ANSV Highlight Procedures & HF After ATR72 Landing Accident
On 2 February 2013, ATR-72-212A (marketed as the 72-500) YR-ATS, sustained substantial damage landing at Roma-Fiumicino Airport, Italy. Of the 50 occupants, seven sustained minor injuries. The aircraft was operated by Romanian operator Carpatair on behalf of Alitalia under an ACMI (aircraft, crew maintenance & insurance) wet lease contract.
Air Traffic had reported the winds at 22 knots gusting to 37 knots from 250 degrees and issued clearance to land on runway 16L.
The Captain (58, ATPL, 18,522 hours total, 3,351 hours on type and the airline’s Chief Pilot), the Pilot Flying, advised he wanted to maintain an approach speed of 130 KIAS. The First Officer (25, CPL, 624 hours total, 15 hours on type plus 36 hours simulator time), the Pilot Monitoring, agreed. The landing gear was extended and 30 degrees flaps selected.
The approach was stable as the aircraft descended through 1000 feet AGL, with the speed being around 130 KIAS fluctuating +/- 10 knots. After the autopilot was disconnected an airspeed of about 125 KIAS was maintained.
The aircraft touched down 2.6 degrees nose down, nose gear first and bounced. While the crew recognised this, neither pilot called for a go-around.
The Captain provided nose down inputs causing the aircraft to sharply touch down a second time on the nose gear, which collapsed, although the aircraft now bounced a third time. The pilots now provided conflicting control inputs. The Captain making further nose down inputs and the First Officer nose up inputs (triggering the interlock to separate left and right flying controls). Consequently the aircraft rolled slightly left and touched down heavily on the left main gear, damaging it, bouncing again with a right bank angle of about 10 degrees, touching down a fifth time causing the collapse of the right main gear.
The aircraft slid for 400m, yawing around 170 degrees before coming to a stop.
Even though the wreckage was 400m in front of an airport fire station it took the airport fire service ten minutes of searching in the dark to find the accident site because they were apparently not familiar with the taxiway designation passed by Air Traffic.
The Safety Investigation
The ANSV concluded that no technical factors contributed to the accident. The ANSV believe wind data transmitted to the crew exceeded the 35 knots demonstrated aircraft crosswind capability of the ATR-72 (though the crosswind gust level component was in fact marginally less) and state the Captain remained confident that he could manage a safe landing nonetheless. This confidence was reinforced as prior aircraft had managed to land safely. The ANSV conclude that in the light of the weather information available, the landing should have been aborted.
The Descent Checklist was read and properly executed by the First Officer according to the ANSV, however, upon the item Landing Briefing the Captain incorrectly stated that this had already been done. The ANSV highlight that this briefing would have been crucial in identifying limits of the approach, reviewed performance data such as approach speed (Vapp), as well as establishing the criteria and procedure for a missed approach.
The ANSV note that the omission of the landing brief led to the acceptance of a Vapp of 130 KIAS and prevented a discussion between the pilots whether landing in Rome or a diversion to the alternate was advisable. The ANSV stating:
…it is reasonable to assume that the First Officer had refrained from pointing out the incorrect approach speed given the considerable difference in experience levels.
ANSV comment that Vapp should have been 118 KIAS. In contrast the Romania Centrul de Investigații și Analiză pentru Siguranța Aviației Civile (CIAS) comment in an Appendix to the report: CIAS also highlight some previous ATR landing occurrences:
- 13 May 2012, EI-REH: Irish AAIU Report
- 16 October 2013 RDPL-34233: Laos AAIC Report
- 11 April 2015 VQ-BMB
- 20 April 2015 PK-WGS
- 20 April 2015 VQ-BLF
UPDATE 9 December 2015: ATR are reported to have pointed out that they have not published, as implied by the CIAS letter, crosswind limits (as they would be required to do in the Flight Manual Limitations section if a crosswind had been reached that was beyond a normal pilot) but simply included the maximum demonstrated crosswind in the normal procedures section of the Flight Manual (as required by certification requirements). They also point out that operators are free to set their own limits in their Operations Manual. However, it is perhaps unreasonable to expect every operator conduct their own scientific flight tests to validate the manufacturer’s ‘non-limiting’ maximum demonstrated crosswind on entry to service.
Rather unhelpfully the ANSV conclude the cause was ‘human factors’. However they go on:
In particular the accident was caused by the improper conduct of landing by the aircraft commander (pilot flying) not in line with standard operating procedures as stated in the aircraft operations manual in the presence of challenging/critical environmental conditions and in the absence of effective cockpit resource management.
The following factors contributed to the accident:
- the absence of an approach briefing
- maintaining an approach speed significantly above computed Vapp
- the conviction of the commander that due to his experience and skills he could still manage a safe landing despite critical winds
- the considerable difference of experience between captain and first officer which possibly inhibited the first officer to express criticism rendering cockpit resource management inefficient
Following the accident the execution of the airport’s accident response plan highlighted several issues which were not executed timely and prevented an effective search and rescue activity for the aircraft and its occupants.
The ANSV has previously issued two recommendations, shortly after the accident, both relating to emergency response:
Issued: 25-MAR-2013 To: ENAC, National Fire Service ANSV-4/132-13/1/A/13 In line with what has already been recommended by ANSV with safety recommendation no. ANSV-13/1836-10/5/A/12, it is recommended to ENAC and the National Fire Department to adopt urgent initiatives deemed most appropriate under the educational profile and representative training to enable staff of firefighters working on Italian airports to have an actual full knowledge of both aviation terminology and of the airport grounds on which it operates, so as to avoid misunderstandings in communications related to disaster relief, to the benefit of timely detection aircraft requiring rescue. Issued: 25-MAR-2013 To: ENAC, ENAV ANSV-5/132-13/2/A/13 ANSV recommends, in general, that the Tower, in directions to be supplied in the activation of the rescue operation, gives references correlated to the GRID-MAP of their airport.
Aerossurance has discussed a Global 6000 crosswind landing accident in the UK and another involving a Dash 8 in Greenland, both of which occurred in 2014. In addition, more background can be found in this NLR crosswind report commissioned by the European Aviation Safety Agency (EASA).
We have previously written an article about a fatal G-IV take-off accident in the US: Gulfstream G-IV Take Off Accident & Human Factors. That accident featured poor use of checklists as did a fatal G-IV landing accident in France: Fatal G-IV Runway Excursion Accident in France – Lessons. However, checklists also need to be well written: S-92A Emergency Landing: MGB Oil Checklist Recommendation. We have also reported on the UK CAA’s release of CRM videos.
UPDATE 10 December 2015: In an unrelated development the UK Air Accidents Investigation Branch (AAIB) published their report into an EasyJet Airbus A319 that overflew a ridge at 488ft after losing awareness of their position while attempting a visual night landing at Bristol in May 2015. The AAIB state that the crew did not update their landing brief following a runway change. Consequently, the First Officer was not fully aware of the Captain’s plan for flying the visual approach and not properly able to monitor it. The AAIB noted thatthe crew did not set the missed-approach altitude until after the go-around commenced, suggesting the landing checklist had not been satisfactorily completed.
UPDATE 13 July 2020: ATR72 Survives Water Impact During Unstabilised Approach
UPDATE 9 December 2020: A Saab 2000 Descended 900 ft Too Low on Approach to Billund