Insights · Aviation Safety · 8 November 2026

Icing, Turbulence and Overspeed Explained: The Aviation Terms in the Palm Beach Learjet Reports

The NTSB has not determined a cause. This guide explains the aviation terms in the public record, from supercooled ice to overspeed to the tail, without drawing conclusions the evidence does not support.

Business jet in flight above a towering storm cloud at high altitude

What the NTSB has and has not said

On Sunday 1 November 2026, a Bombardier Learjet 60, registration N722BD, went down on Military Trail near 10th Avenue North in Greenacres, Palm Beach County, Florida. Nine people were aboard, two crew and seven passengers, and all of them were killed. One man in a car on the ground suffered a leg injury. The National Transportation Safety Board opened investigation ERA27MA026. This article does not try to say what happened. It tries to explain the physical and operational vocabulary that has appeared in news coverage, so that readers can follow the investigation as it develops and judge early claims with care.

The first point is the most important one. As of 8 November 2026, the NTSB has not determined a cause. It has described some evidence and some open questions. NTSB member Tom Chapman said on 2 November: "We do have some evidence that there was some breakup of the aircraft before impact." On 3 and 4 November he confirmed that the left elevator, the tail surface that controls pitch, separated before impact and was found in three pieces between 1.3 and 3.8 miles from the crater. The Board has said it does not yet know whether that separation caused the dive or resulted from the forces of the dive. Weather, the Board says, is "a major focus".

What the public record contains, then, is a timeline drawn from air traffic control audio and flight tracking, a set of statements from the NTSB, and a great deal of commentary. Those three things should be kept apart. Timelines are largely facts that can be checked. NTSB statements are authoritative about what the Board has found so far. Commentary, including from respected former officials and analysts, is informed opinion and should be read as such.

The reported sequence is short. The jet left Austin-Bergstrom International Airport at about 10:37 a.m. Eastern time, bound for North Eleuthera in the Bahamas, and reached a cruise altitude of about 41,000 feet by about 11:00. For roughly 90 minutes nothing notable was recorded. At 12:31 p.m. a controller warned of moderate precipitation five miles away at about 41,000 feet and offered a left or right option. At 12:36 the crew was told to descend to 35,000 feet, and at 12:40 deviations left or right were authorized. At 12:47, at 35,000 feet, the crew reported what was described as "light clear ice" that was manageable, moderate turbulence with altitude deviations of about 700 feet, and an outside air temperature of minus 47 C (minus 53 F). At 12:48:00 the crew asked to climb to 37,000 feet, which was approved. Less than 30 seconds later the crew reported descending. By 12:48:28 a mayday call had been made, and at 12:48:36 the descent rate was reported at about 20,000 feet per minute with an "overspeed condition". The final transmission, at 12:49:17, was "We're out of time."

The rest of this article takes the technical words in that sequence one at a time: altitude, temperature, ice, turbulence, overspeed, and the tail. For each, the aim is to describe how the phenomenon works in general, what it can do to an aircraft in general, and where the evidence in this case does and does not reach. No section should be read as a statement about what these two pilots did or did not do, or about why the aircraft behaved as it did. Those are exactly the questions the investigation exists to answer.

Nine people died. They included Adam Robinson, the founder of Retention.com and RB2B, his wife Helen Sharp, their children Emma and Charlie, the children's nanny Anita Maria Cetina, family friends Rachael Reed and Charmi Fehr, Captain Jeffrey Gregory and First Officer Nathalia Gadelha Filgueiras. The technical discussion that follows is offered with respect for them and for the families, and with awareness that families and neighbours on Military Trail are still living with the aftermath.

Flying at 35,000 feet: the environment

Business jets climb to altitudes that airliners also use, and often higher. There are good reasons. The air is thin, so drag is lower and fuel burn per mile improves. Much, but not all, weather lies below cruise level. And traffic is separated by altitude, so higher levels give a crew more room. In this case the aircraft cruised at about 41,000 feet and was later told to descend to 35,000. These are ordinary flight levels for the type of aircraft.

But altitude changes the physics of flight in ways that matter when something goes wrong. The most useful picture is that thin air gives a wing less to work with. A wing generates lift in proportion to the density of the air passing over it and to the square of the speed. At 35,000 feet the density is a fraction of what it is at sea level, so an aircraft must fly much faster through the air to carry the same weight. The instruments and the crew think in terms of indicated airspeed and Mach number, which are ways of describing speed relative to the air and relative to the speed of sound.

Cold air and the speed of sound

The outside air temperature reported at 35,000 feet was minus 47 C. That is very cold, but it is typical for those levels, and the standard atmosphere models used in aviation describe similarly low temperatures. Cold air has a lower speed of sound than warm air. That matters because a jet can only approach the speed of sound so closely before airflow over parts of the wing and tail begins to behave differently, with shock waves forming and control feel changing. Designers set a maximum operating Mach number to keep a margin from that region.

The narrow margin sometimes called coffin corner

At high altitude two limits move toward each other. On the slow side is the speed at which the wing stalls, meaning it can no longer produce enough lift. That speed, expressed as Mach number, rises with altitude because the air is thinner. On the fast side is the maximum Mach limit, which changes much less. Pilots and writers sometimes call the narrowing gap between them "coffin corner", a dramatic name for a real concept. The honest hedge is this: the width of the margin depends on the specific aircraft, its weight, the temperature, the load factor in manoeuvring, and other conditions, and a modern business jet at 35,000 feet is usually well inside a workable range. The margin is narrower than at lower altitude, and it narrows further in turbulence or a turn, because changes in airspeed and in the load on the wing eat into it.

It would be wrong to claim that this concept explains what happened to N722BD. The NTSB has made no such finding, and nothing in the public record says this aircraft was operating near either end of its envelope before the events at 12:48. The point of describing the narrow margin is simpler: at altitude, small disturbances have less room to be absorbed, which is why airline and business aviation training puts so much weight on managing speed and attitude early in any upset.

It also helps to understand what a crew has in front of it. At cruise, the autopilot flies the aircraft and the pilots monitor, adjust speed, talk to controllers and watch weather radar. A disturbance such as ice accretion, strong turbulence or a sudden change in airspeed can cause the autopilot to disconnect, handing a pilot an aircraft that is already in motion. That handover, in thin air, is one of the situations upset training prepares crews for. We return to it in a later section.

Icing at altitude: supercooled water and what it does

The first officer's reported text message, "I've got turbulence, I've got ice," and the crew's radio report of "light clear ice" at 35,000 feet have drawn attention to icing. A common first reaction is surprise: at minus 47 C, surely everything is frozen solid and there is no liquid water left to form ice? The answer is a useful piece of meteorology, and it is worth explaining carefully.

Supercooled water

Water does not always freeze at 0 C. Very small cloud droplets can stay liquid well below freezing, a state called supercooled. They freeze readily when they strike something, such as the leading edge of a wing. In ordinary clouds, supercooled droplets are mostly found in the temperature range between about 0 C and minus 20 C, and become rarer as temperatures fall, because ice crystals form more readily at lower temperatures. At minus 47 C, liquid droplets would normally be unusual.

Deep convective clouds, the towering thunderstorm type, are the exception. Inside a strong updraft, air rises so quickly that moisture is carried to very high levels before it has time to freeze out. Meteorologists describe how strong updrafts can loft supercooled liquid droplets to very cold levels. The exact amount depends on the storm, and no one can say from the outside how much liquid water was present in the air N722BD flew through. That is a question for the NTSB's meteorology group.

The type of ice also varies. Crews and forecasters use terms such as rime ice, which is opaque and forms from small droplets freezing on impact, and clear ice, which is smooth and forms from larger droplets that spread before freezing. Crystals of ice can also pass through engine inlets and probes, a separate hazard sometimes described as high altitude ice crystal icing. The crew's report mentioned "light clear ice", and descriptions such as "light" and "manageable" refer to the crew's own assessment at that moment.

Effects on lift, drag and the tail

Ice matters because wings and tails are carefully shaped. A thin layer of ice, particularly on the leading edge, changes the shape of the airfoil. In general terms, ice can reduce the maximum lift a wing can produce, raise the speed at which it stalls, and increase drag, so the aircraft needs more thrust to hold speed. Ice can also add weight, though this is usually minor compared with the aerodynamic effects.

The tail is a special case. The horizontal stabiliser and elevator generate a downward force that balances the aircraft in pitch. Because the tail surfaces are thinner than the wings, ice tends to accumulate faster on them, a point made in many icing training materials. If the tail loses effectiveness because of ice, pitch control can change in ways the pilot does not expect. This has been described in accident literature on other aircraft and other events. It is background knowledge only. Nothing published about N722BD says that ice formed on the tail of this aircraft. Mary Schiavo, a CNN analyst and commentator, mentioned possibilities such as an engine flameout or ice on the wings or tail as things to examine, and she placed focus on "the weather, the storm cell, the icing." These are her views as an analyst, not NTSB findings.

Probes, sensors and engines

Aircraft rely on probes to measure airspeed and altitude, usually pitot tubes and static ports. They are heated to prevent ice. If a probe is blocked or provides incorrect data, the airspeed shown to the crew and to automated systems can be wrong. This is a recognised category of event in aviation safety, with well-documented cases on other aircraft types. Ice ingestion into engines is another recognised hazard, which is why engine anti-ice systems exist. The recovery of a portion of one engine on 5 November gives investigators something to examine, but no findings have been announced.

Turbulence and convective weather

A thunderstorm is a heat engine. Warm, moist air rises in updrafts, cools, condenses and releases heat that drives the rise further. Rain and ice fall and drag air down, producing downdrafts. Where these opposing motions meet, the air is violent. The cloud top of a strong storm can extend far above the altitudes where jets cruise, and the turbulence is not limited to the cloud itself: it can extend some distance beyond the visible edge.

According to the timeline, the controller at 12:32 warned of "moderate to extreme precipitation" and described the turn toward Freeport as looking "pretty bad". CNN quoted the controller as saying, "There is moderate to extreme precipitation 11 o'clock, and 25 miles. 100 miles in diameter." The jet flew south around the storm before being cleared to turn east. Meteorological analysis reported in the press indicates the jet passed through a stronger storm cell several minutes before the rapid descent, though it is unclear whether weather contributed.

What radar can and cannot show

Both air traffic control radar and an aircraft's onboard weather radar detect precipitation, not turbulence directly. Heavy rain returns a strong signal, and strong returns often go with strong convection, but the relationship is imperfect. Radar can be attenuated, meaning a heavy storm near the aircraft can partly hide what lies behind it. Storm cells also grow and decay quickly, so the picture is minutes old. This is why deviation around cells is standard practice, and why guidance usually favours generous distances from strong returns. The reported authorisations to deviate left or right at 12:40 fit that practice.

Why a moderate report matters

Turbulence reports use words with specific meanings. Light turbulence causes slight, erratic changes in altitude or attitude. Moderate turbulence is stronger, causing changes in altitude or attitude and variations in airspeed, with the aircraft remaining in positive control; unsecured objects move. Severe turbulence causes large abrupt changes and momentary difficulty holding altitude and attitude. At 12:47 the crew reported moderate turbulence with altitude deviations of about 700 feet. That tells a listener the air was rough and that the aircraft was being moved around vertically, and it is the kind of report that matters for the next aircraft and for controllers. It does not, by itself, tell anyone how the next minute would unfold. Turbulence also changes the load on the airframe and the airspeed the aircraft sees, which links it to the next topic.

Overspeed and high-speed dives

At 12:48:36, the reported transmission included the words "overspeed condition", and CNN's timeline says the crew said this exceeds limits. The phrase has a precise meaning in aviation, and it deserves a plain explanation.

What overspeed means

Every aircraft has operating limits on speed. They are expressed in two ways. At lower altitudes the limit is an indicated airspeed, which relates to the loads on the structure and control surfaces. At higher altitudes the limit is a Mach number, which relates to the behaviour of the airflow as it approaches the speed of sound. Many jets are certified with a maximum operating speed and a higher design dive speed, with a margin built in between them. Exceeding the operating limit is an overspeed. Aircraft usually include warnings, often an audible horn or a flashing indication, to tell the crew when the limit is reached. Beyond the limit, three kinds of concern arise: structural loads can approach design values, particularly if the aircraft is also manoeuvring or in turbulence; airflow effects near the speed of sound can alter control feel and effectiveness; and the aircraft is harder to recover because pulling out of a steep dive imposes high load factors.

These are general statements. The specific limits for a Learjet 60 are set out in its flight manual and type certificate data, and this article does not give numbers for them because none appear in the public record for this accident.

What 20,000 feet per minute means

The timeline reports the descent rate climbing from about 8,000 to 10,000 feet per minute at 12:48:28 and to about 20,000 feet per minute eight seconds later, and about 22,000 feet per minute at 12:49:01. Other outlets report tracking-data descent rates higher, over 30,000 feet per minute at times, so the exact figures should be treated as approximate. To get a sense of scale, 20,000 feet per minute is about 333 feet per second, or about 227 miles per hour straight down, and that is only the vertical component of the aircraft's motion. CNN's graphic showed roughly 500 miles per hour at one point and 274 miles per hour at 8,750 feet. The FAA reported a descent from about 35,250 feet to 8,750 feet in the final minute, about 27,000 feet lost in one minute.

For comparison, a typical jet airliner on a normal descent might lose a few thousand feet per minute. A rate of 20,000 feet per minute is far outside normal operations. It indicates a steep nose-down attitude or a spiral, and it explains why an aircraft can pass from cruise altitude to the ground in about a minute. It does not, on its own, tell anyone why the descent began. A dive may follow structural failure, control surface failure, loss of thrust combined with a stall, or other events, and descent data alone cannot separate them.

Time (ET)Reported eventSource type
12:4735,000 ft, light clear ice, moderate turbulence, minus 47 CCrew radio report
12:48:00Asked to climb to 37,000 ft; approvedATC audio
12:48:28Mayday; descent 8,000 then 10,000 ft/minATC audio and tracking
12:48:36About 20,000 ft/min; "overspeed condition"ATC audio and tracking data
12:49:01About 22,000 ft/minTracking data
12:49:17Final transmissionATC audio

All values are as reported by the outlets listed at the end of this article and are approximate. The NTSB's own reconstruction, when published, will supersede them.

The tail and the elevator

The most specific physical finding announced so far is that the left elevator separated before impact. What does that mean, and why is it hard to interpret?

What an elevator does

On most aircraft the horizontal tail has a fixed or adjustable front part, the horizontal stabiliser, and hinged rear surfaces, the elevators. Moving the elevators changes the downward or upward force the tail produces, which rotates the aircraft nose up or nose down. This is pitch control. In normal flight, the pilot or autopilot makes small elevator inputs continuously. If one elevator were lost or damaged, the loss would change pitch control and balance dramatically, though how an aircraft would respond depends on many details that investigators will establish from the wreckage.

Separation before impact can point two ways

The NTSB's statement is carefully worded: it does not yet know whether the separation caused the dive or resulted from the forces of the dive. Both are physically plausible in principle, and that is the whole difficulty.

Peter Goelz, a former NTSB managing director, told CNN: "My guess is the elevator is a result of the disrupted flight, not the cause of it." Mary Schiavo, the CNN analyst, said parts most likely came off "in this horrific dive". Both are commentators, not the NTSB, and the Board has not endorsed either view. Their remarks are best read as experienced professionals explaining how they would weigh the evidence so far.

What the wreckage can reveal

Investigators can learn a lot from the fracture surfaces on separated parts. In general, structures specialists look at the direction of failure, whether a fracture is the clean, grainy pattern associated with fatigue or the rough, deformed pattern of overload, the sequence in which pieces departed, and witness marks, which are scratches, paint transfers or impact damage showing where one part touched another. The distances matter too. Pieces found 1.3 to 3.8 miles from the crater may have separated in flight, and investigators can compare landing points with wind data. These are analytical methods used in many investigations. Results will take time to prepare, and in some cases the metallurgical laboratory work alone takes weeks.

Crew workload and the first seconds of an upset

Some readers will want to know how a crew experiences an event like this. This section is general human factors background. It makes no statement about what the Captain or the First Officer did, thought or decided. Investigators have not said what role either pilot had in flying the aircraft, and being second in command at 20 is not evidence of anything about the cause. First Officer Nathalia Gadelha Filgueiras earned her pilot licence at 17 through a school aviation academy, and Newsweek reported that her licence showed the rating to fly this Learjet type. Her mother, Camila Corteleti, said of her: "She loved the freedom. It was completely different seeing the world from the top." NASA administrator Jared Isaacman publicly defended her against online criticism, as did one of her flight instructors.

What an upset is

Aviation safety bodies describe an upset as an unintentional event in which an aircraft exceeds the attitudes or speeds normally expected, commonly a steep pitch, a steep bank, or an airspeed inappropriate to the conditions. Upsets in flight have been among the leading categories of fatal accidents for years, according to industry safety reviews. Upsets can begin in many ways: turbulence, icing, a system fault, an automation disconnect, a mistaken input or a combination.

The first seconds

Human factors research describes a few consistent features of sudden, unexpected events. Attention narrows. Warnings and alerts can arrive together: stall warnings, overspeed horns, autopilot disconnect tones and caution messages. Physical cues can be misleading, because the inner ear and the body's sense of motion are unreliable without visual reference, and in cloud or at night the instruments are the only reliable picture. In severe turbulence the instrument panel itself can be hard to read. Crews are trained to expect that the first few seconds are the most difficult, and that a deliberate, practised sequence works better than improvisation.

Workload is also affected by what came before. In the minutes before 12:48, the reported timeline shows a crew coordinating deviations around a large storm, communicating with the controller, managing ice and handling turbulence. That is a busy flight deck, in the way that any flight near convective weather is busy. It is a description of the context, not an evaluation of anyone's performance.

Upset prevention and recovery training

The aviation industry has invested heavily in upset prevention and recovery training. The general approach teaches recognition, then a standard set of steps: disconnect automation, reduce the angle of attack, manage thrust, roll to wings level, and recover carefully without exceeding structural limits. In a high-speed dive, recovery needs a careful pull that avoids overloading the airframe, which is a narrow task if the aircraft is already beyond its normal speed. Simulators cannot reproduce every real-world situation, and no training removes risk entirely. Whether any recovery was possible in this event, and what was attempted, are questions only the investigation can address.

Cockpit voice recorder, flight data, and why recorders sometimes fail

Recorders are the investigator's most direct source on what an aircraft and its crew were doing. Two types are usually discussed. The cockpit voice recorder captures conversations and sounds in the cockpit, such as alerts, switch movements and engine noise. The flight data recorder captures parameters such as airspeed, altitude, heading, control surface positions and engine settings. Some aircraft carry a combined unit. Requirements depend on the type of aircraft and the type of operation, and the details for N722BD are not established in the public record.

What is known is limited. Maintenance records show a cockpit voice recorder was installed, but as of 6 November it had not been found, and the NTSB could not confirm one was aboard. That is a careful statement: not found does not mean it did not exist, and it does not mean it failed. The impact was almost vertical, with a large crater that filled with water from storms, which slowed recovery. Crews drained the water and began excavating on 4 and 5 November, and a portion of one engine was recovered. Debris was also found on rooftops nearby, and the public has been asked to report debris and not touch it.

Why recorders sometimes fail

Recorders are built to survive extreme impacts and fire, yet they can still fail to provide usable information. Causes in past events include physical destruction, damage to the memory module, fire, submersion, power loss before the final moments, and mechanical faults. CNN has noted one example relevant to this class of aircraft: after a Learjet 55 crash in Philadelphia in January 2025, the recorder jammed and had no recording of the fatal flight, according to an NTSB report from March 2026 as reported by CNN. That example is cited only to show that recorder problems are a recognised issue. It says nothing about the unit on N722BD, which is a different aircraft and a different event.

If a recorder is not found, or if it is recovered with no usable data, investigators turn to other sources: air traffic control audio and radar, ADS-B tracking data from services such as FlightRadar24, weather records, maintenance records, engine and wreckage examination, witness accounts and video. The public has been asked to send in dashcam footage, and a dashcam video of the impact exists. The investigation can still reach conclusions without recorders, though it can be slower and the findings may carry more uncertainty on some questions.

How the NTSB investigates

The NTSB is an independent federal agency. It does not assign legal blame. Its mission, in the words of Tom Chapman, is "to understand not just what happened, but why it happened, and to recommend changes to prevent it from happening again." The process follows a recognisable path.

Two features of this process deserve emphasis. First, the NTSB reports to the public in stages, and the early stages are deliberately narrow. Second, parties to the investigation, such as the FAA, the manufacturer and the operator, may supply technical help under the NTSB's direction, but they do not control its conclusions.

What is realistic to expect, and the danger of early theories

In the weeks after a major accident there is a strong appetite for an answer. Social media fills with theories, some from knowledgeable people and many not. A few cautions are worth setting out.

The evidence is incomplete. The cockpit voice recorder had not been found as of 6 November. A theory built on the timeline alone has to guess about everything the wreckage will eventually show.

Visible facts can be consistent with more than one story. Icing, turbulence, overspeed and a separated elevator are each real and each reported. How they relate to one another, and in what order, is the open question. Attributing the accident to any one of them now would be speculation, whatever its source.

Early theories also have costs. They can land on individuals who cannot respond, including the Captain, the First Officer, the operator and the owner. The NTSB has said weather is a major focus, but it has not said weather was the cause. The NTSB found no evidence of a Mar-a-Lago connection, according to a Fox News headline.

A realistic expectation is this. In the near term, expect a preliminary report in about a month with timeline facts, a continuing search for the voice recorder, and possibly further statements about the tail and engine. Expect no probable cause for a year or more. Some questions, such as exactly what happened in the final minute, may remain partly uncertain if recorder data is limited.

Frequently asked technical questions

Can ice form at minus 47 C?

Yes, in principle. Supercooled droplets can be carried far above the freezing level by strong storm updrafts. The crew reported light clear ice at that temperature.

Does an overspeed say why the aircraft descended?

Not by itself. It means a speed limit was exceeded. Why it happened is for the NTSB to establish.

Did the elevator separation cause the dive?

Unknown. The NTSB says it does not yet know whether the separation caused the dive or resulted from it.

Why does the investigation take so long?

Wreckage analysis, laboratory work, data review and a public process take time. CNN reports one to two years for probable cause.

When will we learn more?

A preliminary report is expected in about a month, per CNN, with factual detail rather than a cause.

Can the public help?

Yes. The NTSB asks people to report debris without touching it and welcomes dashcam footage.

Sources and further reading

Facts about the event are drawn from the outlets below and from NTSB statements as reported. Aviation explanations are general background.

Questions about aviation risk in your own household?

Obsidian Helm offers private, by-invitation advisory on technology, security and AI for principals. If a conversation would be useful, we are glad to listen.

Request Your Invitation

Frequently asked

Can ice really form at minus 47 C?

In principle, yes. Strong thunderstorm updrafts can carry supercooled liquid droplets to very high, very cold levels before they freeze. Such droplets freeze on contact with an aircraft surface. The crew of the Learjet reported light clear ice at minus 47 C at 35,000 feet. How much liquid water was present in that air is a question for the NTSB meteorology group, and no finding has been announced.

What does an overspeed condition mean?

Overspeed means an aircraft has exceeded a certified speed limit, expressed as airspeed at lower altitudes or as a Mach number higher up. Beyond the limit, structural loads and airflow effects can become a concern. The crew reported an overspeed condition at about 12:48:36. The reason it occurred has not been determined, and a high speed can be a result of an upset as well as part of its cause.

Did the elevator separation cause the dive?

That is not known. NTSB member Tom Chapman confirmed the left elevator separated before impact and was found in three pieces 1.3 to 3.8 miles from the crater. The NTSB does not yet know whether the separation caused the dive or resulted from the forces of the dive. Commentators such as Peter Goelz and Mary Schiavo have offered views, but they are not NTSB findings.

What does 20,000 feet per minute mean?

It means the aircraft was losing about 333 feet every second, roughly 227 miles per hour straight down, which is far beyond any normal descent. Reported figures vary between outlets and are approximate. The rate shows how quickly altitude was lost, but it cannot by itself show why the descent started, since several different events can produce a steep dive.

Why might a cockpit voice recorder not be found or not work?

Recorders are built to survive severe impacts, yet they can be destroyed, damaged, submerged or jammed. As of 6 November the recorder installed according to maintenance records had not been found, and the NTSB could not confirm it was aboard. CNN noted that after a Learjet 55 crash in Philadelphia in January 2025 the recorder jammed and had no recording.

When will the NTSB name a cause?

Not soon. CNN reports a preliminary report in about a month, around early December 2026, which sets out facts rather than a cause. Determining probable cause usually takes one to two years. Until then, any claim about the cause, including from commentators, is a theory. The NTSB has said weather is a major focus but has not said weather was the cause.

By Invitation Only

The office answers.
The rest is silence.

Tell us, in confidence, what keeps you up. We reply privately, under NDA.

Request Your Invitation
Replies under NDA · Strictly Confidential