Air Safety

Introduction

As part of the Air to Surface phase of the Hornet Conversion Course, No. 2 Operational Conversion Unit (20CU) deployed to Townsville on Monday 16 November 1987. The flying programme to be conducted at Townsville included day strikes, close air support and a night radar navigation and bombing sortie. While conducting the night radar navigation and bombing sortie on 18 November 1987, aircraft A21-104 impacted a hill on Great Palm Island while in the radar bombing pattern, killing the pilot and destroying the aircraft.

Sequence of Events

The sortie consisted of a 40 minute radar navigation sortie which terminated at White Rock, a small island off the southeastern tip of Great Palm Island. From White Rock, which was used as the Initial Point (IP) for the bombing run, the students were to descend to 1,000ft and carry out a radar bombing pass on Cordelia Rocks, 15nm to the south. After the first pass the aircraft was to commence a climbing left turn onto a downwind heading of 020° magnetic for a further racetrack pattern to complete one more bombing pass. The turn from downwind onto attack heading was to be commenced at 10nm from the target. The run-in speed was 480 knots. Speed on the remainder of the pattern was not briefed, but it would have been in the order of 400 to 480 knots.

The briefing for the sortie was conducted at 1830 on the evening of the accident. The briefing covered:

• the directions to ensure clearance from Palm Island using ground map radar, (on a ‘standard pass’ the aircraft would remain approximately 5nm from Palm Island); and

• the need to ensure that the aircraft had passed a westerly heading before commencing descent from 3,000ft to the run-in height of 1,000ft.

One significant fact omitted from the briefing was that the pilots would have to abide by night Visual Flight Rules once the aircraft had descended below safety height, ie they had to maintain visual reference to the ground or water.

Weather in the range area at the time of the accident was: 418th Cumulus base 2,500ft, tops 7-8,000ft, wind 060/15. The moon had not risen and consequently the range area was very dark and there was no significant man-made illumination in the range area.

The Range Safety Officer (RSO) was the first pilot airborne. He was to fly the navigation exercise, conduct the two briefed bombing passes and then remain on the range to act as RSO for the first two students. The duties of the airborne RSO were to ensure that the student had identified the correct target (Cordelia Rocks) and, having done so, to confirm that the student had designated the correct target. He was then to give a release clearance to the student. There was no requirement for the RSO to monitor the students’ position or flight parameters around the pattern.

The RSO conducted the radar navigation exercise onto the range, carried out the two briefed bombing passes, then carried out his RSO duties while the first student conducted his two bombing passes. There were no problems experienced with the first student onto the range. After the first student departed the range, the RSO commenced orbiting east of the target waiting for the second student to enter the range.

Eventually the second student entered the range area, not from overhead White Rock as expected, but on a direct track from the previous turn point, Goold Island, (Way Point 4). This error was almost certainly caused by incorrect use of the navigation system and was not detected by the RSO who was in the air to air radar mode searching for the inbound aircraft at the time. As the student approached overhead the target (Way Point 6) at 5,000ft he called “approaching White Rock (Way Point 5) and descending”.

It was clear from the evidence of the radio transmissions and from the subsequently retrieved Maintenance and Serviceability Data Recording System (MSDRS) data that, while he was overhead Cordelia Rocks, (Way Point 6), the pilot believed he was over White Rock (Way Point 5). At this stage, the RSO was completing an intercept on the student. From overhead the target, the student commenced a right turn to track towards the next way point and rolled out on a heading of 290°, which pointed him directly at Gate Out (Way Point 7). At this stage, it was apparent to both the pilot, who reported an Inertial Navigation System problem, and to the RSO, who was aware that they were approaching the range boundary, that they were not on the

• expected heading of 190° magnetic from White Rock to the target. The RSO told the student to commence a 180° right turn, in order to keep the aircraft within the range boundary, and also advised the student to check that he had the correct Way Point set. The student responded to the call by advising that the problem was now fixed and he rolled out of the right turn and tracked direct to the target on a heading of 060°

After tracking over the target with the RSO in one to one and a half miles radar trail, the pilot turned onto the briefed downwind heading of 020° and climbed to 3,000ft. Because the pilot had turned directly onto downwind from his 060° approach to the target, instead of from the attack heading of 190°, the downwind leg that the aircraft actually flew was some four to five nautical miles west of the briefed downwind leg and therefore closer to Palm Island.

Contrary to the brief, which required the pilots to pass 270° in the turn onto the attack leg before descending, the incident pilot commenced a descending left turn onto the bombing heading at 10nm from the target and levelled at approximately 1,000ft. He continued this level left turn until his aircraft impacted the south eastern tip of Great Palm Island. At the time of the crash, the aircraft was level at 1,000ft above mean sea level with 46° of left bank, at approximately 460 knots. As the downwind leg had been flown intermittently in cloud, the RSO had been flying in air to air radar mode and was not able to monitor the relative position of Great Palm Island.

During the final descent, the RSO had become visual with the student and switched his radar to air to ground mode. This occurred only three to four seconds prior to impact and the RSO had insufficient time to interpret what he saw and warn the student of their proximity to Palm Island. The lack of either moonlight or man-made lighting precluded either pilot from visually acquiring Palm Island. The Coronial Inquest into the accident was concluded in Townsville on 15 september 1988. The Coroner’s finding are consistent with those of the RAAF Board of Inquiry.

Summary

The aircraft impacted the ridge on Great Palm Island as a result of descending below safety height without first meeting the requirements for night Visual Flight Rules.

An error in management of the navigation system by the pilot resulted in a non-standard entry into the application bombing pattern, which caused his downwind leg to be some 5nm closer to Palm Island than the briefed standard pattern would have been and this negated the planned 5nm geographic separation. This fact, coupled with the design of the pattern which required an attack pass at 1,000ft, left the pilot with no margin for error. Therefore, although he turned close to the pre-briefed distance from the target of 10nm, the subsequent turn caused him to track over the south-eastern tip of Great Palm Island where ground impact occurred.

The coronial inquest into the accident has been completed.

Who’s Watching the Pilot?

It was to be a routine Anti Submarine Warfare (ASW) sortie departing from Edinburgh in the early hours of the morning, picking up some fuel from Richmond and then transiting to be on task off the New South Wales coast.

The trip to Richmond went smoothly enough, as did the refuel and departure for waters off the NSW coast. On task, things progressed smoothly for about four hours . . . and then it happened. We were actively prosecuting the sub, my copilot was flying and had just flown on top of a sonobuoy and was turning left to intercept the next ‘fly to’ point. I turned my attention to the pedestal instruments so that I could make the adjustments necessary to provide for proper flight instrument information.

My copilot was utilizing the command steering system. I was not bothered that my attention was being diverted from monitoring where the aircraft was going because he had_ everything under control as he had done for the last 35 minutes or so. Having completed my task (which consumed about 5 seconds of time) I looked at the VSI and to my horror saw 1500ft/min rate of descent. A quick glance at the radar altimeter confirmed our imminent peril; it was reading 140′ and decreasing rapidly. As I yelled at the copilot to watch his altitude, I pulled back on the control column to get the aircraft climbing to a safe altitude. We bottomed at about 70′ ASL; just seconds from death.

How did an incident like this occur? How could a multi-crewed aircraft come within seconds of flying into the water?

During daylight/VMC the minimum operating altitude for a P-3 is 100ft ASL. Often, during an active prosecution, the aircraft descends to 100ft ASL but, in the main, operates at a safer altitude. On this occasion my copilot was attempting to manoeuvre the aircraft at a constant 350ft ASL but due to lack of experience and poor cross-reference, the deviations from altitude often put the aircraft below 300ft ASL. Each time we went below this altitude the Radar Altimeter Warning System (RAWS) announced the fact. This was becoming annoying so my copilot elected to maintain 200ft ASL where the RAWS would not operate. My mistake was in allowing this to continue rather than have the aircraft operate at 500ft ASL. My copilot was an ex P-3B pilot who had just completed a refresher course prior to returning to an operational squadron. He came along on this trip to gain more technical experience. At the critical point he forgot that the command steering system gives only roll information and not roll and pitch. Hence, he allowed the altitude to go nose low during the turn and. this, coupled with a breakdown in cross-reference, sent the aircraft heading towards the ocean at 1500ft/min.

My suNival instincts ensured that, though I had a lot of confidence in the copilot, I still maintained a constant scan of the flight instruments. This scan being interrupted only by my manipulation of the flight director system control panel which is located on the pedestal. I had been proceeding in this manner for the past 35 minutes when, out of the blue, we went on our hair-raising descent.

The moral of this story is that when operating in a multiple crew environment, each crew member at the flight station should monitor what the other is doing. We should be vigilant at all times regardless of the experience level or expertise of the other members. At the time of this incident the flight engineer was concentrating on making some power adjustments. If my scan had been delayed a further 4 seconds, for whatever reason, we would have contacted the water. Complacency can lead to catastrophe and you must condition yourself to guard against it happening to you.