Flight 981 Crash After Prolonged Holding and Missed Approaches in Poor Weather

The event in context

Flydubai Flight FZ981 was a Boeing 737-800 operating from Dubai to Rostov-on-Don, Russia. On 19 March 2016, after a normal cruise, the aircraft encountered poor weather at its destination. The crew made an initial approach, went around after a windshear warning and then entered a holding pattern while waiting for conditions to improve. A later approach also ended in a go-around. During the second go-around, the aircraft entered a steep climb followed by a rapid descent and crashed near the runway. All 62 people aboard were killed.

The weather was a major part of the operational environment: low cloud, rain, turbulence and windshear were reported. The crew were also operating at night and manually flying the approach. The accident therefore developed in an environment with high workload and limited visual references.

The investigation of FZ981 is much more specific than ‘bad weather caused the crash.’ The first approach was discontinued after a windshear warning, and the aircraft then spent a prolonged period holding while conditions were monitored. A later approach produced another windshear warning and another go-around. The second go-around was followed by a large pitch-up, rapid climb, control inputs and then a steep descent. The accident therefore occurred during the transition between an approach configuration and a go-around configuration, not during a simple steady descent.

Reconstructing this flight

During that second approach, another windshear alert prompted a second go-around. The aircraft was configured for the go-around with TOGA thrust, flap changes and landing gear retraction. The airplane climbed rapidly. At approximately 1,900 feet, the flight data recorder showed significant control inputs and changes in stabilizer position.

The airplane continued climbing, then transitioned into a steep descent. A nose-down stabilizer input was recorded, followed by a loss of vertical control. Recovery actions were unsuccessful. The aircraft struck the runway area about 120 metres from the threshold.

The aircraft’s configuration changes are central to the reconstruction. The crew applied go-around thrust, retracted the landing gear and changed the flap configuration. The resulting pitch response was significant. The investigation examined the interaction between control-column inputs and stabilizer trim, the crew’s workload, their perception of the aircraft’s attitude and the absence of a successful recovery from the upset.

What the evidence establishes

The final investigation focused on loss of control during the go-around, including the crew’s control inputs, workload, situational awareness and response to the changing aircraft state. A go-around is normally one of the safest escape maneuvers in aviation, but it also creates a rapid transition in pitch, power, configuration and workload.

The important safety lesson is not simply “bad weather caused the crash.” Weather created the environment, but the accident sequence depended on how the crew managed repeated approaches, holding, configuration changes and the go-around. Repeated attempts can also increase fatigue and pressure to complete an approach, especially when an aircraft has already spent a long time waiting for improvement.

The event became a major training case for upset prevention and recovery, go-around discipline and the management of high-workload approach environments.

The airport environment also matters. Other aircraft had already struggled with the weather, and one flight diverted after multiple unsuccessful approaches. The decision to remain at Rostov was therefore made against a growing body of evidence that the weather was not improving enough to make a normal landing reliable. It is not established that a go-around is dangerous; it is that a go-around must be flown as a new phase of flight, with the aircraft’s pitch, trim, power and configuration actively monitored.

Investigation and findings

The ATC timeline shows how an apparently ordinary arrival can turn into a rapidly changing emergency. Controllers can provide vectors, runway information and weather updates, but the crew still has to decide when the approach is no longer acceptable.

The Holding, Go-Arounds and Final Climb

The reconstructed sequence can be read as a chain of six decision points: Dubai departure → Windshear warning / go-around → Extended holding → Second approach → Second go-around → Steep climb → descent → impact. Each step changed the aircraft’s available options. The important analytical point is not to isolate the final impact from the preceding events; the final outcome was produced by the accumulation of the earlier changes in aircraft state, position, workload and available escape options.

The ATC/radar perspective is most useful when it is synchronized with the aircraft evidence. In this case the critical transition is the move from dubai departure to steep climb → descent → impact. Once the aircraft reached the later stages of the sequence, the crew had fewer safe alternatives than they had at the beginning. That is the operational value of reconstructing the event rather than describing it only from the final crash scene.

How the Approach Sequence Unfolded

The decisive safety issue in this incident was specific to the sequence: Windshear warning / go-around; Extended holding; Second approach. Treating those events as isolated anomalies would miss the way they interacted.

Sources and investigation material

• https://www.ntsb.gov/
• https://www.faa.gov/newsroom/windshear-and-microburst-training-aids

Accident Facts