Explosive volcanic eruptions like the one that devastated Tonga are not restricted to shallow water and can occur at depths of "at least" one kilometre, a study says.
Researchers have discovered that underwater eruptions are much more powerful than first thought, able to shoot volcanic rock into the air at "supersonic" speeds within seconds.
While the Hunga Tonga-Hunga Ha'apai volcano made world headlines last month, underwater eruptions were still largely underestimated.
However a Queensland University of Technology research team say they have changed that perception after their study of volcanic rock or pumice.
QUT researcher Scott Bryan said the presence of pink pumice in the water after a 2012 South Pacific eruption situated 900m below sea level was critical to the study.
He said it indicated magma had been shot out by the underwater Havre volcano and projected into the air at speeds of at least 200 metres per second.
"Previous studies thought the magma gently came out of the sea floor and that deep underwater eruptions weren't able to be explosive," Prof Bryan told AAP.
"But our study shows that Havre was so powerful it was able to punch through almost a kilometre of ocean water to get hot pumice into the air to oxidise and get that colour.
"What we are seeing is that 1km of ocean water depth is not stopping an eruption from being explosive, that they are not restricted to shallow water like what we saw in Tonga.
"They can still be explosive at least down to a kilometre water depth and potentially more."
Prof Bryan said the study showed the potential dangers of underwater volcanoes could not be underestimated, a timely warning in the wake of the Tonga eruption.
The Hunga Tonga-Hunga Ha'apai volcano triggered a tsunami that killed three people, destroyed villages and resorts, blanketed the nation's capital in ash and knocked out communications.
Prof Bryan said the remote Havre eruption, situated 1200km away from the Tonga volcano, would have been "much more powerful".
"The interesting thing is that the Tonga eruption itself started on December 20," he said.
"When we hit the big explosion on January 15 it had already been erupting for a few weeks.
"We tend to measure eruptions by the amount of magma that comes out and this (Havre) would have been a lot bigger than the Tonga eruption.
"That (Tonga) was a singular nuclear bomb blast going off while this (Havre's) power and energy would have been sustained for many minutes if not some hours."
The QUT-led study has been published in the Communications Earth and Environment journal.
