On November 1, 2026, at about 12:50 p.m., a 2004 Learjet 60 (N722BD) crashed near Military Trail and 10th Avenue North in Greenacres, Palm Beach County, Florida. Aurora Aviation, Inc., doing business as Upper Jets, was operating it under Part 135 on a charter from Austin-Bergstrom to North Eleuthera, Bahamas. All nine aboard, two crew and seven passengers, were killed in a near-vertical, high-speed impact that left a large crater in the roadway. A man in a vehicle on the ground suffered a leg injury. The NTSB is leading the investigation (ERA27MA026). Expect a preliminary report in about a month and probable cause in more than a year.
According to the NTSB, at about 12:47 p.m., level at FL350, the crew reported light clear ice, which they called manageable, and moderate turbulence with altitude excursions of about 700 feet. OAT was −47°C. They reported picking up precipitation, requested FL370, and received it. Less than 30 seconds later, they reported descending. Then came the Mayday and a report of an overspeed at 20,000 fpm. A later call reported 22,000 fpm and a 60-degree bank. About 35 seconds after the first Mayday, the last words were "We're out of time." The FAA says the airplane fell from 35,250 to 8,750 feet in about a minute. ADS-B peaked at −32,640 fpm, the largest value that data field can encode, so the actual rate may have been higher.
The radio calls are tough to listen to. In less than a minute, the crew went from routine reports to knowing they weren't going to make it. Let's look at what appears to have happened.
WHAT THE WRECKAGE TELLS US SO FAR
The NTSB recovered most of the left elevator in three pieces between 1.3 and 3.8 miles from the impact site. Parts don't travel that far at ground impact, so it came off in flight. Dashcam video appears to show the horizontal stabilizer missing during the final plunge. When asked about the video, NTSB Member Tom Chapman said that was possible. The NTSB hasn't determined whether the separation started the upset or resulted from the dive loads. Chapman noted that at those speeds the elevator might not have helped even if it had stayed attached. Some outside investigators think the short debris distances point to a separation low in the descent rather than at cruise altitude. Weather remains a central focus.
THE HIGH-ALTITUDE ENVELOPE
I've no time in a Lear, but high altitude can be unforgiving in any jet. Coffin corner is the narrowing band between the low-speed buffet (stall) and the high-speed Mach buffet. The higher you go, the narrower it gets. FL350 is well below the Lear 60's 51,000-foot ceiling, so in smooth air there's a workable margin. But ice, weight, bank, and turbulence eat into that margin from the bottom, and gusts and updrafts push you toward MMO from the top. Climbing to FL370 gives up some of whatever is left.
A wing stalls when it exceeds its critical angle of attack: the airflow separates from the upper surface, and lift falls off sharply. Indicated stall speed rises with weight and load factor. At altitude, it also rises because compressibility lowers the wing's maximum lift coefficient. Ice makes all of this worse. It reshapes the airfoil, triggers early separation, adds weight and drag, and can restrict the controls. That's not what you want when the margin is already thin, and the Lear 60's relatively high wing loading leaves little room once an upset begins.
One detail deserves a closer look. Supercooled water essentially can't exist below about −40°C, so clear ice at −47°C is unusual. It suggests a strong updraft carrying liquid water to altitude, which is exactly what a rapidly building cell does. These storms also carry dense concentrations of ice crystals. The crystals rarely stick to a cold airframe, but they can affect engines and block pitot probes, and then your airspeed indications can't be trusted.
THE STORM AND THE RADAR PICTURE
I've spent more than 40 years deliberately staying clear of thunderstorms. Few things in flying are more nerve-racking than not knowing how close you are to a bad cell. Airborne radar is very good at showing you where not to go, provided tilt, gain, and mode are set right, and you're scanning well ahead instead of reacting to what's on the nose. This airplane was required to have that radar. Under 14 CFR 135.175, a large transport-category airplane carrying passengers must have approved weather radar, and when thunderstorms may reasonably be expected along the route, an IFR flight can't begin unless it's working. The questions are what the radar showed, how the crew was using it, and whether their read of the weather matched what they were flying into.
The cell was growing fast. A post-accident analysis of NEXRAD data by MyRadar meteorologist Matthew Cappucci estimates its top rose from just under 33,000 feet at 12:36 p.m. to about 48,300 feet by 12:56 p.m. That's more than 15,000 feet in roughly 20 minutes, or about 750 fpm on average. The analysis puts the airplane entering the cell at about 12:48, while it was still building. You don't count on outclimbing that, especially in an iced-up jet at FL350.
Published readings put reflectivity near the airplane at about 34 dBZ, with a core of roughly 52 dBZ below. On the old NWS VIP scale, 52 dBZ is Level 5, "intense," the same word the AIM uses in its 20-mile rule. Moderate returns at your altitude over an intense core below show exactly what airborne radar can understate, because ice reflects far less energy than liquid water. If you scan level, you may see a modest return while the storm's real energy sits below you and rises toward you. That's why you tilt down. It's also why the AIM warns against using data-linked NEXRAD alone for tactical avoidance: the picture can be older than it looks, and this cell was changing by the minute.
The priority is early lateral avoidance with plenty of room. Climbing over or ducking under a cell once you're already in turbulence or precipitation is a riskier play, especially with thin performance margins. If you're inside, hold turbulence penetration speed, keep inputs and altitude changes small, and get out as smoothly as you can. The crew called the ice manageable and asked to climb almost in the same breath, which shows how fast things go downhill inside a building cell.
This is why you don't play around with these things. More than once, I've turned off course to avoid a storm without an ATC clearance. Sometimes the frequency was too congested to get a word in. Several times over the Indian Ocean, radio coverage was poor, or the storm static made talking to anyone impossible. The pilot in command is the final authority, and the regulations support that. 14 CFR 91.3 lets you deviate as needed to meet an emergency. 14 CFR 91.123 requires you to tell ATC as soon as possible, and you may be asked for a written report. I'd rather write that report than have someone else write mine.
THE STANDARD OF CARE
The FAA's guidance has been around a long time. AC 00-24C and AIM 7-1-27 advise pilots to:
- avoid by at least 20 miles any thunderstorm identified as severe or giving an intense radar echo, especially under the anvil;
- circumnavigate the entire area if it has 6/10 or greater thunderstorm coverage;
- regard as extremely hazardous any thunderstorm with tops of 35,000 feet or higher;
- not fly under a thunderstorm, or land or take off in the face of an approaching one; and
- not use datalinked NEXRAD as the sole means of negotiating a path through a thunderstorm area.
The AIM adds that no path between strong or very strong echoes separated by 20 miles or less may be considered free of severe turbulence.
None of this is a regulatory mandate under Part 91 or Part 135, but it carries real weight. Training programs and operator manuals build it in, and Part 135 crews are bound by their manuals. The NTSB measures crew decisions against it, and when the evidence supports it, the courts do too. The prohibition on careless or reckless operation in 14 CFR 91.13 applies to Part 135 flights too.
POSSIBLE SCENARIOS
Based on what's been reported, possible contributing factors include rapid ice accretion, ice crystals affecting the engines or the airspeed indications, severe vertical currents driving the airplane into an overspeed or Mach tuck, and structural overload that tore the elevator free.
Mach tuck is the nose-down pitching moment of the transonic range. Shock waves form on the wing and move aft; the center of pressure moves back with them, and shock-induced separation can rob the elevator of authority just when you need it. Let it go past MMO, and the dive can get too steep to recover.
These scenarios aren't mutually exclusive. A stall or a Mach tuck, the two sides of coffin corner, could have started the dive, and the tail could then have failed under the loads that followed. Or turbulence could have taken the tail first, and if it did, recovery was impossible from that moment. Either way, at a 60-degree bank and more than 20,000 fpm in an overspeed, the airplane was almost certainly past saving.
FOR THE AVIATION LAWYERS
The case will turn on several factual questions:
- the radar picture and how the crew used it;
- the weather information available before and during the flight;
- crew training and experience;
- the status of the ice protection;
- the tail's condition and maintenance history;
- the sequence once the upset began;
- the airplane's proximity to the cell; and
- what ATC knew and passed along, which the NTSB has said it will examine.
Maintenance records show a CVR was installed, but whether it survived is unknown. Part 135 generally requires an FDR only with ten or more passenger seats, so we can't assume there was one.
Several threshold issues are worth flagging early:
- Evidence. Under 49 U.S.C. § 1154(b), NTSB reports, including probable-cause findings, can't be used in a civil damages action. Factual reports are treated differently (49 CFR 835.2). CVR recordings and transcripts have their own statutory limits on discovery.
- GARA. At 22 years old, the airplane is past GARA's 18-year repose period for claims against the manufacturer. But the clock restarts for any part replaced within 18 years, such as elevator or stabilizer components. There's also a knowing-misrepresentation exception, and GARA doesn't bar claims by people who weren't aboard, such as the injured motorist.
- Owner versus operator. The airplane was registered to a Delaware LLC and flown under someone else's certificate. 49 U.S.C. § 44112 generally shields an owner or lessor that doesn't have possession or operational control.
- Treaties. Whether the Montreal Convention or the Warsaw system governs the passenger claims depends on each country's treaty status and each passenger's itinerary. A round trip beginning and ending in the U.S. can qualify as international carriage. The Bahamas signed Montreal in 1999 but reportedly has not ratified it. Don't assume either answer.
- ATC. Claims against the United States arising from air traffic control go through the Federal Tort Claims Act, which has a two-year administrative claim deadline.
This remains a preliminary assessment. Probable cause is a long way off, and anyone, pilot or lawyer, who forms firm conclusions now risks being wrong. The accident is a stark reminder that thunderstorms demand early, conservative avoidance, especially at high altitude, where the margins for recovery can disappear in seconds.
