Scientists are about to get an unprecedented look at enigmatic earthquakes that may help forecast the next massive rupture in our most dangerous fault zone.
An international team of researchers are deploying 50 instruments just off the North Island's East Coast, where they'll spend two years on the sea floor gathering data on the tiniest of movements.
The project will make New Zealand's big-risk Hikurangi Subduction Zone (HSZ) one of the planet's most studied sites for mysterious "slow-slip" earthquakes.
Only discovered in recent times, these events – essentially quakes in slow motion - have been linked with some of the planet's most devastating natural disasters.
They've been found to precede Japan's 9.1 Tohuku earthquake and tsunami in 2011, the 8.1 Iquique earthquake in Chile in 2014, and a 7.2 shake off the coast of Mexico the same year.
Only last year, researchers reported how the slowest quake ever recorded - lasting 32 years - eventually led to the calamitous 1861 Sumatra earthquake in Indonesia.
It's thought our own subduction zone – where the Pacific Plate dives, or subducts, westward beneath the North Island – is similarly capable of generating "megathrust" quakes as large as 9.0, posing a tsunami risk to swathes of our coastline.
Headline projections in one EQC-commissioned report estimated worst-case scenario impacts from a one-in-500-year event could include 33,000 fatalities, 27,000 injuries and $45b worth of property loss.
More recent research has indicated a 26 per cent chance of an event 8.0 or larger striking beneath the lower North Island within the next 50 years – and scientists increasingly believe that understanding slow-slip quakes is key to forecasting major tremblors.
Along the subduction zone, these silent events have been shown to slowly unfold at shallow depths off the coast of Gisborne and Hawke's Bay, and at deeper levels observed off the Manawatū and Kāpiti regions, releasing pent-up energy equivalent to a magnitude 7.0 earthquake.
Scientists have located them in specific areas where the zone was shifting from being "stuck" beneath the southern North Island, to an area where it was "creeping" further north – around Gisborne and Hawke's Bay.
But much remains unknown.
GNS Science geodetic scientist Dr Laura Wallace said the new programme – also involving researchers from University of Tokyo, Kyoto University, and Tōhoku University in Japan, and Columbia University and University of Rhode Island in the US - would help test some possible explanations.
These include the areas they occur in regularly reaching some form of threshold after being constantly loaded with stress by plate motion, or being primed by a build-up of water around the fault zone.
"To test the hypothesis about fluid movement, for instance, we'll have seismometers that can help us understand the properties of earthquakes that are happening during slow-slip events," Wallace told the Herald today.
While significant events have been recorded in roughly five-year cycles – the most recent being a pair of sequences last year – Wallace said reasonably large ones tended to occur near Gisborne at least once every two years.
"So, I'm hoping over this two-year period we'll be able to observe at least one larger event, and maybe two or three smaller ones."
Of the 50 instruments being deployed from aboard Niwa's RV Tangaroa, most are equipped with seafloor pressure sensors capable of picking up vertical movement on the seafloor down to a centimetre.
Ultimately, the team expect their wider array will help detect shifts as small as half a centimetre - offering our sharpest view yet.
"Another really exciting thing is we're also going to have some oceanographic sensors that are able to measure current and changes in oceanographic variations off Gisborne," Wallace said.
"That's really going to help us bring down the noise level and help detect that much finer upward or downward movement of the sea floor, during slow-motion earthquakes."
What the team found at the HSZ – where more than $70m in funding has been poured into a series of scientific efforts over the past decade – might also have crucial implications elsewhere in the world.
"I suspect the kind of things that we see off Gisborne are actually much more common at other subduction zones than we realise," she said.
"We also know there are several subduction zones around the world with fairly similar characteristics to this one – so we might expect similar behaviours to occur at those."
It was partly its accessibility – similar subduction zone sections lay more as far as 150km offshore in most of other parts of the world – that made the HSZ a global observatory for these events.
"And this is really the first experiment of its kind."
A separate programme, which Wallace is collaborating on, is drawing on troves of seismic data to tease out the relationship between slow-slip events and earthquake swarms, with a hope of revealing tell-tale signals.
Another study has turned to another intriguing indicator to help gauge the size of the next HSZ cataclysm: tiny creatures that lived tens of millions of years ago.
Victoria University's Dr Carolyn Boulton has been leading a team of earthquake scientists studying a rocky bluff on the Hungaroa fault, located on the margins of the zone.
Layers of limestone, mudstone, and siltstone on the bluff near Tora, about 35km southeast of Martinborough, provided handy proxy for what was happening in the subduction zone offshore.
Similar rocks to those on the bluff had been deposited on the seafloor between 35 and 65 million years ago, but their location makes it difficult to study them.
Instead, scientists could look at the rocks on land to tell them more about what's happening under the sea.
"The rocks all contain calcite from ancient single-celled marine organisms, mainly foraminifera, such as plankton," Boulton said.
"We've found calcite from those tiny organisms can affect movement in the subduction zone.
"Just think, these tiny, long-dead organisms can affect how two huge tectonic plates interact mechanically."
If the calcite in the rocks was able to dissolve — like sugar in tea — the fault could be weak, and slide easily without an earthquake.
However, if the calcite couldn't dissolve, the fault might instead lock up - and store energy that might be released in a large quake.
"Calcite dissolves faster when it's highly stressed and when temperatures are cooler," she said.
"It dissolves more easily at low temperatures - say, room temperature. But it gets harder to dissolve as temperature goes up - say, deeper in the earth.
"In the subduction zone, temperature increases more slowly than on land - by only around 10C per km.
"So, the fault is really sensitive to what calcite, those shells of old dead marine organisms, is doing.
"The amount and behaviour of calcite from these organisms is a big piece of the puzzle of how large the next earthquake might be."
Boulton said observations at Tora showed the shallow part of the subduction zone could accommodate plate movements by slipping slowly - or by slipping quickly in large and damaging quakes.
"What we really want to know is: Are there slow-slip events out there we haven't detected
"Are the rocks moving without earthquakes, or are they truly locked up? That will help tell us what might happen in the next earthquake."
Meanwhile, in a separate EQC-funded study, scientists are setting up seismometers across 19 spots in Southland to fill in what could be a big blind-spot in New Zealand's earthquake hazard.
"Historically, we think that Southland is a low seismicity area, but the distance between Geonet sensors is over 100km, so a lot of the smaller earthquakes go unrecorded," said the team's leader, Dr Jack Williams of Otago University.
"These new sensors will be roughly 30km apart, which will give us a much better idea of active fault lines in Southland."
While past seismic activity was often judged by signs at the surface, in Southland, these indicators might not tell the full story.
"We can see a lot of scarring in the Central Otago landscape from prehistoric earthquakes because it is dry, but the weather in Southland is very wet, and so the evidence from past earthquakes is removed from the landscape much more quickly," he said.
"These regions are similar in that we expect there are active faults concealed beneath these plains, but we cannot necessarily identify them from looking at the landscape alone.
"The Canterbury landscape did not give us a lot of clues where the active faults were, but these days you can see land uplifted all over that region.
"Clearly, the surface may not always tell you what goes on below."
The new projects come as New Zealand's hazard from shaking in future earthquakes was upgraded in a freshly-updated model, with implications for building design.
Much of the South Island is set to plunge below 0C tonight and tomorrow.