Let’s start with the basics

Cosmic radiation are high-energy particles originating from the sun, but also from far more extreme objects in the universe, such as supernovae, black holes and other exotic sources. The sun is a source of natural radiation we receive from space. Its activity follows an approximately 11‑year cycle with alternating periods of peak and quiet phases. We are currently moving out of a recent maximum, which is why we observed strong solar activity in the spring of 2024 and again around 11 November 2025, including exceptional displays of polar lights at unusually low latitudes.

Solar activity and cosmic rays

During periods of high solar activity, the Sun can emit intense streams of high‑energy particles (referred to as cosmic rays). These particles interact with the Earth’s atmosphere and magnetic field, which usually protect us. However, during strong solar events, some of these particles can penetrate deeper and even reach the Earth’s surface.

Solar activity can:

• Disturb HF radio communications

• Degrade satellite signals

• Damage sensitive electronic equipment

Aircraft are more exposed than people on the ground because we operate at high altitudes, where the atmosphere offers less shielding, and often closer to polar regions, where the magnetic protection is weaker.

Cosmic rays can be compared to X‑rays used in medical imaging: they pass through most materials they encounter. When they pass through an aircraft, they can affect onboard electronics in three main ways:

1. Transient error (single event upset):

A temporary “bug” occurs when data transfer is blocked or momentarily corrupted. The system can usually cope with this and resumes normal operation once the event is over. Satellites are equipped with detectors for such events, so we know they occur regularly.

2. Minor component damage:

A component may suffer a small, localised degradation. The system can often work around it, for example after a reset, but the fault may reappear later.

3. Permanent damage:

The part is irreversibly damaged and becomes unserviceable, requiring replacement.

Space weather has been monitored in accordance with ICAO requirements since 2021. Airbus has required its equipment suppliers to take the effects of cosmic rays into account in their designs since 2007.

Let’s look into category 1: Single event upsets or Bit Flips

In all our computers – from laptops to flight computers – information is stored as bits: zeros and ones. A bit flip is an error where such a 0 suddenly turns into a 1, or vice versa. Often, this isn’t a software bug but the result of cosmic radiation.

When such a particle strikes a sensitive transistor in memory or a processor, we call it a single event upset (SEU). Most of the time this goes unnoticed or gets corrected by error detection and ECC memory. But sometimes a single wrong bit has a visible effect: a video game that suddenly glitches, a supercomputer that crashes, or – as in the well-known Belgian election case – a candidate suddenly receiving exactly 4,096 (2¹²) extra votes because one bit flipped in the counting system.

For aviation, this is particularly relevant. At cruising altitude the dose of cosmic radiation is much higher than on the ground, which increases the likelihood of bit flips. Critical aircraft systems also run continuously and are highly automated. In rare cases, a bit flip can lead to incorrect sensor data or unexpected behaviour of a flight computer, as was investigated in incidents similar to Qantas flight QF72. This is where it becomes interesting:

QF72 was an Airbus A330 operating a scheduled flight from Singapore to Perth on 7 October 2008. At cruise, FL370 over the Indian Ocean, the autopilot suddenly disconnected, followed by a cascade of ECAM warnings and conflicting stall and overspeed alerts. Shortly afterwards, the aircraft entered a sudden, uncommanded nose‑down pitch; a few minutes later a second, less severe dive followed. More than one hundred passengers and crew were injured when they were thrown against the ceiling or cabin interior.

The investigation showed that one of the three Air Data Inertial Reference Units (ADIRU 1) briefly sent “spikes” on several parameters, including an absurdly high angle of attack (AOA). The Flight Control Primary Computer (FCPC) interpreted these erroneous AOA values as if the aircraft were in a deep stall and automatically triggered the high‑AOA protection: the system commanded a strong nose‑down input even though the pilots themselves had not initiated a dive. A design limitation in the A330/A340 control laws made it possible for repeated spikes from a single ADIRU to result in a real control input.

The ATSB then examined possible triggers for this data corruption: software bugs, hardware faults, electromagnetic interference and also secondary particles from cosmic radiation capable of causing a bitflip. No single, definitive cause was identified; many scenarios were judged to be (highly) unlikely, while a single‑event upset (SEU) due to cosmic radiation remained a plausible explanation without a precise probability.

The current JetBlue A320 incident 

On October 30th, a JetBlue Airbus A320‑200 operating a flight from Cancun to Newark, was en route at FL350, about 70 NM west‑southwest of Tampa, when the aircraft experienced an in‑flight upset lasting approximately 4–5 seconds before the autopilot corrected the trajectory. This likely occurred during an ELAC (Elevator & Aileron Computer) switch change. Ten people on board were injured. The aircraft then descended rapidly, briefly leveled at around 20.000 ft, and continued down for an approach into Tampa. The crew reported flight control problems.

Investigators found that the ELAC B hardware running software version L104 could be affected by solar flares. In a worst‑case scenario, this vulnerability might cause the elevators to move unexpectedly, potentially bringing the aircraft close to or beyond its structural limits.

Similarity between QF72 and Jetblue?

The full investigation is still ongoing. Although Airbus itself is referring to cosmic radiation as a possible factor, there is no official causality (yet) between the Jetblue incident and bit flips. Through this Airworthiness Directive, airlines are instructed to address the issue by either:

• Reverting the flight‑control software to a previous, stable version (Airbus estimates this fix takes about three hours per aircraft), or

• Replacing the affected ELAC hardware with units already running the earlier software version.

It is estimated that approximately 1,000 aircraft will require this hardware change.

The essential role of pilots

We can’t completely prevent bit flips – the universe remains “hostile” to computers – but thanks to alert pilots, the risk stays manageable.  This incident, and others like it, remind us that pilots remain critical safety gatekeepers on board the aircraft. Rather than focusing solely on the fact that around 70% of incidents and accidents are attributed to “human error”, we should recognise that human crews routinely correct countless technological issues, thereby preventing many more incidents and accidents than they cause.

This is one more strong argument against reduced‑crew operations: in a highly automated, complex system exposed to phenomena such as space weather, the human element is not a weakness but a vital layer of resilience.