K'gari (Fraser Island) is known for its iconic beaches that stretch as far as the eye can see, freshwater lakes and ancient dunes.
Jutting out from the south-east Queensland coast on Butchulla country, it is the largest sand island in the world.
Over the years, people have speculated about when, and how, this UNESCO World Heritage-listed island was formed.
"We knew the dunes on the island were old … but we had no idea when [they] formed," said Jamie Shulmeister of the University of Canterbury.
Now, a team of scientists led by Daniel Ellerton of Stockholm University and Professor Shulmeister have provided the first snapshot of the island's geological age.
Working closely with the Butchulla community, they dated sand dunes on the east coast of K'gari and nearby Rainbow Beach in the Cooloola National Park.
"The earliest ages suggest K'gari is almost 900,000 years old," Dr Ellerton said.
K'gari's rise appears to have coincided with changes in sea levels caused by a dramatic shift in the climate about 1 million years ago, the scientists report in Nature Geoscience.
The findings also suggest the island played a key role in the establishment of another iconic, but poorly understood, wonder: the Great Barrier Reef (GBR).
East coast sand conveyer belt
Every year, around 500,000 square metres of sand is transported up the east coast of Australia from rivers such as the Hawkesbury and Hunter.
"It's the reason why we have beautiful sandy beaches in New South Wales and Queensland," Dr Ellerton said.
As it is pushed north by south-easterly trade winds and waves, it builds around Stradbroke and Moreton islands and eventually reaches K'gari.
The island sits on a pivotal point on the south-eastern Queensland coast, close to the edge of the continental shelf.
As the sand hits the island's northerly point, it slips off the edge of the shelf into deep water.
But if K'gari didn't exist, the researchers argue, the sand would have continued further north and stopped coral growing on what is now the southern and central parts of the reef.
"What [K'gari] is doing is essentially forming a seawall," Dr Ellerton said.
"The sand would have been transported to the north-west [by the trade winds], but now it is actually going to the north-east because of K'gari."
Despite a favourable climate for millions of years, there were no reefs at all in the southern part of the GBR until about 650,000 to 700,000 years ago, Professor Shulmeister said.
"Just before they appear, we've got the formation of K'gari."
How do they know how old the island is?
Establishing the age of ancient landscapes, especially those made of sand that contains very little carbon, is very challenging.
To start with, the researchers used a technique known as optically stimulated luminescence, which dates the time that individual grains of sand last saw light.
The ancient sands along the east coast of Australia are perfect for this technique, Professor Shulmeister said.
"By the time you get to south-east Queensland, there's nothing in there except for a few heavy minerals and these gorgeous quartz grains."
Normally, the technique yields ages of up to 200,000 years, but samples taken from Rainbow Beach — just south of K'gari — and the Fraser Island Cliffs returned dates concentrated between 700,000 to 800,000 years, with the oldest around 1.2 million.
To make sure the dates were accurate, they turned to another technique known as paleomagnetism, which analyses the alignment of iron nanoparticles in relation to the Earth's magnetic poles.
Around 773,000 years ago, Earth's magnetic poles flipped so particles dug up from layers below what is known as the Matuyama/Brunhes boundary lie in a different direction to those above.
This kind of analysis is very difficult to do with crumbly sand, but the cliffs of Rainbow Beach also contain a harder substance called ferricrete — iron that has washed out of the sand and set like concrete.
While samples from the top two-thirds of the cliffs had normal magnetisation, particles in the bottom third were reversed.
"So we knew that 773,000 years was real. There were one or two samples that weren't perfect, but the bulk of the [samples] work really well," Professor Shulmeister said.
How a dramatic shift in climate shifted sand
Layers in the cliffs, particularly those around Rainbow Beach that were the same age and part of the same system as those on K'gari, reveal how the island developed.
"Rainbow Beach is very unique," said Dr Ellerton, who sampled the area for his PhD at the University of Queensland.
"You get to see the oldest material at the bottom all the way to the youngest material that's at the top and you can just stand there and see it for the entire length of the beach."
While the bottom of the cliff is dominated by vast sheets of rippled sand known as transverse dunes, the top two-thirds contain large parabolic-shaped dunes like those on the island today.
Up until around 1 million years ago, Earth cycled through glacial periods every 40,000 years in line with variations in its orbit, causing sea levels to rise and drop by 60–80 metres.
But suddenly at a point known as the Middle Pleistocene transition, the glacial cycle changed to 100,000 years, and sea level oscillations doubled to more than 120 metres.
"That's the same climate system that we currently know of and what we experience in the modern time period," Dr Ellerton said.
The team believe this unexplained transition may have triggered the formation of K'gari.
According to their hypothesis, the sudden drop of sea levels exposed sand stored on the continental shelf that was then bulldozed up into massive sheets as the sea rose again.
"You've got this giant storage system of sand on the outer parts of the shelf, and then you basically drop the sea level not by 75 metres, but by 120 metres," Professor Shulmeister said.
"The erosion caused by sea-level rise will free up all that sand that's been in storage and huge volumes of sand get pushed back onto the coast."
The island grew in pulses over hundreds of thousands of years as the sea levels returned to modern levels after each period of glaciation.
As time went on, there was less sand on the continental shelf to be pushed up, swamping everything in its path.
"That's when we start to see parabolic dunes," Dr Ellerton said.
"Parabolic dunes tell you that the landscape had to have been at least somewhat vegetated."
Old dunes were eroded and blown inland through gaps in vegetation, producing the parabolic shape.
"If you look on K'gari today, some of the parabolic dunes are 10 kilometres long," Dr Ellerton said.
"They are eroding and moving old material successively over time, as opposed to what we see at the base of the island and Rainbow Beach."
Major changes to Queensland
Oceanographer Helen Bostock said the rise of large sand islands along the east coast was fascinating.
"These are a relatively recent feature of east Australia," said Dr Bostock, who was not involved in the research.
"It blows my mind these have been formed from very old sand … that has been constantly reworked."
While the reworking means the dates in the study jump around a bit, Dr Bostock said the research was very impressive and provided more insight into how the coastline of our tectonically quiet continent was shaped 1 million years ago by global changes in climate.
"This time period has not been well studied in Australia," said Dr Bostock, who is involved with the million-year ice core project in Antarctica to understand why the climate shifted so dramatically.
"It's actually very hard to get continuous records that go back 1 million years."
While sand islands don't provide a consistent record like ice cores, Dr Bostock said they still gave an interesting insight into the period.
"It's great to see how Australia responded to this climate transition, which was clearly significant enough to cause some pretty major changes to Queensland."
But, she said, more work was needed to understand when and how different parts of the Great Barrier Reef formed.
"We need to find evidence of the fact that the sand used to make it further north before K'gari became this wall to block it."
Dr Bostock is currently analysing coral cores to pin down the age of the northern and central parts of the reef.
"If the sand was the main thing preventing it from developing in the south, we could have had a Great Barrier Reef developing further north," she said.
"There's a lot of work still to do, but I think this is an interesting hypothesis we can test going forward."
