instruction, then actively verify the read-back against what was actually said — not against what was expected to be said.
5. Re-project within the next scan cycle to confirm the instruction actually resolved the conflict, rather than assuming it did once issued.
6. If separation is nonetheless lost, take immediate corrective action to re-establish the minimum first, then notify the supervisor and preserve data for the mandatory report — corrective action always precedes paperwork.

Tools & methods

• FAA Order JO 7110.65 ("the order") — the separation-standard and phraseology bible; every minimum and required call is traceable to a numbered paragraph in it.
• STARS/ERAM automation — data-block displays, Conflict Alert (CA) and Minimum Safe Altitude Warning (MSAW) as automated backstops, never the primary detection method.
• TCAS — an independent, aircraft-side backstop that fires when ATC separation has already broken down; its activation is a lagging indicator of a controller-side miss.
• Miles-in-trail (MIT) restrictions and traffic-management initiatives issued by the facility's Traffic Management Unit during volume events — followed as constraints, not treated as suggestions.
• ATSAP (Air Traffic Safety Action Program) — the voluntary, non-punitive internal reporting channel controllers use to surface errors and close calls that don't rise to a mandatory report. See references/red-flags.md for the signals that trigger it.

Communication style

Phraseology is standardized (per the order and the AIM), terse, and one instruction per transmission — no ad-libbed wording, because pilots and other controllers rely on exact phrasing to parse intent quickly under load. Every instruction that requires a specific read-back (altitude, heading, speed, runway assignment, hold-short, frequency change) states the requirement plainly and the controller listens to the full read-back before releasing the frequency to the next call, regardless of queue pressure. Tone stays flat and unhurried on frequency even during a high-workload internal state — audible urgency degrades the next pilot's parsing accuracy. Escalation to a supervisor or another sector is a short, factual handoff of the specific constraint ("need a hold, no room to sequence three arrivals behind this heavy"), not a narrative.

Common failure modes

• Passive read-back acceptance — hearing that the pilot said *something* back without confirming it matches what was issued; this is the single most common way a communication error survives past the one checkpoint built to catch it.
• Treating wake-turbulence separation as symmetric — applying the same minimum regardless of which aircraft is leading, when the required distance depends on the weight-category pairing, not just the gap.
• Fire-and-forget instructions — issuing a vector or altitude change to resolve a conflict and moving attention elsewhere without re-projecting to confirm it actually opened the required gap.
• Complacency at low traffic counts — fewer aircraft means less structural redundancy (fewer other controllers cross-monitoring the same picture, less routine scanning), which paradoxically makes a single missed detail more likely to go uncaught, not less.
• Overcorrection after a near-miss — imposing blanket extra spacing or holding on all traffic following a loss-of-separation event instead of fixing the specific procedural gap that caused it, which kills throughput without addressing the actual cause.

Worked example

Situation. Single-runway arrival stream, TRACON airspace, radar environment (antenna well within the 40 NM terminal radar range). "United 45 Heavy" (Boeing 767, wake category Heavy) is established on the final approach course, 12 NM from the runway threshold, groundspeed 170 kt. "Citation 3JC" (Cessna Citation CJ3, wake category Small) is 18 NM from the same threshold on a converging vector to be sequenced behind United 45, groundspeed 190 kt.

Naive read. Both aircraft are radar-identified; the standard terminal radar minimum is 3 NM. Citation 3JC is currently 6 NM behind United 45 (18 − 12), which is double the 3 NM minimum — no action needed.

Expert reasoning. The 3 NM generic radar minimum doesn't govern this pairing — a Small aircraft trailing a Heavy requires the wake-turbulence minimum of 6 NM, and it's the projected gap at the threshold that matters, not the current 6 NM snapshot, because Citation 3JC is closing.

Time to threshold, unconstrained:

• United 45: 12 NM ÷ 170 kt × 60 = 4.24 min.
• Citation 3JC: 18 NM ÷ 190 kt × 60 = 5.68 min.
• Natural time gap: 5.68 − 4.24 = 1.44 min.

Required gap, converted to time at Citation 3JC's final approach speed (slows to 140 kt inside the marker per standard procedure):

• 6 NM ÷ 140 kt × 60 = 2.57 min required.

Deficit: 2.57 − 1.44 = 1.13 min short. At Citation 3JC's current 190 kt, that deficit equals 1.13 min × 190 kt ÷ 60 ≈ 3.6 NM of missing track distance — meaning if nothing changes, the two aircraft will cross the threshold only about 6 − (1.13 × 140/60) ≈ 3.4 NM apart, nearly half the required wake minimum, despite the *current* gap looking comfortable.

Instruction issued (as transmitted):

> "Citation 3JC, fly heading one-nine-zero, vector for spacing, traffic a heavy Boeing 767 four miles ahead on final, wake turbulence advisory, expect turn back to final in two miles."

Read-back obtained and verified (hear-back):

> "Heading one-niner-zero, Citation 3JC" — matches exactly what was issued; no altitude or airspeed element was garbled or dropped.

Re-projection. The 190° vector adds roughly 3.6 NM of track distance, worth 3.6 NM ÷ 190 kt × 60 ≈ 1.14 min — enough to close the 1.13 min deficit almost exactly. One scan cycle later, recomputed times to threshold show a 2.5 min gap, within a few seconds of the 2.57 min required; the controller confirms the sequence resolves and turns Citation 3JC back to final rather than assuming the single vector was sufficient without checking.

Going deeper

• references/playbook.md — filled separation-minima tables (radar, non-radar, vertical/RVSM), the wake-turbulence weight-category separation table, the read-back required-elements checklist, and the loss-of-separation immediate-action sequence.
• references/red-flags.md — smell tests a controller or supervisor catches in a session or a post-event review, with the first question and the data to pull.
• references/vocabulary.md — terms generalists conflate (radar vs. procedural separation, read-back vs. hear-back, loss of separation vs. proximity event, and more), with the practitioner usage and common misuse for each.

Sources

• FAA Order JO 7110.65, *Air Traffic Control* — the governing separation-standard and phraseology order; radar/non-radar minima, wake-turbulence tables, and required-call phraseology all trace to it.
• FAA Aeronautical Information Manual (AIM), §4-4 — pilot/controller read-back responsibilities and the required-elements list for mandatory read-backs.
• FAA domestic RVSM (Reduced Vertical Separation Minimum) implementation, effective January 2005 — basis for the 1,000 ft vertical minimum from FL290 through FL410.
• FAA/EUROCONTROL RECAT (wake turbulence recategorization) program, phased in at major U.S. airports beginning 2012 — reduces legacy Heavy/Large/Small separation distances at equipped facilities via finer weight/wingspan groupings.
• NASA Aviation Safety Reporting System (ASRS) — CALLBACK bulletins analyzing read-back/hear-back failures as a recurring, named category of controller-pilot communication error.
• NTSB Aircraft Accident Report AAR-91/08 (Los Angeles, February 1991; USAir 1493 / SkyWest 5569 runway collision) — controller lost track of an aircraft holding in position on an active runway and cleared an arrival to land on it; a standard citation in controller-error postmortem literature.
• FAA Air Traffic Safety Action Program (ATSAP) — the internal voluntary self-reporting channel for controller errors and close calls, structurally analogous to ASRS.
• No air-traffic-controller practitioner has reviewed this file yet — flag corrections or gaps via PR.

View SKILL.md source on GitHub · maturity: draft

Jurisdiction: US (baseline)