Turbulence injuries in US commercial aviation cluster in the enroute phase of flight — level cruise at altitude, where the seatbelt sign is most often dark. In NTSB data covering 2008 to 2023, 113 of 136 turbulence accidents, or 83.1%, occurred enroute.
What the data does not show is how many of those occupants were unbuckled. No public dataset measures seatbelt compliance by flight phase, so the link between cruise-phase injuries and unfastened belts remains a strong inference, not a proven cause.
Somewhere over the Pacific, the seatbelt sign clicks off. The cabin comes alive — people stretch, queue for the lavatory, dig through overhead bins. It is the most ordinary moment of a long-haul flight, and it is also, according to a decade of US accident data, the moment when turbulence injuries are most likely to happen.
That is not a guess about human behavior. It is what the NTSB’s turbulence record shows about flight phase. Across 136 turbulence accidents involving US Part 121 carriers between 2008 and 2023, 113 — 83.1% — happened enroute, in level flight at altitude rather than during takeoff, climb, or landing.
The obvious explanation is that this is when passengers unbuckle. The problem is that no one has measured it.
Phase-by-phase seatbelt compliance has never been published, which means the standard advice to keep your belt fastened rests on an inference the data supports but does not prove. This article separates the two: what the record shows, and what it only suggests.
Where turbulence injuries actually happen
Turbulence is not an occasional hazard in US commercial flying. The NTSB tracks it as a distinct accident category for Part 121 carriers, and the injuries it produces are serious ones.
The phase breakdown is where it gets interesting. The NTSB’s 2009–2018 data classifies 40 accidents as enroute–descent and 34 as enroute–cruise, with a further 56 accidents and 61 serious injuries falling into a combined descent-and-approach grouping. Those categories do not stack — the combined figure already absorbs some of the individual phases, so adding them together would double-count.
| Flight phase | Accident count | Serious injuries |
|---|---|---|
| Enroute–descent | 40 | Not verified in the reviewed source material |
| Enroute–cruise | 34 | Not verified in the reviewed source material |
| Descent through approach | 56 combined | 61 combined |
| Source: National Transportation Safety Board (NTSB) | ||
The injuries themselves are not minor. They are the kind that come from being thrown against a cabin surface or the floor, not from a gentle jolt.
The measurement nobody has taken
Here is where most coverage of this topic takes a shortcut. It reports the enroute concentration, notes that the seatbelt sign is usually off during cruise, and concludes that unbuckled passengers are the cause.
That conclusion is reasonable. It is also unverified.
The NTSB records accident phase in one column and whether an injured person was belted in another. Nowhere does the study compute that the enroute share equals the proportion of unbelted passengers. The reviewed official material turned up no public dataset that measures seatbelt compliance by phase at all.
The windows do not line up either. The enroute share comes from a 136-accident record spanning 2008–2023, while the injury counts come from a 2009–2018 study. Neither measures compliance, and whether the two can be read together is unresolved.
One documented case shows how it plays out. On October 19, 2011, a passenger on a scheduled American Airlines flight was walking back from the lavatory during climb when the plane hit moderate turbulence that had not been forecast, and he went down in the aisle with a fractured ankle.
The NTSB logged a single serious injury and named non-compliance with the lit seatbelt sign as probable cause. What the report captures is one incident — not evidence that non-compliance drives the wider cruise-phase pattern.
Superdeals cover departures from North America, Europe and Australia.
Flight deals most people never see
Air Traveler Club monitors thousands of routes for pricing anomalies that traditional search engines miss. With ATC Superdeals members save up to 80% on each flight vs. regular fares.
Too good to be true? Check it out yourself. It's free.
Delivered weekly. Unsubscribe anytime.
Why flight attendants get hurt more than passengers
The asymmetry in the injury data is stark. Flight attendants accounted for 97 of the 123 serious injuries logged from 2009 to 2018; passengers accounted for the remaining 26.
The reason is duty, not carelessness. Cabin crew are required to be on their feet — securing the cabin, checking on passengers, moving through the aisle — in conditions where a seated passenger has the option to stay buckled. The job puts them in the aisle precisely when the aircraft is most likely to lurch.
That reframes the seatbelt instruction. It is not a rule aimed at people who don’t know better. It protects passengers because they, unlike crew, can choose to stay strapped in.
The sign is a permission slip, not a safety clearance
In 14 CFR Part 121 operations, the fasten-seat-belt sign belongs to the pilot in command. Regulations require it to be lit for surface movement, takeoff, and landing — and at any other moment the pilot judges necessary. While it is on, passengers must buckle up, and FAA guidance goes further, advising them to stay belted whenever they are seated.
A dark sign, in practice, is permission to move around. It is not a clearance.
That distinction matters because of how turbulence is detected — or isn’t. ATC radar cannot pick up turbulence directly, and clear-air turbulence forms at high altitude where the precipitation echoes weather radar relies on are absent.
Forecasts and pilot reports can flag risk, but they cannot guarantee a quiet cabin. An encounter can still arrive in ordinary cruise with the sign dark. That is the operational exposure — it does not prove anyone aboard was unbelted, but it explains how a cruise-phase encounter comes with no warning and no instruction to sit down.
What this means for you
The practical takeaway is grounded in accident data rather than airline boilerplate: keep your belt loosely fastened whenever you are seated, even when the sign is off. The cruise phase is the longest stretch of a long-haul Asia-Pacific flight and the one where the sign is most often dark — the exact window the data flags.
Set expectations about who gets hurt, too. Crew are injured far more often because their job requires standing. For a passenger, the belt is the one variable you control.
A caveat on scope: every figure here comes from US Part 121 operations. The datasets do not cover Asia-Pacific carriers or routes, so applying them to a flight on a foreign carrier is an analogy, not a measurement.
Reporting by
Key terms
- NTSB
- The National Transportation Safety Board is the US agency that investigates civil aviation accidents and publishes the findings. Its Part 121 turbulence study covers 2009–2018 and recorded 123 serious injuries, 26 to passengers and 97 to flight attendants. For this article, the agency matters because its accident reports are the place where the phase of each turbulence event is recorded, which is what makes the enroute pattern visible at all.
- Part 121
- Part 121 is the section of US federal aviation regulations that governs scheduled air-carrier operations — the airlines most travelers fly. It sets the rules for those carriers, including the requirement that the fasten-seat-belt sign be lit for surface movement, takeoff, and landing, and whenever the pilot in command considers it necessary. Every figure in this article comes from Part 121 flying, which is why the findings describe scheduled US airlines rather than private or business aviation.
- Enroute
- Enroute is the phase of flight between climb and descent, when the aircraft is in level cruise at altitude. It is the longest phase of most flights and the one where the seatbelt sign is most often switched off. In this article, enroute is the phase that matters most, because it is the stretch where the sign is dark and the compliance data is missing.
- Clear-air turbulence
- Clear-air turbulence is turbulence that occurs in cloudless sky, with no visible warning to the flight crew or the cabin. It forms at high altitude, where the precipitation echoes that weather radar uses as an indirect cue are absent, so radar cannot detect it directly. That is why the cruise-phase exposure in this article is hard to design around — the warning the cabin would need is the one the instruments cannot give.
Questions? Answers.
When do turbulence injuries happen most often?
The NTSB’s phase data for 2009–2018 classifies 40 accidents as enroute–descent and 34 as enroute–cruise, with 56 accidents and 61 serious injuries in a combined descent-and-approach grouping. Across the broader 2008–2023 record, 113 of 136 turbulence accidents — 83.1% — occurred enroute.
Does the NTSB data prove that unbuckled passengers cause cruise-phase turbulence injuries?
No. The NTSB records accident phase and injury seatbelt status separately, and no public dataset measures phase-specific seatbelt compliance. The enroute concentration cannot be converted into a non-compliance percentage.
How often do serious turbulence injuries occur each year?
The NTSB counted 123 serious turbulence injuries across 2009–2018, an average of about 12.3 per year across the US Part 121 system. No verified denominator exists for a per-flight or per-passenger rate.
Is it safe to walk around the cabin when the seatbelt sign is off?
The sign is the pilot’s call under 14 CFR Part 121, and a dark sign is permission to move around, not a safety clearance. ATC radar cannot detect turbulence directly, so an unexpected encounter can arrive with no warning.
