© British Broadcast CorporationSuch a wave today would flood Wall Street and the Long Island Expressway
A huge wave crashed into the New York City region 2,300 years ago, dumping sediment and shells across Long Island and New Jersey and casting wood debris far up the Hudson River.

The scenario, proposed by scientists, is undergoing further examination to verify radiocarbon dates and to rule out other causes of the upheaval.

Sedimentary deposits from more than 20 cores in New York and New Jersey indicate that some sort of violent force swept the Northeast coastal region in 300 BC.

It may have been a large storm, but evidence is increasingly pointing to a rare Atlantic Ocean tsunami.

Steven Goodbred, an Earth scientist at Vanderbilt University, said large gravel, marine fossils and other unusual deposits found in sediment cores across the area date to 2,300 years ago.

The size and distribution of material would require a high velocity wave and strong currents to move it, he said, and it is unlikely that short bursts produced in a storm would suffice.

"If we're wrong, it was one heck of a storm," said Dr Goodbred.

Landslide or asteroid?

The origin of such a tsunami is also under debate. An undersea landslide is the most likely source, but one research group has proposed that an asteroid impact provided the trigger.

In 300BC, barrier beaches and marsh grass embroidered the coast, and Native Americans walked the shore.

Today, a wave of the proposed size would leave Wall Street and the Long Island Expressway awash with salt water.

© British Broadcast CorporationAtlantic tsunamis are rare, but could be triggered by submarine landslides
An Atlantic tsunami was rare but not inconceivable, said Neal Driscoll, a geologist from Scripps Institution of Oceanography, who is not associated with the research. But verifying one that is 2,000 years old is tricky.

Earthquakes, underwater landslides, or a combination of the two were the most frequent Atlantic tsunami triggers, said Professor Driscoll.

The 1929 Grand Banks tsunami, in Newfoundland, which killed more than two-dozen people and snapped many transatlantic cables, was set in motion by a submarine landslide set off by an earthquake.

Dr Goodbred imagines that the New York wave was on the Grand Banks scale - three to four meters high and big enough to leap over the barrier islands; but that it did not reach the magnitude of the 2004 Sumatran tsunami.

The evidence is buried under metres of sediment in New York and New Jersey.

High-speed wave

Dr Goodbred first proposed the link between the layers of unusual debris found in sediment cores and a tsunami while studying shellfish populations in Great South Bay, Long Island.

He extracted many mud cores with incongruous 20cm layers of sand and gravel.

Their age matched that of wood deposits buried in the Hudson riverbed and marine fossils in a New Jersey debris flow in cores gathered by other researchers.

The fist-sized gravel he found in Long Island would require a high velocity of water - well over a metre per second - to land where it did, said Dr Goodbred.

Among the fossils and shells sandwiched in the organic black mud of Sandy Hook Bay, New Jersey, Marine Geologist Cecilia McHugh of Queens College, City University of New York, discovered mud balls made from red clay that matched iron-rich sediments found onshore.

The balls form their spherical shape only through vigorous reworking, said Dr McHugh, and they do not form in small storms.

"I didn't think much about it until we dated the deposit and came up with the same date that Steve did on Long Island," she said.

It prompted her to check cores extracted from the upper continental slope 200km offshore.

She discovered a 2,200-year-old layer of sand and mud, on top of sedimentary layers 8,000 to 14,000 years old.

Dr McHugh says such relatively young debris is not found that far out on the slope, and the date is close to that of the New York and New Jersey samples.

Age of a storm

The age and nature of the material make tsunami verification a challenge.

The radiocarbon dates of the debris are accurate to within a century, said Dr Goodbred. But the only evidence that a dramatic event took place thousands of years ago is common coastal debris - wood, sand, shells and rock.

Researchers must discern whether it was strewn by a tsunami or a hurricane, or another large storm, such as a "nor'easter", said Professor Driscoll.

Unusual layers in sediment cores may be a sign of an ancient tsunami
"Understanding the origins of these deposits can be difficult," he added.

While tsunamis can occur in any ocean, they are most common in the Pacific and Indian Oceans where continental plates collide.

There, large undersea earthquakes are relatively common.

In the Atlantic, where the plates spread, tsunamis are rare, which means Atlantic tsunamis are not well studied, said Bruce Jaffe, of the United States Geological Survey.

There is little research on tsunami debris in the variety of northeast coastal environments - riverbeds, marine bays - where the New York debris layers were found. There are few modern analogues to compare them with for identification, he said.

"Grand Banks is the only unequivocal tsunami in the Atlantic on the Northeast coast because there were eye-witness accounts and the deposits matched that of other modern tsunamis," said Dr Jaffe.

To rule out the possibility of a severe storm, said Professor Driscoll, tsunami groups should collect more core samples to see whether the distribution of the debris is consistent.

Dr Goodbred said teams were planning to do just that. And this would confirm that the deposits are not quirks of local geology.

'Circumstantial evidence'

The researchers would also repeat carbon dating on cores to verify ages, said Dr Goodbred, but he has a hunch the tsunami theory will win out.

"We're building a case of circumstantial evidence that is getting harder and harder to ignore," he said.

While many geologists say a submarine landslide is the likely trigger of a tsunami, a group led by Columbia University geologist Dallas Abbot thinks a space impactor may have set off the massive wave.

© British Broadcast CorporationSome researchers suspect a space impactor may have been involved
Her team discovered material in the New Jersey and Hudson River cores dated to 2,300 ago, and believe it to be meteoritic in nature. This includes carbon spherules, shocked minerals, and nanodiamonds, which are produced under extreme pressures and temperatures.

"We didn't find the typical shocked quartz, but that is usual for a water impact," said Dr Abbott.

She theorised that an asteroid landed in the water off the coast of New York and New Jersey, either creating the wave directly or triggering a submarine landslide. No crater has yet been found.

Many geologists and other scientists remain sceptical of the asteroid evidence so far; but proof of an asteroid impact is not necessary to build the case for a massive wave.

As Dr Goodbred pointed out: "The tsunami story stands on its own without the impact."