On your own – airline flying in the South Pacific before GPS

The island of Nauru, Latitude South 003.28.3, Longitude East 166.55.0, lies on the Equator where it circles over the Central Pacific. In 1976, when this story starts, it took just 25 minutes to ride around the island on my trusty Honda 50.

An NDB and DME were the only navigation aids on the island. Both sometimes failed without warning. Local fishermen rely on the aerodrome beacon light for navigation in case the ocean currents drift them out of sight of the low-lying atoll. At night the beacon attracts sea gulls like moths around a flame.

F28 with Honda 50
The F28 parked at Nauru with the author’s Honda 50 in the foreground.

In those days, Nauru’s runway was 5600 feet long and, with the ocean at both ends, there were no over-run safety zones. Air Nauru crews drinking at the bar of the island’s Menen Hotel agreed it was best to keep going if an engine failed within 15 knots of V1. An Australian-designed T-VASIS lay submerged under long grass surrounding the sealed runway. It too was unreliable, susceptible to vandals breaking the lights after dark. For the weary crews arriving from all points of the compass en route to Melbourne or Hong Kong, the VASIS was indispensable.

I was a passenger on an Air Nauru F28 that had just landed from Melbourne with stops at Sydney, French New Caledonia, and the Solomon Islands. Recruited by Air Nauru, I was on the island to receive a briefing on the route structure over which I would fly for the next 10 years. The airline had two F28s and two B737-200s. The latter were equipped with a gravel protect kit and low-pressure tyres for operations from coral airstrips and among the destinations served by Air Nauru were Hong Kong, Taipei, Manila, Truk, Ponape, Guam, Okinawa, the Solomon Islands, the Gilbert Islands, and Fiji.

After the comfortable air conditioning of the F28, I was struck by the equatorial heat and sweating humidity as I walked toward the small airport terminal. Hearing the sound of an approaching jet, I turned around and there was my first Boeing 737 as it turned on final for runway 12. Twin smoke trails from its Pratt & Whitney JT8D-17 engines curled down toward the ocean breakers that surrounded the atoll and seconds later, with a spurt of blue smoke from tortured tyres, the Boeing smacked down hard on the white-painted 1000 feet markers barely 30 metres from the adjacent road that parallels the runway.

Boeing 737
A Boeing 737-200 flashes over Nauru.

It was a well-executed short field landing and all the more impressive for the shattering roar of full reverse thrust reverberating through the wartime fibro houses among the coconut trees that surrounded the airstrip. Now that, I thought, is a real man’s aeroplane – and one day I would fly it. As I was to find out later, the F28 was a kiddy-cart when compared to the sheer brute force of the 737.

A year passed and, with new-found skills on the lovely little F28, it was time to go to New Zealand where Air Nauru had arranged for crews to undergo type rating training on the Boeing 737-100 of the National Airways Corporation. Mid-winter in Christchurch was a marked change to life in the tropics, but we were well rugged-up, as the frost sparkled on the green parks of that beautiful city. English-style pubs abounded, with their friendly atmosphere and cosy fireplaces. After daily lectures, we gathered at the bar, beers in hand, and admired the rosy-cheeked girls from nearby Christchurch University.

In late 1977 I obtained my command on the 737. There followed many happy years flying Pacific air routes, navigating with combinations of Doppler, Omega or INS. One regular route was Nauru to Faleolo airport at Apia, the capital of Western Samoa. The over-water distance to Apia, position South 13.49.9 and West 171.59.8, was 1485 nm. The short runways at Nauru and Apia meant fuel uplift was critical for such a long distance, but fortunately over that route winds were generally light giving a flight time of under four hours.

The only airport on the direct track was Funafuti atoll in the Ellis Islands, roughly halfway between Nauru and Apia. The short coral airstrip had a reputation of being slippery from sea spray. During the Pacific campaign in WW2, Funafuti was home to squadrons of B24 Liberators and B25 Mitchells. From there, these bombers would fly to attack Japanese-held installations at Nauru, Tarawa and the far-flung Marshall Islands.

Nauru runway 12
The runway at Nauru didn’t leave a lot of room for error.

The nearest alternate for Nauru was Tarawa atoll, 391 nm at North 01.22.7 and East 173.09.0. There were no runway lights at Tarawa. That left Majuro atoll 526 nm NNE of Nauru as the only available night alternate.

When weather was poor at Nauru, departing crews would take the precaution of asking the Nauru FSO to come on duty three hours early in order to monitor our HF frequency of 13261, 8867 and 5643 from ETD Samoa for Nauru. With marginal weather at Nauru for our ETA, we needed full tanks (15,400 kgs) out of Samoa to allow sufficient reserves to divert from overhead Nauru for Majuro. If, however, we committed to an NDB instrument approach at Nauru and missed out, there would be barely enough fuel to get to our daylight alternate of Tarawa before last light. It was therefore imperative to have reliable weather reports for our arrival back at Nauru.

The trip from Nauru to Samoa was uneventful apart from isolated thunderstorms near Funafuti. The only reason that we would contemplate landing at Funafuti would be if an engine failed half way between Samoa and Nauru. Our radar picked up the storms at 180 miles and we diverted 50 miles off track to get around them.

Faleolo airport with 737
Operations at Faleolo Airport in Western Samoa required careful weight calculations.

At Faleolo airport, on the northern coast of Samoa, we were relieved to have only a dozen passengers waiting for our return to Nauru, thus allowing full tanks. The runway was 5800 ft long, with plantations of coconut trees directly under the take-off flight path further restricting the takeoff weight. With the late afternoon temperature at 31 degrees Celsius, obstacle clearance requirements limited the take-off weight to 50 tonnes. With full tanks, the available payload was around 3100 kgs. Clearly this route was not a money-spinner – but neither were the majority of Air Nauru routes, most of which were highly subsidized by the Nauru government.

Operating between 31,000 ft and 35,000 feet at ISA+15 temperature and Mach No. 0.74, the planned fuel consumption from Samoa to Nauru, was 10 tonnes. Diversion fuel from Nauru to Majuro was 3.6 tonnes. That left 2.4 tonnes for contingency and holding and you can see that we needed over fifteen tonnes and not a drop less.

After the full thrust take off from Faleolo airport, a climbing left hand turn soon has the aircraft abeam the islands of Manono and Apolima on the port side. Meanwhile, Tahiti in the Society Islands lies invisible over the horizon 1400 miles behind us, with the Fiji islands 500 miles away just behind our left wing tip. Given the choice I would rather be going to Tahiti than Nauru any day!

Looming 20 miles ahead is the mysterious volcanic island of Savai’i. At centre stage is Mount Silisli, at 7000 ft, the largest of several extinct volcanoes that rise above green jungle covered slopes.

On the northern slopes of Mount Silisli can be seen stark evidence of the lava flow that wiped out villages a long time ago. Legend has it that some villagers sheltered in a small church at the base of the mountain and prayed as the lava devoured everything in its path. Miraculously, the lava flow parted around the church at the last second and those inside the church were saved. Today, the parting of the lava flow on each side of the church can still be seen.

Radar display
Savai’i Island shows up on the radar – note the large shadow caused by Mt. Silisli.

At 33,000 ft the TAS is 435 knots, Mach No. 0.735, and fuel flow 2650 kgs per hour. The first officer passes position reports by HF to Nadi Oceanic Control at each five degrees of longitude. Earlier, we were relieved to make contact on HF with Nauru Flight Service 1200 miles up the track. The operator warned us that there was rain and low cloud over the island with visibility down to 2000 metres. We noted the time and asked him to tell us when the visibility had improved. To further cheer us up, he added that the electrical power to the runway lights and VASIS had failed and that efforts were being made to bring the standby generator on line. With the aircraft getting lighter as fuel was used, the decision was made to climb to 35,000 feet and reduce speed to Long Range Cruise of Mach 0.725, giving a lower fuel flow of 2400 kgs per hour.

It was now time to contact Honolulu on HF to obtain the latest weather for Majuro, our alternate. Fortunately the weather there is good – meaning isolated cumulo-nimbus clouds with associated rain showers. Majuro is blessed with an 8000-foot sealed runway on a narrow isthmus between a wide lagoon and the ocean. The MDA for the NDB approach is quite low at 370 feet – reflecting lack of significant obstacles; that is if you don’t count the masts of freighters or yachts at anchor in the lagoon.

Heavy showers
Heavy showers frequently sent the weather below minimums at Nauru.

Not long after our first HF contact with Nauru, the FSO called again, saying the rain had stopped and he could see the windsock, the local measuring stick of 1000 metres visibility. The first officer started his stopwatch and we noted the time on the flight plan. Long experience in these areas revealed that heavy rain over Nauru would occur in 15-minute cycles. There would be low cloud and rain below the NDB minima then 15 minutes later the weather would clear. The cycle would repeat.

Although the circling MDA for Nauru was only 750 feet, it was lack of forward vision in heavy rain that was the Boeing’s Achilles heel. Despite noisy yet efficient windscreen wipers in moderate rain and the use of rain repellant, it is difficult to see clearly through the windscreen in tropical downpours. Erroneous indications caused by refraction on water-covered windscreens add to the difficulty of judging height and angle of approach.

The sun was setting as our radar displayed the thunderstorms that we had earlier encountered near Funafuti. The radar gain control was changed from automatic to maximum gain, enabling the radar to pick up minute reflections from ice crystals present in the storm tops of thunderstorms that can reach 50,000 ft over the Pacific. Once these tiny echoes are seen on radar, the tilt control can be used to study the depth of the storm.

As the INS display ticked over through 180 East Longitude, the FSO on Nauru reported more rain obliterating his sight of the windsocks at both ends of the runway. Of course, it could be that it was too wet for him to venture outside his control tower for a better look and that it was from inside his tower the weather looked grim. We lived in hope! Stopwatches were pressed and a note made on the flight plan. Jepp computers were shuffled to calculate a PNR on Nadi at our seven o’clock position at 600 miles and falling behind us at six miles a minute.

Thunderstorms over Pacific
Storms in the South Pacific often grow to well over 50,000 ft.

Not long afterwards, the ADF needle swung overhead Funafuti with the INS showing spot on. Fuel flow was down to 2300 kgs per hour after 90 minutes of flight and in another hour or so, the decision would have to be made to either descend into Nauru for an instrument approach or divert at cruising altitude for Majuro. The flight deck lighting was turned low and we peered ahead looking for storm tops illuminated by the rising moon. In cirrus cloud and if the radar was unable to pick out the tops of embedded storms, sometimes a keen eye would pick out a lightning flash at our level.

Meanwhile reports from the Nauru FSO indicated that if we could hang around long enough at low level ready to land when a gap in the weather occurred, there was a good chance we could avoid a diversion. A few weeks earlier on this same route, the crew of a 737 had been unable to make contact on HF with Nauru from the time that the aircraft had departed Samoa. Experience had shown that either the FSO had slept in, or he did not know the ETA of the aircraft. As a fall-back plan, the crew of the inbound Boeing would call Nadi on HF and ask the operator to urgently telephone the Menen Hotel switchboard at Nauru 1200 miles to the west. Contact made, the hotel operator would rush upstairs to the Air Nauru crew room, and tell any off-duty pilots or engineers to drive immediately to the airport and either wake up the FSO or get the police to track him down. Either way, HF communications with Nauru would be established within half an hour.

Tracking 292 degrees from Funafuti, we tried without success to pick up NDB signals from the atolls of Nanomea, Arorae and Tabiteua North. These radio beacons served former wartime airstrips now used by DH Herons of Air Tungaru, the national airline of the Gilbert Islands. These navaids seldom worked because of shortage of fuel shipped in by sea to run the diesel generators.

NDB tower
NDBs don’t get much respect, but in the vast Pacific Ocean, they were all pilots had to go on.

It was dark outside and a quick flash of the landing lights revealed we still were in cloud. Nauru came back on air and said the rain was heavy, but that the runway lights and NDB were serviceable. The NDB, call sign NI and frequency 355 kHz, could often be picked up at 300 miles – a comforting feeling when it is the only instrument approach procedure available. Nearing Nauru, we needed an update on the latest Majuro weather. This time, Honolulu could not be raised on HF and so Plan B was implemented, based on local knowledge.

At Honolulu, the USAF had a military base called Hickham Air Force Base; the same Hickham AFB that was destroyed during the Japanese attack on Pearl Harbour in December 1941. Lockheed C5A Galaxies and C141 Starlifters ranging west from USA to Honolulu, Guam, Clark AFB and Japan, operated on dedicated military HF frequencies. Air Nauru crews knew those frequencies and a call to Hickham elicited a quick response.

We would simply call the USAF and ask them to relay the latest Majuro weather. It worked every time, including tonight. Majuro was still wide open. We tuned to Nauru NDB, watching the needle as it wavered in the general direction only to be seduced by thunderstorm activity off to our starboard. It seemed a good time to crosscheck our INS position against a long range plotting chart on which I had drawn our planned track. It is all too easy to enter a wrong coordinate when using the INS and the best way to prevent that is to have both pilots crosscheck entries. But even then, a rushed departure can cause momentary distraction in entering geographical coordinates, particularly on this route where we cross the 180-degree line of longitude 550 miles from Samoa. The plotting chart crosscheck confirmed we were still on track for Nauru.

Banaba Island
Ocean Island (now called Banaba), is a lonely atoll in the Pacific.

Leaving the comfort of Nadi CTA and at 170 degrees east longitude, we entered the mysteriously named region of uncontrolled airspace over the Pacific called The Unnamed FIR. We were now responsible for our own separation from other aircraft – assuming we knew of their presence, which we seldom did. On radar I picked up a tiny echo 140 miles off our starboard wing. A check of the plotting chart revealed it to be Ocean Island, a lonely atoll with a population of around 200 people. Ocean Island was made of pure phosphate deposits, but was mined out by British phosphate interests and is now no more than a barren rocky atoll. Japanese forces occupied the island during the war and murdered most of its inhabitants.

At 150 miles from Nauru, contact was made with flight service on VHF. Our radar showed heavy clouds surrounding the atoll, although according to the FSO, the rain had ceased temporarily. We had received enough weather reports en route to safely assume that there would be clear patches every ten minutes or so and the decision was made to descend. Rather than fly overhead the NDB and conduct a time (and fuel) consuming NDB instrument approach, we decided on a straight-in approach to runway 30 from over the water. The FSO reported the surface wind as calm although our INS revealed a 20-knot downwind component at altitude.

Passing 10,000 feet we slowed from 300 to 250 knots just as the FSO reported a heavy shower blotting his view of the runway end. The radar screen was covered in red echoes as sheets of water beat against the radome. Radar range is significantly reduced in heavy rain, hiding storms that could be otherwise evaded. Surprisingly there was little turbulence signifying that it was “soft” rain from thick wet nimbus clouds rather than “hard” rain and hail experienced in cumulo-nimbus.

Nevertheless, it would have been nice to see the island on radar just in case the DME packed up at the wrong moment. At 3000 feet we slowed down at idle thrust from 250 knots to 210 knots — minimum flaps up speed. At 12 miles final, and in cloud, flaps were extended and the landing gear lowered.

South Pacific map
The distances are vast, and the IFR alternates are few.

I began to have doubts about the reported surface wind of calm because the INS showed the presence of a significant tailwind component of 15 knots. With a short runway wet from rain and absence of safe over-runs, Nauru was no place to land fast and risk skidding off the end.

We broke out of ragged cloud at 1000 feet only to find the VASIS was not working. The FSO didn’t warn us of that one. With the INS still indicating 15 knots tailwind and a rate of descent confirming the higher ground speed, I made the decision to convert the straight-in approach to a circling approach for the opposite end – runway 12.

As the 737 was already in the circling approach configuration at flap 15, landing gear down and 150 knots IAS, it was just a question of which way to circle – left or right circuit. I flew a left circuit (easier to keep an eye on the runway) and after confirming with the FSO that the runway was clear (no beer bottles, wandering animals or people crossing the runway), full flap was extended and the speed reduced on final to Vref of 115 knots. An INS ground speed reading of 105 knots confirmed that the decision to circle and land into wind was the right one. True to form, the lights on both windsocks were out – someone had shot them out with a catapult.

The touchdown was close to the 1000-foot markers and as the speed brake lever whipped back on wheel spin-up, full reverse thrust was pulled. The bucket shaped reversers on the 737-200 are very effective and we only needed gentle braking at 80 knots to pull up well before the end of the runway. After backtracking, we had just turned into the parking area when the heavens opened and visibility reduced to less than 200 metres in blinding rain. With brakes parked, wheel chocks were inserted by rain-soaked ground staff who then rapped on the side of the fuselage signifying that it was OK to release the brakes for cooling purposes.

When down on fuel and nowhere else to go, survival may depend on the ability of the pilot to land safely despite heavy rain restricting vision through the windshield. Fortunately, the pilots’ side windows can be opened in flight — after first depressurizing. Because of the aircraft design, there is an area of relative calm air over the open window. Forward visibility can be maintained by looking out of the open window using care to stay clear of the air-stream. Good old Boeing – they thought of everything.

I waved to the FSO looking down upon us from his control tower situated on the roof of the airport terminal building just yards away. He would meet us later to pick up a box of fruit that his relatives in Samoa had given to us, and which we had stashed at the back of the seats in first class. Rather than put the box in the cargo hold, I thought they would make nice emergency rations in case we ditched. As a new crew took over the Boeing for the next leg to Guam, the crew bus took the first officer and me back to our accommodation at the rat-infested Menen Hotel. It had been a long day for us all.

Tags from the story

7 Comments

  • John,
    The success story is chilling enough,but what about those flights that didn’t end well? With the thin margins you described, at least a few airliners must have gone into the drink with no fuel or because of weather or mechanical troubles. I hope you will give us additional insights and lessons in future articles.

  • This is a great story. It dredged up an incident in my “way back machine” when I was a crew member on a C-54 flying out of Andersen AFB, Guam. It was the late ’50s and we were flying round robins from Guam to Japan to Okinawa. We made a normal trip to Yokota (empty) and left for Naha to pick up about 30 passengers. Halfway to Naha we lost #3 engine and as we approached the ADIZ called for a GCI/GCA letdown. No way; radar was off the air for maintenance. Give us weather conditions. The entire island is socked in; 1 mile visibility, 800 feet. By the time we got close the fog had lifted to 1000 feet, 2 miles. The pilot put that bird down on 3 engines with no problem. We got a mag installed and the next day left for Guam. Years later flying a 172 into some small airports in East Texas I remembered to keep my alternate in mind. The storms coming in off the Gulf of Mexico caused me to sit it out in several pilots lounges. Good memories.

  • Mate that is some serious flying!

    Thanks for sharing! In my current job I often get told it’s the hardest job in aviation but reading your blog renews my faith in “true” piloting!

    In our world of automation, the skills that make flying great are disappearing. Thank Goodness for the pacific, one of the true final frontiers of the world!

    Keep the stories coming my friend!

  • I loved flying the Pacific.
    Flew with “Air Mike” out of Guam to everywhere.
    Going in to Kosrae the local native on the company paid Unicom was asked for local traffic. The reply was “Yes, yes! We have a C-130 tomorrow!”

Leave a Reply

Your email address will not be published. Required fields are marked *