Original subtitles

A New Year's Eve tragedy.

A seaplane with six on board lost control.

This is big news.

A multi-millionaire and his family are killed.

Initial reports about the crash site are puzzling.

He's not where he's supposed to be.

Investigators quickly gather witness statements.

It took off normally, like any other flight.

And scrutinize the pilot's background.

This guy was a competent pilot.

With no significant clues the team turns to recovered wreckage We had a

completely functional aircraft at the time of impact But something buried deep

inside the plane provides investigators with their biggest lead.

This was in the cabin I felt we may have an answer in our grasp

Five British tourists have started celebrating New Year's Eve at a

waterside restaurant at Cottage Point, Australia.

And how was lunch?

It was great.

Gareth Morgan is the pilot for the 20-minute flight back to Sydney Harbour.

Have you enjoyed your Australian visit so far?

Very much.

Take the front seat.

Get some good shots from up there.

A born athlete, raised in Vancouver, Canada.

Gareth Morgan has come to Australia to fly float planes.

My dad's friend, he owned an airplane, got me hooked, so I just thought, hey,

this looks like a great career.

Gareth did not like to be bored.

He liked to be challenged and have variety in his life.

And a pilot's career, particularly a pilot on float planes, gave that to him

in spades.

Richard Cousins is the CEO of one of the largest food companies in the UK.

He's on holiday with his fiancée, her daughter and his two sons.

Buckled in?

The trip to Cottage Point attracts a high-end crowd.

Celebrities, business tycoons and even close relatives of the royal family.

Seven months ago, Pippa Middleton... the sister of the Duchess of Cambridge, and

her husband, James Matthews, took the same trip during their honeymoon in

Australia.

Welcome back, folks.

Pleasure having you on board with me again.

The flight is operated by Sydney Seaplanes, which runs a small fleet of

agile float planes.

Today, Gareth Morgan is flying a de Havilland Beaver.

It's a very gentle, easy, relaxed airplane to fly. I call it the big comfy

couch.

You can take large payloads. You can get into tight places, confined areas.

It's very versatile.

The Beaver is powered by a single 450-horsepower radial engine. The iconic

aircraft is at work all over the world.

We're about ready for takeoff here, folks.

I went over the safety regulations on the way over. Does anyone need a

refresher?

All good.

We'll have you back in Sydney in a flash.

Morgan takes the plane towards the designated takeoff area in Cowan Creek.

We are deciding, okay, where are the hazards?

Where are my obstacles?

Which direction is the wind coming from?

We are considering all those variables.

Cowan Creek traffic.

This is Float Beaver, November Oscar Oscar.

We are taxiing to the takeoff point for a northeast departure to Rose Bay.

Most locations where a float plane is operating, you're in uncontrolled

airspace, so you're not talking to a controller.

You are just making blind calls on the radio to tell all aircraft what your

intentions are in the area.

Any conflicting traffic, please advise.

November Oscar Oscar.

Traffic is clear. They can depart.

A takeoff on water is executed in two stages.

You're going to be plowing through the water until it kind of crests over that

bow wave that you're creating with the floats.

Once you're on that step, you're watching your attitude, keeping your

directional control with your rudder and your ailerons.

Racing at 60 miles per hour, the float plane takes only 15 seconds to lift off.

You'll see Cowan Point down there.

In the moments after takeoff, the plane must gain altitude quickly.

The hills surrounding the bay are more than 400 feet high.

What makes it different for pilots flying these kind of aircraft is just

being open to constant changes.

The variables are always changing.

Gareth Morgan has to decide the best way to gain the altitude he needs to clear

the surrounding hills.

He can continue forward, climbing along the length of the channel.

He can also make a U-turn and head back over the waterway from where he just

took off.

A pilot chooses their departure path based on experience, knowledge of the

area, comfort level with the aircraft.

Typically, you want to go out into the most open area possible.

For the pilot, this is the eighth flight during a busy day shuttling passengers.

That there's cow in, um, Comox Creek.

Say again?

There's Comox Creek.

At just 130 feet, the floatplane stops climbing.

They got up.

Oh.

Morgan needs to gain altitude.

He's headed towards a dead end.

You have to have a point of making a decision.

Can I make it or can I not make it?

If you cannot make it, do you have space to get out of that situation?

Less than two minutes after taking off from Cottage Point.

The float plane crashes into Jerusalem Bay and sinks.

A New Year's Eve tragedy.

Six people feared dead in a seaplane crash on the Hawkesbury River.

By the time rescuers arrived on the scene, the aircraft had disappeared

under the water. This is big news.

Investigators from the ATSB, the Australian Transport Safety Bureau,

begin their first assignment of the new year. I'll handle the press, try to find

some witnesses.

Someone must have seen this.

The media attention was huge.

It was on every television that I turned on.

The search for answers intensifies as police dive on the crash site, joined on

the surface by investigators from the Australian Transport Safety Bureau.

All five passengers, including CEO Richard Cousins, are dead. So is the

pilot Gareth Morgan.

It was a day that drastically changed our lives.

It was a devastating moment and we still live with it every day.

My first thoughts went to those who had just lost loved ones in such a tragic

accident.

And then it was changed gears and went straight into the business side of

things. What did we know and how we were going to approach the investigation?

OK, great.

How deep is the plane?

The ATSB begins its investigation into the crash that took the lives of six

people. Police are saying it's probably under 14 metres.

It's right here.

In Jerusalem Bay.

Recovering all the wreckage is crucial.

There was no radio contact with the pilot, the flight wasn't tracked on

radar and the plane didn't carry a key source of information.

Unfortunately, we find with accidents involving smaller aircraft, it is very

unlikely that they will have some form of recording device on board.

Until the wreckage can be recovered...

Witness statements are all there is to go by.

Let's just walk through where you were and what you saw.

We consider witness interviews what we call perishable evidence in that we try

to get that done as soon as possible.

Well, it took off normally.

It just seemed like any other flight.

We utilised aircraft models so that they can show us rather than tell us of what

their observations were.

And where were you?

Near the creek here.

We also use maps to try and get a sense of what they saw and where they were

positioned Investigators quickly discover that although several people

witnessed parts of the fatal flight.

No one saw all of it. Thank you so much No smoke no one had any

unusual noises.

Witness A says he takes off normally

and flies down Cowan Creek.

Witness B sees him climbing.

No obvious issues.

The initial witnesses we interviewed described the flight path of an aircraft

steady, no erratic movements on the southern bank of Jerusalem Bay.

They then described the aircraft doing a steep turn before the nose dropped and

the aircraft impacted the river.

We've got a gap in the flight path.

It starts off here.

somehow ends up here.

Investigators can't confirm the plane's entire flight path.

The more we spoke to the witnesses, the more questions it did raise.

Do you have the flight chart?

As investigators review the flight chart, they make a key discovery.

He's not where he's supposed to be.

The plane ended up in Jerusalem Bay.

a narrow dead-end body of water west of the take-off area.

We established that there was no operational reason for the aircraft to

be operating in Jerusalem Bay.

With little to go on, the biggest piece of evidence is on the bottom of the bay.

Investigators will have to retrieve it.

The lift of the wreckage was quite a nerve-wracking event for my team.

We had the area cordoned off to vessels except for emergency services.

The police divers first went down and secured the main fuselage.

The team delicately lifts the aircraft to the surface.

We were very careful, especially with the main fuselage, which carries a lot

of water. We needed to be patient to drain the water from it before we lifted

it onto the barge.

The full recovery of the wreckage of the Beaver seaplane provides hope for the

investigation.

They got everything.

ATSB investigator Leon Campbell is tasked with dissecting what's left of

the plane and searching for signs of mechanical failure.

The nose is creased.

Let's check the flight controls.

The aircraft was delivered to the hangar inverted and we decided to keep it in

that orientation, one for safety because it was quite stable and also it afforded

us good access to the flight controls.

Did the plane crash because of an engine failure?

Well, the blade is damaged.

We are looking for signatures if it was providing power at impact with the

terrain or water.

Slight forward bending here.

There's a mid-span bend too.

It's a double bend.

Investigators find a pattern they recognise.

So the propeller was turning normally.

The engine was working.

The wreckage examination determined that we had a completely functional aircraft

during the flight.

And at the time of impact, everything appeared normal.

Still without any answers, investigators scour every inch of the plane, inside

and out, looking for clues.

We noted that the forward cabin roof area was filled with mud.

To ensure that we're not missing any vital evidence, I had the unenviable

task of going through this mud.

Duncan.

This was in the cabin.

Wow.

Great job.

They find something completely unexpected.

The discovery of a camera could change the course of the investigation.

The camera was a crucial piece of evidence that we could hopefully gain

some more insights as to what occurred during the flight.

Maybe there's something valuable on the memory card.

But the camera was underwater for four days.

The compact flash card was damaged.

We treated this card just like we treat the memory from a flight data recorder

that is damaged.

The creasing damage on the nose and the fuselage is fairly significant.

Agreed. While work begins on the data card, investigators turn their focus

back to the wreckage.

Deformation increasing of the fuselage and the pontoons can give us an

indication of the speed, orientation and angle of entry into the water.

Investigators measure how much the force of impact bent the plane's nose.

Deformation angle is 25 degrees upward.

Normally...

It's only 12.

So, the impact bent the nose upward 13 degrees.

Correct. So maybe we're looking at a stall.

It's possible.

With the engine operating and eyewitnesses reporting the plane in a

steep right turn, investigators consider whether the pilot lost lift over the

wings, stalled the plane and fell from the sky.

Stalls at low altitude are dangerous because there's no time to recover.

It says here that the angle for a stall is 12 degrees or more if the flaps were

set to climb out.

The flap actuator was extended so the flaps were set to climb.

Investigators conclude the float plane stalled before hitting the water.

We had a pilot who was in Jerusalem Bay at low altitude conducting a steep turn.

This raised questions with the investigation team on what was happening

in the cockpit.

So why would a professional float plane pilot go into a dead end and then stall

the plane?

Good question.

Investigators want to know more about the pilot who crashed the float plane

into Jerusalem Bay, Australia.

The majority of his time is on float planes.

More than 9,000 hours.

He was very experienced.

This is Float Beaver November Oscar Oscar.

Looks like he started with Sydney Seaplanes in May of 2017.

He also had a stint with them between December 2011 and April

2014.

535 hours on the plane that crashed.

So he knew that plane.

Hey, look here.

He had an incident only a few days before the crash.

The pilot was involved in an incident while he was flying a Cessna 208 float

plane. Was it his fault?

No, I don't think so.

It's just a rough landing in Rose Bay.

Traffic was busy and waves were high.

The event was essentially a bad luck event beyond the control of the pilot.

This guy was a competent pilot.

So what are we missing here?

It was New Year's Eve.

Not a normal day.

What are you getting at?

Was he celebrating a bit early?

I'll check on the status of the toxicology report.

I'll make some calls.

One of the things that we do look at is whether there was the use of any

substances such as alcohol or any of the medications they may have been taking.

Can you think back?

What was he up to while the passengers were having lunch?

He had a lunch break at the cafe at Cottage Point.

Happy New Year, my friend.

We were aware that the pilot had been required to move his beaver while

another aircraft entered the area to onboard passengers.

I gotta go.

I gotta move my plane.

Okay. Did Gareth Morgan have a drink between flights?

Alright. Alright, thank you.

No one saw Gareth having a drink.

Toxicology report confirms that as well.

We knew his integrity.

We knew that he would have had the well-being of his passengers first and

foremost in his mind at all times.

A month into the investigation and the ATSB can't find any problems with the

pilot or the plane involved in the crash at Jerusalem Bay.

Nobody knew what had happened.

Something went terribly wrong in that cockpit.

And we do not know what.

Investigators place their hopes on the camera found in the wreckage and the

data on the memory card inside.

This was meticulous work.

The controller chip contained 144 solder points, which needed to be

reconnected under the microscope.

Thank you.

Looks like we got something.

The lab is sending the images now.

We were successful in recovering over 350 images from the compact flashcard.

Investigators analyse the pictures, hoping they see something that might

help them solve the case.

Looks like they're taxiing.

Oh, it's taken from the passenger side next to the pilot.

The next 22 pictures were taken during the taxi takeoff and climb.

Nine were taken while the plane was airborne.

Taken over 39 seconds.

That's the last one.

Too bad that's all there is.

I think there's more we can do with these.

When we had the photos from the accident flight, we decided to see if we could

rebuild the flight path of the aircraft using these images.

Time, 15, 11, 45.

Direction forward.

Estimated altitude.

98 feet they compiled data from the photographs about the plane's position

the camera angle and the time of day okay that's all of them let's process

this so how's it going strung together the

images provide investigators with a key portion of the plane's flight path using

camera tracking software we were able to get a better idea of the aircraft's

flight path but also have a more accurate understanding of the altitude

of the aircraft.

The plane is at 98 feet and turning right and stops climbing for some

reason.

That's odd.

We could see that the aircraft was in a banked turn, but despite being in the

climb, flat setting, the aircraft was losing altitude.

The photo analysis tells investigators that the plane was descending, but the

complete flight path remains an educated guess.

Well, the good news is that matches the flight path from our eye witnesses.

But it doesn't go much beyond that.

You're right.

We need to come at this from a different perspective.

At that point, because we had no obvious reason as to why the accident occurred,

we were looking at all the other possible hypotheses.

Maybe he had a seizure.

Or heart failure.

Well, that's a possibility.

His medical records indicate...

No history of headaches, blackouts, dizziness. I'm not seeing anything.

Investigators shift their focus back to the pilot, Gareth Morgan.

What about the cardiology?

This isn't normal.

Some of the ECGs had signs of sinus bradycardia, which is a slower-than

-normal heart rate. His resting heart rate is 50 BPM.

He's like a trained athlete.

It's an abnormality, but it's not a problem.

He was a very fit individual, and there was nothing in his medical history that

could have led to medical incapacitation.

I thought we'd have more answers by now.

Almost two years into the investigation...

The ATSB prepares a final report without any firm conclusions about what caused

the crash.

The investigation team were disappointed and frustrated that they could not

provide an answer to the families.

Duncan Bosworth.

Yeah, well, incapacitation was always a possibility.

And pretty far down that road, it came up empty.

While the report was being prepared, we have an internal review.

This included our aviation medical specialists.

No, he must have tested for that.

I'll get back to you, all right?

The aviation specialists raised the fact that carbon monoxide could have been an

issue during this accident.

When aviation fuel is burned, exhaust from the engines contains a highly

poisonous carbon monoxide gas, which can cause short-term symptoms.

Long-term exposure can have a severe impact on health and can be fatal.

They tested for carbon monoxide, right?

Here's the pathologist's report.

I said I thought it must have been examined because we had the toxicology

report and nothing had been raised about carbon monoxide.

Kerry?

It's not here.

I could not find evidence of carbon monoxide testing during the autopsy.

I have a few questions for you.

Is there any reason you didn't test for carbon monoxide poisoning?

Any chance you still have the original samples?

I see.

Oh, you do?

Well, that's great.

That's when they went off and did the additional testing to assess whether

carbon monoxide was an issue or not.

26 months after the crash, Gareth Morgan's blood sample is tested for

carbon monoxide.

Look at this.

The results provide investigators with their biggest lead yet.

I felt we may have an answer in our grasp.

Morgan's carbon monoxide levels were at 11%.

Some of the passengers also tested positive for CO poisoning.

Toxicology results confirm that the pilot and two passengers of the ill

-fated floatplane had higher than normal levels of carbon monoxide in their

blood.

We were re-energized because we may have found that missing piece of evidence.

According to this study, the pilot would have been suffering from

neurobehavioural and cognitive side effects.

We did a lot of research to see what that 11% in particular for the pilot,

how that would have affected his performance.

You'll see cow in Cold Rocks Creek.

Was Gareth Morgan's carbon monoxide level high enough to incapacitate him

and cause him to crash the plane?

At 11%, he would have had feelings of nausea, dizziness, possibly a headache.

It's hard to know with those symptoms if it's carbon monoxide poisoning or

something else.

The side effects of 11% carbon monoxide poisoning are so insidious it would have

been hard for Gareth Morgan to be aware of what was happening to him.

Quite often the cognitive effects can be quite subtle and they may often be

overlooked or even dismissed by the pilot.

How much carbon monoxide does he need to breathe in to reach 11%?

The safety literature says there would have to be a concentration of 80 parts

per million in the cabin for a sustained period.

That's an awful lot of carbon monoxide in the aircraft.

So, what's the source?

And how did it get into the cabin?

We knew the source of carbon monoxide in an aircraft would be the exhaust system

or a heater in the aircraft.

There was no heater, so our focus was on the engine exhaust system.

Well, this piece seems OK.

When an engine burns fuel, the hot gases are directed through the exhaust

manifold and out of the tailpipe, preventing them from entering the

plane's cabin.

Take a look at this.

See those cracks?

Investigators take a closer look at pieces of the cracked manifold to

determine if the damage occurred before or after the crash.

There's some oxidation here.

Rust on the fractures in the exhaust manifold tells investigators that the

crack developed prior to the accident.

We found the leak.

After we discovered there was an exhaust leak we had to determine how the carbon

monoxide got into the cabin to affect the occupants.

What's the most likely point of entry?

Right here.

The firewall.

Investigators try to determine how carbon monoxide entered the float

plane's cockpit, incapacitating the pilot.

The firewall exists between the engine accessory bay and the cabin to prevent

fumes getting into the cabin.

Zoom in.

The team examines the plane's firewall.

It's pretty banged up.

They focus on two panels which allow access to the engine for maintenance.

Here's what we found.

They discover the left access panel is missing one of its four required bolts.

The right panel is missing two.

That means there were three holes in the main firewall on the day of the flight.

It's a pathway for exhaust to travel from the engine bay into the cabin of

the Beaver.

Several pre-existing cracks in the exhaust system released the poisonous

gas into the engine bay.

Tiny bolt holes allowed the gas to leak into the cabin through the main

firewall.

So we have the source and the point of entry.

The discovery leaves investigators with a burning question.

How long has the dangerous problem existed?

The last work performed on the parts was done in early 2017.

That's months before the accident.

That means Sydney Sea Planes pilots have flown the same Beaver float plane

hundreds of times without being affected by the carbon monoxide leak.

And it also means Gareth Morgan flew the plane dozens of times himself without a

problem.

So why did it affect him this time?

One of the key questions we were asking ourselves was,

Why did the accident happen on this flight?

ATSB investigators examine pilot logbooks to understand what made the

fatal flight of a Sydney seaplane different from others.

He's flying the same plane, day after day.

It has the same problem.

But nothing happens until noon is in.

I can't figure it out.

What was his schedule on the day of the crash?

Well, he's making flights all day long, so...

The carbon monoxide is going to be built up in the system.

An individual's carbon monoxide levels will increase with intensity depending

on the duration.

We had a pilot who was conducting multiple flights throughout the day with

only short intervals between each flight.

There's something else.

As investigators reconstruct Gareth Morgan's schedule, they remember they

have to add one more piece to the puzzle.

An hour before he took off from Cottage Point, Gareth Morgan was asked to move

his plane off the dock to make room for another plane.

I gotta go.

I gotta move my plane.

Maybe Morgan was exposed to even more carbon monoxide gases while he moved the

plane. Exactly.

How long did that take?

27 minutes.

That's almost as much as an extra flight.

That's a lot more time in a plane.

Morgan had to manoeuvre his plane away from the dock and around the bay, while

another plane picked up passengers.

There's still a missing link.

These holes don't seem big enough to let in enough exhaust.

Why is that exhaust flowing into the cabin?

Maybe you had the windows open during the flight?

Good point.

That would create enough suction for the carbon monoxide to flow in.

Most pilots have their windows down to get some ventilation into the cabin.

Investigators revisit pictures they collected from witnesses.

This was taken earlier in the day.

Looks like he cracked the door during the taxi into Cottage Point to get some

air.

The open door could have created a draft that sucked the exhaust into the cabin

through holes in the firewall.

But is a short taxi really long enough to be dangerous?

Hard to say.

Here's a short video from one of the witnesses during the time the pilot

moved the plane.

The windows are closed, but check out the door.

An extra 27 minutes in the plane with the door ajar.

That likely elevated the level of carbon monoxide in the pilot system

The investigation team

had an hypothesis about how the door ajar may have exacerbated the passage of

carbon monoxide We needed to test this theory

Okay

initiating test number one

With the door closed.

Investigators recreate the conditions of Gareth Morgan's plane to see if

concentrations of carbon monoxide were high enough to incapacitate him.

We needed to use an exemplar Beaver aircraft with a simulated exhaust leak

and excess panel bolts removed to test our hypotheses.

They begin the test with the windows and doors closed.

First, we had to set a baseline level with the engine running.

We then removed the access panel bolts and simulated an exhaust leak by feeding

the exhaust from the engine directly into the engine bay.

55 parts per million.

There was no dangerous level of carbon monoxide in the cabin at that stage.

OK.

Let's clear the cabin and start again.

To protect against carbon monoxide poisoning, investigators agree to stop

the test if the levels reach 120 parts per million.

We then tested with the DORA jar.

It has to get to 70 parts per million.

And we saw that the levels increased.

Within minutes, carbon monoxide in the cabin climbs to deadly levels.

We're at 144 ppm in here. Way too high. Let's shut it down.

Prolonged exposure to carbon monoxide levels greater than 144 ppm is dangerous

to the occupants.

The tests are conclusive.

We did it.

It was a relief to see that our theory may have been proved.

and confirmed that carbon monoxide could enter the cabin.

Investigators worry the same thing could happen again.

They tracked down other Beaver float planes that were serviced by the same

company that maintained the Sydney Seaplanes fleet.

There's more than one plane with the same problem.

We inspected the access panels on three other Beaver aircraft.

And we were surprised to find that there was one bolt missing from each of those

panels.

It's unbelievable.

It was the same in every one of them.

They have to be fixed.

Immediately.

The missing bolts are replaced in all the planes where they're missing.

As investigators try to determine how the accident could have been prevented,

they're left with one last question.

Was this thing even working?

Carbon monoxide detectors provide pilots with the opportunity to detect exposure

before they become incapacitated.

Was the carbon monoxide detecting equipment on the Beaver floatplane

functioning?

So how does this one work?

It's supposed to turn a different colour when carbon monoxide is present.

It's sun bleached.

There's no way it was working.

When we examined the carbon monoxide detector from this aircraft, it was a

beige colour, which meant it was unserviceable and not capable of

detecting carbon monoxide.

You know, the detectors aren't even mandatory.

Not in Australia.

Not anywhere.

The ATSB's final report highlights the critical importance of audible carbon

monoxide detectors.

As a pilot myself, I strongly encourage all pilots to carry active carbon

monoxide detectors every time they go out flying.

I have carbon monoxide detectors in my home. They're very loud and they're not

expensive.

This should be mandatory in all small planes.

The crash at Jerusalem Bay raises awareness of the deadly threat carbon

monoxide poses to aviators It

was fulfilling to be able to identify important safety issues in the hope

that we can prevent such a tragic accident from occurring again in the

future But it was also rewarding to provide the family and friends some

degree of resolution

I've had many people, particularly after his death, say the wildest experiences

they had as passengers with Gareth.

He really touched people with his humbleness and his sense of

dedication to his profession.

And that's very much the human being that Gareth was.