On 18 July 2005, an Embraer 145 being operated by Swiss Air Lines on a scheduled passenger flight from Zurich to Nuremberg left the 2700 metre runway during the landing roll at destination in normal daylight visibility by means of an intentional high speed attempt to turn to one side when it became apparent that the aircraft would not stop before the end of the runway. The aircraft departed the runway tail first during a ground loop of approximately 200° to the left and eventually came to a stop 30 metres from the centreline with the main landing gear on the grass. None of the 19 occupants was injured and there was only slight damage to the aircraft. The position of the aircraft upon stopping is shown in the diagram below reproduced from the official report which also indicates the trajectory of the aircraft as the ground loop was entered (landing direction right to left).
Aircraft position on coming to rest. Source: BFU Report EX005-0/05
An investigation was carried out by the German BFU. It was found that the aircraft commander had been PF for the sector and that the ILS approach to landing runway 28 had been flown in turbulence and intermittent rain and had been stabilised at the Company-prescribed check height in Instrument Meteorological Conditions (IMC) of 1000ft agl. However, airspeed had subsequently increased to 150 knots by the time the runway threshold was crossed compared to an applicable Reference Speed (Vref) of 127 knots, and touchdown on the wet surface did not occur until between a third and half way down the runway with a moderate crosswind from the right.
In view of the fact that there was no clear information about the runway condition at the time of the accident, the investigation calculated the un-factored landing distances for 3mm water depth – 2,010 metres - and for 5mm water depth - 1,960 metres, both well within the 2700 metres of available runway length but both greater than the distance remaining beyond the actual late touchdown point
Flight Data Recorder (FDR) data showed that brake pressures commensurate with use of the parking brake had occurred late in the landing roll and the runway surface there showed evidence of reverted rubber aquaplaning which results from locked wheels – the parking brake has no anti skid pressure by pass system so this is unsurprising.
The data showed that 32 seconds after touchdown at a groundspeed ay 52 knots, the observed ground loop had commenced taking the aircraft heading from 280° left to end at 080°.
The investigation noted that the runway surface was of asphalt for the first 1768 metres and them concrete for the remaining 932 metres. FDR data provided indications of applied brake pressure which indicated that dynamic aquaplaning may have begun as the aircraft reached the concrete section of the runway.
It was noted that the aircraft was not fitted with thrust reversers, which are “the sole effective means of speed reduction under aquaplaning conditions”. It was noted too that the late landing had also been “soft” and that this had facilitated aquaplaning over the final one third of the runway.
It was found that “the airport did not provide the crew with adequate information on the runway surface condition” although it was considered that the crew “could have deduced the likelihood of reduced braking conditions on a wet runway.
The Cause of the runway excursion was determined by the investigation as “aquaplaning on a rain-soaked runway surface”. Four “additional factors” were identified:
- the crew had insufficient information on the actual runway surface condition
- the approach flown was unstable due to gusty wind
- the speed flown was 20 kt above Vref
- the aircraft was above the target height on crossing the runway threshold, and the consequential late touchdown.
No Safety Recommendations were made
The Final Report was published in July 2010 and may be seen in full at SKYbrary bookshelf: BFU Report EX005-0/05